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/W2775441981	https://journal.unnes.ac.id/nju/index.php/kemas/article/download/9488/6564	English	null	Counseling Model Development Based on Analysis of Unwanted Pregnancy Case in Teenagers	Kemas	2,017	cc-by	4,837	 Correspondece Address: Gedung F5 Lantai 2, Jurusan IKM, FIK, Universitas Negeri Semarang Email : efa.nugroho@gmail.com Abstract Teenegers who experience unwanted pregnancies are increasing. The number of client that access the Unwanted Pregnancy counseling services in IPPA Central Java in the year 2006 (94 clients), 2007 (91 clients), 2008 (95 clients), 2009 (68 clients), 2010 (157 clients), 2011 (98 clients), and 2012 (83 clients). Related to that data, 31% of clients were referred Haid Induction (HI), 3% refer shelter, and 66% continue the pregnancy and there were not identified. This research cunducted in 2014 used quantitative- qualitative approach which aimed to develop a model of counseling after mapping the case. Respondents were 5 Unwaned Prenancy counseling clients selected based on the final decision of counseling. From the research developed counseling model for teenage which should have an easy procedure, complete services, opening hours accordingly, no discrimination, respect privacy, provide pro choice, and low prices. Services consists of counseling, contraception, safe abortion, treatment of STIs, information center counseling and HIV testing, gynecological, prenatal and postnatal services, as well as the services of victims of gender based violence and sexual abuse. (UWP). UWP occurs due to several factors such as sociodemographic factors (poverty, sexually active and lack of contraception usage, mass media), disharmonized family (family relationships), developmental status (lack of thought about the future, want to experience, attention-seeking), usage and abuse of drugs (Rosyeni, 2013).Other factors that contribute include incomplete and inaccurate knowledge about the process of pregnancy and the prevention method, contraceptive failure, and being a rape victim (Kusmiran, 2011). Increasingly permissive attitudes would also affect teenagers’ sexual behavior that could lead to UWP (Azinar, 2013). UWP has physical, psychological and social impact. Pregnant KEMAS 13 (1) (2017) 137-144 Efa Nugroho1,2, Zahroh Shaluhiyah1, Cahya Tri Purnami1, Kristawansari2 Efa Nugroho1,2, Zahroh Shaluhiyah1, Cahya Tri Purnami1, Kristawansari2 1Prodi Magister Promosi Kesehatan, Universitas Diponegoro 2Jurusan Kesehatan Masyarakat, Universitas Negeri Semarang 1Prodi Magister Promosi Kesehatan, Universitas Diponegoro 2Jurusan Kesehatan Masyarakat, Universitas Negeri Semarang Info Artikel Article History: Submitted April 2017 Accepted July 2017 Published July 2017 Keywords: Model; Counseling; Unwanted Pregnancy DOI http://dx.doi.org/10.15294/ kemas.v13i1.9488 UWP coun Methodsh This study consists of 3 steps. First step is a descriptive study with quantitative approach, the second step is qualitative approach, and the last step is development of a model based on the findings from the first and second study. In the first study, quantitative approach was used to picture an event that occurred. Descriptive design was chosen to picture the events studied and then described the events as it happened, hence intervention and variable manipulation was not needed. Quantitative approach is an approach that produces result from statistical procedure or other mathematical processes. We want to know the research problem without manipulating research setting, understand the situation as it is happened, without direct contact. This study would try to analyze the data characteristics and describe the UWP cases in Semarang based on respondents characteristic (age, educational status, and job), parents’ job, sex partner characteristic (age, educational status, and job), sociodemographic status (origin and living status), and UWP case trend in Semarang, using secondary data from UWP clients in PKBI Central Java. The second step used a qualitative approach with case study design to gain in-depth information about the experiences and behaviors of those who experienced UWP, counseling process, until the decision-making process. The third step of this study was developing a counseling model based on the findings of the first and second studies.hi Based on the Indonesian Teenage Survey on Reproductive Health (SKRRI) 2002-2003, it was recorded that teenagers who had a friend who already had sex at the age of 14-19 years comprised 34.7% of women and 30.9% of men, while those who have friends who already had sex at the age of 20-24 years comprised 48.6% of women and 46.5% of men. h The number of UWP clients who accessed PILAR PKBI counseling service in Central Java varies from year to year. There were 94 people in 2006, 91 people in 2007, 95 people in 2008, 68 people in 2009, 157 people in 2010, 98 people in 2011, and 83 people in 2012. Counseling data in PILAR PKBI is still limited because the instrument used for counseling guidance could only explore the characteristics of the client but is unable to explore the data of the spouse, parents, and the sociodemographic status of the client (Hermawan, 2012). Introduction Reproductive health issue has been under the spotlight ever since the issue was raised at the International Conference on Population and Development (ICPD), in Cairo, Egypt, in 1994 (Panchauri, 2013). Population control policies then shifted to a broader range, including reproductive health needs for both men and women throughout the life cycle, including their reproductive rights, gender- equality and equity, women empowerment and the handling of gender-based violence, and also men’s responsibilities pertaining to reproductive health. Risky sexual intercourse among adolescents would lead to Unwanted Pregnancy pISSN 1858-1196 eISSN 2355-3596 pISSN 1858-1196 eISSN 2355-3596 Efa Nugroho, Zahroh Shaluhiyah, Cahya Tri Purnami & Kristawansari / Counseling Model Development Based On Analysis develop counseling model based on analysis of unwanted pregnancy (UWP) cases in Semarang, 2) To analyze cases of UWP among adolescent in Semarang in the year 2008-2012, 3) To analyze the counseling process of UWP cases at PKBI Central Java and, 4) To develop UWP counseling model. teenage students would face responses from two parties. The first is from the school, who usually responded unfavourably such as by expelling the pregnant student. Expelled students would lose the opportunity to work and, leaving the options of working as a single parent or going through an unplanned early marriage (Pachauri, 2011; Gyan, 2013). UWP coun Methodsh UWP counseling conducted by PKBI is not optimal because the number of adolescents who accessed it was low, and there is no follow up after a decision was taken. Based on the description above, this study conducted a data mapping of UWP clients who had accessed counseling services in PKBI. Mapping or analysis of specific social situations is often considered as a method to encourage the participation of people who experienced the same situation to develop a specific community development program by establishing the understanding between group members on the issues. According to the data, in 2010 31% of UWP clients were referred for abortion, 3% were referred into shelters, while 66% had no record of their final decision. This can be caused by the limitation in the existing counseling model in PKBI. Therefore, we develop UWP counseling model that was started by mapping or case analysis in PKBI Central Java. The population in the first step is clients of PILAR PKBI Central Java from 2008 to 2012. In the second study, a qualitative approach was used; hence we conducted in-depth interviews to counselors, clients, and parents of clients as informants. The primary data in this study was obtained from in-depth interviews with respondents consisting of 5 UWP counseling clients selected after considering the final decision post-counseling, 2 counselors The purpose of this research are: 1) To 138 KEMAS 13 (1) (2017) 137-144 triangulation respondents, and 2 parents of the clients. who experienced UWP from the year 2008- 2012 was late adolescents, not early teenage. The definition of adolescents used in this study were all adolescents aged 15-24 years old, while adolescents selected as respondents were those who had out of marriage pregnancy or UWP and was counseled in PILAR PKBI Central Java. who experienced UWP from the year 2008- 2012 was late adolescents, not early teenage.hi The definition of adolescents used in this study were all adolescents aged 15-24 years old, while adolescents selected as respondents were those who had out of marriage pregnancy or UWP and was counseled in PILAR PKBI Central Java. The secondary data was obtained from adolescent unwanted pregnancy (UWP) case information in Semarang who were clients of PKBI Central Java from 2008-2012 which had were then analysed and presented as descriptive data analysis. UWP coun Methodsh Qualitative data which was the result of interview with counselor, UWP client in PKBI Central Java, and parents were analyzed using Miles and Huberman analysis. This analysis uses three components of analysis, namely Data Reduction, Data Presentation, and Conclusion Withdrawal. The counseling model development was conducted after the quantitative and qualitative research data were analyzed. The data was used to develop a model appropriate for the needs and expectations of counselors, clients, and parents. Based on the age of respondents, UWP mostly occurs in adolescence. Pregnancy in adolescence increases the risk of maternal and infant death 2-4 times higher than mothers aged 20-25 years (Tusiime, 2015). The other risk is when the pregnant adolescent decides to end UWP through abortion or IH. Teenagers who experience UWP at school age are at risk to be expelled, hence threatening their welfare because of lack of education. Today, not many schools are willing to enroll students who have experienced UWP. However, advocacy related to such issue are ongoing (Nasution, 2012).h Results and Discussion Dating is normal a behaviour, however nowadays the relationship between man and woman is viewed more openly. It is important to realize that dating among teenagers would be an influence and initiative to have sexual intercourse. The place of origin is the respondent’s identity according to the address where he/she lives. The highest UWP cases based on place of origin from 2008-2012 was respondents who lived in Ngaliyan with 14 cases from 2008-2012.h There were 3 respondents from rural areas and 2 respondents from urban areas. This shows that sexual behavior did not happen only in the urban area. One of the cause is the rapid development of technology, information, and social media. Children or teenagers can easily access pornographic media either from electronic or printed source. This increase is not followed by increased knowledge on reproductive health (Speizer, 2015). Dating is considered as the enterance into deeper relationship. Sexual intercourse before marriage is considered as a form of closeness between two people in love. Sometimes teenagers could be carried away to have sexual intercourse with their boyfriend in the absence of clear restriction in dating, or it was done unaware or unplanned (Leerlooijer, 2013; Hewageegana, 2014). In this study, all of the respondents said that the one who initiated to have sexual intercourse is their boyfriend. They believed their boyfriend would be their husband someday, while the boyfriends thought that The result of the study showed that the majority of respondents’ sexual partner that resulted in UWP was their boyfriend. This happened from 2008 to 2012 with the highest Table 3. Results and Discussion The place of origin is the respondent’s identity according to the address where he/she lives. The highest UWP cases based on place of origin from 2008-2012 was respondents who lived in Ngaliyan with 14 cases from 2008-2012. There were 3 respondents from rural areas and 2 respondents from urban areas. This shows that sexual behavior did not happen only in the urban area. One of the cause is the rapid development of technology, information, and social media. Children or teenagers can easily access pornographic media either from electronic or printed source. This increase is not followed by increased knowledge on reproductive health (Speizer, 2015). The result of the study showed that the majority of respondents’ sexual partner that resulted in UWP was their boyfriend. This happened from 2008 to 2012 with the highest number of cases in 2011 (36 cases). Similar to the study’s responden five respondents said that sexual relati with their boyfriend was the cause of unw pregnancy. As social creatures living in m civilization, teenagers cannot be pro from dating dan having sexual interc Dating is normal a behaviour, ho nowadays the relationship between ma woman is viewed more openly. It is imp to realize that dating among teenagers be an influence and initiative to have intercourse. Dating is considered as the ent into deeper relationship. Sexual inter before marriage is considered as a of closeness between two people in Sometimes teenagers could be carried a have sexual intercourse with their boyfri the absence of clear restriction in datin was done unaware or unplanned (Leer 2013; Hewageegana, 2014). In this study, all of the respo said that the one who initiated to have intercourse is their boyfriend. They b their boyfriend would be their hu someday, while the boyfriends though number of cases in 2011 (36 cases). number of cases in 2011 (36 cases). In this study, 1 respondent was a senior high school student, 2 respondents were in college, and 2 respondents were already working. Increased mobility of adolescents and the higher chance for teenagers to study, work, and live separately from their parents were one of the factors that increase the incidence of sex before marriage.h Similar to the study’s respondents, the five respondents said that sexual relationship with their boyfriend was the cause of unwanted pregnancy. As social creatures living in modern civilization, teenagers cannot be prohibited from dating dan having sexual intercourse. Results and Discussion Based on the secondary data, the age frequency distribution of respondents who experienced an unwanted pregnancy event (UWP) from 2008-2012 was at least 11 years old in 2012, 13 years old in 2009, while 15 years old in the years of 2008, 2010, and 2011. The maximum age of respondents who experienced a UWP from 2008-2012 was 24 years old. The age category of the majority of respondents The level of education were elementary school, junior high, senior high, and diploma/ undergraduate. The majority of UWP clients in PILR PKBI Central Java from 2008-2012 had an education level of senior high while those having other levels of education in was 68% in 2008, 65.5% in 2009, 84.2% in 2010, 69.4% in 2011, and 55.3% in 2012. Table 1. Frequency Distribution of Cases of UWP Based on Age of Informant Details Year 2008 2009 2010 2011 2012 Total 25 100 29 100 19 Minimum 15 13 15 15 11 Maximum 24 24 24 24 24 Mean 19,08 19,62 18,95 18,64 18,30 Source : Secondary data from the result of the study Table 1. Frequency Distribution of Cases of UWP Based on Age of Informant Table 2. Frequency Distribution of UWP Cases Based on Education Level Table 2. Frequency Distribution of UWP Cases Based on Education Level Category Year 2008 2009 2010 2011 2012 f % f % f % f % f % No School 0 0 0 0 0 0 0 0 0 0 Elementary 0 0 0 0 0 0 2 5,6 1 3,7 Junior High 6 24 6 20,7 3 15,8 3 8,3 9 33,3 Senior High 17 68 19 65,5 16 84,2 25 69,4 15 55,6 Diploma/college 2 8 4 13,8 0 0 6 16,7 2 7,4 Total 25 100 29 100 19 100 36 100 27 100 Source : Secondary data from the result of the study 139 Efa Nugroho, Zahroh Shaluhiyah, Cahya Tri Purnami & Kristawansari / Counseling Model Development Based On Analysis Table 3. Frequency Distribution of Unwanted Pregnancy Cases by Place of Origin Year In this study, 1 respondent was a senior high school student, 2 respondents were in college, and 2 respondents were already working. Increased mobility of adolescents and the higher chance for teenagers to study, work, and live separately from their parents were one of the factors that increase the incidence of sex before marriage. Results and Discussion Based on study result, PILAR PKBI counseling clinic, Jawa Tengah, already achieved some criteria of youth friendly sevices. From the providers’ side, PILAR PKBI counseling clinic at Central Java had clinical staffs who had received special training to respect dan be friendly with teenagers and to respect their privacy. Duration of counseling in the clinic was also deemed sufficient for developing an interaction between the counselor and the client. From the clinical facilities and program design aspect, the clinic was easily accessible because it was near to the city, Jl. Jembawan Raya No.8 – 12 Semarang. The counseling fee was affordable. There was no gender discrimination, both teenage boys or girls were well-recieved. The service hour was sufficient and appropriate to teenagers needs. The interior and exterior design of the clinic provided comfort and ensure that their privacy would be safeguarded and also had a special transit room for media communication, education, and information for teenagers. PILAR PKBI counseling clinic, Central Java, also provided service for various teenage reproductive health issues such as unwanted pregnancy counseling, STI service, and HIV – AIDS. It can be formulated that the role of counselor are as preventive that helps clients maintain or prevent the occurence of problems and curative/corrective which is helping clients solve the problem at hand (King, 2013; Wieler, 2016). Preservative, helps people improve the situation from problematic into a good state (problems solved) and that state of good maintained for as long as possible. Developmental, helping client maintain a good situation and, whenever possible, improve the situation, so that it no longer cause problem for them (Desirae, 2007; Hurlock, 2009). According to International Planned Parenthood Federation (IPPF), the following are mandatory services in adolescent-friendly clinics. First, counseling services. A truly adolescent-friendly clinic should provide adequate counseling to its clients. Teenagers can choose their counselor, male or female, and couselor should be willing to discuss about health, friendship, dating, and sexual relationship. Some aspects that need to be improved in PILAR PKBI counseling clinic, Central Java, were the ease to refer to a hospital when needed. This could be done by cooperating with other health facilities in Semarang City or other cities in Jawa Tengah. Teenagers’ involvement in designing and further developing the counseling clinic is also needed because those who will access this services are teenagers. Routinely providing group discussion on teenagers’ issues could also be done. Results and Discussion Frequency Distribution of Unwanted Pregnancy Cases by Place of Origin Place of Origin Year 2008 2009 2010 2011 2012 f % f % f % f % f % Banyumanik 2 8 3 10,3 2 10,5 0 0 1 3,7 Candisari 1 4 2 6,9 3 15,8 3 8,3 0 0 Gajah Mungkur 1 4 3 10,3 1 5,3 3 8,3 0 0 Gayamsari 3 12 0 0 0 0 3 8,3 4 14,8 Genuk 0 0 2 6,9 0 0 1 2,8 4 14,8 Gunung Pati 1 4 1 3,4 0 0 1 2,8 1 3,7 Mijen 0 0 1 3,4 1 5,3 4 11,1 0 0 Ngaliyan 3 12 6 20,7 1 5,3 9 25 3 11,1 Pedurungan 2 8 3 10,3 1 5,3 4 11,1 3 11,1 Semarang Barat 1 4 1 3,4 1 5,3 1 2,8 2 7,4 Semarang Selatan 2 8 0 0 1 5,3 1 2,8 2 7,4 Semarang Tengah 1 4 0 0 0 0 0 0 2 7,4 Semarang Timur 0 0 2 6,9 1 5,3 4 11,1 2 7,4 Semarang Utara 2 8 3 10,3 0 0 1 2,8 0 0 Tembalang 2 8 2 6,9 1 5,3 1 2,8 1 3,7 Tugu 4 16 0 0 0 0 0 0 1 3,7 Total 25 100 29 100 19 100 36 100 26 100 Source : Secondary data from the result of the study Table 3. Frequency Distribution of Unwanted Pregnancy Cases by Place of Origin 140 KEMAS 13 (1) (2017) 137-144 used by many teenagers. used by many teenagers. his girlfriend would not necessarily be his wife someday because they were still exploring. If they were not matced, the boyfriends would easily leave his girlfriend even if they already had sexual intercourse. Then women were the ones experiencing a huge loss. It is not easy ot build a reproductive health service which is easily accessible to teenagers. The problems encountered include lack of adolescent-friendly medical staff, lack of funds to create an ideal service, and barriers of value from the surrounding community due to their cultural construction (Ceylan, 2009). There were also policy barriers from the goverment, such as difficulty to obtain permits to establish a clinic for teenagers. Despite the constraints, establishing programs that meet teenagers’ needs for information and reprodructive health services should be a positive challenge (Situmorang, 2011). Results and Discussion Improving access to information, consultation, and services provided by the clinic could improve the quality of service. Improving promotive efforts related to counseling services is needed because there were many teenagers who do not know about the services. Promotional efforts can be done through socia lmedia that is currently being Second, contraceptive services. Access to contraceptive is a mandatory service in adolescent-friendly services. However, contraceptive services cannot be provided without proper regulations. Before accessing the service, the client should be counseled to provide a good understanding of how to use contraception, why contraception is necessary, and the risks of sexual behavior. After recieving counseling services, the clinic need to provide contraceptives such as birth control pills, condom, injections, and/or IUD. If needed, the clinic should have emergency contraceptions. 141 Efa Nugroho, Zahroh Shaluhiyah, Cahya Tri Purnami & Kristawansari / Counseling Model Development Based On Analysis Third, safe abortion services. Safe abortion services is still a controversial issue, but it is a serious problem. According to World Health Organization (WHO) data, 11 – 14% of maternal deaths in Indonesia are caused by unsafe abortions. Many women feel compelled to have an abortion because they had an unwanted pregnancy. This happens because of lack of access to contraceptive services, and the absence of education that warns teenagers about the risk of sexual behavior. Therefore, safe abortion should also be provided by clinics. Beside. Adolescent-friendly clinics should provide the choice of abortion methods, pre– and post–abortion counseling to clients, to ensure that clients make the right decision and would not be traumatized afterwards. virus, as well as information about condom to prevent potential HIV dissemination from the client to others. Sixth, Sixth, gynecological services. Gynecologist or women reproductive health specialist must be present in every clinc. In addition to providing various genital and breast examination, a good clinic should provide pap smear test or other methods to detect cervical cancer. Seventh, Prenatal and Postnatal services. Prenatal (before delivery) and postnatal (after delivery) services must be provided by reproductive health clinic and adolescent- friendly clinic. In addition, the clinic also needs to provide an acurate and affordable pregnancy test. Eighth, service for sexual- and gender- based violence victim. Besides identifying the victim of gender- and sexual-based violence, clinics should be able to handle or introduce victims to other parties who can handle violence cases; such as psychologists, authorities, etc. Results and Discussion that should be available at clinics are counseling services, contraceptive services, safe abortion, STI and RTI care, counseling information center dan HIV testing, gynecologist, prenatal and postnatal services, and services for sexual- and gender-based violence victim. and gender-based violence victim. References Azinar, Muhammad. 2013. Perilaku Seksual Pranikah Berisiko Terhadap Kehamilan Tidak Diinginkan. Jurnal Kemas, 8 (2) : 153- 160 Ceylan, Ali., et al. 2009. Post abortion family planning counseling as a tool to increase contraception use. BMC Public Health, 9 : 20 Desirae, Domenico & Jones, Karen. 2007. Adolescent Pregnancy in America: Causes and Responses. The Journal for Vocational Special Needs Education, 30 (1) : 4-12hf Gyan, Carles. 2013. The Effects of Teenage Pregnancy on the Educational Attainment of Girls at Chorkor, a Suburb of Accra. Journal of Educational and Social Research, 3 (3) : 53- 60 With the development of a comprehensive counseling model, clients would be able to find the most feasible alternative solution, hence solving the problmes of unwanted pregnancy and unsafe abortion. Hermawan D.Y. 2012. Informasi Kasus (Infus) PILAR PKBI Jawa Tengah. Semarang: PKBI Jawa Tengah Results and Discussion t Fourth, sexually transmitted infections (STIs) and reproductive tract infections (RTIs) care. STIs and RTIs are serious, but easily treated with the appropriate medications and therapies. In addition to providing tests for STIs and RTIs, a comprehensive clinic should provide at least one method of treating STIs and RTIs, while providing condom as contraception tool that can prevent STIs and RTIs. t In addition, referral services for teenage- friendly unwanted pregnancy counseling was developed in this study. Developing adolescent- friendly counseling requires cooperation from many parties. A figure of developing adolescent-friendly counseling referral service model is shown below: Fifth, VCT service for HIV. An adolescent-friendly clinics should provide counseling before and after HIV test, laboratory test to find out if the client is infected with HIV Figure 1. Counseling Model Based on Unwanted Pregnancy Case Analysis STIs HIV HIV - HIV + Peer Support Group Treatment Client Violence in courtship Unwanted Pregnancy STis and HIV Test Counseling Counseling (with parents) Counseling LBH APIK PPT Seruni Medical Surgical Home Shelter Complete by Self Law Support Haid Induction (Abortion) Continue No Risk Risk Education Test Counseling (with parents) Figure 1. Counseling Model Based on Unwanted Pregnancy Case Analysis 142 KEMAS 13 (1) (2017) 137-144 Based on Figure 1, it can be seen that after the client arrives, counseling services will be given in the presence of their parents. After counseling, the client will decide whether to choose menstrual induction (MI) or continue the pregnancy. If the client chooses menstrual induction, then the client will receive two alternative methods, medical abortion (MA) or surgical abortion (SA) in PKBI center clinic. If the client chooses to continue the pregnancy, there are two alternatives which is continuing at home or at a shelter. After the baby is born, there are also two alternatives, cared by the family or submitted to a shelter. For clients in need of legal assistance, there are NGOs that could provide assistance i.e. APIK Legal Aid Institute (LBH APIK) and Intergrated Service Center (PPT Seruni). that should be available at clinics are counseling services, contraceptive services, safe abortion, STI and RTI care, counseling information center dan HIV testing, gynecologist, prenatal and postnatal services, and services for sexual- and gender-based violence victim. Conclusion Unwanted adolescent pregnancy cases in Semarang City during 2008 – 2012 were 55 cases in average, and the average of client’s age was 19 years. The average level of education among cases of unwanted adolescent pregnancy was senior high school, amounting to more than 55% every year. The higest case of unwanted pregnancy came from Ngaliyan with a total of 14 cases. More than 78% of the teenagers had sexual intercourse with their boyfriend.h Hewageegana, Neelamani Rajapaksa., et al. 2014. A quantitative exploration of the sociocultural context of teenage pregnancy in Sri Lanka. BMC Pregnancy and Childbirth, 14 : 394 Hurlock, E. B. 2009. Psikologi Perkembangan: Suatu pendekatan Sepanjang Rentang Kehidupan. Jakarta: Erlangga King, Rachel., et al. 2013. Pregnancy comes accidentally like it did with me : reproductive decisions among women on ART and their partners in rural Uganda. BMC Public Health, 11 : 530 The unwanted pregnancy counseling process at PKB was performed by trained counselors with experience in unwanted pregnancy counseling. The main reason why clients accessed the counseling service at PKBI was for abortion. On average, clients had 1-2 sessions of counseling before making a decision. The follow-up after decision-making is still limited. Kusmiran, E. 2011. Kesehatan Reproduksi Remaja dan Wanita. Jakarta: Salemba Medika Leerlooijer, Joanne N., et al. 2013. Qualitative evaluation of the Teenage Mothers Project in Uganda: a community-based empowerment intervention for unmarried teenage mothers. BMC Public Health, 13 : 816 Nasution, Sri L. 2012. Pengaruh Pengetahuan Tentang Kesehatan Reproduksi Remaja Terhadap Perilaku Seksual Pranikah Remaja di Indonesia. Widyariset, 15 (1) : 75-83 According to the result of interviews and adolescents-friendly services standards compiled by the National PKBI, Unwanted Pregnancy Counseling Model for teenagers should have easy procedures, comprehensive services, appropriate service time, free of discrimination, respecting teen privacy, providing an open option, and cheap. Services Pachauri & Santhya. 2011. Reproductive Choices for Asian Adolescents: a focus on contraceptive behavior. IPSRH, 28 (4) Pachauri, Saroj & Santhya, K.G. 2013. Reproductive Choices for Asian Adolescents: A Focus on Contraceptive Behavior. International Family 143 Planning Perspectives, 28 (4) :186–195 Rosyeni, Yeni & Dariah, Isti. 2013. Hubungan Pengetahuan dan Sikap Remaja Putri dengan Experience: Longitudinal Examination of Urban Areas In Three African Countries. BMC Pregnancy and Childbirth, 15 : 294 Planning Perspectives, 28 (4) :186–195 Experience: Longitudinal Examination of Urban Areas In Three African Countries. Conclusion BMC Pregnancy and Childbirth, 15 : 294 Planning Perspectives, 28 (4) :186–195 Rosyeni, Yeni & Dariah, Isti. 2013. Hubungan Pengetahuan dan Sikap Remaja Putri dengan Kehamilan Remaja di Puskesmas Cipageran Cimahi Utara Tahun 2010. Jurnal Kesehatan Kartika, 3 (2) : 35-42 Tusiime, Suzan. et al., 2015. Prevalence of Sexual Coercion and its Association with Unwanted Pregnancies Among Young Pregnant Females in Kampala, Uganda: a facility based cross- sectional study. BMC Women’s Health, 15 : 79 Situmorang, Agustin. 2011. Pelayanan Kesehatan Reproduksi Remaja di Puskesmas: Isu dan Tantangan. Jurnal Kependudukan Indonesia, 6 (2) : 21-32 y Wieler, Elizabeth Wall., et al. 2016. Teenage Pregnancy: the Impact of Maternal Adolescent Childbearing and Older Sister’s Teenage Pregnancy on a Younger Sister. BMC Pregnancy and Childbirth, 16 : 120 Speizer, Ilene S. and Lance, Peter. 2015. Fertility Desires, Family Planning Use and Pregnancy 144
https://openalex.org/W2529921966	https://europepmc.org/articles/pmc5065587?pdf=render	English	null	From basic mechanisms to clinical applications in heart protection, new players in cardiovascular diseases and cardiac theranostics: meeting report from the third international symposium on “New frontiers in cardiovascular research”	Basic research in cardiology	2,016	cc-by	11,922	10 Center for Cardiovascular Research, John A. Burns School of Medicine, University of Hawaii, Honolulu, USA Basic Res Cardiol (2016) 111:69 DOI 10.1007/s00395-016-0586-x Basic Res Cardiol (2016) 111:69 DOI 10.1007/s00395-016-0586-x MEETING REPORT & Derek J. Hausenloy derek.hausenloy@duke-nus.edu.sg From basic mechanisms to clinical applications in heart protection, new players in cardiovascular diseases and cardiac theranostics: meeting report from the third international symposium on ‘‘New frontiers in cardiovascular research’’ Burns School of Medicine, University of Hawaii, Honolulu, USA 11 Department of Cardiology, Aarhus University Hospital, Skejby, Aarhus N, Denmark 12 Department of Biomedical Sciences, University of Padova, Padua, Italy 13 Experimental Renal and Cardiovascular Research, Department of Nephropathology, Institute of Pathology, Friedrich-Alexander-Universita¨t Erlangen-Nu¨rnberg, Nuremberg, Germany 14 Institut fu¨r Laboratoriumsmedizin, Ludwig-Maximilians- Universita¨t, Munich, Germany 15 Institute of Human Genetics, Friedrich-Alexander-Universita¨t Erlangen-Nu¨rnberg, Nuremberg, Germany 16 Department of Pharmacology and Pharmacotherapy, Semmelweis University, Budapest, Hungary 17 Pharmahungary Group, Szeged, Hungary 18 Department of Cardiology, Sarawak Heart Centre, Sarawak, Malaysia 19 Institute for Vascular Signalling, Centre for Molecular Medicine, Goethe-University, Frankfurt, Germany 20 D. Swarovski Research Lab, Department of Visceral, Transplant Thoracic Surgery, Medical Univ Innsbruck, Innsbruck, Austria 11 Department of Cardiology, Aarhus University Hospital, Skejby, Aarhus N, Denmark 11 Department of Cardiology, Aarhus University Hospital, Skejby, Aarhus N, Denmark 1 Institute of Biochemistry, Medical School, Justus-Liebig University, Giessen, Germany 1 Institute of Biochemistry, Medical School, Justus-Liebig University, Giessen, Germany 12 Department of Biomedical Sciences, University of Padova, Padua, Italy 12 Department of Biomedical Sciences, University of Padova, Padua, Italy 2 Cardiovascular and Metabolic Disorders Program, Duke- National University of Singapore, 8 College Road, Singapore 169857, Singapore 13 Experimental Renal and Cardiovascular Research, Department of Nephropathology, Institute of Pathology, Friedrich-Alexander-Universita¨t Erlangen-Nu¨rnberg, Nuremberg, Germany 3 National Heart Research Institute Singapore, National Heart Centre Singapore, Singapore, Singapore 14 Institut fu¨r Laboratoriumsmedizin, Ludwig-Maximilians- Universita¨t, Munich, Germany 4 Department of Microbiology, Kazan Federal University, Kazan, Russian Federation 4 Department of Microbiology, Kazan Federal University, Kazan, Russian Federation 15 Institute of Human Genetics, Friedrich-Alexander-Universita¨t Erlangen-Nu¨rnberg, Nuremberg, Germany 5 Centro de Biotecnologı´a-FEMSA, Tecnolo´gico de Monterrey, Monterrey, NL, Mexico 5 Centro de Biotecnologı´a-FEMSA, Tecnolo´gico de Monterrey, Monterrey, NL, Mexico 16 Department of Pharmacology and Pharmacotherapy, Semmelweis University, Budapest, Hungary 6 Research Unit, Hospital of Santa Cristina, Research Institute Princesa, Autonomous University of Madrid, Madrid, Spain 6 Research Unit, Hospital of Santa Cristina, Research Institute Princesa, Autonomous University of Madrid, Madrid, Spain 17 Pharmahungary Group, Szeged, Hungary 7 Department of Vascular Biology, Institute for Stroke and Dementia Research, Klinikum der Ludwig-Maximilians- Universita¨t, Munich, Germany 7 Department of Vascular Biology, Institute for Stroke and Dementia Research, Klinikum der Ludwig-Maximilians- Universita¨t, Munich, Germany 18 Department of Cardiology, Sarawak Heart Centre, Sarawak, Malaysia 19 Institute for Vascular Signalling, Centre for Molecular Medicine, Goethe-University, Frankfurt, Germany 8 Munich Cluster for Systems Neurology (SyNergy), Munich, Germany 20 D. From basic mechanisms to clinical applications in heart protection, new players in cardiovascular diseases and cardiac theranostics: meeting report from the third international symposium on ‘‘New frontiers in cardiovascular research’’ Hector A. Cabrera-Fuentes1,2,3,4,5 • Julian Aragones6 • Ju¨rgen Bernhagen7,8 • Andreas Boening9 • William A. Boisvert4,10 • Hans E. Bøtker11 • Heerajnarain Bulluck2,3,33 • Stuart Cook2,3 • Fabio Di Lisa12 • Felix B. Engel13 • Bernd Engelmann14 • Fulvia Ferrazzi15 • Pe´ter Ferdinandy16,17 • Alan Fong18 • Ingrid Fleming19 • Erich Gnaiger20 • Sauri Herna´ndez-Rese´ndiz2,3,35 • Siavash Beikoghli Kalkhoran33,34 • Moo Hyun Kim21 • Sandrine Lecour22 • Elisa A. Liehn23 • Michael S. Marber24 • Manuel Mayr25 • Tetsuji Miura26 • Sang-Bing Ong2,3 • Karlheinz Peter27 • Daniel Sedding28 • Manvendra K. Singh2,3 • M. Saadeh Suleiman29 • Hans J. Schnittler30 • Rainer Schulz31 • Winston Shim3 • Daniel Tello6 • Carl-Wilhelm Vogel32 • Malcolm Walker33 • Qilong Oscar Yang Li6 • Derek M. Yellon33,34 • Derek J. Hausenloy2,3,33,34 • Klaus T. Preissner1,4 Received: 14 August 2016 / Revised: 2 October 2016 / Accepted: 4 October 2016 / Published online: 14 October 2016 The Author(s) 2016. This article is published with open access at Springerlink.com Received: 14 August 2016 / Revised: 2 October 2016 / Accepted: 4 October 2016 / Published online: 14 October 2016 The Author(s) 2016. This article is published with open access at Springerlink.com Abstract In this meeting report, particularly addressing the topic of protection of the cardiovascular system from ischemia/reperfusion injury, highlights are presented that relate to conditioning strategies of the heart with respect to molecular mechanisms and outcome in patients’ cohorts, the inﬂuence of co-morbidities and medications, as well as the contribution of innate immune reactions in cardiopro- tection. Moreover, developmental or systems biology & Derek J. Hausenloy derek.hausenloy@duke-nus.edu.sg 1 Institute of Biochemistry, Medical School, Justus-Liebig University, Giessen, Germany 2 Cardiovascular and Metabolic Disorders Program, Duke- National University of Singapore, 8 College Road, Singapore 169857, Singapore 3 National Heart Research Institute Singapore, National Heart Centre Singapore, Singapore, Singapore 4 Department of Microbiology, Kazan Federal University, Kazan, Russian Federation 5 Centro de Biotecnologı´a-FEMSA, Tecnolo´gico de Monterrey, Monterrey, NL, Mexico 6 Research Unit, Hospital of Santa Cristina, Research Institute Princesa, Autonomous University of Madrid, Madrid, Spain 7 Department of Vascular Biology, Institute for Stroke and Dementia Research, Klinikum der Ludwig-Maximilians- Universita¨t, Munich, Germany 8 Munich Cluster for Systems Neurology (SyNergy), Munich, Germany 9 Department of Cardiovascular Surgery, Medical School, Justus-Liebig-University, Giessen, Germany & Derek J. Hausenloy derek.hausenloy@duke-nus.edu.sg 10 Center for Cardiovascular Research, John A. Introduction Despite the progress in our understanding of cellular sig- naling mechanisms in cardiomyocytes and the cellular communication within the cardiovascular system as well as new treatment options for cardiovascular diseases, ischemic heart disease remains the leading cause of death and disability worldwide [14, 30, 32, 40, 44, 71, 72]. Usually, such events are manifested by acute thrombotic occlusion of a main coronary artery at preselected sites of disturbed blood ﬂow or at ruptured atherosclerotic lesions [18, 56]. Although percutaneous coronary intervention (PCI) is the treatment of choice for reducing the size of a myocardial infarction (MI), the induced reperfusion may not only lead to the recovery of ischemic cardiac tissue but From basic mechanisms to clinical applications in heart protection, new players in cardiovascular diseases and cardiac theranostics: meeting report from the third international symposium on ‘‘New frontiers in cardiovascular research’’ Swarovski Research Lab, Department of Visceral, Transplant Thoracic Surgery, Medical Univ Innsbruck, Innsbruck, Austria 9 Department of Cardiovascular Surgery, Medical School, Justus-Liebig-University, Giessen, Germany 12 3 69 Page 2 of 13 Basic Res Cardiol (2016) 111:69 69 also brings about the paradoxical phenomenon of myocardial ‘‘ischemia/reperfusion injury’’ (IRI). approaches bear great potential in systematically uncov- ering unexpected components involved in ischemia– reperfusion injury or heart regeneration. Based on the characterization of particular platelet integrins, mitochon- drial redox-linked proteins, or lipid-diol compounds in cardiovascular diseases, their targeting by newly developed theranostics and technologies opens new avenues for diagnosis and therapy of myocardial infarction to improve the patients’ outcome. During the recent 3rd International Symposium on ‘‘New Frontiers in Cardiovascular Research’’ (Singapore), basic researchers and clinicians discussed new biomedical developments as well as new players in cardiovascular diseases, novel targets, and respective interventional strategies, particularly in the area of heart failure. The same holds true for mechanisms of action of the emerging class of atypical chemokines that was focused on at the sym- posium. Based on novel methods using single-chain anti- bodies, induced pluripotent stem cells, or miRNAs, novel drugs and technologies, summarized as ‘‘Theranostics’’, were presented and heavily discussed. In essence, the meeting covered heterogenous and unrelated intra- as well as extracellular molecular targets, which are all linked to the development or prevention of cardiovascular diseases, not only reﬂecting the complexity of the biological system but also indicating the variety of possible interventional approaches that can be helpful or even lifesaving as car- dioprotective strategies. Keywords Cardiomyocyte signaling pathways Cardioprotection Cardiovascular disease Co- morbidities Drug targeting Endothelial permeability Extracellular RNA (eRNA) Heart regeneration Induced pluripotent stem cells Ischemia–reperfusion injury Lipid metabolism MicroRNAs (miRNAs) Mitochondria Remote ischemic conditioning 29 Bristol Heart Institute, University of Bristol, Bristol Royal Inﬁrmary, Bristol, UK 21 Department of Cardiology, Dong-A University Hospital, Busan, Korea 21 Department of Cardiology, Dong-A University Hospital, Busan, Korea 22 Hatter Institute and MRC Inter-University Cape Heart Unit, Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa 23 Institute for Molecular Cardiovascular Research, RWTH University Hospital, Aachen, Germany 24 Department of Cardiology, The Rayne Institute, St Thomas’ Campus, King’s College London, London, UK 25 The James Black Centre, King’s College, University of London, London, UK 26 Department of Cardiovascular, Renal and Metabolic Medicine, Sapporo Medical University School of Medicine, Sapporo, Japan 27 Baker IDI Heart and Diabetes Institute, Melbourne, Australia 28 Department of Cardiology and Angiology, Hannover Medical School, Hannover, Germany 35 Department of Cardiovascular Medicine, National Institute of Cardiology, Ignacio Chavez, Mexico, D.F., Mexico Ischemia/reperfusion injury and cardiac conditioning: basic mechanisms Myocardial IRI has been investigated in experimental animals, and attenuation of this injury can be obtained by pharmacological and mechanical conditioning strategies, including ischemic pre- and postconditioning and ‘‘remote ischemic conditioning’’ (RIC) [36, 45]. sinus blood) became substantially decreased [15]. More- over, in rats undergoing the RIC procedure, RNase1 sig- niﬁcantly rose in animals receiving buprenorphine or isoﬂurane, when compared with RIC without these drugs, implying a signiﬁcant contribution of the RIC-dependent endothelial RNase1 release for improving the outcome of cardiac IRI. Yet, the exact mechanism of RNase1-induced cardioprotection still remains to be uncovered, although split products of RNA, such as nucleotides or nucleosides, may be valuable candidates to confer cardioprotection. It is also possible that medical preconditioning exhausts the conditioning capacity of mammalians, such that RIC may not promote additional preconditioning anymore. p p g y ‘‘Macrophage migration inhibition factor’’ (MIF) family proteins feature overlapping properties with protein-type alarmins, such as high mobility group binding protein-1 (HMGB-1) or RNase1. MIF is known to be a pleiotropic inﬂammatory cytokine with chemokine-like functions and as such is a prototypical member of the emerging class of atypical or chemokine-like function proteins [95]. MIF is also a major pro-atherogenic factor, but counter-intuitively exhibits protective activities in the heart, damaged by IRI [6, 65]. Interestingly, the cardioprotective effects of MIF are predominant in the early reperfusion phase [74] and are ampliﬁed by post-translational modiﬁcations, such as S-nitrosylation and N-oxidation. In line with these obser- vations, exogenously administered (‘‘unmodiﬁed’’) recombinant MIF was unable to convey RIC in a Lan- gendorff heart model [78]. Moreover, MIF produced in the later phase and exacerbated IRI through promoting inﬂammatory leukocyte inﬁltration and activation [20, 76]. The role of novel MIF family member MIF-2/D-DT is currently unclear, as its knockout in a mouse model led to an exacerbation of infarct size, while its levels in patients undergoing coronary artery bypass grafting are predictive of ischemia–reperfusion outcome parameters, such as acute kidney injury [92], clearly necessitating the application of additional models on this and to-be-discovered family members, as well as on the MIF/MIF-2 double knockout mouse. The characterization of the underlying mechanisms and the contributing components in RIC will be a major chal- lenge for future translational work in the cardiovascular ﬁeld to reduce infarct size and improve clinical outcomes in patients with ischemic heart disease [41, 80]. Ischemia/reperfusion injury and cardiac conditioning: basic mechanisms During tissue injury and upon exposure of tissue towards hypoxia as a source of tissue stress, large amounts of intracellular components as alarming compounds are released that dis- seminate into the circulation or remain at the site of cell activation/damage and trigger inﬂammation by activating the innate immune system [109]. In particular, nuclear proteins, such as histones, heat-shock proteins, or ampho- terin [56, 79], as well as nuclear DNA [99], mitochondrial DNA [7, 103, 110], ribosomal RNA [86, 87, 111], and miRNAs, become liberated in isolated or complexed form, such as in association with exosomes or microvesicles [14, 15, 17, 25]. The innate immune system senses the signals released by the necrotic cells and activates inﬂammatory pathways to neutralize such danger signals [75]. In this regard, the endogenous extracellular RNA/RNase system appears to provide new alarmins, primarily present following tissue stress (such as in hypoxia) and cell dam- age: extracellular RNA (eRNA), together with Tumor Necrosis Factor a (TNF-a), serve as damaging factors in experimental IRI, whereas adminstration of RNase1 had a major therapeutical impact by signiﬁcantly reducing the infarct area and preventing cardiomyocyte death [16, 17]. Moreover, in a small pilot clinical study, patients under- going cardiac surgery received initial RIC (4 9 5 min limb ischemia) or a sham procedure. The RIC protocol signiﬁ- cantly increased plasma levels of endogenous vascular RNase1 (derived from endothelial cells) with the conse- quence that circulating eRNA (from arterial and coronary Ischemia/reperfusion injury and cardiac conditioning: basic mechanisms Burns School of Medicine, University of Hawaii, Honolulu, USA 24 Department of Cardiology, The Rayne Institute, St Thomas’ Campus, King’s College London, London, UK 33 The Hatter Cardiovascular Institute, University College London, London, UK 25 The James Black Centre, King’s College, University of London, London, UK 34 The National Institute of Health Research University College London Hospitals Biomedical Research Centre, London, UK 26 Department of Cardiovascular, Renal and Metabolic Medicine, Sapporo Medical University School of Medicine, Sapporo, Japan 35 Department of Cardiovascular Medicine, National Institute of Cardiology, Ignacio Chavez, Mexico, D.F., Mexico 27 Baker IDI Heart and Diabetes Institute, Melbourne, Australia 27 Baker IDI Heart and Diabetes Institute, Melbourne, Australia 28 Department of Cardiology and Angiology, Hannover Medical School, Hannover, Germany 28 Department of Cardiology and Angiology, Hannover Medical School, Hannover, Germany 12 3 Page 3 of 13 69 Basic Res Cardiol (2016) 111:69 69 Page 3 of 13 69 cardiac surgery. Yet, experimental animal models as well as clinical studies have presented diverging results of ischemic conditioning in cardiac surgery and PCI or interventions following acute MI [27, 28, 38, 41, 72]. IRI can manifest as reperfusion-induced arrhythmias, myocar- dial stunning, microvascular obstruction, and cardiomy- ocyte death, being the main cause of reperfusion-induced myocardial lethality. While the former two situations are more or less self-terminating, the third situation occurs in more than 50 % of patients and is associated with capillary damage, microthrombosis, and cardiomyocyte swelling. Moreover, oxidative stress, calcium overload, and irre- versible hypercontracture of cardiomyocytes contribute to lethal myocardial IRI [33, 35]. Myocardial IRI has been investigated in experimental animals, and attenuation of this injury can be obtained by pharmacological and mechanical conditioning strategies, including ischemic pre- and postconditioning and ‘‘remote ischemic conditioning’’ (RIC) [36, 45]. cardiac surgery. Yet, experimental animal models as well as clinical studies have presented diverging results of ischemic conditioning in cardiac surgery and PCI or interventions following acute MI [27, 28, 38, 41, 72]. IRI can manifest as reperfusion-induced arrhythmias, myocar- dial stunning, microvascular obstruction, and cardiomy- ocyte death, being the main cause of reperfusion-induced myocardial lethality. While the former two situations are more or less self-terminating, the third situation occurs in more than 50 % of patients and is associated with capillary damage, microthrombosis, and cardiomyocyte swelling. Moreover, oxidative stress, calcium overload, and irre- versible hypercontracture of cardiomyocytes contribute to lethal myocardial IRI [33, 35]. Ischemia/reperfusion injury and cardiac conditioning: basic mechanisms While myocardial IRI is associated with an increased death of cardiomyocytes, brief cycles of ischemia and reperfu- sion (termed ‘‘ischemic conditioning’’) appear to protect the heart from acute MI and IRI [5, 27, 72]. This phe- nomenon was discovered 30 years ago and stimulated intense research and clinical trials to understand the mechanisms of myocardial IRI and cardioprotection in patients presenting with ST elevation MI or undergoing 21 Department of Cardiology, Dong-A University Hospital, Busan, Korea 22 Hatter Institute and MRC Inter-University Cape Heart Unit, Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa 23 Institute for Molecular Cardiovascular Research, RWTH University Hospital, Aachen, Germany 24 Department of Cardiology, The Rayne Institute, St Thomas’ Campus, King’s College London, London, UK 25 The James Black Centre, King’s College, University of London, London, UK 26 Department of Cardiovascular, Renal and Metabolic Medicine, Sapporo Medical University School of Medicine, Sapporo, Japan 27 Baker IDI Heart and Diabetes Institute, Melbourne, Australia 28 Department of Cardiology and Angiology, Hannover Medical School, Hannover, Germany 22 Hatter Institute and MRC Inter-University Cape Heart Unit, Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa 30 Institute of Anatomy and Vascular Biology, Westfalian- Wilhelms-University, Mu¨nster, Germany 30 Institute of Anatomy and Vascular Biology, Westfalian- Wilhelms-University, Mu¨nster, Germany 31 Institute of Physiology, Justus-Liebig University, Giessen, Germany 23 Institute for Molecular Cardiovascular Research, RWTH University Hospital, Aachen, Germany 32 Department of Pathology, John A. Burns School of Medicine, University of Hawaii, Honolulu, USA 32 Department of Pathology, John A. Ischemia/reperfusion injury and cardiac conditioning: clinical applications The translation of the RIC procedure to patients has been challenging, because improvement of clinical outcome has been observed only in a minority of studies with RIC and ﬁndings are not consistent [30, 41, 80]. The explanation for the difﬁculties in translating the beneﬁcial ﬁndings in experimental animals to humans remains unclear and may include confounding factors, such as age, co-morbidity, co- medication and anesthetic regimen in procedures requiring 12 3 69 Page 4 of 13 Basic Res Cardiol (2016) 111:69 69 CKD that is directly related to insufﬁcient activation of Akt at the time of reperfusion. Chronic treatment with an ery- thropoietin receptor ligand prevented CKD-induced enlargement of myocardial infarct size by restoration of Akt-mediated signaling possibly via normalized malate- aspartate shuttle ﬂux in cardiomyocytes [69]. These promising data may provide clues for deciphering other co- morbidities and their mechanistic relations to the metabolic status in the context of cardiovascular diseases (Fig. 1). CKD that is directly related to insufﬁcient activation of Akt at the time of reperfusion. Chronic treatment with an ery- thropoietin receptor ligand prevented CKD-induced enlargement of myocardial infarct size by restoration of Akt-mediated signaling possibly via normalized malate- aspartate shuttle ﬂux in cardiomyocytes [69]. These promising data may provide clues for deciphering other co- morbidities and their mechanistic relations to the metabolic status in the context of cardiovascular diseases (Fig. 1). general anesthesia [10, 23, 49]. The intensity of the con- ditioning procedure also seems important and depends on the number and duration of the conditioning cycles, thereby deﬁning the efﬁcacy of protection by a speciﬁc algorithm [36, 45]. The indicated intervention can be initiated already in the ambulance during transportation to the PCI table and increases myocardial salvage leading to improved left ventricular function and outcome when used as an adjunct to primary PCI [90]. In the clinical setting, RIC was found to be effective in patients with diabetes mellitus and other cardiovascular risk factors as well [91, 105]. With regard to the multi-ethnic population in a devel- oping country like Malaysia, a pool of cardiac patients was drawn from both urban and rural areas for the analysis of acute coronary syndrome, being the leading cause of mortality. A striking feature was found in that the majority of patients were presented at a younger age group com- pared with similar populations in developed countries. Ischemia/reperfusion injury and cardiac conditioning: clinical applications While early intervention for ST-elevation MI has more relevance in urban areas with well-staffed care centres [96], substantial data from the voluntary ‘‘National Cardiovas- cular Disease Registry’’ (launched in 2006 in Malaysia) resulted in the improvement of risk stratiﬁcation mecha- nisms for patients from rural areas [3]. Such strategies appear to serve as a role-model and can be relevant to the population at risk in other developing countries as well. Yet, in two recent multicentre double-blind randomised controlled clinical trials (RIP-Heart, ERICCA) [64, 105], neutral results were reported upon application of RIC in patients, whereby the type of pre-operative medication appears to have a major impact on the outcome as well [29, 37]. Interestingly enough, both neuronal and vascular factors seem to play an important role in promoting RIC, as demonstrated in experimental animal and human studies [36, 57, 73]. Nevertheless, interventional cardiological procedures in acute angioplasty for ST-segment elevation MI seem to hold the best potential at present. Despite the successful basic and clinical ischemic- reperfusion research for three decades, no deﬁnitive car- dioprotective drugs are available yet. The lack of suc- cessful translation of experimental results into the clinical setting may be due to (1) the lack of proper co-morbidity experimental models as well as (2) the existence of hypothesis-driven-biased approaches to ﬁnd molecular targets. Indeed, major cardiovascular co-morbidities, such as hypertension, hyper-lipidemia, diabetes, and their co- medications, interfere with most of the known cardiopro- tective mechanisms [23]. Moreover, cardiovascular co- morbidities as well as the available conditioning proce- dures affect the global myocardial gene expression proﬁle at the transcriptional level and the ﬁne-tuning regulators of translation, such as miRNAs [97, 98]. Thus, the compre- hensive analysis of the cardioprotective gene expression ﬁngerprint at the transcription and protein level in normal, protected, and in co-morbid probands may lead to the identiﬁcation of novel molecular targets for cardioprotec- tion by an unbiased non-hypothesis-driven approach. 123 New players in cardiovascular diseases The integrity and functionality of the monolayer vascular endothelium as a lively and dynamic barrier between the ﬂowing blood and the underlying tissues, including the heart, depend on the natural implementation of distinct, ultralarge protein complexes at cell–cell borders, charac- teristic for adherens-, tight- und gap-junctions [81]. Newly described, highly dynamic structures, termed ‘‘Junction- associated intermittent lamellipodia’’ (JAIL), which are controlled by the WAVE-WASP/ARP2/3 protein complex, are driven by actin ﬁlament rearrangement to provide small plasma membrane protrusions that preferentially appear at junctional sites of endothelial cells with a low level of cadherin-5 (VE-cadherin) [1, 2, 84]. Thrombin as a pro- inﬂammatory mediator blocks JAIL formation and thus increases endothelial permeability. Such dynamic pro- cesses were made visible by stimulated emission depletion (STED) ﬂuorescence microscopy and 3D reconstruction ‘‘structured illumination microscopy’’ (SIM). Since these structures are also disturbed by ﬂuid shear stress and other stimulants, such as hypoxia, it remains to be analyzed in which way JAIL plays a role in, e.g., ischemia-driven disintegration of the endothelium in small and large cardiac vessels or during myocardial IRI. Like other co-morbidities, chronic kidney disease (CKD) is known as a major risk factor of cardiovascular events and mortality after MI [104]; yet, the contribution of CKD to myocardial IRI remains unclear. In systematic studies, the inﬂuence of CKD on cytoprotective signaling was analyzed using a rat model of CKD, particularly addressing the functional role of Akt-phosphorylation. The results indicate that CKD suppresses Akt-activation upon reperfusion with the disruption of protective signaling to mitochondria, associated with infarct size enlargement. Moreover, the impact of CKD on infarct size depends on the severity of An emerging new concept relates to innate immunity- related protective mechanism of the heart. In the context of 123 Page 5 of 13 69 Basic Res Cardiol (2016) 111:69 Theranostics Potential Target (Remote) Ischemic Conditioning RNase1 Dietary Melatonin Modified MIF Single-chain Antibodies Against Platelet GPIIb/IIIa Akt-mediated signalling Hippo signalling Cardiac-Myosin-BP Induced Pluripotent Stem Cells Ischemia/Reperfusion Injury Extracellular RNA Kidney Disease; Diabetes) Immunothrombosis Cardiac-specific MicroRNAs Soluble Epoxid-Hydrolase ROS / Connexin-43 Co-morbidities (e.g. Chronic g. 1 Potential new targets d theranostics in cardio- tection. The basic chanisms, preclinical models d some clinical applications several cardio-destructive hologies and cardiovascular eases (red box) are discussed the text. New players in cardiovascular diseases Existing and novel agonistic procedures as well the related theranostics een box), both in vitro and in perimental models, were nd to promote cardio- tection on different lecular levels, particularly proving the functional status cardiomyocytes (ROS ctive oxygen species, MIF crophage migration ibition factor, GP coprotein, BP binding tein) Potential Target Ischemia/Reperfusion Injury Extracellular RNA Kidney Disease; Diabetes) Immunothrombosis Co-morbidities (e.g. Chronic Fig. 1 Potential new targets and theranostics in cardio- protection. The basic mechanisms, preclinical models and some clinical applications of several cardio-destructive pathologies and cardiovascular diseases (red box) are discussed in the text. Existing and novel antagonistic procedures as well as the related theranostics (green box), both in vitro and in experimental models, were found to promote cardio- protection on different molecular levels, particularly improving the functional status of cardiomyocytes (ROS reactive oxygen species, MIF macrophage migration inhibition factor, GP glycoprotein, BP binding protein) Cardiac-specific MicroRNAs Soluble Epoxid-Hydrolase ROS / Connexin-43 mechanism of microvascular thrombosis has remained enigmatic. It has been assumed that the failure to assign a clear pathogenetic role to microvascular thrombosis in many diseases is due to difﬁculties in its detection and in the inability to assess the efﬁcacy of antithrombotic treat- ments in the clinical situation. It has recently been shown that during systemic bacterial infections, microvascular thrombosis under certain conditions acts as an instrument of intravascular immunity [22, 62]. In organs, such as the liver and spleen, ﬁbrin-rich microthrombi support the containment and elimination of Escherichia coli inside blood vessels and thereby prevent tissue invasion and dissemination of the pathogens. This mechanism has been termed ‘‘Immunothrombosis’’. Immunothrombosis is sug- gested to form a major biological basis of pathological microvascular and macrovascular thrombosis (especially deep vein thrombosis), together with the physiological mechanism arresting bleeding (haemostasis). Immunothrombosis is a transient process as it appears to be normally resolved within 2 days. Pathological forms of microvascular thrombosis during infections, such as dis- seminated intravascular coagulation, are likely caused by an excessive activation of immunothrombosis and/or by its impaired resolution [82]. Most probably, the formation of microvascular thrombi under non-infectious conditions might equally be able to protect the intravascular com- partment from damage as caused by immune complexes, circulating cell fragments, or endothelial damage. This would indicate that the beneﬁcial nature of microvascular thrombosis may also apply to non-infectious conditions. New players in cardiovascular diseases Hence, the failure to document a pathological role for ischemic heart disease, TNF-a, a major player of the immune system, initiates the induction of a cardioprotec- tive signaling pathway [89] that involves the activation of the signal transducer and activator of transcription 3 (STAT-3) [70], designated as the SAFE (‘‘Survivor Acti- vating Factor Enhancement’’) pathway [50, 53]. Toll-like receptor 4 (TLR4), sphingosine-1 phosphate, and activation of speciﬁc miRNAs are involved in this pathway as well [48]. In particular, TLR4 may trigger the activation of the SAFE pathway to promote cell survival following myocardial IRI [68]. Dietary melatonin, given at a con- centration found in red wine, was demonstrated as respective trigger to confer cardioprotection [52] but also to prevent pulmonary hypertension via the activation of the SAFE pathway [59]. Based on the fact that high-density lipoproteins (HDL) can also activate the SAFE pathway, sphingosine-1 phosphate was identiﬁed as predominant component of HDL to protect against myocardial IRI [94], ultimately regulating the mitochondrial functions of car- diomyocytes [13]. With the detailed analysis of various lipoprotein subfractions (Lipoprint), new insights into their composition and functionality in patients suffering from cardiovascular diseases are now available, and this may provide a personal cardiac risk calculator. So far, microvascular obstruction has been linked to many vascular diseases, including stroke, myocardial infarction, thrombotic microangiopathies, infections, and cancer [39, 85]. Although the pathogenetic relevance of microangiopathies, such as haemolytic-uremic syndrome or disseminated intravascular coagulation, during sepsis has clearly been demonstrated, the pathogenetic 12 3 3 69 Page 6 of 13 69 Page 6 of 13 69 Basic Res Cardiol (2016) 111:69 largely protected against the development of type 2 dia- betes and the associated hypertension when fed a high fat diet. This occurs at the expense of the liver, as sEH controls the expression of key enzymes involved in lipid metabo- lism. Thus, inhibitors of sEH that increase epoxide, but decrease diol levels have potential for the treatment of the metabolic syndrome/type 2 diabetes (inﬂuencing choles- terol homeostasis) and its cardiovascular complications [60]. In which way the level of sEH may affect the pro- cesses during myocardial IRI as they relate to the distur- bance of vascular integrity or the dysfunction of cardiomyocytes remains to be studied (Fig. 1). microvascular thrombosis under several pathological con- ditions could potentially, at least in part, be related to the fact that it is host-protective under those conditions. New players in cardiovascular diseases Following atherectomy, drug-eluting stents allow a deﬁned local application of anti-proliferative agents and other drugs to reduce neointimal formation and restenosis. Despite the tremendous success of drug-eluting stents using unselective cytostatic substances, their efﬁcacy and safety need further improvement. However, strategies to achieve both of these goals are currently lacking. Likewise, myocardial remodeling and regeneration rely on myocar- dial capillary density and thus on effective neovascular- ization after MI. Yet, the mechanisms underlying myocardial angiogenesis and its regulation by miRNAs are not well deﬁned. Comprehensive screens were established to analyze the expression of non-coding RNAs during the development of neointimal hyperplasia in established mouse models of atherogenesis, and particular miRNAs that are regulated in a lesion-, time- and cell-speciﬁc manner were identiﬁed. As an example, inhibition of miRNA-92a appeared to be safe as an effective treatment to prevent neointima formation as well as to improve re- endothelialization at the same time [19]. In addition, miRNA-146a was upregulated in the ischemic myocardium of mice following ligation of the left anterior descending artery in a time-dependent manner. In vitro, the overex- pression of miRNA-146a signiﬁcantly attenuated endothelial cell proliferation, migration, and abolished endothelial capillary network formation. In contrast, knock-down of miRNA-146a markedly augmented endothelial cell proliferation, migration, network forma- tion, and sprouting (unpublished data). Mechanistically, NOX4, NOTCH1, and nRAS were identiﬁed and validated as direct targets of microRNA-146a in endothelial cells (unpublished data). In vivo, antagomirs against miRNA- 146a signiﬁcantly enhanced angiogenesis and re-vascular- ization in the infarcted myocardium, accompanied by preserved cardiac function and a markedly reduced infarct size (unpublished data). It is expected that additional car- diac-speciﬁc miRNAs become characterized that would serve as attractive targets for future therapeutic interven- tions in the treatment of ischemic heart disease. Cardiac repair and regeneration sEH-deﬁcient mice are also 123 123 Basic Res Cardiol (2016) 111:69 Page 7 of 13 69 69 strong case can be made for the importance of the gap junction protein connexin 43 in the context of myocardial IRI and cardioprotection. Apart from being present at gap- junctions along cardiomyocyte cell borders, connexin 43 is also located at mitochondria and is involved in mitochon- drial respiration, ATP generation, and mitochondrial potassium inﬂux [8, 9, 11, 12, 55, 67, 83]. Blockade of connexin 43-formed channels reduces myocardial IRI, but, at the same time, also abolishes cardioprotection induced by ischemic preconditioning. Another redox-linked protein is p66SHC, which translocates into mitochondria where it catalyzes electron transfer from cytochrome c to oxygen resulting in ROS production [34]. Deletion of p66SHC, however, does not reduce infarct size in mice in vivo undergoing 30 min ischemia and 120 min reperfusion (unpublished data), but p66SHC contributes to vascular abnormalities related to diabetes and aging. On the other hand, ROS formation might contribute to self-endogenous defense against mild myocardial IRI [21], whereas p66SHC knockout does not affect endogenous cardiopro- tection (unpublished data). describing heart development and a transcriptomic dataset describing induced cardiomyocyte proliferation were merged and are currently being experimentally validated, eventually leading to novel candidate cytokine genes for cardiac regeneration. Hippo signaling has been implicated in cardiac devel- opment and regeneration after myocardial injury. Genetic deletion of upstream Hippo signaling kinases (Mst1/2, Lats2, or Salvador) leads to an expansion of ventricular myocardium due to increased cardiomyocyte proliferation [31]. Global deletion of Yap results in the early embryonic lethality due to defects in multiple tissues, including yolk sac vasculogenesis, chorioallantoic fusion, and body axis elongation. However, Taz mutant mice are viable but develop glomerulocystic kidney disease and pulmonary disease. Genetic deletion of Yap and Taz both leads to the early embryonic lethality suggesting functional redun- dancy. Despite the studies described above, a role for Yap and Taz in the epicardium during coronary vasculature development has not been explored. Formation of the coronary vasculature is a complex and precisely coordi- nated morphogenetic process that begins with the forma- tion of epicardium. The epicardium gives rise to many components of the coronary vasculature, including ﬁbrob- lasts, smooth muscle cells, and the endothelium. Cardiac repair and regeneration The understanding of developmental and regenerative processes of heart biology and cardiomyocyte proliferation and the underlying mechanisms may lead to their reacti- vation in diseased hearts postnatally and may have a great potential to protect or improve heart function. It is known that the mammalian heart loses its ability to regenerate, largely due to the fact that after birth cardiomyocytes fail to undergo cytokinesis. Instead, they exit the cell cycle after karyokinesis resulting in bi-nucleated cells, and cardiac tissue further expands by hypertrophy. Moreover, the efﬁciency in inducing cardiomyocyte proliferation decreases proportionally to cardiomyocyte age [54, 107]. Thus, an understanding how cardiomyocyte cytokinesis is regulated during development might provide new clues towards cardiac regeneration. Interestingly enough, car- diomyocyte centrosome integrity is lost shortly after birth in mammals. This is coupled with the relocalization of various centrosome proteins to the nuclear envelope. Consequently, postnatal cardiomyocytes are unable to undergo ciliogenesis, and the nuclear envelope adopts the function as cellular microtubule organizing center [108]. Loss of centrosome integrity is associated with, and can promote, cardiomyocyte G0/G1 cell cycle arrest, suggest- ing that centrosome disassembly is developmentally uti- lized to achieve the post-mitotic state in mammalian cardiomyocytes. In contrast, adult newt and zebraﬁsh maintain the ability to regenerate their hearts through proliferation of cardiomyocytes, which also retain centro- some integrity. Based on these novel results, underlying the post-mitotic state of mammalian cardiomyocytes, potential mechanisms of heart regeneration in zebraﬁsh and newts are testable that may provide clues for the regeneration of mammalian hearts. In addition, using systems biology approaches [24], novel regulators of cardiac development and regulatory networks were identiﬁed, integrating large- scale expression datasets. As an example, the joint analysis of a high-resolution temporal expression data set Besides the well-known bioactive lipid mediators derived from arachidonic acid by cytochrome-P450-cat- alyzed reactions, other polyunsaturated fatty acids can be metabolized to epoxides and then diols by the actions of cytochrome P450 enzymes followed by soluble epoxide hydrolase (sEH). These metabolites appear to be ignored in their physiological relevance in the cardiovascular system, although they become generated in high concentrations. Deletion of sEH signiﬁcantly delayed angiogenesis in the retina, a phenomenon associated with activation of the Notch signaling pathway [43]. Cardiac repair and regeneration While Hippo signaling mediators Yap and Taz are expressed in proepicardial and epicardial cells, a combination of genetic and pharmacological approaches that inhibited Hippo sig- naling mediators Yap and Taz also impaired epicardial epithelial-to-mesenchymal transition (EMT) as well as a reduction in epicardial cell proliferation and differentiation into coronary endothelial cells. As a conclusion, Yap and Taz control epicardial cell behavior, in part by regulating Tbx18 and Wt1 expression. These ﬁndings show a role for Hippo signaling in epicardial cell proliferation, EMT, and cell fate speciﬁcation during cardiac organogenesis [88]. It has been shown that the mitochondrial protein NDUFA4L2 plays an essential role in decreasing oxygen consumption and ROS production through inhibition of respiratory chain-complex I [93]. However, even though NDUFA4L2-induced mitochondrial repression has been proven by several research groups [51, 66], its physiolog- ical role remains unknown. It appears that the ‘‘hypoxia- inducible factor’’ (HIF)-NDUFA4L2 axis acts as one of the major pathways for cellular adaptation towards hypoxia via mitochondrial activity suppression. Along this line, NDU- FA4L2 protein was highly expressed in heart tissue, whereby the cardiac fetal tissues exhibited highest levels when compared with tissues of adult mice. Finally, since the fetal heart is one of the sites that present higher levels of NDUFA4L2, a speciﬁc cardiac knockout mouse line was successfully generated by breeding a conditional NDU- FA4L2 exon 2-ﬂoxed mice line with a with a heart-speciﬁc CRE (Nkx 2.5) transgenic line. However, these mice are not embryonically lethal and the role of cardiac NDU- FA4L2 in adulthood warrants further investigation. Cardiac mitochondria and cardioprotection Reactive oxygen species (ROS) at high levels do play an adverse role in myocardial IRI, but contribute to endoge- nous cardioprotection at lower concentrations [35]. More- over, the aforementioned conditioning protocols appear to recruit complex signaling cascades of activation of car- diomyocyte sarcolemmal receptors, intracellular enzymes, as well as ROS and nitrosative species, to gain mitochon- drial stabilisation and ﬁnally to protect against cell death. The following questions remain to be addressed: What are the relevant sources of ROS in the cytosol and the mito- chondria of cardiomyocytes? Which proteins/enzymes do contribute to ROS formation at different levels and how does the ROS-induced ROS release work in detail? A Diagnosis of cardiac damage and new theranostics The diagnosis of ‘‘Non-ST elevation myocardial infarc- tion’’ (NSTEMI) is dominated by the need to document an elevation in cardiac troponins I or T above the population- deﬁned 99th centile. Yet, these biomarkers are released only slowly, thereby delaying the rule-in or the rule-out criteria for NSTEMI. The latest ESC guidelines attempt to circumvent this obstacle by adopting a ‘rule-out’ troponin value signiﬁcantly below the 99th centile and a ‘rule-in’ 12 3 69 Page 8 of 13 Basic Res Cardiol (2016) 111:69 69 value well above the 99th centile [77]. However, this leaves the majority of patients in an undeﬁned diagnostic window, requiring repeated testing and further observation. A new cardiac biomarker, the ‘‘Cardiac myosin-binding protein C’’ (cMyC), had been recently introduced, which constitutes an abundant sarcomeric protein with a unique cardiac isoform that is released upon cardiac damage or iatrogenic MI much faster than the troponins [4, 26]. Using newly generated highly speciﬁc monoclonal antibodies, an ultrasensitive assay to measure cMyC in biological samples was established with a detection limit of 0.4 ng/l and intra- and inter-series precision around 10 % [61]. In ambulatory patients, cMyC concentration in plasma (median 12.23 ng/ l) was linearly correlated with troponin levels, and both parameters showed a similar dependence on age, renal function, or left ventricular activity. In another patient cohort with aortic stenosis, cMyC levels were strongly related to ﬁbrosis (detected by cardiac magnetic resonance imaging) and clinical outcome. It is expected that cMyC will be introduced into the clinics as new diagnostic parameter, e.g., after spontaneous, type 1 MI [46] as well as in other groups of cardiac patients to obtain fast and reli- able results on the degree of cardiac damage. conformational change upon platelet activation, the exposed characteristic epitopes can be used as thrombus- speciﬁc targets. Following the screening by phage-display of PCR-cloned human single-chain antibodies, highly speciﬁc integrin antagonists were generated. These were coupled by genetic, chemical, or biological approaches to obtain various fusion products, which are available either as therapeutic drugs or in combination with contrast par- ticles. Using these approaches, various anticoagulants, anti- platelet drugs, or ﬁbrinolytics were speciﬁcally targeted to the thrombus site to achieve effective thrombolysis and prevention of emboli without any bleeding complications in mice [42]. Acknowledgments The herein cited work was supported as follows: HACF is funded by a Startup Grant of the ‘‘Excellence Cluster Car- dio-Pulmonary System’’ (ECCPS) from the German Research Foun- dation (DFG, Bonn, Germany) and the ‘‘Peter und Traudl Engelhorn- Stiftung’’ (Weilheim, Germany). Part of the work presented by HACF, WAB, and KTP is supported by the Russian Government Program for competitive growth of Kazan Federal University, Kazan (Russian Federation). JA is supported by Grants from the Ministerio de Economı´a y Competitividad (SAF2013-46058-R), Comunidad de Madrid/Fondo Social Europeo (S2010/BMD-2542 ‘‘Consepoc-CM’’) and Red de Cardiovascular (RD12/0042/0065). JB is supported by Deutsche Forschungsgemeinschaft (DFG) Grants BE 1977/9-1, SFB1123 (Project A03), by DFG within the framework of the Munich Cluster for Systems Neurology (EXC 1010 SyNergy), and by Else Kro¨ner-Fresenius Stiftung (EKFS) Grant 2014_A216. HB is Diagnosis of cardiac damage and new theranostics Moreover, these procedures allow the design of targeted ‘‘theranostic compounds’’, such as ultrasound micro-bubbles, magnetic resonance nanoparticles, or posi- tron emission tomography tracers, for thrombus detection with high sensitivity and speciﬁcity in the respective imaging modality [101]. Compounds are underway that allows the combination of detection and imaging of thrombi with concomitant effective treatment together with the monitoring of success or failure of therapy [100] (Fig. 1). Human-induced pluripotent stem cell (iPSC)-derived cardiomyocytes present a tremendous opportunity for the study of cardiac arrhythmias in vitro. The characterization of cellular models for major subtypes of inherited chan- nelopathy revealed that these defects were caused by dys- functional potassium and sodium channels that contribute to the Long QT Syndrome (LQTS) 1, LQTS2, and LQTS3 [63]. In particular, the correction of the trafﬁcking of KCNH2 (LQTS2) potassium channel through intracellular mechanisms restored hERG currents and reduced arrhyth- mia in LQTS2 patient-derived cardiomyocytes, also doc- umenting the usefulness of iPSC-cardiomyocytes in LQTS2 modeling and drug testing. Moreover, iPSC-car- diomyocytes were used as a tool to evaluate the cardiac toxicity of topical drugs [58]. These applications document the powerful iPSC technology as value creation for understanding the pathomechanisms of cardiac arrhyth- mias, but also for drug testing and toxicology research. Besides diagnostic markers, there is also need for new prognostic biomarkers. With the rise in obesity and its associated metabolic complications, many more patients will be considered at high risk of cardiovascular disease. Circulating miRNAs have recently evolved as novel play- ers in the ﬁeld of medicine [102]. Platelets contain and release miRNAs and are a major source of abundant miRNAs in plasma and in serum. There is a striking cor- relation of miRNAs with platelet activation markers in the general population and platelet-derived miRNAs in plasma correlate with indices of platelet function in patients on dual anti-platelet therapy [47]. Moreover, platelet miRNAs appear to alter their function, most probably by inﬂuencing gene expression in megakaryocytes [47]. Since thrombus formation is a key event in triggering the clinical mani- festations of atherosclerotic disease, it could be informative to assess platelet activation in the context of cardiovascular risk. Similarly, circulating angiogenic miRNAs have been linked to the onset and progression of retinopathy in patients with type 1 diabetes [106]. Diagnosis of cardiac damage and new theranostics SBO is supported by a Khoo Post- doctoral Fellowship Award (Duke-NUS-KPFA/2016/0010) from the Estate of Tan Sri Khoo Teck Puat, Singapore. This work is supported by a German Research Foundation (Cluster of excellence REBIRTH) Grant to DS. MKS is supported by funds from Duke-NUS Medical School Singapore, Goh foundation and Singapore National Research Foundation (NRF) fellowship (NRF-NRFF2016-01). This work is supported by the German Research Council DFG, INST 2105/24-1 and SCHN 430/6-2 to HS. RS is supported by the German Research Foundation (DFG, Bonn, Germany) Schu843/9-1. This work was supported by funds from the National Medical Research Council (NMRC/BNIG/1074/2012), Goh Foundation/Duke-NUS Medical School (GCR/2013/008, GCR/2013/010 and GCR/2013/011), Sin- gHealth Foundation (SHF/FG569P/2014 and SHF/FG630S/2014), Biomedical Research Council Singapore (BMRC13/1/96/19/686A), and Singapore National Research Foundation (NRF) Competitive Research Program (NRF-CRP-2008-02) to WS. MW acknowledges funding support from the Cardiometabolic Board of the Biomedical Research Centre at UCL, funded by the NIHR UK. summary of three years report. Int J Cardiol 165:161–164. doi:10.1016/j.ijcard.2011.08.015 4. Baker JO, Tyther R, Liebetrau C, Clark J, Howarth R, Patterson T, Mollmann H, Nef H, Sicard P, Kailey B, Devaraj R, Redwood SR, Kunst G, Weber E, Marber MS (2015) Cardiac myosin- binding protein C: a potential early biomarker of myocardial injury. Basic Res Cardiol 110:23. doi:10.1007/s00395-015- 0478-5 5. Bell RM, Botker HE, Carr RD, Davidson SM, Downey JM, Dutka DP, Heusch G, Ibanez B, Macallister R, Stoppe C, Ovize M, Redington A, Walker JM, Yellon DM (2016) 9th Hatter Biannual Meeting: position document on ischaemia/reperfusion injury, conditioning and the ten commandments of cardiopro- tection. Basic Res Cardiol 111:41. doi:10.1007/s00395-016- 0558-1 6. Bernhagen J, Krohn R, Lue H, Gregory JL, Zernecke A, Koenen RR, Dewor M, Georgiev I, Schober A, Leng L, Kooistra T, Fingerle-Rowson G, Ghezzi P, Kleemann R, McColl SR, Bucala R, Hickey MJ, Weber C (2007) MIF is a noncognate ligand of CXC chemokine receptors in inﬂammatory and atherogenic cell recruitment. Nat Med 13:587–596. doi:10.1038/nm1567 7. Bliksoen M, Mariero LH, Torp MK, Baysa A, Ytrehus K, Haugen F, Seljeﬂot I, Vaage J, Valen G, Stenslokken KO (2016) Extracellular mtDNA activates NF-kappaB via toll-like receptor 9 and induces cell death in cardiomyocytes. Basic Res Cardiol 111:42. doi:10.1007/s00395-016-0553-6 8. Boengler K, Dodoni G, Rodriguez-Sinovas A, Cabestrero A, Ruiz-Meana M, Gres P, Konietzka I, Lopez-Iglesias C, Garcia- Dorado D, Di Lisa F, Heusch G, Schulz R (2005) Connexin 43 in cardiomyocyte mitochondria and its increase by ischemic preconditioning. Diagnosis of cardiac damage and new theranostics Acknowledgments The herein cited work was supported as follows: HACF is funded by a Startup Grant of the ‘‘Excellence Cluster Car- dio-Pulmonary System’’ (ECCPS) from the German Research Foun- dation (DFG, Bonn, Germany) and the ‘‘Peter und Traudl Engelhorn- Stiftung’’ (Weilheim, Germany). Part of the work presented by HACF, WAB, and KTP is supported by the Russian Government Program for competitive growth of Kazan Federal University, Kazan (Russian Federation). JA is supported by Grants from the Ministerio de Economı´a y Competitividad (SAF2013-46058-R), Comunidad de Madrid/Fondo Social Europeo (S2010/BMD-2542 ‘‘Consepoc-CM’’) and Red de Cardiovascular (RD12/0042/0065). JB is supported by Deutsche Forschungsgemeinschaft (DFG) Grants BE 1977/9-1, SFB1123 (Project A03), by DFG within the framework of the Munich Cluster for Systems Neurology (EXC 1010 SyNergy), and by Else Kro¨ner-Fresenius Stiftung (EKFS) Grant 2014_A216. HB is To provide an efﬁcient antithrombotic therapy in the context of cardiovascular diseases, the detection and elimination of thrombi and emboli are the major challenge in clinical practice. Here, the site-directed molecular imaging and the associated therapeutic targeting of plate- let-speciﬁc antigens is a highly promising approach. An established target in the therapy of cardiovascular disease is the integrin aIIb-b3 (GP IIb/IIIa, CD41/CD61), the highly abundant ﬁbrinogen receptor on the activated pla- telet surface. Since this integrin undergoes a 123 Page 9 of 13 69 Basic Res Cardiol (2016) 111:69 supported by the Danish Council for Strategic Research (11-115818), Trygfonden, NovoNordisk Fonden. WAB is funded by NIH Grants HL075677 and HL081863. PF is supported by Grants from the Hungarian Scientiﬁc Research Fund (OTKA ANN 107803, OTKA K-105555). This work was supported by the Interdisciplinary Centre for Clinical Research Erlangen (IZKF projects J42 to FF and F3 to FBE), and the Emerging Fields Initiative Cell ‘‘Cycle in Disease and Regeneration (CYDER)’’ (Friedrich-Alexander-University Erlangen- Nu¨rnberg, to FBE). IF is funded by the Deutsche Forschungsge- meinschaft (SFB 1039/A6). SL is funded from Winetech, National Research Foundation. This study was supported by the Interdisci- plinary Centre for Clinical Research IZKF Aachen (junior research group to EAL). MMarber is supported by Grants from the Medical Research Council (UK) (G1000737), Guy’s and St Thomas’ Charity (R060701, R100404), British Heart Foundation (TG/15/1/31518), and the UK Department of Health through the National Institute for Health Research Biomedical Research Centre award to Guy’s and St Thomas’ NHS Foundation Trust. Conﬂict of interest HEB is a shareholder of CellAegis Inc. 10. Boengler K, Schulz R, Heusch G (2009) Loss of cardioprotec- tion with ageing. Cardiovasc Res 83:247–261. doi:10.1093/cvr/ cvp033 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://crea tivecommons.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. 11. Boengler K, Stahlhofen S, van de Sand A, Gres P, Ruiz-Meana M, Garcia-Dorado D, Heusch G, Schulz R (2009) Presence of connexin 43 in subsarcolemmal, but not in interﬁbrillar car- diomyocyte mitochondria. Basic Res Cardiol 104:141–147. doi:10.1007/s00395-009-0007-5 12. Boengler K, Ungefug E, Heusch G, Leybaert L, Schulz R (2013) Connexin 43 impacts on mitochondrial potassium uptake. Front Pharmacol 4:73. doi:10.3389/fphar.2013.00073 13. Brulhart-Meynet MC, Braunersreuther V, Brinck J, Montecucco F, Prost JC, Thomas A, Galan K, Pelli G, Pedretti S, Vuilleumier N, Mach F, Lecour S, James RW, Frias MA (2015) Improving reconstituted HDL composition for efﬁcient post-ischemic reduction of ischemia reperfusion injury. PLoS One 10:e0119664. doi:10.1371/journal.pone.0119664 Diagnosis of cardiac damage and new theranostics Cardiovasc Res 67:234–244. doi:10.1016/j.car diores.2005.04.014 9. Boengler K, Ruiz-Meana M, Gent S, Ungefug E, Soetkamp D, Miro-Casas E, Cabestrero A, Fernandez-Sanz C, Semenzato M, Di Lisa F, Rohrbach S, Garcia-Dorado D, Heusch G, Schulz R (2012) Mitochondrial connexin 43 impacts on respiratory complex I activity and mitochondrial oxygen consumption. J Cell Mol Med 16:1649–1655. doi:10.1111/j.1582-4934.2011. 01516.x Compliance with ethical standards Conﬂict of interest HEB is a shareholder of CellAegis Inc. References doi:10.1111/bph.13478 22. Engelmann B, Massberg S (2013) Thrombosis as an intravas- cular effector of innate immunity. Nat Rev Immunol 13:34–45. doi:10.1038/nri3345 39. Heusch G, Kleinbongard P, Skyschally A (2013) Myocardial infarction and coronary microvascular obstruction: an intimate, but complicated relationship. Basic Res Cardiol 108:380. doi:10. 1007/s00395-013-0380-y 23. Ferdinandy P, Hausenloy DJ, Heusch G, Baxter GF, Schulz R (2014) Interaction of risk factors, comorbidities, and comedi- cations with ischemia/reperfusion injury and cardioprotection by preconditioning, postconditioning, and remote conditioning. Pharmacol Rev 66:1142–1174. doi:10.1124/pr.113.008300 40. Heusch G, Libby P, Gersh B, Yellon D, Bohm M, Lopaschuk G, Opie L (2014) Cardiovascular remodelling in coronary artery disease and heart failure. Lancet 383:1933–1943. doi:10.1016/ S0140-6736(14)60107-0 p 24. Ferrazzi F, Bellazzi R, Engel FB (2015) Gene network analysis: from heart development to cardiac therapy. Thromb Haemost 113:522–531. doi:10.1160/TH14-06-0483 41. Heusch G, Rassaf T (2016) Time to give up on cardioprotection? A critical appraisal of clinical studies on ischemic pre-, post-, and remote conditioning. Circ Res 119:676–695. doi:10.1161/ CIRCRESAHA.116.308736 25. Fischer S, Cabrera-Fuentes HA, Noll T, Preissner KT (2014) Impact of extracellular RNA on endothelial barrier function. Cell Tissue Res 355:635–645. doi:10.1007/s00441-014-1850-8 42. Hohmann JD, Wang X, Krajewski S, Selan C, Haller CA, Straub A, Chaikof EL, Nandurkar HH, Hagemeyer CE, Peter K (2013) Delayed targeting of CD39 to activated platelet GPIIb/IIIa via a single-chain antibody: breaking the link between antithrombotic potency and bleeding? Blood 121:3067–3075. doi:10.1182/ blood-2012-08-449694 26. Giannitsis E, Katus HA (2015) A long way to translation: will cMyC survive? Basic Res Cardiol 110:24. doi:10.1007/s00395- 015-0479-4 27. Hausenloy DJ, Baxter G, Bell R, Botker HE, Davidson SM, Downey J, Heusch G, Kitakaze M, Lecour S, Mentzer R, Mocanu MM, Ovize M, Schulz R, Shannon R, Walker M, Walkinshaw G, Yellon DM (2010) Translating novel strategies for cardioprotection: the Hatter Workshop Recommendations. Basic Res Cardiol 105:677–686. doi:10.1007/s00395-010-0121- 4 43. Hu J, Popp R, Fromel T, Ehling M, Awwad K, Adams RH, Hammes HP, Fleming I (2014) Muller glia cells regulate Notch signaling and retinal angiogenesis via the generation of 19,20- dihydroxydocosapentaenoic acid. J Exp Med 211:281–295. doi:10.1084/jem.20131494 44. Ibanez B, Heusch G, Ovize M, Van de Werf F (2015) Evolving therapies for myocardial ischemia/reperfusion injury. J Am Coll Cardiol 65:1454–1471. doi:10.1016/j.jacc.2015.02.032 28. Hausenloy DJ, Botker HE, Engstrom T, Erlinge D, Heusch G, Ibanez B, Kloner RA, Ovize M, Yellon DM, Garcia-Dorado D (2016) Targeting reperfusion injury in patients with ST-segment elevation myocardial infarction: trials and tribulations. References 1. Abu Taha A, Schnittler HJ (2014) Dynamics between actin and the VE-cadherin/catenin complex: novel aspects of the ARP2/3 complex in regulation of endothelial junctions. Cell Adh Migr 8:125–135. doi:10.4161/cam.28243 1. Abu Taha A, Schnittler HJ (2014) Dynamics between actin and the VE-cadherin/catenin complex: novel aspects of the ARP2/3 complex in regulation of endothelial junctions. Cell Adh Migr 8:125–135. doi:10.4161/cam.28243 14. Cabrera-Fuentes HA, Alba-Alba C, Aragones J, Bernhagen J, Boisvert WA, Botker HE, Cesarman-Maus G, Fleming I, Gar- cia-Dorado D, Lecour S, Liehn E, Marber MS, Marina N, Mayr M, Perez-Mendez O, Miura T, Ruiz-Meana M, Salinas-Este- fanon EM, Ong SB, Schnittler HJ, Sanchez-Vega JT, Sumoza- Toledo A, Vogel CW, Yarullina D, Yellon DM, Preissner KT, Hausenloy DJ (2016) Meeting report from the 2nd International Symposium on New Frontiers in Cardiovascular Research. Protecting the cardiovascular system from ischemia: between 2. Abu Taha A, Taha M, Seebach J, Schnittler HJ (2014) ARP2/3- mediated junction-associated lamellipodia control VE-cadherin- based cell junction dynamics and maintain monolayer integrity. Mol Biol Cell 25:245–256. doi:10.1091/mbc.E13-07-0404 2. Abu Taha A, Taha M, Seebach J, Schnittler HJ (2014) ARP2/3- mediated junction-associated lamellipodia control VE-cadherin- based cell junction dynamics and maintain monolayer integrity. Mol Biol Cell 25:245–256. doi:10.1091/mbc.E13-07-0404 3. Ahmad WA, Ali RM, Khanom M, Han CK, Bang LH, Yip AF, Ghazi AM, Ismail O, Zambahari R, Hian SK (2013) The journey of Malaysian NCVD-PCI (National Cardiovascular Disease Database-Percutaneous Coronary Intervention) Registry: a 3. Ahmad WA, Ali RM, Khanom M, Han CK, Bang LH, Yip AF, Ghazi AM, Ismail O, Zambahari R, Hian SK (2013) The journey of Malaysian NCVD-PCI (National Cardiovascular Disease Database-Percutaneous Coronary Intervention) Registry: a 12 3 69 Page 10 of 13 69 Basic Res Cardiol (2016) 111:69 bench and bedside. Basic Res Cardiol 111:7. doi:10.1007/ s00395-015-0527-0 Robertson S, Xenou M, Clayton T, Yellon DM (2016) Effect of remote ischaemic preconditioning on clinical outcomes in patients undergoing coronary artery bypass graft surgery (ERICCA study): a multicentre double-blind randomised con- trolled clinical trial, Southampton. doi:10.3310/eme03040 15. Cabrera-Fuentes HA, Niemann B, Grieshaber P, Wollbrueck M, Gehron J, Preissner KT, Boning A (2015) RNase1 as a potential mediator of remote ischaemic preconditioning for cardiopro- tectiondagger. Eur J Cardiothorac Surg 48(5):732–737. doi:10. 1093/ejcts/ezu519 30. Hausenloy DJ, Yellon DM (2015) Targeting myocardial reper- fusion injury—the search continues. N Engl J Med 373:1073–1075. doi:10.1056/NEJMe1509718 16. References Cabrera-Fuentes HA, Preissner KT (2014) Abstract 20396: Induction of ischemia–reperfusion injury by extracellular RNA: a case for tumor necrosis factor (TNF-a)—shedding. Circulation 130:A20396 31. Heallen T, Zhang M, Wang J, Bonilla-Claudio M, Klysik E, Johnson RL, Martin JF (2011) Hippo pathway inhibits Wnt signaling to restrain cardiomyocyte proliferation and heart size. Science 332:458–461. doi:10.1126/science.1199010 Science 332:458–461. doi:10.1126/science.1199010 17. Cabrera-Fuentes HA, Ruiz-Meana M, Simsekyilmaz S, Kostin S, Inserte J, Saffarzadeh M, Galuska SP, Vijayan V, Barba I, Barreto G, Fischer S, Lochnit G, Ilinskaya ON, Baumgart-Vogt E, Boning A, Lecour S, Hausenloy DJ, Liehn EA, Garcia-Do- rado D, Schluter KD, Preissner KT (2014) RNase1 prevents the damaging interplay between extracellular RNA and tumour necrosis factor-alpha in cardiac ischaemia/reperfusion injury. Thromb Haemost 112:1110–1119. doi:10.1160/TH14-08-0703 32. Heusch G (2013) Cardioprotection: chances and challenges of its translation to the clinic. Lancet 381:166–175. doi:10.1016/ S0140-6736(12)60916-7 33. Heusch G (2016) The coronary circulation as a target of car- dioprotection. Circ Res 118:1643–1658. doi:10.1161/CIRCRE SAHA.116.308640 34. Heusch G (2015) Mitochondria at the heart of cardiovascular protection: p66shc-friend or foe? Eur Heart J 36:469–471. doi:10.1093/eurheartj/ehu409 18. Cochain C, Zernecke A (2015) Macrophages and immune cells in atherosclerosis: recent advances and novel concepts. Basic Res Cardiol 110:34. doi:10.1007/s00395-015-0491-8 35. Heusch G (2015) Molecular basis of cardioprotection: signal transduction in ischemic pre-, post-, and remote conditioning. Circ Res 116:674–699. doi:10.1161/CIRCRESAHA.116.305348 19. Daniel JM, Penzkofer D, Teske R, Dutzmann J, Koch A, Bie- lenberg W, Bonauer A, Boon RA, Fischer A, Bauersachs J, van Rooij E, Dimmeler S, Sedding DG (2014) Inhibition of miR-92a improves re-endothelialization and prevents neointima forma- tion following vascular injury. Cardiovasc Res 103:564–572. doi:10.1093/cvr/cvu162 36. Heusch G, Botker HE, Przyklenk K, Redington A, Yellon D (2015) Remote ischemic conditioning. J Am Coll Cardiol 65:177–195. doi:10.1016/j.jacc.2014.10.031 37. Heusch G, Gersh BJ (2016) ERICCA and RIPHeart: two nails in the cofﬁn for cardioprotection by remote ischemic conditioning? Probably not! Eur Heart J 37:200–202. doi:10.1093/eurheartj/ ehv606 20. Dayawansa NH, Gao XM, White DA, Dart AM, Du XJ (2014) Role of MIF in myocardial ischaemia and infarction: insight from recent clinical and experimental ﬁndings. Clin Sci (Lond) 127:149–161. doi:10.1042/CS20130828 38. Heusch G, Gersh BJ (2016) The pathophysiology of acute myocardial infarction and strategies of protection beyond reperfusion: a continual challenge. Eur Heart J. doi:10.1093/ eurheartj/ehw224 21. Di Lisa F, Giorgio M, Ferdinandy P, Schulz R (2016) New aspects of p66Shc in ischemia reperfusion injury and cardio- vascular diseases. Br J Pharmacol. References Eur Heart J. doi:10.1093/eurheartj/ehw145 45. Johnsen J, Pryds K, Salman R, Lofgren B, Kristiansen SB, Botker HE (2016) The remote ischemic preconditioning algo- rithm: effect of number of cycles, cycle duration and effector organ mass on efﬁcacy of protection. Basic Res Cardiol 111:10. doi:10.1007/s00395-016-0529-6 29. Hausenloy DJ, Candilio L, Evans R, Ariti C, Jenkins DP, Kol- vekar S, Knight R, Kunst G, Laing C, Nicholas JM, Pepper J, 12 3 Basic Res Cardiol (2016) 111:69 Page 11 of 13 69 69 46. Kaier TE, Anand A, Shah AS, Mills NL, Marber M (2016) Temporal relationship between cardiac myosin-binding protein C and cardiac troponin I in type 1 myocardial infarction. Clin Chem 62:1153–1155. doi:10.1373/clinchem.2016.257188 61. Marjot J, Liebetrau C, Goodson RJ, Kaier T, Weber E, Heseltine P, Marber MS (2016) The development and application of a high-sensitivity immunoassay for cardiac myosin-binding pro- tein C. Transl Res 170(17–25):e11–e15. doi:10.1016/j.trsl.2015. 11.008 47. Kaudewitz D, Skroblin P, Bender LH, Barwari T, Willeit P, Pechlaner R, Sunderland NP, Willeit K, Morton AC, Armstrong PC, Chan MV, Lu R, Yin X, Gracio F, Dudek K, Langley SR, Zampetaki A, de Rinaldis E, Ye S, Warner TD, Saxena A, Kiechl S, Storey RF, Mayr M (2016) Association of MicroRNAs and YRNAs with platelet function. Circ Res 118:420–432. doi:10.1161/CIRCRESAHA.114.305663 62. Massberg S, Grahl L, von Bruehl ML, Manukyan D, Pfeiler S, Goosmann C, Brinkmann V, Lorenz M, Bidzhekov K, Khandagale AB, Konrad I, Kennerknecht E, Reges K, Holden- rieder S, Braun S, Reinhardt C, Spannagl M, Preissner KT, Engelmann B (2010) Reciprocal coupling of coagulation and innate immunity via neutrophil serine proteases. Nat Med 16:887–896. doi:10.1038/nm.2184 48. Kelly-Laubscher RF, King JC, Hacking D, Somers S, Hastie S, Stewart T, Imamdin A, Maarman G, Pedretti S, Lecour S (2014) Cardiac preconditioning with sphingosine-1-phosphate requires activation of signal transducer and activator of transcription-3. Cardiovasc J Afr 25:118–123. doi:10.5830/CVJA-2014-016 63. Mehta A, Sequiera GL, Ramachandra CJ, Sudibyo Y, Chung Y, Sheng J, Wong KY, Tan TH, Wong P, Liew R, Shim W (2014) Re-trafﬁcking of hERG reverses long QT syndrome 2 phenotype in human iPS-derived cardiomyocytes. Cardiovasc Res 102:497–506. doi:10.1093/cvr/cvu060 49. Kleinbongard P, Neuhauser M, Thielmann M, Kottenberg E, Peters J, Jakob H, Heusch G (2016) Confounders of cardio- protection by remote ischemic preconditioning in patients undergoing coronary artery bypass grafting. Cardiology 133:128–133. doi:10.1159/000441216 64. References Meybohm P, Bein B, Brosteanu O, Cremer J, Gruenewald M, Stoppe C, Coburn M, Schaelte G, Boning A, Niemann B, Roesner J, Kletzin F, Strouhal U, Reyher C, Laufenberg-Feld- mann R, Ferner M, Brandes IF, Bauer M, Stehr SN, Kortgen A, Wittmann M, Baumgarten G, Meyer-Treschan T, Kienbaum P, Heringlake M, Schon J, Sander M, Treskatsch S, Smul T, Wolwender E, Schilling T, Fuernau G, Hasenclever D, Zacharowski K, Collaborators RIS (2015) A multicenter trial of remote ischemic preconditioning for heart surgery. N Engl J Med 373:1397–1407. doi:10.1056/NEJMoa1413579 50. Lacerda L, McCarthy J, Mungly SF, Lynn EG, Sack MN, Opie LH, Lecour S (2010) TNFalpha protects cardiac mitochondria independently of its cell surface receptors. Basic Res Cardiol 105:751–762. doi:10.1007/s00395-010-0113-4 51. Lai RK, Xu IM, Chiu DK, Tse AP, Wei LL, Law CT, Lee D, Wong CM, Wong MP, Ng IO, Wong CC (2016) NDUFA4L2 ﬁne-tunes oxidative stress in hepatocellular carcinoma. Clin Cancer Res 22:3105–3117. doi:10.1158/1078-0432.CCR-15- 1987 65. Miller EJ, Li J, Leng L, McDonald C, Atsumi T, Bucala R, Young LH (2008) Macrophage migration inhibitory factor stimulates AMP-activated protein kinase in the ischaemic heart. Nature 451:578–582. doi:10.1038/nature06504 52. Lamont K, Nduhirabandi F, Adam T, Thomas DP, Opie LH, Lecour S (2015) Role of melatonin, melatonin receptors and STAT3 in the cardioprotective effect of chronic and moderate consumption of red wine. Biochem Biophys Res Commun 465:719–724. doi:10.1016/j.bbrc.2015.08.064 66. Minton DR, Fu L, Mongan NP, Shevchuk MM, Nanus DM, Gudas LJ (2016) Role of NADH dehydrogenase (ubiquinone) 1 alpha subcomplex 4-like 2 in clear cell renal cell carcinoma. Clin Cancer Res 22:2791–2801. doi:10.1158/1078-0432.CCR- 15-1511 53. Lecour S, James RW (2011) When are pro-inﬂammatory cytokines SAFE in heart failure? Eur Heart J 32:680–685. doi:10.1093/eurheartj/ehq484 67. Miro-Casas E, Ruiz-Meana M, Agullo E, Stahlhofen S, Rodri- guez-Sinovas A, Cabestrero A, Jorge I, Torre I, Vazquez J, Boengler K, Schulz R, Heusch G, Garcia-Dorado D (2009) Connexin43 in cardiomyocyte mitochondria contributes to mitochondrial potassium uptake. Cardiovasc Res 83:747–756. doi:10.1093/cvr/cvp157 54. Leone M, Magadum A, Engel FB (2015) Cardiomyocyte pro- liferation in cardiac development and regeneration: a guide to methodologies and interpretations. Am J Physiol Heart Circ Physiol 309:H1237–H1250. doi:10.1152/ajpheart.00559.2015 55. Li X, Heinzel FR, Boengler K, Schulz R, Heusch G (2004) Role of connexin 43 in ischemic preconditioning does not involve intercellular communication through gap junctions. J Mol Cell Cardiol 36:161–163 68. References Simsekyilmaz S, Cabrera-Fuentes HA, Meiler S, Kostin S, Baumer Y, Liehn EA, Weber C, Boisvert WA, Preissner KT, Zernecke A (2014) Role of extracellular RNA in atherosclerotic plaque formation in mice. Circulation 129:598–606. doi:10. 1161/CIRCULATIONAHA.113.002562 74. Pohl J, Hendgen-Cotta UB, Rammos C, Luedike P, Mull E, Stoppe C, Julicher K, Lue H, Merx MW, Kelm M, Bernhagen J, Rassaf T (2016) Targeted intracellular accumulation of macro- phage migration inhibitory factor in the reperfused heart medi- ates cardioprotection. Thromb Haemost 115:200–212. doi:10. 1160/TH15-05-0436 88. Singh A, Ramesh S, Cibi DM, Yun LS, Li J, Li L, Manderﬁeld LJ, Olson EN, Epstein JA, Singh MK (2016) Hippo signaling mediators yap and taz are required in the epicardium for coro- nary vasculature development. Cell Rep 15:1384–1393. doi:10. 1016/j.celrep.2016.04.027 89. Skyschally A, Gres P, Hoffmann S, Haude M, Erbel R, Schulz R, Heusch G (2007) Bidirectional role of tumor necrosis factor- alpha in coronary microembolization: progressive contractile dysfunction versus delayed protection against infarction. Circ Res 100:140–146. doi:10.1161/01.RES.0000255031.15793.86 75. Preissner KT, Boisvert WA, Hausenloy DJ (2015) Surﬁng on the cardiovascular frontier wave. Thromb Haemost 113:439–440. doi:10.1160/TH15-01-0086 76. Rassaf T, Weber C, Bernhagen J (2014) Macrophage migration inhibitory factor in myocardial ischaemia/reperfusion injury. Cardiovasc Res 102:321–328. doi:10.1093/cvr/cvu071 90. Sloth AD, Schmidt MR, Munk K, Kharbanda RK, Redington AN, Schmidt M, Pedersen L, Sorensen HT, Botker HE, Inves- tigators C (2014) Improved long-term clinical outcomes in patients with ST-elevation myocardial infarction undergoing remote ischaemic conditioning as an adjunct to primary percu- taneous coronary intervention. Eur Heart J 35:168–175. doi:10. 1093/eurheartj/eht369 77. RofﬁM, Patrono C, Zamorano JL (2016) CardioPulse: what’s new in the 2015 European Society of Cardiology Guidelines for the management of acute coronary syndromes in patients pre- senting without persistent ST-segment elevation. Eur Heart J 37:206–207. doi:10.1093/eurheartj/ehv654 78. Rossello X, Burke N, Stoppe C, Bernhagen J, Davidson SM, Yellon DM (2016) Exogenous administration of recombinant mif at physiological concentrations failed to attenuate infarct size in a Langendorff perfused isolated mouse heart model. Cardiovasc Drugs Ther. doi:10.1007/s10557-016-6673-2 j 91. Sloth AD, Schmidt MR, Munk K, Schmidt M, Pedersen L, Sorensen HT, Botker HE, Investigators C (2015) Impact of cardiovascular risk factors and medication use on the efﬁcacy of remote ischaemic conditioning: post hoc subgroup analysis of a randomised controlled trial. BMJ Open 5:e006923. doi:10.1136/ bmjopen-2014-006923 79. References Saffarzadeh M, Cabrera-Fuentes HA, Veit F, Jiang D, Scharf- fetter-Kochanek K, Gille C, Rooijakkers SHM, Hartl D, Preissner KT (2014) Characterization of rapid neutrophil extracellular trap formation and its cooperation with phagocy- tosis in human neutrophils. Discoveries 2:e19. doi:10.15190/d. 2014.11 j p 92. Stoppe C, Rex S, Goetzenich A, Kraemer S, Emontzpohl C, Soppert J, Averdunk L, Sun Y, Rossaint R, Lue H, Huang C, Song Y, Pantouris G, Lolis E, Leng L, Schulte W, Bucala R, Weber C, Bernhagen J (2015) Interaction of MIF family proteins in myocardial ischemia/reperfusion damage and their inﬂuence on clinical outcome of cardiac surgery patients. Antioxid Redox Signal 23:865–879. doi:10.1089/ars.2014.6243 80. Sardar P, Chatterjee S, Kundu A, Samady H, Owan T, Giri J, Nairooz R, Selzman CH, Heusch G, Gersh BJ, Abbott JD, Mukherjee D, Fang JC (2016) Remote ischemic preconditioning in patients undergoing cardiovascular surgery: evidence from a meta-analysis of randomized controlled trials. Int J Cardiol 221:34–41. doi:10.1016/j.ijcard.2016.06.325 93. Tello D, Balsa E, Acosta-Iborra B, Fuertes-Yebra E, Elorza A, Ordonez A, Corral-Escariz M, Soro I, Lopez-Bernardo E, Per- ales-Clemente E, Martinez-Ruiz A, Enriquez JA, Aragones J, Cadenas S, Landazuri MO (2011) Induction of the mitochon- drial NDUFA4L2 protein by HIF-1alpha decreases oxygen consumption by inhibiting Complex I activity. Cell Metab 14:768–779. doi:10.1016/j.cmet.2011.10.008 81. Schnittler H (2014) Between sealing and leakiness: molecular dynamics of the endothelium to maintain and regulate barrier function. Cell Tissue Res 355:481–483. doi:10.1007/s00441- 014-1838-4 j 94. Theilmeier G, Schmidt C, Herrmann J, Keul P, Schafers M, Herrgott I, Mersmann J, Larmann J, Hermann S, Stypmann J, Schober O, Hildebrand R, Schulz R, Heusch G, Haude M, von Wnuck Lipinski K, Herzog C, Schmitz M, Erbel R, Chun J, Levkau B (2006) High-density lipoproteins and their con- stituent, sphingosine-1-phosphate, directly protect the heart against ischemia/reperfusion injury in vivo via the S1P3 lysophospholipid receptor. Circulation 114:1403–1409. doi:10. 1161/CIRCULATIONAHA.105.607135 82. Schulz C, Engelmann B, Massberg S (2013) Crossroads of coagulation and innate immunity: the case of deep vein throm- bosis. J Thromb Haemost 11(Suppl 1):233–241. doi:10.1111/jth. 12261 83. Schulz R, Gorge PM, Gorbe A, Ferdinandy P, Lampe PD, Leybaert L (2015) Connexin 43 is an emerging therapeutic target in ischemia/reperfusion injury, cardioprotection and neuroprotection. Pharmacol Ther 153:90–106. doi:10.1016/j. pharmthera.2015.06.005 95. Tillmann S, Bernhagen J, Noels H (2013) Arrest functions of the MIF ligand/receptor axes in atherogenesis. Front Immunol 4:115. doi:10.3389/ﬁmmu.2013.00115 84. References Nduhirabandi F, Lamont K, Albertyn Z, Opie LH, Lecour S (2016) Role of toll-like receptor 4 in melatonin-induced car- dioprotection. J Pineal Res 60:39–47. doi:10.1111/jpi.12286 69. Nishizawa K, Yano T, Tanno M, Miki T, Kuno A, Tobisawa T, Ogasawara M, Muratsubaki S, Ohno K, Ishikawa S, Miura T (2016) Chronic treatment with an erythropoietin receptor ligand prevents chronic kidney disease-induced enlargement of myocardial infarct size. Hypertension 68(3):697–706. doi:10. 1161/HYPERTENSIONAHA.116.07480 56. Liehn EA, Cabrera-Fuentes HA (2015) Inﬂammation between defense and disease: impact on tissue repair and chronic sick- ness. Discoveries 3(1):e42. doi:10.15190/d.2015.34 57. Lim SY, Yellon DM, Hausenloy DJ (2010) The neural and humoral pathways in remote limb ischemic preconditioning. Basic Res Cardiol 105:651–655. doi:10.1007/s00395-010-0099-y 70. O’Sullivan KE, Breen EP, Gallagher HC, Buggy DJ, Hurley JP (2016) Understanding STAT3 signaling in cardiac ischemia. Basic Res Cardiol 111:27. doi:10.1007/s00395-016-0543-8 58. Lu J, Wei H, Wu J, Jamil MF, Tan ML, Adenan MI, Wong P, Shim W (2014) Evaluation of the cardiotoxicity of mitragynine and its analogues using human induced pluripotent stem cell- derived cardiomyocytes. PLoS One 9:e115648. doi:10.1371/ journal.pone.0115648 71. Organization WHO (2014) Global status report on noncommu- nicable diseases (online) 72. Pickard JM, Botker HE, Crimi G, Davidson B, Davidson SM, Dutka D, Ferdinandy P, Ganske R, Garcia-Dorado D, Giricz Z, Gourine AV, Heusch G, Kharbanda R, Kleinbongard P, MacAllister R, McIntyre C, Meybohm P, Prunier F, Redington A, Robertson NJ, Suleiman MS, Vanezis A, Walsh S, Yellon DM, Hausenloy DJ (2015) Remote ischemic conditioning: from experimental observation to clinical application: report from the 8th Biennial Hatter Cardiovascular Institute Workshop. Basic Res Cardiol 110:453. doi:10.1007/s00395-014-0453-6 59. Maarman G, Blackhurst D, Thienemann F, Blauwet L, Butrous G, Davies N, Sliwa K, Lecour S (2015) Melatonin as a pre- ventive and curative therapy against pulmonary hypertension. J Pineal Res 59:343–353. doi:10.1111/jpi.12263 60. Mangels N, Awwad K, Wettenmann A, Dos Santos LR, Fromel T, Fleming I (2016) The soluble epoxide hydrolase determines cholesterol homeostasis by regulating AMPK and SREBP activity. Prostaglandins Other Lipid Mediat. doi:10.1016/j.pros taglandins.2016.05.003 12 3 69 Page 12 of 13 69 Page 12 of 13 Basic Res Cardiol (2016) 111:69 73. Pickard JM, Davidson SM, Hausenloy DJ, Yellon DM (2016) Co-dependence of the neural and humoral pathways in the mechanism of remote ischemic conditioning. Basic Res Cardiol 111:50. doi:10.1007/s00395-016-0568-z 87. References Seebach J, Taha AA, Lenk J, Lindemann N, Jiang X, Brinkmann K, Bogdan S, Schnittler HJ (2015) The Cell BorderTracker, a novel tool to quantitatively analyze spatiotemporal endothelial junction dynamics at the subcellular level. Histochem Cell Biol 144:517–532. doi:10.1007/s00418-015-1357-8 96. Tiong WN, Fong AY, Sim EU, Chan HC, Ong TK, Chang BC, Sim KH (2012) Increased serum levels of interleukin-6 and von Willenbrand Factor in early phase of acute coronary syndrome in a young and multiethnic Malaysian population. Heart Asia 4:146–150. doi:10.1136/heartasia-2012-010131 85. Serebruany VL, Cherepanov V, Cabrera-Fuentes HA, Kim MH (2015) Solid cancers after antiplatelet therapy: conﬁrmations, controversies, and challenges. Thromb Haemost 114:1104–1112. doi:10.1160/TH15-01-0077 97. Varga ZV, Giricz Z, Bencsik P, Madonna R, Gyongyosi M, Schulz R, Mayr M, Thum T, Puskas LG, Ferdinandy P (2015) Functional genomics of cardioprotection by ischemic condi- tioning and the inﬂuence of comorbid conditions: implications in target identiﬁcation. Curr Drug Targets 16:904–911. doi:10. 2174/1389450116666150427154203 86. Simsekyilmaz S, Cabrera-Fuentes HA, Meiler S, Kostin S, Baumer Y, Liehn EA, Weber C, Boisvert WA, Preissner KT, Zernecke A (2014) Response to letter regarding article ‘‘Role of extracellular RNA in atherosclerotic plaque formation in mice’’. Circulation 130:e144–e145. doi:10.1161/CIRCULATIONAHA. 114.012346 12 3 Basic Res Cardiol (2016) 111:69 Page 13 of 13 69 69 98. Varga ZV, Zvara A, Farago N, Kocsis GF, Pipicz M, Gaspar R, Bencsik P, Gorbe A, Csonka C, Puskas LG, Thum T, Csont T, Ferdinandy P (2014) MicroRNAs associated with ischemia–reperfusion injury and cardioprotection by ischemic pre- and postconditioning: protectomiRs. Am J Physiol Heart Circ Physiol 307:H216–H227. doi:10.1152/ajpheart.00812. 2013 105. Yellon DM, Ackbarkhan AK, Balgobin V, Bulluck H, Deelc- hand A, Dhuny MR, Domah N, Gaoneadry D, Jagessur RK, Joonas N, Kowlessur S, Lutchoo J, Nicholas JM, Pauvaday K, Shamloll O, Walker JM, Hausenloy DJ (2015) Remote ischemic conditioning reduces myocardial infarct size in STEMI patients treated by thrombolysis. J Am Coll Cardiol 65:2764–2765. doi:10.1016/j.jacc.2015.02.082 99. Vogel B, Shinagawa H, Hofmann U, Ertl G, Frantz S (2015) Acute DNase1 treatment improves left ventricular remodeling after myocardial infarction by disruption of free chromatin. Basic Res Cardiol 110:15. doi:10.1007/s00395-015-0472-y 106. Zampetaki A, Willeit P, Burr S, Yin X, Langley SR, Kiechl S, Klein R, Rossing P, Chaturvedi N, Mayr M (2016) Angiogenic microRNAs linked to incidence and progression of diabetic retinopathy in type 1 diabetes. Diabetes 65:216–227. doi:10. 2337/db15-0389 100. References Wang X, Gkanatsas Y, Palasubramaniam J, Hohmann JD, Chen YC, Lim B, Hagemeyer CE, Peter K (2016) Thrombus-targeted theranostic microbubbles: a new technology towards concurrent rapid ultrasound diagnosis and bleeding-free ﬁbrinolytic treat- ment of thrombosis. Theranostics 6:726–738. doi:10.7150/thno. 14514 107. Zebrowski DC, Becker R, Engel FB (2016) Towards regener- ating the mammalian heart: challenges in evaluating experi- mentally induced adult mammalian cardiomyocyte proliferation. Am J Physiol Heart Circ Physiol 310:H1045–H1054. doi:10. 1152/ajpheart.00697.2015 101. Wang X, Hagemeyer CE, Hohmann JD, Leitner E, Armstrong PC, Jia F, Olschewski M, Needles A, Peter K, Ahrens I (2012) Novel single-chain antibody-targeted microbubbles for molec- ular ultrasound imaging of thrombosis: validation of a unique noninvasive method for rapid and sensitive detection of thrombi and monitoring of success or failure of thrombolysis in mice. Circulation 125:3117–3126. doi:10.1161/CIRCULATIONAHA. 111.030312 108. Zebrowski DC, Vergarajauregui S, Wu CC, Piatkowski T, Becker R, Leone M, Hirth S, Ricciardi F, Falk N, Giessl A, Just S, Braun T, Weidinger G, Engel FB (2015) Developmental alterations in centrosome integrity contribute to the post-mitotic state of mammalian cardiomyocytes. Elife 4:e05563. doi:10. 7554/eLife.05563 109. Zernecke A, Preissner KT (2016) Extracellular ribonucleic acids (RNA) enter the stage in cardiovascular disease. Circ Res 118:469–479. doi:10.1161/CIRCRESAHA.115.307961 102. Willeit P, Skroblin P, Kiechl S, Fernandez-Hernando C, Mayr M (2016) Liver microRNAs: potential mediators and biomarkers for metabolic and cardiovascular disease? Eur Heart J. doi:10. 1093/eurheartj/ehw146 110. Zhang YS, Liu B, Luo XJ, Li TB, Zhang JJ, Peng JJ, Zhang XJ, Ma QL, Hu CP, Li YJ, Peng J, Li Q (2015) Nuclear cardiac myosin light chain 2 modulates NADPH oxidase 2 expression in myocardium: a novel function beyond muscle contraction. Basic Res Cardiol 110:38. doi:10.1007/s00395-015-0494-5 103. Yang XM, Cui L, White J, Kuck J, Ruchko MV, Wilson GL, Alexeyev M, Gillespie MN, Downey JM, Cohen MV (2015) Mitochondrially targeted Endonuclease III has a powerful anti- infarct effect in an in vivo rat model of myocardial ischemia/ reperfusion. Basic Res Cardiol 110:3. doi:10.1007/s00395-014- 0459-0 111. Zimmermann-Geller B, Koppert S, Fischer S, Cabrera-Fuentes HA, Lefevre S, Rickert M, Steinmeyer J, Rehart S, Umscheid T, Schonburg M, Muller-Ladner U, Preissner KT, Frommer KW, Neumann E (2016) Inﬂuence of extracellular RNAs, released by rheumatoid arthritis synovial ﬁbroblasts, on their adhesive and invasive properties. J Immunol 197:2589–2597. doi:10.4049/ jimmunol.1501580 104. References Yano T, Miki T, Tanno M, Kuno A, Itoh T, Takada A, Sato T, Kouzu H, Shimamoto K, Miura T (2011) Hypertensive hyper- trophied myocardium is vulnerable to infarction and refractory to erythropoietin-induced protection. Hypertension 57:110–115. doi:10.1161/HYPERTENSIONAHA.110.158469 123 12
https://openalex.org/W4226430669	https://dergipark.org.tr/tr/download/article-file/1662874	Turkish	null	Covid -19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim Mi!	Hacettepe Üniversitesi Edebiyat Fakültesi dergisi	2,022	cc-by	8,373	Öz Covid-19 salgını, küresel sağlık krizi olması sebebiyle tıp alanının konusu olduğu akla ilk gelse de sebep ve sonuçlarıyla güçlü sosyolojik boyutlara sahiptir. Çalışmada küresel bir felaket olması ile ilk kez deneyimlenen, etkisi uzun yıllar süreceği tahmin edilen, toplumsal tüm alanlara olumsuz etkisi ile geleceğe yönelik belirsizlikler içeren, Covid-19 salgınını anlamak için Sosyolog Ulrick Beck’in risk toplumu teorisi incelenmiştir. Beck, yayınlandığından beri büyük ilgi gören teorisinde, sanayi toplumunun risk toplumuna dönüşüm sürecini, risk toplumu olmanın ne anlama geldiğini, bu süreçte toplumda yaşanan değişimleri betimlemiştir. Çalışmada Aralık 2019’dan beri küresel düzeyde yaşanan Covid-19 salgını; Beck’in risk toplumu teorisiyle benzeşen ve ayrılan yönleriyle ilişkilendirilerek incelenmiştir. Covid-19 salgınının zoonotik kökeninden başlayıp sosyal, siyasal, ekonomik çoklu etkileri salgın dönemiyle ilişkilendirerek anlatılmıştır. Risk toplumu teorisinin, Covid-19 salgınını açıklamaya yardımcı bir yaklaşım mı olduğu tartışılmıştır. Teoriye göre Covid-19 salgını imal edilmiş bir risk olarak kabul edilebilir. Covid-19 salgınının sosyal, siyasal, ekonomik etkileri risk toplumu teorisiyle uyumluluk göstermektedir. y y p y y g Anahtar sözcükler: Beck, risk toplumu, Covid-19 salgını, sosyal eşitsizlik, belirsizlik, modernleşme Doktora Öğrencisi, Cumhuriyet Üniversitesi, Edebiyat Fakültesi, Sosyoloji Bölümü. E-posta: ozge_celik18@hotmail.com, ORCID: 0000-0001-8354-6057. Hacettepe Üniversitesi Edebiyat Fakültesi Dergisi Hacettepe University Journal of Faculty of Letters Haziran/June 2022 – 39(1), 198-211 doi:10.32600/huefd.903406 Hakemli Makaleler – Refereed Articles Geliş Tarihi / Received: 25.03.2021 Kabul Tarihi / Accepted: 20.11.2021 Hacettepe Üniversitesi Edebiyat Fakültesi Dergisi Hacettepe University Journal of Faculty of Letters Haziran/June 2022 – 39(1), 198-211 doi:10.32600/huefd.903406 Hakemli Makaleler – Refereed Articles Geliş Tarihi / Received: 25.03.2021 Kabul Tarihi / Accepted: 20.11.2021 Hacettepe Üniversitesi Edebiyat Fakültesi Dergisi Hacettepe University Journal of Faculty of Letters Haziran/June 2022 – 39(1), 198-211 doi:10.32600/huefd.903406 Hacettepe Üniversitesi Edebiyat Fakültesi Dergisi Hacettepe University Journal of Faculty of Letters Haziran/June 2022 – 39(1), 198-211 doi:10.32600/huefd.903406 Hakemli Makaleler – Refereed Articles Geliş Tarihi / Received: 25.03.2021 Kabul Tarihi / Accepted: 20.11.2021 Covid -19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim Mi! Covid -19 Pandemic: A Short Visit to Beck's Risk Community or Permanent Residential! Özge ÇELİK Öz Giriş Modernliğin temsilcisi ve en genel ifadesi olan rasyonellik/akılcılık 21. Yüzyılda hiç de rasyonel olmayan bir problemle karşı karşıya; Covid-19 salgını. 21. Yüzyıl toplumunun karşılaşacağı problemin olsa olsa robot-insan çatışması gibi problemler fütüristlerce tahmin edilirken, daha ilkel bir problemle karşı karşıya gelmiş bulunmaktadır. İlkel bir problem olarak nitelendirilme sebebi, günümüz toplumları için hijyen teknolojisindeki gelişme ve sağlık alanındaki ilerlemelere bağlı olarak virüslerin küresel boyutta bir salgın halini alması çok uzak bir ihtimal olarak görülmesidir. Ancak Aralık 2019 itibariyle bu uzak ihtimal gerçekleşti. Etkileri ile insanlığı tahminlerin ötesinde küresel bir belirsizliğe sürükledi. İçinde bulunulan durumu tüm yönleriyle anlatan bir risk teorisi henüz olmasada, Giddens, Bauman’a ait teoriler üzerinde de etkisi bulunan, ilk risk toplumu teorisi üzerinden Covid-19 salgınının nasıl bir risk olduğu açıklanmaya çalışılacaktır. Salgının dünyayı sürüklediği belirsizliği anlamak ve açıklamak için ilk risk toplumu teorisinden yararlanmanın faydalı olacağı öngörülmüştür. Ulrich Beck, 1986 yılında yayınladığı “Risk Toplumu; Başka Bir Modernliğe Doğru” sanayi toplumunun başka bir modernliğe risk toplumuna geçtiğini anlatan bir teori kitabıdır. Çalışmada bu kitap referans alınmıştır. Bu çalışmada Beck’in risk toplumu teorisi ile Aralık 2019’dan beri küresel olarak yaşanan Covid-19 salgınının getirdiği yaşam deneyimlerinin benzer ve farklı yönleri karşılaştırılacaktır. Ayrıca aşağıdaki sorular makalede ele alınacaktır. • Zoonotik enfeksiyonların (hayvandan insanlara bulaşan hastalıklar) yayılmasını ve dünyayı etkisini anlamada sosyolojik yaklaşım nasıl yardımcı olabilir? • Beck, modernite ve risk arasındaki ilişkiyi açıkladığı sosyal teorisi bağlamında Covid-19 salgını ve modernlik ilişkisi kurulabilir mi? • Risk toplumu, Covid-19 salgınının sosyal, kültürel ve siyasi yansımalarını anlamaya ve açıklamaya yardımcı bir yaklaşım mı? Covid-19 salgını, Beck’in risk toplumu teorisi üzerinden toplumsal ilişkiselliğinin anlaşılması amaçlanmaktadır. Risk toplumu teorisi ile Covid-19 salgını sürecinde toplumların yaşadığı deneyimler arasındaki benzerlikler örneklendirilecektir. Literatürde risk toplumu ve Covid-19 salgınını bir arada ele alan çalışmalar; “Covid-19 Salgını Özelinde Ulrich Beck’in “Risk Toplumu” ile Anthony Giddens’ın “Geç Modernite” Kavramlarını Yeniden Düşünmek’’ (Ersöz, 2020), çalışmada Covid-19 salgınının küresel belirsizlik ve insanlar üzerinde korku yarattığı sonucuna ulaşılmıştır. “Risk Toplumu Bağlamında Covid-19 Haberlerine Yönelik Bir İnceleme’’ (Akgül ve Can, 2021), çalışmada Covid-19 salgın döneminde Ulrich Beck’in Risk Toplumu bağlamında iki haber sitesinin bir aylık haberleri nitel ve nicel içerik analizi yapılarak salgınla ilgili haberlerin nasıl sunulduğu incelenmiştir. Abstract Although the Covid-19 pandemic is a global health crisis, it comes to mind first that it is the subject of medicine, but it has strong sociological dimensions with its causes and consequences. In the study, the risk society theory of Sociologist Ulrick Beck was examined in order to understand the Covid-19 pandemic, which was experienced for the first time as a global disaster, its effect is expected to last for many years, and contains uncertainties for the future with its negative impact on all social areas. In his theory, which has attracted great attention since its publication, Beck described the transformation process of the industrial society into a risk society, what it means to be a risk society, and the changes experienced in the society in this process. Covid-19 pandemic experienced globally since December 2019; It has been examined in relation to the similar and divergent aspects of Beck's theory of risk society. Starting from the zoonotic origin of the Covid-19 pandemic, the multiple social, political and economic effects were explained by associating it with the epidemic period. It has been discussed whether the risk society theory is an approach that helps explain the Covid-19 pandemic. According to the theory, the Covid-19 pandemic can be considered as a manufactured risk. The social, political and economic effects of the Covid-19 pandemic are in line with the risk society theory. Keywords: Beck, risk society, Covid-19 pandemic, social inequality, uncertainty, modernization. 198 Covid-19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim mi! 20. Yüzyıl Sonundan Günümüze Doğa, Salgın, Toplum ve Sosyoloji Covid-19 virüsü SARS ve MERS ile aynı koronavirüs ailesinden olduğu için benzer özelliklere sahiptir. Ancak genetik farklılıklar bulunmaktadır. Covid-19 salgınının enfekte ettiği insan sayısı iki ay içerisinde SARS ve MERS’i geçmiştir. Bunun sebebi Covid-19’un son derece bulaşıcı olması sonucu virüsün bireyler arasında kolaylıkla bulaşabilir olmasıdır (Akkuş, 2020, s. 3). Tıpçılar, kimyagerler, biyologlar vd. laboratuvarlarda elde ettikleri bilgileri dünyayla paylaşmaları sonucu virüsün bulaşıcılığının yüksek olduğu, bu nedenle önlem alınmazsa hızla enfekte olan insan sayısının artacağını ve yüksek sayıda can kayıpları ile karşılaşılabileceğini belirtmiştir. Bunun üzerine yerel, bölgesel, küresel önlemler kendi dinamiği içerisinde belirlenmiş ve hızla uygulanmaya başlanmıştır. Doğa bilimleri alanları üyeleri büyük bir özveriyle hastahanelerde hastaları iyileştirmeye çalışmış, diğer yandan tedavi, aşı ve ilaç çalışmalarına başlamıştır. Sosyal bilimler ise salgının insanlara ve topluma olası etkilerini keşfederek, anormal hal alan toplumsal yaşamın en az zararla sonlandırma çabası içerisine girmiştir. Riskin bilimselleştirilmesi sonucu olan sigorta kavramı ile geleceğin belirsizliğinden kurtulmak modern dönem özelliğidir. Aşının bulunacağına olan güven, akla olan güvenin ve yine onun belirsizliği ortadan kaldıracağı anlamına gelmektedir. Modern bir inanç halidir (Maldonado, 2020, s. 4). Salgınlar, modern dünyada gelişen tıbla birlikte insanların toplumsal hayatında belirleyiciliği kalmadığı sanılan bir olgudur. Modern insan, planlarında, düşüncesinde salgın hastalık için önlem almak gibi bir uğraş içerisinde olmamaktadır. Kuşkusuz bunda etken bilime olan güvendir. Ancak salgınlar modern toplumlar öncesinde tarihin seyrini değiştirecek kadar güce sahiptir. Toplumların geleceğini düşünürken hesaba katması gereken bir etkiye sahip olmuştur. Beck, risk toplumunun bilgi sosyolojisi olduğunu belirtmiştir. Toplum ve bilgi arasındaki ayrılmalar, birleşmeler, doğru sanılanlar, yanlış olduğu bilindiği halde doğru gibi davranmalar hepsi bu sosyolojinin içinde bulunmaktadır. “Büyüyen servet karşısında, küresel ölçekte görünmeyen yoksullaşma var’’ (Beck, 2019, s. 80). Toprak, su, hava, ormanlar, okyanuslar, gıda içindeki zehirli etken maddeler vd. nesnel bir şekilde ölçümlerde görüldüğü gibi artık gözlede görülebiliyor. İhtiyacımız olan temel kaynaklar çoğalıp bollaşmıyor, azalıyor, fakirleşiyor. Bilimler, risklerin ortaya çıkmasında çok büyük paya sahip olduklarından, risklere tepki göstermenin uzağındadır. Hatta riskleri meşrulaştırırlar. Artan modern riskler ve tehditler karşısında artık başarısız olduklarını bilmektedirler. Bilim gizli yan etkiler ve kanıtlanmamış bağlantılar diye önemsizleştirdiği, meşrulaştırdığı şeyler hasta insanlar, hastalıklar şeklinde toplumda belirti gösteriyor. Bu belirtiler bilimsel olarak kanıtlanana kadar insanların bilimsel cahillikle çatışmak zorunda kaldığını belirtmektedir (Beck, 2019, s. 87). Beck, sosyologların kimyasal, biyolojik veya tıbbi risk formüllerine başvurarak eleştirel sosyolojik araştırmalar yapabileceğini belirtir. Örtük içeriklerin içeriğine tam vakıf olunamasa da bir çare bir yol olarak önermiştir (Beck, 2019, s. 127). 20. Yüzyıl Sonundan Günümüze Doğa, Salgın, Toplum ve Sosyoloji 19.yüzyıl sosyal teorileri doğa toplum karşıtlığı içermekteydi. Doğa, hükmedilmesi gereken, insan için tahsis edilmiş, keşfedilmesi gereken bir yabancı olarak görülmekteydi. Doğa üzerinde yapılan tüm değişimler insanlar için meşru kabul edilmekteydi. 20. Yüzyıl sonuna gelindiğinde ise bunun böyle olmadığı, doğanın evrensel tahribatının sadece doğa tahribatı olmadığını gösterdi. Doğal koşulların ihlali toplumsal, ekonomik, siyasi bir hal aldı ve insanlık için bir tehdide dönüşmesi sebebiyle günlük yaşamı etkiler oldu. Küresel ısınma, iklim değişikliği, suların kirlenmesi, ozon tabakasındaki değişimler, çevre kirlilikleri, seller, virüsler, güvensiz gıda potansiyel bir tehdit olarak cılız seslerle ifade edilirken 2019 Aralık itibariyle Covid-19 pandemisi gerçekleşti. Doğa-insan dengesindeki değişimlerin şuan toplumsal, ekonomik, tıbbi ve siyasi bir meseleye döndüğünü, ortaya çıktığı günden beri insanların bir gününün dahi eskisi gibi geçmediğini görmekteyiz. Doğa tehditleri, kökenleri ve sonuçları bakımından toplumsal sorunlardır. İnsanların doğa ile kurduğu yanlış bağın ve ilişkinin sonuçlarıdır (Beck, 2019, s. 122-123). Hayvanlardan insanlara virüs bulaşı ile başlayan zoonotik hastalık ve salgınların Covid-19 ile ilk olmadığını tarihi incelediğimizde rahatlıkla görebiliriz. Virüs salgınları dünya tarihine yön veren etkileriyle 199 Özge ÇELİK eski zamanlardan beri varlığını sürdürmüştür. Antik Çağda dahi bilinen Veba, Ortaçağ’da 1347-1353 yıllarında Avrupa nüfusunun üçte birinin kaybedilmesine sebep olmuştur. İspanyol Gribi (1914), Ebola (1976), HIV (1980), Hantavirüs (1993), SARS (2003), MERS (2012) ve COVİD-19 (2019) en bilinen etkili olan virüs salgınlarıdır. Dünya Sağlık Örgütü 1980 yılında çiçek hastalığından insanların kurtulduğunu ilan etsede 19.yy.’da hastalığın 300 milyon insanı öldürdüğü tahmin edilmektedir. Avrupa’lı kaşiflerin taşıdığı çiçek hastalığıyla Amerika’nın yerli nüfusunun %90’ının öldüğü tahmin edilmektedir. Virüsler varlığını her zaman sürdürdü ancak modern toplumda bilim sayesinde virüslere yaklaşım, değişim ve dönüşümler yaşadı. Çiçek hastalığı, 19.yüzyılda büyük ölüm oranlarına sebep olurken, 20.yy. sonlarına doğru insan üzerindeki olumsuz etkisi tamamen yok edildi. İnsanlığın akıl ve bilim ile virüs ve hastalıklarla başedebileceğini anlaması sonucu insanlar salgınları günlük kaygı ve endişelerinin çok ötesine konumlandırdı. g g yg Covid-19 salgınının başlangıcı Çin’de sebebi bilinmeyen zatürre vakaları 31 Aralık 2019’da Dünya Sağlık Örgütü’ne bildirilmesiyle başlamıştır. Vakaların Vuhan şehrindeki deniz ürünleri ve vahşi hayvanların satıldığı bir pazarla temaslı olduğu tespit edilmiştir. Virüs kaynağı olarak yarasa ve ticareti yasak olan pangolin hayvanının taşıyıcısı olduğu iddia edilmiştir. Virüsün ana kaynağı henüz kesin olarak belirlenememiştir. Virüs hızla diğer ülkelere de yayılmış ve 11 Mart 2020’de Dünya Sağlık Örgütü virüsü pandemi olarak tanımlamıştır. 11 Mart tarihinde Türkiye’de yurtdışı temaslı ilk vakaya rastlanmıştır. p ş y y ş y ş SARS (2003) ve MERS (2012) günümüze en yakın pekçok ülkeyi etkileyen ve can kayıplarına sebep olan salgınlardır. Risk Toplumu Teorisi Beck, endüstri dönemi sonrasını post kavramı ile anlatmaktadır. Sonrası, geç, ötesinde anlamları ile kullanılan post kavramının modern çağın 1970’ler sonrasını anlatmak için gereksinim hissettiren, değişimleri teorisinde anlatmıştır. Modernliğin, kendinden önceki toplumsallığı eleştiri süreci Aydınlanma ile hızlanmış, rasyonelliğin pozitivizme evrilmesi ile modernliğin kendisi eleştirilmeye başlanmıştır. Beck, sosyal teorisi modernlikten kopuşla birlikte sanayi toplumundan nasıl risk toplumuna geçilmiş olduğunu anlatmaktadır. İnsanı, sanayi toplumundan bir risk toplumu yaratmanın öznesi ve nesnesi olarak görmektedir. Bu geçişte modernlik ile sanayi toplumu, sanayi toplumu ile risk toplumu arasında beliren çatışmaları değerlendirmektedir. Modernitenin yeni halinde riskin dağılımının nasıl olduğu, nasıl önlenebileceği, ortaya çıktığında risk konumlarının ne olacağını ele almıştır. O hem modernliğe sıkı sıkı tutunanlara hem de modernliğin olumsuz temsilleriyle birlikte görüp ondan kurtulmak isteyenlere karşı koyma bakış açısını benimsemiştir (Beck, 2019, s. 8). Beck’in tavrı, modernliğin savunucusu ve tamamen bir eleştirmeni olmaktan ziyade düşünümsel bir modernliği ele alır. Kendi üzerine eleştirel düşünce geliştiren bir modernliktir. Asıl mesele olarak Beck, dönemin eleştirellikten uzaklaşmasını, irdelemeden tastik etmenin tehlikeli olduğunu, dönemi yansıtan bu belirsizliklerin sosyolojik yaklaşımla nasıl anlaşılacağını görmektedir. Bu teoride modernleşme 19. yüzyılda tarım toplumunu ortadan kaldırıp sanayi toplumunu yerine getirdiyse artık bu sürecin bir devamı olarak yeni bir toplumsal biçim ortaya çıkmakta olduğunu ifade etmektedir (Beck, 2019, s. 9). Risk kavramı, modern dönem öncesinde de bulunmaktaydı. Küresel değil, kişisel risklerdi. Cesaret ve macera içerirdi. Bu dönemde ise ileri teknoloji ile yeryüzündeki tüm hayatın kendini yok etmesi tehlikesi riski tanımlıyor (Beck, 2019, s. 24). ) 19. Yüzyıl’da yönetici statüsündekilerin ayrıcalıkları ve dinin toplumsal hayattaki etkisi nasıl azaldıysa, güven sarsıldıysa günümüzde de sanayi toplumundaki bilim ve teknolojiye olan güven sarsılmaktadır. Modernleşmenin hedefleri, sanayi toplumu içerisinde müphem bir hal almıştır. Klasik modernleşme ile dönüşlü modernleşme arasındaki ayrımlar daha uzun yıllar süreceğe benzemekle birlikte, teoride bugünkü yansımalarına yer verilmiştir. 19.yy. da başlayan ve 20.yy. son otuz yılına kadar süren modern toplum olma özelliğini tüm özellikleriyle yansıtmaktaydı. Çalışma düzeni, bilim ve teknoloji yaklaşımı, üretim, ekonomik gelişim ve demokrasi modern toplumun özellikleri olarak artık tarihin sonuna (Fukuyama) gelindiği bile söylenmiştir. Toplumsal değişimlerle son halini alan modernitedeki değişimler sosyolojik olarak biraz daha ilerleme biraz daha modernleşme, sanayileşme sayılmaktaydı. Aile, iş, fabrika, meslek, ücretli emek, sınıf değişmedi denilmesine rağmen, olgular içindeki kurulan ilişkilerin de değişmediği varsayıldı (Beck, 2019, s. 10-11). Aynı kurumlar modernleşmenin başlangıcında ve sonunda varlığını devam ettirmekle beraber iç dinamikleri piyasa ekonomisine entegre bir yaşam içinde büyük değişimler geçirmiştir. 20. Yüzyıl Sonundan Günümüze Doğa, Salgın, Toplum ve Sosyoloji Beck, sanayi toplumunun üretim anlayışı ve halka sunuşunun sosyal bilimler tarafından eleştirel olarak ele alınması gerektiğini düşünmektedir. Eleştirel bakış olmadığı sürece insanlığı çok daha zor yaşam şartları beklemektedir. 200 Covid-19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim mi! Risk Toplumu Teorisi Geniş aile, çekirdek aileye; iş yeri, küresel iş organizasyonlarına evrildi; sanayi yanında hizmet sektörü ortaya çıktı, tam zamanlı iş; yarı zamanlı veya uzaktan çalışmaya bölündü, bireyselleşme arttı, meslekler içerisinde kadın istihdamı nispeten arttı, meslekler çeşitlendi, alt sınıf ve üst sınıflar arasındaki gelir dağılımı eşitsizliği yükseldi. 19. Yüzyıl’da yönetici statüsündekilerin ayrıcalıkları ve dinin toplumsal hayattaki etkisi nasıl azaldıysa, güven sarsıldıysa günümüzde de sanayi toplumundaki bilim ve teknolojiye olan güven sarsılmaktadır. Modernleşmenin hedefleri, sanayi toplumu içerisinde müphem bir hal almıştır. Klasik modernleşme ile dönüşlü modernleşme arasındaki ayrımlar daha uzun yıllar süreceğe benzemekle birlikte, teoride bugünkü yansımalarına yer verilmiştir. 19.yy. da başlayan ve 20.yy. son otuz yılına kadar süren modern toplum olma özelliğini tüm özellikleriyle yansıtmaktaydı. Çalışma düzeni, bilim ve teknoloji yaklaşımı, üretim, ekonomik gelişim ve demokrasi modern toplumun özellikleri olarak artık tarihin sonuna (Fukuyama) gelindiği bile söylenmiştir. Toplumsal değişimlerle son halini alan modernitedeki değişimler sosyolojik olarak biraz daha ilerleme biraz daha modernleşme, sanayileşme sayılmaktaydı. Aile, iş, fabrika, meslek, ücretli emek, sınıf değişmedi denilmesine rağmen, olgular içindeki kurulan ilişkilerin de değişmediği varsayıldı (Beck, 2019, s. 10-11). Aynı kurumlar modernleşmenin başlangıcında ve sonunda varlığını devam ettirmekle beraber iç dinamikleri piyasa ekonomisine entegre bir yaşam içinde büyük değişimler geçirmiştir. Geniş aile, çekirdek aileye; iş yeri, küresel iş organizasyonlarına evrildi; sanayi yanında hizmet sektörü ortaya çıktı, tam zamanlı iş; yarı zamanlı veya uzaktan çalışmaya bölündü, bireyselleşme arttı, meslekler içerisinde kadın istihdamı nispeten arttı, meslekler çeşitlendi, alt sınıf ve üst sınıflar arasındaki gelir dağılımı eşitsizliği yükseldi. Beck, modernliği birinci modernlik ve ikinci modernlik olarak ayırmaktadır. Birincisi modernliğin toplumsal düzen olarak kurulumu, ikincisi modernliğin toplumsal düzeni her an bozacak bir hal almasını anlatmaktadır. Modernliğin karanlık yüzü etkilerini göstermeye başladı denilebilir. Modernleşme kendi kendisini konu ve sorun ettiği, dönüşlü bir hale geldi. Tekno-ekonomik gelişmenin getirdiği sorunlar üzerine düşünmek ve çözmek gerekmektedir. Toplumsal servet üretimine, ileri modernlikte toplumsal risk üretimi eşlik etmektedir. İlk modernlikte refah bölüşümü ikinci modernlikte risk bölüşümüne doğru bir değişim vardır. Bu değişimin ilk basamağını insani ve teknolojik gelişim yanı sıra hukuk ve sosyal güvence sisteminin düzenli hale gelmesiyle temel ihtiyaçları (gıda vb.) elde etmek insanlar için sorun olmaktan çıkmıştır. İkinci basamak teknoloji ile birlikte üretimdeki küresel azami artış tehlikenin daha önceki dönemlerde olmadığı kadar gerçekleşme ihtimalini ortaya çıkarmıştır (Beck, 2019, s. 21). Marx ve Weber, sanayi toplumundaki üretilen serveti toplumsal açıdan eşitsiz ve meşru bölüşümünün nasıl olduğu ile ilgili çözümlemeler yapmıştır. İleri modernliğin getirdiği risk toplumunun sorunsallaştırdığı ise servet değil, riskler ve tehlikelerdir. Risk Toplumu Teorisi Bu modernleşmenin riskleri, tekno-ekonomik gelişime bağlı olarak 201 Özge ÇELİK insan eliyle üretilmiştir. Sanayide gerçekleşebilecek bir felaket toplumları hesapta olmayan zorluklara ve sıkıntılarla karşı karşıya getirecek hale gelmiştir. Nitekim Çernobil nükleer kazasının etkileri kazanın olduğu üretim yerinde, ülkede kalmamış küresel etkileri olmuştur. Modernitenin temsilcisi rasyonelliğe ve bilime olan güven, risklerin hesaplanamaz doğası karşısında kutuplardaki buzlardan daha hızlı erimektedir. Yüksek riskli sanayilerin küreselleşmesiyle, hesaplama skalasına girmeyen risklerde dünyanın her yerine yayılmış durumdadır (Beck, 2019, s. 22-26). Beck, teorisini öncelikle beş tez halinde ortaya konulabileceğini savunmuştur. 1. Küresel ölçekte insanlığın varlığını tehdit eden insan yapımı, sonuçları itibariyle telafi 1. Küresel ölçekte insanlığın varlığını tehdit eden insan yapımı, sonuçları itibariyle telafi edilemez boyutlara varabilecek büyük risklere sahip bir toplumuz. 1. Küresel ölçekte insanlığın varlığını tehdit eden insan yapımı, sonuçları itibariyle telafi edilemez boyutlara varabilecek büyük risklere sahip bir toplumuz. 2. Risklerin paylaşımından bir eşitsizlik mevcut, toplumsal risk konumları bazı insanlar için daha fazla riskten etkilenme anlamına gelecektir. 3. Riskler, çok farklı ve fazla olduğundan onunla ilgilenmek yeni bir iş alanı ve bunun ü ekonomik sömürü getirecektir. 4. Riskler, her zaman kaybettirir. Riskler sosyal teori içerisinde doğuşundan sonuçlar incelenmelidir. 5. Riskler, siyasal alana taşınma özelliğine sahiptir. Sağlık sorunları yanında sosyo-ekonomik etkileriyle anormalin normal düzen olması potansiyeli taşırlar (Beck, 2019, s. 27-29). Riskler için artık kültürel eleştiri ile yapılan teşhisler yeterli kalmıyor, eleştirilerin gerçekleşmiş hali ile karşı karşıya gelmiş bulunmaktayız. Risk üretimi, teknik-ekonomik gelişimin önünde gitmektedir. Riskler, örtük yan etkiler olarak ilk başlarda tölere edilip, kabullenilmesi sağlanıyordu. Riskler küreselleştikçe ve toplumun eleştirilerini gördükçe, örtük yan etkilerin üstü açılmaya, kamusal olarak görülmeye sosyal ve siyasi tartışma gündemi olmaya başladı. Modernleşmenin risklerinin sonuçları, insanlar, hayvanlar ve bitkiler için geri dönülemez, telafi edilemez yaşam tehditleri oluşturmaktadır (Beck, 2019, s. 13-14). Modern toplum kurumları, piyasa ekonomisi ile (bilim, askeriye, iş dünyası, devlet gibi) geleceği öngörme için yıllarca çabalamasının sonucu belirsiz, öngörülemez bir toplum içinde kendini ironik bir şekilde bulması oldu. İnsanlık içinde olduğu tehlikelerinden bildiklerimizi biliyoruz, peki ya bilmediklerimiz! Modern toplum, rasyonel olduğu için riskler hesaplanabilir, öngörülebilir ve kontrol edilebilir olarak kabul edildi. Ancak artık riskler insan üretimidir. İnsanlığın ürettiği imal edilmiş belirsizlik, güvensizlik sonucu nerede, ne zaman patlak vereceği bilinmeyen bir hal almıştır. Rasyonellik temelli modern toplumu üzerine inşa edilmiş irrasyonellikler gibi hayal edilebilir. İklim değişikliği, küresel ısınma, doğal afetler küresel mali krizler, virüs hastalıkları (deli dana hastalığı, domuz gribi, kuş gribi vb.) insanlar tarafından üretilen risklerdir. Riske verilen tepkiler; inkar, ihmal, dönüşüm şeklinde sıralanır (Beck, 2019, s. Covid-19 Salgını Modernlik İlişkisi Doğanın yok edilmesi, çevrenin kirletilmesi doğa bilimleri kategorisine dahil edilen konular olarak görülüyor ancak sosyal, kültürel ve siyasi anlamlara sahip olduğu görülmelidir. Örnek; anne sütünde, DDT gibi bitki koruma ilaçlarında bulunan zehirli maddelere rastlanmaktadır (Beck, 2019, s. 30). Çevre sorunlarının insan sağlığı ve toplumsallığı üzerindeki olumsuz etkileri ilk kez sanayi toplumu ile ortaya çıkmıştır. Riskin toplumsallaşması söz konusudur. Modern düzenin doğa üzerindeki tahakkümünün insanlar üzerinde etkileri görülmeye başlandığı dönem tarihsel olarak ilk kez modern dönemde yaşanmıştır. Modernlik tehlikeleri sanayideki aşırı üretimle bağlantılıdır. Geçmişteki tehlikeler ise hijyen teknolojisindeki yetersizlikle açıklanabilmekteydi. Artık insanlar, hayvan ve bitkiler üzerinde küresel ve modern sebeplerden kaynaklı risk ve tehlikeler bulunmaktadır (Beck, 2019, s. 25). Beck’in risk toplumu teorisiyle Covid-19 salgınını ilişkilendirileceğinden öncelikle salgının modernlik kaynaklı mı yoksa sadece insan-hayvan etkileşimi sonucu olarak tarihin farklı dönemlerindeki salgınlarla (veba, sars, mers, vd.) benzer mi olduğuna bakılacaktır. g ( ) ğ İnsanın ekonomik kalkınmasının gizli maliyeti artık vahşi yaşamdan virüs bulaşmasıdır. Virüsler, insandan daha fazla ve her ortamda bulunmaktadır. Bizler ise bozulmamış yerlere gittikçe daha fazla virüsle maruz kalma ihtimalimizi arttırmaktayız. Virüslerin daha kolay bulaştığı habitatlar yaratıyoruz ve sonra salgın olduğunda bunlara şaşırıyoruz (Vidal, 2020, s. 6). Salgın riski oluşturacak patojenler vahşi yaşamda duruyordu, insanoğlu doğaya yaptığı plansız müdahalesiyle virüslerle temasını artırdı. Böylede devam ederse daha pekçok salgın çeşidiyle tanışmak zorunda kalacağız. Covid-19 salgınının kaynağı deniz ürünleri ve vahşi hayvanların satıldığı ıslak pazar, bu pazarlardan dünyanın başka yerlerinde de var. Covid-19, salgınların başlangıcı olabilir. Örnek Lagos/Nijerya’daki ıslak pazar patlamayı bekleyen nükleer bir bomba gibi ancak bu pazarlar aynı zamanda milyonlarca fakir insanın gıda kaynağı olma rolünüde üstlenmektedir (Vidal, 2020, s. 10). Virüs riski her zaman doğada mevcuttu, modern yaşamla virüslerin dünyasına insanlar ziyaretlerini sıklaştırdıkça onlardan daha fazla etkilenmeye başlamış olabileceği söylenebilir. Ancak bu etkileşimden zararlı çıkan insanlar olmaktadır. Virüslerle insan arasında doğa vardı, insan doğayı aştıkça virüslere yaklaştı. Covid-19 salgını risk toplumunun mu ürünü olduğuna bakıldığında, Beck’in kastettiği modern risklere; etkisinin hesaplanamaz olması, sebep ve sonuçlarının tek bir alanla sınırlı olmaması potansiyelinin belirsiz olması ve tazminin ve telafisinin çok güç olmasıyla benzemektedir. Covid-19 virüsünün insan yapımı olmaması onu risk toplumu ürünü olmadığını düşündürebilir. Ancak modern yaşam şeklimize virüs dahil olduğunda, emniyet, önlem, sigorta kavramlarının geçersizliği risk toplumu özelliklerini taşıdığının göstergesi kabul edilebilir. Eski zamanlarda da virüs kaynaklı salgınların olduğu düşünüldüğünde moderniteye bağlı bir sonuç olarak almakta doğru olmayacaktır. Ancak Covid-19 salgınının kısa sürede yayılması modern dönem özelliğidir. Risk Toplumu Teorisi 355-356). Covid-19 salgınına İngiltere, ABD, Almanya ve İran’ın tepkileri önce inkar olmuş, maske ile gezmeye gerek olmadığı halka duyurulmuş, nüfusta sürü bağışıklığı oluşması gibi düşüncelerle başlarda önlem alınmamıştır. Daha sonra ise artan vaka sayısı ile salgına verilen tepki değişmiş-dönüşmüş kısıtlamalar ve zorunluluklar getirilmiştir. Modern toplumun kendi ürettiği risklerle başa çıkmak için daha fazla çabalaması onu risk toplumu haline getirmiştir. Risk toplumunda küresel risklerin üç özelliği: 1. Mahalsizleşme: Sebepleri ve sonuçları tek bir mahalle sınırlı değildir. 2. Hesaplanamazlık: Sonuçları hesaplanamazdır. 3. Telafi edilemezlik: Güvenli olmayan sonuçların geri dönülemez hale gelmesidir. İklim değişikliği, telafi edilemez hal aldığında geç kalınmış olacaktır. 3. Telafi edilemezlik: Güvenli olmayan sonuçların geri dönülemez hale gelmesidir. İklim değişikliği, telafi edilemez hal aldığında geç kalınmış olacaktır. Riskler, mekânsal olarak ulus devlet sınırlarını tanımaz. Zamansal olarak ne zaman ortaya çıkacağı ve ortaya çıktığında etkilerinin ne kadar süreceği belirlenemez. Toplumsal olarak etkilerinin neler olabileceği, sonuçlarının nerede sonlanacağı ve neleri etkileşime geçireceğini tahmin etmek imkânsıza yakındır (Beck, 2020, s. 357-358). 202 Covid-19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim mi! Covid-19 Salgını Modernlik İlişkisi Modernitenin getirdiği yaşam tarzı insanı ev dışındaki sosyal organizasyonlarla bağlantısını arttırdığından bulaş oranı yüksek ve küreseldir. ğ y Eski zamanlardaki virüs yayılımına göre şimdi küreselleşme etkisiyle dünyayı sarması daha hızlı bir hal almıştır. Küresel riskler (finansal krizler, iklim değişikliği vb.) arasında virüsler unutuldu ama hiçbir zaman kaybolmadı. Covid-19 virüsü yüksek teknoloji içeren üretim tesisinde ortaya çıkıp yayılsaydı modernlik riski olarak kabul edilebilirdi. Ancak doğada virüsler her zaman bulunmaktadır. Islak pazarlarda salgın virüsleri barındırmaktadır. Üretim tesislerinin katkı maddelerini içermemektedir. Küreselleşme ile artan sosyal ilişkiler ve nüfus, şehirleşme sisteminin virüsün bulaştırıcılığını arttırdığı da göz ardı edilemez. Toplum ve doğa eskiden olduğundan birbirine daha yakın konumdadır. Salgınlar tahmin edilen risklerdendir (Maldonado, 2020, s. 5-6). Covid-19, küreselleşme çağının ilk salgını olarak tanımlanabilir. Pandemiler, modernleşmenin temel risklerinden görülebilir. Bilimsel ve endüstriyel gelişim, sonuçları ekolojik ve çevresel bozulmalarla virüslerin farklı türler açısından yayılımını arttırmasına sebep teşkil eder (Ferreira, Sá, Martins, ve Serpa, 2020, s. 8). Kentleşme, doğal kaynakların yoğun kullanımı, küresel nüfus hareketliliği ve arazi kullanımındaki değişiklikler nedeniyle geçen yüzyıla göre salgın hastalık ihtimali yükselmiştir. Artan nüfus ve hareketliliğine bağlı olarak salgın riskinin artarak devam etmesi beklenmektedir (Galanakis, 2020, s. 6). 203 Özge ÇELİK Covid-19 salgını; doğal, sosyal, ekonomik ve siyasal sistemlerin bozulması ile ortaya çıkan karmaşa ile bağlantı gösterir. Doğal habitatın yok edilmesi, türlerin yok olması, hayvanların kötü yakalanması, pazarlanması ve tüketilmesi, koruma için ise uygulamaların geçersiz kılınması ya da geciktirilmesi, hükümetler ve şirketlerce zararlı kanıtların göz ardı edilmesi Covid-19 salgınının olmasını kolaylaştırdı. Beck’de zararsız yan etkiler ve kanıtlanmamış bağlantılar ile riskin görünürlüğünün azaltıldığını belirtir. Küresel iklim krizi, fosil yakıt tüketimi, ormanların tahrip edilmesi, hayvan yetiştiriciliğinin çiftliklere taşınması, sera gazı emisyonu ile meydana gelen kriz kaynakları ile beslendikçe potansiyel salgın riskinin meydana gelmesini kolaylaştırmaktadır. Ulaşılması zor alanlarda yolların açılması, insanlar ve yaban hayatı arasındaki teması teşvik etmekte ve avlanma yaban hayvan eti tüketimini kolaylaştırmaktadır (Gordo ve vd. 2020, s. 2). , ) Beck, hava, su ve gıda maddelerindeki kirletici unsurların artışı rakamlarla ve bilimsel unsurlarla anlatıldığını ifade eder. İnsanlar için temel ihtiyaç olan bu maddelerin kirlenmesinin sahip olduğu sosyal, kültürel ve siyasal anlamları ise farkedilmez. Hava, su ve gıda maddelerindeki tehdit edici unsurların artması bir yoksullaşmayı içinde barındırmaktadır. Bu yoksullaşma temel ihtiyaç maddelerinin miktarındaki azalmayı ve azalan maddelere ulaşmak için verilecek daha fazla emek anlamına gelmektedir. Çift yönlü bir yoksullaşmayı içinde barındırmaktadır (Beck, 2019, s. 30). Beck’in risk toplumunda yaşamanın getirdiği yoksullaşma Covid-19 salgını sürecinde küresel bir şekilde yaşanmaktadır. Covid-19 Salgını Modernlik İlişkisi Gıda ürünlerinin fiyatları küresel bir şekilde yükselmiştir. Risk toplumunda riskler tehlike olarak ortaya çıktığında temel ihtiyaç maddelerinin üretimi, sevkiyatında yaşanan aksaklıklar ile iklim değişikliğine bağlı üretimdeki azalmayla birleşince fiyatlar artmıştır. İklim değişikliği sayılar ve rakamlarla anlatılırken, Covid-19 salgını etkisiyle bu değişimin sosyal, kültürel ve siyasal etkileri somut olarak raporlarda anlatılmaktadır. y y p Birleşmiş Milletler Gıda ve Tarım Örgütü'nün 04.02.2021 tarihli Ocak raporunda Gıda Fiyat Endeksinin Ocak ayında ortalama 113.3 puanla Aralık 2020'ye göre yüzde 4,3 artış göstererek Temmuz 2014'ten bu yana en yüksek seviyesine ulaştığını yayınlamıştır (Birleşmiş Milletler Gıda ve Tarım Örgütü [FAO], 2021). FAO, 04.03.2021 raporuna göre, Şubat ayında da küresel gıda fiyatları üst üste dokuzuncu ayda da artmıştır. Yaygın olarak ticareti yapılan gıda ürünlerinin uluslararası fiyatlarındaki değişim bir önceki aya göre 2,4 artarak Şubat 2021 ayında ortalama 116.0 ortalamaya ulaşmıştır (FAO, 2021). FAO, politika analizlerini desteklemek ve salgının gıda, tarım, fiyatlar, gıda güvenliği konusunda değerlendirmesi ile salgının insanların gıda güvenliği ve geçim kaynakları için yaratacağı panikten kaçınmak için mesaj yayınlamıştır. Yayınlanan mesajda, salgının başta yoksullar ve en savunmasızlar olmak üzere herkes üzerinde yaratabileceği şok riskini azaltmak için ülkelere gıda tedariğinin canlı tutulmasını önermiştir. Covid-19 salgınının etkilerine özellikle insani krizleri olan ülkelerin maruz kaldığını belirtmiştir. Salgın nedeniyle ülkelerin kendi ihtiyaçları artsa da ülkeler arası insani yardımın öneminin de arttığını, salgın sınırları tanımadığından tek bir yerde kontrol edilmeden bırakılırsa, tüm insan topluluğu risk altında kalmaya devam edeceğini belirtmiştir (FAO, 2021). Covid-19 Salgını ve Çoklu Etkileri Beck, risk bölüşümü ile servet bölüşümünü farklı noktalardan karşılaştırarak aynı ve ayrı yanlarını ele almıştır. Sınıflı toplum ile risk toplumu arasında geniş benzerlikler görmüştür. Risk ve servetin sınıfsal modellere sadık bir şekilde dağıldığını belirtir. Servet, toplumun refah seviyesi daha yüksek sınıfında birikirken, riskler alt sınıfta birikmektedir. Riskler sınıflı toplumu güçlendirmekte ve pekiştirmektedir. Yoksulluk; güvenlikten uzak olmayı, risk bolluğunu getirir. Zenginler; gelir, güç ve eğitim bakımından sermayesi güçlü olanlar, risklerden korunmak için güvenlik satın alabilirler. Sınıfa özgü risk bölüşümüyle sınıflar arası karşıtlık, zıtlaşma güçlenir (Beck, 2019, s. 47). Toplumun her iki sınıfı için riskin anlamı farklılıklar içermektedir. Sağlık Bakanlığı’nın “Hayat Eve Sığar” uygulaması ile Covid-19 virüsü hakkında bilgilendirmek, yönlendirmek, yoğunluğu hakkında bilgi vererek riski azaltmak için kullanılan mobil uygulamadır. Vakaların yoğun olduğu yerler kırmızı ile gösterilmektedir. Yoğun olmayan yerlerde mavi ve yeşil renkler kullanılmaktadır. Uygulama vatandaşların yaşadıkları bölgede ya da gitmek istedikleri yerdeki risk durumu hakkında bilgi vermektedir. Uygulama vaka sayıları ile birlikte farklı bilgilerde sunmuştur. Aynı şehir 204 Covid-19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim mi! içerisinde çok kısa mesafeler arasında renk değişimleri görülmüştür. Bu renk değişimlerinin sebebine bakıldığında farklı gelir düzeyleri değişimleri ile paralellik gösterdiği farkedilmiştir. İstanbul’un yaşam kalitesi düşük, konut fiyatları düşük, çok katlı ve insanların gelir düzeyinin düşük olduğu Bağcılar, Bahçelievler, Sultangazi, Şirinevler ve Zeytinburnu ilçeleri Covid-19 vakalarının en yoğun olduğu yerler olmuştur. Şekil 1. “Hayat Eve Sığar” uygulaması (Onedio, 2020) Şekil 1. “Hayat Eve Sığar” uygulaması (Onedio, 2020) Harita İstanbul’un Ataköy ve Şirinevler semtlerine ait vaka haritasıdır. İki semt arasından otoban geçmektedir. Ataköy semti, gelir düzeyi yüksek, az katlı binaların ve yeşil alanların olduğu bir yerleşimdir. Haritada mavi-yeşil renklerle gösterilmiştir. Vaka sayılarının düşük olduğunu göstermektedir. Şirinevler semti ise gelir düzeyi düşük, çok katlı apartmanların bulunduğu kalabalık nüfusa sahiptir. Haritada kırmızı renkle gösterilmiştir. Büyük şehirlerdeki yerleşim şekline bağlı olarak çok katlı, sık nüfus barındıran ve daha düşük bir gelir düzeyine sahip yerlerde yaşayanlar virüs kapma riskine karşı daha savunmasızdır (Bozkurt ve Sayın, 2020, 206). İki semt mesafe olarak yakın olmasına rağmen gelir düzeyi ve yaşam standardı olarak birbirinden çok uzak konumlardadır. Sınıf farkı etkisi risk paylaşımı etkisinin en somut örneğidir. Servet tepede birikirken, riskler alt sınıflarda birikmektedir. Sınıfa bağlı olumsuz etkilenme, riskle başa çıkma, onlardan kaçınma ya da yaşandığında telafi etme imkanları çeşitli mesleklere ve eğitim seviyelerine göre eşitsizlikler bölüşülmektedir. Finansal imkanlar, sosyal haklar insanların mülkiyet çeşitliliğini ve sahipliliğini farklılaştırmaktadır. Covid-19 Salgını ve Çoklu Etkileri İkinci bir ev sahibi olanlar veya tatillere çıkma imkanı olanlar riskle baş etme ve telafi etme noktasında ikinci bir şansa sahip olabilmektedir (Beck, 2019, s. 47-48). Covid-19 salgını sürecinin başlangıcında medyada sık sık haber olan futbolcu, aktör, aktrislerin ve zenginlerin ada satın alıp ailesiyle bu adalara yerleştikleri, ormanlık alanlarda büyük evlere yerleştikleri yer aldı. Finansal imkanlara sahip olanlar risklerden kaçınma ve başa çıkmakta yoksullara göre çok iyi imkanlara sahiptir. Risk bölüşümündeki eşitsizlik küresel şekilde yaşanmıştır. Ülkemizde İstanbul, Ankara gibi büyük şehirlerde yaşayan insanlardan Ege ve Akdeniz tatil beldelerinde özelliklede müstakil evlere sahip olanlar ikinci evlerine gitmiştir. Bu hareketlilik yollarda trafik oluşturmuştur. İnsanlar bir kaçış halinde daha güvenli gördüğü alanlara gitmeyi tercih etmiştir. Tatil beldelerinde tatil mevsimi dışında hiç yaşanmayan nüfus artışı yaşanmıştır. Modernleşmenin riskleri arttıkça, sosyal eşitsizliklerin arttığını belirtse de Beck, risklerin eninde sonunda onları yaratanları ve kar edenleride etkileyeceğinden demokratik olduğunuda belirtir. Endüstriyel küreselleşmeye bağlı olarak hava kirliliği, nükleer bir kaza, salgın hastalıklar vb. etkileri ayrım gözetmeksizin toplumun tüm sınıfları tarafından deneyimlenmektedir. Aşırı sanayileşme, yüksek üretim 205 Özge ÇELİK gücü herşeyi tehlikeli hale getirir ve sakınmanın mümkün hale gelmediği bir hal almaya başlar. Riskler yayılırken, sosyal yaşamda bumerang etkisi yaratır. Emniyette olamamanın etkisi zenginler ve iktidar sahiplerinide etkisi altına almaya başlar. Risk toplumunun yaratıcı modernleşme aktörleri kar elde ettikleri bu düzende bir noktada zarar görenler konumunda yerlerini alırlar (Beck, 2019, s. 49-51). Uluslararası şirketler ve güçlü devletler yüksek teknoloji içeren üretim ağları ile ön plana çıkmaktadır. Covid-19 salgının, küçük ve orta ölçekli işletmeler ve üreticiler iş yeri kapatmalarıyla mali krize girmiş, iflasa ulaşan sonuçlar meydana gelmiştir. Covid-19 salgının boyutlarının artması küresel üretim ve tüketim zincirindeki aksaklıkla birleşince büyük ekonomiler içinde ekonomik durgunluk ve küçülme, sektörel bazda iflaslar yaşatmıştır. Kar elde edenlerde salgınla birlikte zarar görenler konumunu almıştır. Tablo 1. IMF Büyüme Verileri (yıllık % değişim) 2020 2021 2022 Dünya -3,5 5,5 4,2 Gelişmiş Ülkeler -4,9 4,3 3,1 ABD -3,4 5,1 2,5 Euro Alanı -7,2 4,2 3,6 Gelişmekte olan Ülkeler -2,4 6,3 5,0 Türkiye 1,2 6,0 3,5 Tablo 1. IMF Büyüme Verileri (yıllık % değişim) Tablo 1 dünya, bazı ülkeler ve bazı bölgelerin 2020 ekonomik küçülme, büyüme oranları ve 2021- 2022 tahminleri verilmiştir (Türkiye İş Bankası, 2021). Covid-19 küresel salgını karşısında ekonomilerin olumsuz etkilendiğini IMF resmi olarak açıklamıştır. IMF’nin gelecek yıllar için salgınla mücadelenin başarılı olacağı varsayımı ile 2021-2022 yılları için büyüme tahminleri yapmaktadır. Salgın, ekonomik büyüme üzerinde olumsuz etkiye sahip olsada, aşının bulunmasının olumlu etkiye sahip olduğu söylenebilir. Covid-19 Salgını ve Çoklu Etkileri Ülkemizde imalat ve sanayi sektörleri üretime alınan tedbirlerle devam ederken kısıtlamalar nedeniyle hizmet sektöründe yavaşlama olmuştur. Ekonomik belirsizlik ise varlığını devam ettirmektedir. Virüsün mutasyona uğrama ihtimali ve aşının uygulanmasındaki sürenin uzaması belirsizliğin sürmesine neden olmaktadır (Türkiye Cumhuriyeti Merkez Bankası, 2021). Ayrıca ekonomisi en büyük ülkeler yani üretim gücünün çoğunluğunu elinde bulunduran aynı zamanda en çok kirletici olan ülkelerinde ekonomileri salgın sürecinden olumsuz etkilenmiştir. Toplumsal yansımaları olmuştur. Üretim kapasitesi çok kısıtlı olan ülkelerin ekonomileri salgından etkilendi; ama riskin üreticisi, kar elde eden devlet ve şirketlerde salgından olumsuz etkilendiler. Covid-19 Salgınında Toplumsallaşma ve Belirsizlikler Beck, risk konumları ile sınıf konumlarının aynı olmadığını farklı dinamikleri olduğunu belirtir. Sınıflı toplumlar mülküyet sahibi olup olmayan şeklinde ayrılır ancak risk toplumunda mülk değersizleşip hayatta kalmak önemli hale geldiğinden toplumda kimlikler, yukarıdaki onlar- aşağıdaki bizler, halini aldığını, çekişmeli bir yapı oluşturduğunu belirtir. Riske maruz kalanlar ve riskten henüz etkilenmeyenler şeklinde ayrıma gidilecektir (Beck, 2019.s. 55). Toplumda Covid-19 aşısı yaptıranlar arttıkça, aşı olanlar ve olmayanlar ayrımı yapılacağı, aşı pasaportu uygulamasının gündeme geleceği tahmin edilmektedir. Toplum içerisinde aşı olan ve olmayanlar ayrımı mülkiyetten bağımsız, risk toplumuna özgü yeni bir dinamik ortaya şimdiden çıkarmış olduğunu söylemek yanlış olmayacaktır. Bu dinamik riskin boyutuna göre canlı bir süreçtir, yeni dinamikler üretmeye devam edecektir. Klasik sanayi toplumunda bireyler kendi istikbali ve isteklerinin kaygısıyla meşgul olmaktadır. Geliri, arabası, ailesi, tatili, evi, giyimi, konforu, lüksü vs. Risk toplumunda ise bu kaygılarla meşguliyet ortadan kalkar. Covid-19 salgını bireyin hayatının sonlanmasına dahi sebep olacağından sanayi toplumunun bu kaygıları pandemide arka planda kalır. Sınıflı toplum yapısında eşitlik ideali temelli iken risk toplumunda herkes emniyette olmak istemektedir. İyiye ulaşma yarışı yerini kötüyü önlemeye bırakmıştır (Beck, 2019, s. 71-72). 2020 yılı TÜİK Yaşam Memnuniyet Araştırmasına göre bireylerin mutluluk kaynağının %70.9 206 Covid-19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim mi! ile sağlıklı olmak olduğu tespit edilmiştir (Türkiye İstatistik Kurumu, 2020). Önceki yıllarda da sonuç bu orana çok yakındır. Sağlıklı olmak birincil mutluluk kaynağı olduğuna göre küresel bir sağlık krizinin bireyler üzerinde yaratacağı baskı ve kaygının onları olumsuz etkilemesi, mutsuz etmesi çok yüksek bir ihtimaldir. Sağlıklı kalmak, virüse yakalanmamak, yakalansa da çabuk atlatıp eski sağlığına kavuşmak daha önemsenir olmuştur. Beck, sanayi toplumunun itici güç söyleminin; açım. Risk toplumundaki itici güç söyleminin; korkuyorum! İfadesi olduğunu belirtir. Endişe ortaklığının dayanışma getirdiğini, ihtiyaç ortaklığı yerini endişe ortaklığı aldığını belirtir (Beck, 2019, s. 71-72). Salgın herkesi inşa ettiği ortak bir kimlikte buluşturur. Çünkü hepsi riskle karşı karşıyadır. Ortak paylaşılan kader duygusu mevcuttur. Kutuplaşmanın azaltılması için daha uygun bir ortamdır (Bavel vd. 2020, s. 464). Emniyette olamamak, korku ve kaygı toplumun oluşturduğu alışkanlıkları, kültürü, yaşam şeklini kademe kademe silikleştirir. Emniyette olmak için normal zamanda yapılan kültürel deneyimler askıya alınır. Covid-19 virüsünün yayılımını engellemek amacıyla kültürel bir unsur olan düğün törenleri kısıtlandı. Beraberlik ve dayanışma duygusu veren bireylerin kolektif bir şekilde hayatta mutluluk paylaşım aracı olan düğün merasimleri korkular nedeniyle normale dönene kadar anormal seyirde devam edecektir. Ölüm olgusu ve ritüelleride, insanların bir araya gelmesine vesile olmasından kısıtlanan kültürel unsurlardan biri oldu. Covid-19 Salgınında Toplumsallaşma ve Belirsizlikler Cenaze törenlerine katılım ve cenaze işlemleri rutinleri kısıtlamalar kapsamında değişimler yaşamıştır. Cenaze törenine katılamayanların bu ayrılık anlarını yaşayabilmesi için teknoloji vasıtasıyla uzakta olan yakınlara uzaktan cenaze katılımları yapıldığı medyaya yansıyan görüntülerle görülmüştür. Farklı toplumlarda aynı davranışlar geliştirilmiştir. Aynı risk kaderine verilen aynı tepkiler diyebiliriz. Sosyal mekanizmalar, bireylerin doğru olanı yapmasında kılavuzluk eder. Yanlış yapmasından kaçınmasını sağlar. Salgında endişe ortaklığı, toplumsal bir olgu ve davranış değişikliklerine neden olduğu için bir güç olmuştur. Yüzyıllarca birikerek gelen gelenekler virüs ile uygulanmaz hale gelmiştir (Bavel vd. 2020, s. 465). İnsanları bir araya getiren toplumsal unsurlar (düğün, cenaze, sünnet, ulusal ve dini bayramlar, vs.) normal dönemde pozitif anlamlara sahiptir. Hatta bu paylaşımlara katılmayan bireylere de toplum olumsuz anlamlar atfeder hatta dışlar. Sosyal normlara bağlılık ve uygunluk önemsenir. Covid-19 salgını bu anlayışı tam tersine çevirmiş durumdadır. Kamu, bireylerin bu tarz bir araya gelmemeleri konusunda medya aracılığıyla ve uygulamalarıyla uyarmaktadır. Artık makbul olan toplumsal ve kültürel birliktelikler için bir araya gelmek değil bir arada bulunmamak makbul olmuştur. Bireyler içinde bulunduğu kültürün üreticisi ve uygulayıcısı olarak, ilişkilerini içinde bulundukları kültür dünyasıyla anlamlandırmaktadır. Kültürlerini uygulayamayan ve deneyimlemeyen bireylerin dünya anlamlandırmasında yaşadıkları engellenmişlik ve belirsizliğin yeni kollektif bir kaygı haline evrilmesi olasıdır. Zorla yakınlık saldırganlık için bir risk faktörü oluşturur. Aile içi ve yakın akrabalarla ani zorla yakınlık, izolasyon ve karantina ile birleşince duygusal eğilimlerin patlak vereceği bir hal alabilir (Bavel vd. 2020, s. 466). Evde karantinada kalmak, sosyo-ekonomik olumsuzluklar, Covid-19 hastası olmak ve bu süreçle başetmeye çalışmak insanların psikolojilerini ve yakın ilişkilerini olumsuz etkilemiştir. Karantina ve kısıtlama uygulaması ile haftasonları ve akşam 21:00’dan sonra sokağa çıkma yasakları getirilmiştir. 65 yaş üstüne devamlı sokağa çıkma yasağı uygulanmıştır. Salgınla mücadele kapsamında öne çıkan sosyal mesafe kavramı yerine fiziksel mesafe kavramını kullanmak bireyler üzerinde algı farkı barındırır. Çünkü sosyal mesafe anlamlı etkileşimlerin sonlandırılması anlamı içermektedir. Fiziksel mesafe kavramı ise insanlar fiziksel olarak ayrı olsalar bile sosyal bağlantılarının mümkün olduğu anlamı barındırır (Bavel vd, 2020:462; Ferreira vd, 2020: 12). Bireylerin alıştığı toplumsallaşmanın getirdiği kollektif bilinç ve bir arada olmakla ortaya çıkabilecek dayanışma, güven gibi duyguların bir anda ortadan kalmış olması da onların toplumsal benliklerinde yabancılaşma getirdiğini söyleyebiliriz. Çevremizdeki insanlardan akraba cenazelerine katılamadığı için kendilerini suçlu hissetmiş olduklarını duymak, toplumsal normlara uymamanın kendini kötü hissettirdiğini duymak yüksek bir olasılıktır. Bu hisler bireylerin davranışlarının sosyal normlara göre düzenlenmesi nedeniyle anormal olan bu dönemde kendilerini kültürel bir alana konumlayamaması olabilir. Belkide ileride salgın dönemi Durkheim’in anomi (normsuzluk) kavramı referans alınarak anlaşılmaya çalışılacaktır. Covid-19 Salgını ve Sosyal Eşitsizlik Riskler, içerisinde her zaman potansiyel bir unsur barındırırlar. Bu riskin meydana gelmesi, sonuç ve zararını somut bir şekilde yaşatmış olması onu bitirmez. Yeni şartlar içerisinde yine potansiyelini var etmeye devam eder. Riskler bu yönüyle hem gerçek hem de gerçek dışıdır. 20. Yüzyılda tehlike olarak görülen risklerden bazıları suyun kirlenmesi, ormanların yok olması, küresel ısınma, buzulların erimesi, yeni hastalık türleri, virüsler vb. 21. Yüzyılda gerçekleşmiştir. Risk toplumsallaştı, risk küresel bir şekilde herkesi tehdit eden, her mekanda varlığını sürdürmesine rağmen, üzerinde toplumsal düşünme ve düşünce geliştirme eşit seviyede olmadı (Beck, 2019, s. 32). Covid-19 salgını aşırı sanayileşme ve küreselleşme ile beklenen risklerden biridir. Bugün ise bu risk gerçekleşmiştir. Riskler yaşa, cinsiyete, mesleğe, işin niteliğine, eğitime, gelire, beslenme alışkanlıklarına vs. göre farklı kişiler için farklı anlamlar taşımaktadır. Risklerin tehdit ve tehlikelerinin aslında ne kadar da toplumsal olduğuna yakından bakalım. Salgın, yeni öngörülebilir riskleri de beraberinde getirmiştir. Covid-19 kısıtlamaları kapsamında özellikle emek gücü ile çalışan kişilerin (işyerlerinin kapanması) işten çıkarılması, işsizlik; eğitimde dijitalleşme ile ekipmanları olmadığı için eğitimden faydalanamayan öğrencilerin eğitimdeki fırsat eşitsizliğinin sonuçları; karantina nedeniyle sosyal yaşamdaki ani değişimlere bağlı olarak bireylerde psikolojik dalgalanmalar; toplumsal değişimler alış-verişin dijitalleşmesi, maske, mesafe, dini ritüellerde kısıtlama, yasaklar, zorluklar toplumda huzursuzluklar yaratmıştır. Covid-19 salgınından olumsuz etkilenen başlıca kesimler yaşlılar, gençler ve kadınlardır. Yaşlı insanlar, yasaklar nedeniyle evlerinde sürekli bulunmaları gerekir ve sosyal medya kullanma oranları düşük olduğundan dış dünya ile bağlantıları daha zayıftır. Gençler; eğitimlerinde aksama yaşamaları nedeniyle geleceğe yönelik sosyal sermaye birikimleri gerilemiştir. İstihdam sayısında azalma nedeniyle iş bulma süreçleri uzamıştır. Kadınlar; okulun online olması nedeniyle çocukların evde bakımı sürekli hale gelmiş ve ev işçiliği artmıştır (Douglas, Katikireddi, Taulbut, McKee, ve McCartney 2020, s. 2). y ) Beck, risk toplumunda işsiz kalma riskinin; vasıfsızlarda vasıflılardan daha fazla olduğunu belirtmiştir. Aynı sanayi tesisinde çalışan vasıflı ve vasıfsız çalışanın yaşadığı stres, radyasyon ve sağlığına zarar verecek risklerin bölüşümü eşitsiz bir şekilde iş bölümüne bağlı olarak farklılaşmaktadır. Bu nedenle risklerin tolerans eşiği yukarılara çekilmektedir (Beck, 2019, s. 47). Covid-19 salgını sürecinde emek yoğun çalışanlar iş yerlerinin kapanma veya işten çıkarılma nedeniyle işsiz kalmış; beyaz yakalı, vasıflı çalışanlar bu durumdan onlar kadar etkilenmemiştir. Salgın döneminde işe gitmek zorunda kalan hiçbir birikimi ve sosyal güvencesi olmayanlar, günlük yevmiye ile çalışanlar, kargocular, market çalışanları, temizlik çalışanları, belediye hizmet çalışanları, sağlık çalışanları, şoförler, gıda ve ilaç sektöründe çalışanlar, hizmet sektöründe çalışanlar; kısaca farklı farklı sektörlerde emek gücüyle çalışanların ‘‘iş yeri sağlık için riskli işe gitmeyeceğim’’ demek gibi bir seçenekleri olmamıştır. Covid-19 Salgınında Toplumsallaşma ve Belirsizlikler Salgından önce, çocukluktan beri başlayan sosyalleşme ile toplum içerisinde nerde ne yapacağını öğrenmiş olan birey aidiyet duygusuyla kültürünün hem uygulayıcısı ve hemde devam ettiricisiydi. Şuan ise bu aidiyeti yeterli seviyede hissedemiyor çünkü alıştığı kültürüne uygun toplumsallaşmayı gerçekleştiremiyor. 207 Özge ÇELİK İçinde bulunduğu toplumun kültürünü yaşamaktan ziyade endişe ve korkunun dayattığı yeni davranış ve uygulamalara uymak zorunda kalıyor. Salgın sonrası bireylerin sosyalleşmesinde salgın dönemindeki sosyal ilişkilerde mesafeli olmanın etkileri ve farklılıklarına tanık olabiliriz. Yaşanan kültürel şokun etkilerinin talep ve uygulamalara evrildiğini görebiliriz. İçinde bulunduğu toplumun kültürünü yaşamaktan ziyade endişe ve korkunun dayattığı yeni davranış ve uygulamalara uymak zorunda kalıyor. Salgın sonrası bireylerin sosyalleşmesinde salgın dönemindeki sosyal ilişkilerde mesafeli olmanın etkileri ve farklılıklarına tanık olabiliriz. Yaşanan kültürel şokun etkilerinin talep ve uygulamalara evrildiğini görebiliriz. Dünya çapında 10 farklı ülkeden (Birleşik Krallık, Amerika Birleşik Devletleri, Avustralya, Almanya, İspanya, İtalya, İsveç, Meksika, Japonya ve Güney Kore) 700 katılımcı ile Mart ve Nisan 2020 tarihleri arasında Covid-19 Risk Algısı, nicel bir araştırma yapılmıştır. Bu ülkeler kültürel ve coğrafi çeşitliliği nedeniyle seçilmiştir. Risk algılama endeksi tüm ülkelerde yüksek çıkmıştır. Riskin, insanların deneyimlerine, değerlerine ve kurumlara olan güvenine dayanarak sosyal olarak belirlendiği görülmüştür. Ülkelerin sosyo-kültürel özelliklerine göre risk algılama oranları farklı çıkmıştır. Risk algılamasında kültürel çeşitlilik olduğu bulgulanmıştır. Covid-19 virüsü ile doğrudan kişisel deneyime sahip olanlar (testi pozitif çıkanlar veya virüse yakalandıklarını düşünenler) ve daha toplum yanlısı (daha az bireyselleşmiş toplumlar) dünya görüşüne sahip olanların risk algılamaları daha yüksek çıkmıştır (Dryhurst vd. 2020, s. 1003). Covid-19 Salgını ve Sosyal Eşitsizlik Bu durumların bir sonucu ve göstergesi olarak büyük şehirlerde toplu taşımaların salgın sürecinde de normal dönemi aratmayacak dolulukta olduğu medya aracılığıyla görüldü. Sayın ve Bozkurt’un yaptığı araştırmada kendini düşük gelir grubunda tanımlayanların 208 Covid-19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim mi! ekonomik ve sosyal kaygıları ile kötümserlik oranı katılımcıların %90’ına yakındır. Çünkü salgın sürecinde işlerini kaybetmiş ya da kaybetme ihtimali olan bireylerdir. Beyaz yakalı, lisans-yüksek lisans derecesinde eğitim alan orta-üst gelir grubuna dahil bireyler gelecek konusunda iyimser ve işsizlik konusunda daha az kaygı duydukları tespit edilmiştir (Sayın ve Bozkurt, 2020, s. 207). yg y p ( y ) Eşitsizlik arttıkça sağlık eşitsizliğide arttı. Hayat pahalılığı arttıkça, gelir düzeyi düşük olan bireyler için hayat zorlaşıyor. Yarı zamanlı çalışma artarken, işsiz kalma ve iş yeri kapatmalarıda yükseliyor. Bu durum toplumda yeni bir kriz oluşturuyor (Nytimes, 2020). Amerika şehir merkezlerinde yaşayan Afro- Amerikan, Latin Amerikan ve Amerika yerlileri, beyaz olmayan topluluklar, Covid-19 salgınında daha çok olumsuz etkilendi. Covid-19 salgını kaynaklı ölümlerde genel nüfusa göre Afro- Amerikanlarda daha yüksek seyretti. Irksal eşitsizlik ve sağlık eşitsizliği salgında daha belirginleşti. Covid-19’un eşitsizlik etkisi üzerinde sosyal mesafe uygulamayı engelleyen faktörlerde bulunmaktadır. Statüsü ve ekonomik geliri düşük market çalışanı, sağlık hizmeti çalışanı, toplu taşıma çalışanları olmaları nedeniyle enfekte olmaya daha yakınlar. Geçim imkanları kısıtlı ve gelir düzeyleri düşük olduğu içinde evde kalma ayrıcalığına sahip değildirler (Van, 2020, s. 1245). ğ ( ) Kontrol etmek, kontrol ettiğiniz şey kontrolünüzden çıkmadığı sürece anlamlıdır. Kontrolden çıktığı an o artık sizi kontrol etmeye başlar. Covid-19 gibi kontrolden çıkan bir salgında, eşitliksiz bir toplumda bireyler güvence ile korunabilir. Birey, işsiz, evsiz kaldığında, hasta olduğunda bu problemlerinin çözülebileceğini ya da karşılanabileceğini bilmesi gerekir. Belirsizlik kontrol edilemediğine göre önlem alınmalıdır. Sosyal eşitsizlikler arasındaki kırılma noktaları sosyal güvence ile güçlendirilmelidir. Salgın küresel olmasına rağmen çözümler yereldir. Ülkeler kendi ekonomik güçlerini sosyal politikalarla tabana yaydığı oranda başa çıkmaya çalışmaktadır. Küresel risklerin etkisi nedeniyle Batı ülkeleri daha otoriter görünüm sergilesede, küresel tehditler karşısında başarısızlar. Güçlü başarısız devletler halini alıyorlar (Ferreira vd., 2020, s. 7). İ ( ) İnsanların ölümle yüzyüze gelmesi insanlık için ilk değil. Bilim ve rasyonelliğin getirdiği ölçülebilirlik, tahmin edilebilirlik ve kontrol etmenin en yüksek olduğu dönemde salgın nedeniyle insanların ölümle yüzyüze gelmesi kaygı vericidir. Kontrol ettiğimizi düşündüğümüz pekçok şeyide kontrol edemediğimizi fark ettik. Modernleşmenin mutlu insan-mutlu toplum amacı; doğabilimlerinin metodları ile küresel mutsuz bir toplum halini aldı denilebilir. Covid-19 Salgını ve Sosyal Eşitsizlik Salgın sürecinde mutlu ve güvenli bir topluma ulaşmak ise daha fazla kontrol, dijital takip anlamına gelmektedir. Beck, risklerin kontrolünün gündelik hayatta baskı yaratabileceği ifade eder. Covid-19 salgın sürecinde artan dijitalleşme, HES kodu, aşı pasaportu gibi uygulamalar bireylerin özgürlük alanını kısıtladığına dair eleştiriler almaktadır. Belirsizlik arttıkça da daha fazla kontrol etme isteğine ihtiyaç duyuluyor. Salgın krizinin getirdiği belirsizliğin aşılması için bilgi ve iletişim teknolojilerinin kontrolü, bireylerin sürekli elektronik gözetimi, teknokratik otoriterlikle birlikte demokrasiye yönelik risklerin oluşumuna yol açabilir (Ferreira vd, 2020, s. 12). Covid-19 küresel salgını, sosyal ve kültürel değişimlerle günümüz toplumlarının risk, bilişim ve gözetim toplumu halini almasına vesile oldu (Bayhan, 2020, s. 821). Salgının bireyleri kontrol ederek son verileceği düşünülüyor belkide sosyal eşitsizlik ve sınıfsal farklar bu kadar yüksek olmasaydı, salgın daha kolay kontrol edilebilirdi. Risk toplumunda geleceği bu yaklaşımla planlamak daha güvenilir bir toplum inşa edebilir. Riskler, işsizliği tehdit ettiğinden, toplumda huzursuzluk yaratıp, insanları (gençler) sokağa dökebilir (Beck, 2019, s. 46). Pekçok ülkede salgın nedeniyle kısıtlama ve kapatmalar nedeniyle protestolar gerçekleşmiştir. Batı ülkeleri, diğer ülkeler ve ülkemizde şiddeti farklı olmakla birlikte sokağa çıkan ve taleplerini bu alanlarda dile getiren insanlar ve topluluklar görülmeye başlanmıştır. Beck, belirsizliklerin insanları nihayetinde sokağa dökeceğini teorisinde belirtmektedir. Laboratuvarlarda sadece bilim çevrelerinin ilgilenmesi gereken bir olgu gibi bakılan risklerin etkisinin toplumsal olarak bir etkiye ve tepkiye dönüşebileceğini öngörmüştür. Bu öngörüsünün Covid-19 salgın sürecinde ortaya çıktığı görülmüştür. g Risk toplumları, katılımcı demokratik bir gelişme içinde bulunurlar. Çünkü riskler ülke sınırlarını aştığından riski yok etmek için dünya toplumu olarak hareket etmek gereklidir. Ulus devlet sınırları, askeri ittifak ve ekonomik birliklerin riski defetmede yetersiz kalacağı söylenebilir. Uygarlık yol açtığı felaketi ortadan kaldırmak için sınırların ötesinde çözümler bulmak zorundadır. Çözüm; uzlaşı içeren konferanslar ile ortak bir yararda buluşup tehlikeyi ortadan kaldırmaktır (Beck, 2019, s. 68-69). Covid-19 salgını 209 Özge ÇELİK sürecinde ulus devletler kendi içinde ekonomilerini ayakta tutmaya ve virüsü sonlandırmaya çalışsa da uluslarüstü platformlarda görüşmeler yapılmakta ve çözümler üretilmeye çalışılmaktadır. Küresel Covid-19 salgınını sonlandırmak için aşının tedariği ve ülkelere dağılımının nasıl olacağı gündemdeki en sıcak konudur. sürecinde ulus devletler kendi içinde ekonomilerini ayakta tutmaya ve virüsü sonlandırmaya çalışsa da uluslarüstü platformlarda görüşmeler yapılmakta ve çözümler üretilmeye çalışılmaktadır. Küresel Covid-19 salgınını sonlandırmak için aşının tedariği ve ülkelere dağılımının nasıl olacağı gündemdeki en sıcak konudur. Neoliberal devletin girdiği ortaklık biçimleri, küresel bir şekilde sermayeyi korumak için ve meşrulaştırmak için devleti araç olarak kullandı. Riskler, neo-liberal sermaye, devlet koalisyonu ile kurulan iktidarı altüst etmektedir. (Beck, 2019, s. 363). Covid-19 Salgını ve Sosyal Eşitsizlik Neoliberal politikaların hızlı ekonomik büyüme getireceği ve kazanımların yoksullar dahil herkese yayılacağına dair inanç Covid-19 salgını ile kendisine bir neden daha ekledi. Çünkü, salgın en fazla yoksulları etkiledi. g y Ekonomisi güçlü olmayan ülkeler ekonomi temelli iken, ekonomisi güçlü sanayi ülkeleri büyümenin sürekliliğini garantileme çabası içerisindedir. Covid-19 salgınında güçlü ülkelerde en basit korunma ekipmanlarının, maske, eldivenin olmadığı ve tedariğinde sıkıntı yaşandığı görülmüştür. Risk toplumunda mülkün ve paranın değersizleştiğini belirten Beck’in teorisine bir örnektir. Maske virüsten korurken para virüs karşısında değersizleşmiştir. Para, maskenin tedariği için önemli bir faktördür. Başka bir ülkeden satın alınabilir. Ancak yaşanan felaket diğer ülkelerden maske tedariği yapamayacak boyutlara ulaştığında para maskeden değersiz olmaya devam edecektir. Sonuç Risk toplumu ile ilgili ilk özellik modern toplumun insan eliyle üretilmiş risklerinin insanlara zarar verir hale gelip felaket olarak yaşanmasıdır. Covid-19 salgınının zoonotik virüs kaynaklı olması sebebiyle, insan ürünü imal edilmiş bir risk olmadığı görüşü ile birlikte; nüfus artışı, küreselleşme ile artan sosyal ilişkiler, şehirleşme, doğaya yapılan plansız müdahale, küresel ısınma, fosil yakıt tüketimi, ormanların yok edilmesi, yaban hayatla temasın arttırılması, vahşi hayvanların kötü koşullarda satılması ve tüketilmesi vb. çoklu nedenlerle salgının meydana gelmesinin beklenen imal edilmiş bir risk olduğu kabul edilmektedir. Beck, doğa tehditlerinin kökenleri ve sonuçlarını insanların doğa ile kurduğu yanlış ilişkinin sonucu olarak gördüğünden, toplumsal sorunlar olarak kabul eder. Teoriye göre Covid-19 salgını da imal edilmiş bir risk olarak kabul edilebilir. Beck’in risk toplumu teorisindeki günümüz toplumunun neden risk toplumu olarak nitelemek gerektiğine yönelik beş tezi de Covid-19 salgını sürecinde yaşanan toplumsal örneklerle çalışmada desteklenmiştir. Risk toplumu teorisine göre küresel riskin üç ayırt edici özelliği olan mahalsizleşme, hesaplanamazlık, telafi edilemezlik özellikleri Covid-19 küresel salgınında görülmektedir (Ersöz, 2020, s. 536). Yine risk toplumu alt ayrıntıları olarak ele alındığında riskler; mekânsal olarak sınır tanımazlar, zamansal olarak belirsizdir ve toplumsal olarak karmaşık problemler üretir. Covid-19 küresel salgını bu üç alt özelliğide taşımaktadır. Risk toplumu, Covid-19 salgınının sosyal, kültürel ve siyasi yansımalarını anlamak için yararlı bir teoridir. Çünkü Beck’in teorisinden yapılan alıntılardan hareketle verilen güncel örnekler gösteriyor ki küresel salgında yaşananlar teorinin uygulaması diyebileceğimiz örneklerdir. g y g y ş yg y ğ Neo-liberal politikaların ürünleri ilerleme, refah, ekonomik büyüme, bilimsel rasyonalizm vb. ile kurduğumuz ilişki insanın özgürlüğünü alıp risk ve zarara vermiş gibi durmaktadır. Modern sanayi toplumunun ekonomi, ilerleme, bilim ile amacı; kirletilmiş hava, su, gıda, hasta insanlar ve belirsizlik içinde kaygılı insanlar olduğunu düşünemeyiz. Ancak geçen yıllarla birlikte dengeler değişti. İnsan kendi eliyle özgürlüğünü bu sefer otoriteye, iktidara değil gözüyle bile göremediği risklere teslim etti. Riskler insanları yöneten ve onların nasıl yaşamaları gerektiği hakkında hiç demokratik olmayan bir tutumla gücünü arttırmaya devam etmektedir. Bunu yaparken teknolojinin gücünden yararlanmayı da ihmal etmemektedir. İnsanlar daha çok gözlenir, daha çok testlerden geçer olmuştur. g g Covid-19 küresel bir salgın olduğu için bireyler küresel bir risk kaderinin içindedir. Salgına karşı aldığımız önlem ve verdiğimiz tepkiler evrensel bir benzerlik içermektedir. Virüs hayvandan insana geçmiş olsa da salgını insanlığın toplumsallığı şekillendirecektir. İnsanların kararları, onların risk toplumunun kısa bir ziyaretçisi mi yoksa kalıcı sakinimi olacağını belirleyecektir. 210 Covid-19 Salgını: Beck’in Risk Toplumuna Kısa Bir Ziyaret Mi Yoksa Kalıcı Yerleşim mi! Kaynakça Akgül, M., ve Can, E. N. (2020). Risk toplumu bağlamında Covid-19 haberlerine yönelik bir inceleme. OPUS Uluslararası Toplum Araştırmaları Dergisi, 17(33), 535-564. Akkuş, M. (2020). Koronavirüs ve Covid-19. Aksaray Üniversitesi Tıp Bilimleri Dergisi, 1(2), 15-20. Bayhan, V. (2020). Küresel salgının anatomisi: İnsan ve toplumun geleceği. M. Özer, A. Özer, C. Korkut (Yay. haz.), Küresel Salgının Sosyolojisi ve Yeni Normal Dönemde Paradigma Dönüşümü içinde (ss. 815-836). Türkiye Bilimler Akademisi. Beck, U. (2019). Risk toplumu: Başka bir modernliğe doğru. (K. Özdoğan ve B. Doğan, Çev. 3.Bs.). İstanbul: İthaki Yayıncılık. Beck, U. (2019). Risk toplumu: Başka bir modernliğe doğru. (K. Özdoğan ve B. Doğan, Çev. 3.Bs.) Birleşmiş Milletler Gıda ve Tarım Örgütü. (2021). Yeni Koronavirüs (COVID-19), http://www.fao.org/2019-ncov/en/ Erişim Tarihi: 24.02.2021. Birleşmiş Milletler Gıda ve Tarım Örgütü. (2021). Gıda Fiyat Endeksi http://www.fao.org/worldfoodsituation/foodpricesindex/en/. Erişim Tarihi: 27.02.2021. Gıda ve Tarım Örgütü. (2021). Gıda Fiyat Endeksi Şubat Raporu, http://www.fao.org/turkey/news/detail 8903/. Erişim Tarihi: 13.03.2021. Birleşmiş Milletler Gıda ve Tarım Örgütü. (2021). Gıda Fiyat Endeksi Şubat Raporu, http://www.fao.o news/tr/c/1378903/. Erişim Tarihi: 13.03.2021. Douglas, M., Katikireddi, S. V., Taulbut, M., McKee, M., ve McCartney, G. (2020). Mitigating the wider health effects of covid- 19 pandemic response. Bmj, 369. 1-6 https://www.bmj.com/content/369/bmj.m1557 adresinden erişildi. Ersöz, A. G. (2020). Covid-19 salgını özelinde Ulrich Beck’in “Risk Toplumu” ile Anthony Giddens’ın “Geç Modernite” kavramlarını yeniden düşünmek. Electronic Turkish Studies, 15(6). 525-538. Dryhurst, S., Schneider, C. R., Kerr, J., Freeman, A. L., Recchia, G., Van Der Bles, A. M., ... ve van der Linden, S. (2020). Risk perceptions of COVID-19 around the world. Journal of Risk Research, 23(7-8), 994-1006. https://www.tandfonline.com/doi/full/10.1080/13669877.2020.1758193 adresinden erişildi. Ferreira, C. M., Sá, M. J., Martins, J. G., ve Serpa, S. (2020). The COVID-19 contagion–pandemic dyad: A view from social sciences. Societies, 10(4), 77. https://www.mdpi.com/2075-4698/10/4/77 adresinden erişildi. Galanakis, C. M. (2020). The food systems in the era of the coronavirus (COVID-19) pandemic crisis. https://www.mdpi.com/2304-8158/9/4/523 adresinden erişildi. Gordo, O., C. ve Antó, J. (2020). COVID-19: The disease of the anthropocene. Environmental Research, Volume https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7227607/ adresinden erişildi. Koronavirüs Eşitsizliği Derinleştirirken Eşitsizlik Yayılmasını Kötüleştiriyor, https://www.nytimes.com/2020/03/15/world/europe /coronavirus-inequality.html.Erişim Tarihi: 02.03.2021. Maldonado, A., M. (2020). COVID-19 as a global risk: Confronting the ambivalences of a socionatural threat, 10 (4), 92. doi: 10.3390 / soc10040092 Sayın, V. ve Bozkurt, V. (2020). Pandemide küresel kentte yaşamak - İstanbul'un Covid-19 deneyimi. Demirbaş D., Bozkurt V., Yorgun S. (eds.). Covıd-19 Pandemisinin Ekonomik, Toplumsal ve Siyasal Etkileri, İstanbul Üniversitesi Yayınevi, E-ISBN: 978-605-07-0729-8 DOI: 10.26650/B/ SS46.2020.005. Kaynakça Sınıf Farkının Etkisi: Koronavirüs Yoksul Semtler ve İlçelerde Daha Hızlı Yayılıyor, https://onedio.com/haber/sinif-farkinin-etkisi- koronavirus-yoksul-semtler-ve-ilcelerde-daha-hizli-yayiliyor-913987. Erişim Tarihi: 01.03.2021. Türkiye İstatistik Kurumu (TÜİK). (2020). Yaşam Memnuniyeti Araştırması, https://data.tuik.gov.tr/Bulten/Index?p=Yasam- Memnuniyeti-Arastirmasi-202037209#:~:text=T%C3%BCrkiye'nin%20%48%2C2,y%C4%B1l%C4%B1nda%20%14%2C 5%20oldu. Türkiye İş Bankası. (2021). Haftalık Bülten Şubat 2021/5. https://ekonomi.isbank.com.tr/contentmanagement/Documents/tr02_haftalik/2021/EFB_20210201.pdf. Erişim Tarihi: 01.03.2021. Türkiye Cumhuriyeti Merkez Bankası. (2021) Enflasyon Raporu 2021- I (28 Ocak 2021) https://www.tcmb.gov.tr/wps/wcm/connect/TR/TCMB+TR/Main+Menu/Yayinlar/Raporlar/Enflasyon+Raporu/2021/Enflas yon+Raporu+2021+-+I. Erişim Tarihi: 01.03.2021./ Van Bavel, J. J., Baicker, K., Boggio, P. S., Capraro, V., Cichocka, A., Cikara, M., ... & Willer, R. (2020). Using social and behavioural science to support COVID-19 pandemic response. Nature human behaviour, 4(5), 460-471. https://www.nature.com/articles/s41562-020-0884-z. adresinden erişildi. Van Dorn, A., Cooney, R. E., & Sabin, M. L. (2020). COVID-19 exacerbating inequalities in the US. Lancet (London, England), 395(10232), 1243. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7162639/ adresinden erişildi. Vidal, J. (2020). Destroyed habitat creates the perfect conditions for coronavirus to emerge. Scientific American, 18. 1-13 https://dtnetwork.org/wp-content/uploads/2020/04/covid-19-related-to-destruction-of-planet.pdf. adresinden erişildi. Vidal, J. (2020). Destroyed habitat creates the perfect conditions for coronavirus to emerge. Scientific American, 18. 1-13 https://dtnetwork.org/wp-content/uploads/2020/04/covid-19-related-to-destruction-of-planet.pdf. adresinden erişildi. 211
https://openalex.org/W2883691184	https://www.biorxiv.org/content/biorxiv/early/2018/07/19/372623.full.pdf	English	null	Extreme allelic heterogeneity at a<i>Caenorhabditis elegans</i>beta-tubulin locus explains natural resistance to benzimidazoles	bioRxiv (Cold Spring Harbor Laboratory)	2,018	cc-by	26,277	. CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Extreme allelic heterogeneity at a Caenorhabditis elegans beta-tubulin locus 1 explains natural resistance to benzimidazoles 2 3 Steffen R. Hahnel⧺,1, Stefan Zdraljevic⧺,1,2, Briana C. Rodriguez1, Yuehui Zhao3, Patrick T. McGrath3, an 4 Erik C. Andersen1,2,4 * 5 6 1. Department of Molecular Biosciences, Northwestern University, Evanston, IL 60208, USA 7 2. Interdisciplinary Biological Sciences Program, Northwestern University, Evanston, IL 60208, USA 8 3. School of Biology, Georgia Institute of Technology, Atlanta, Georgia 30332 9 4. Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL 60611, USA 10 ⧺ Equal contribution 11 * Corresponding author 12 13 14 Erik C. Andersen 15 Assistant Professor of Molecular Biosciences 16 Northwestern University 17 Evanston, IL 60208, USA 18 Tel: (847) 467-4382 19 Fax: (847) 491-4461 20 Email: Erik.Andersen@Northwestern.edu 21 22 23 24 25 Steffen R. Hahnel, steffen.hahnel@northwestern.edu, ORCID 0000-0001-8848-0691 26 Stefan Zdraljevic, stefanzdraljevic2018@u.northwestern.edu, ORCID 0000-0003-2883-4616 27 Briana C. Rodriguez, briana.rodriguez@northwestern.edu 28 Yuehui Zhao, yzhao349@gatech.edu 29 Patrick T. McGrath, patrick.mcgrath@biology.gatech.edu, ORCID 0000-0002-1598-3746 30 Erik C. Andersen, erik.andersen@northwestern.edu, ORCID 0000-0003-0229-9651 31 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under Extreme allelic heterogeneity at a Caenorhabditis elegans beta-tubulin locus 1 explains natural resistance to benzimidazoles 2 3 Steffen R. Hahnel⧺,1, Stefan Zdraljevic⧺,1,2, Briana C. Rodriguez1, Yuehui Zhao3, Patrick T. McGrath3, a 4 Erik C. Andersen1,2,4 * 5 6 1. Department of Molecular Biosciences, Northwestern University, Evanston, IL 60208, USA 7 2. Interdisciplinary Biological Sciences Program, Northwestern University, Evanston, IL 60208, USA 8 3. School of Biology, Georgia Institute of Technology, Atlanta, Georgia 30332 9 4. Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL 60611, USA 10 ⧺ Equal contribution 11 * Corresponding author 12 13 14 Erik C. Andersen 15 Assistant Professor of Molecular Biosciences 16 Northwestern University 17 Evanston, IL 60208, USA 18 Tel: (847) 467-4382 19 Fax: (847) 491-4461 20 Email: Erik.Andersen@Northwestern.edu 21 22 23 24 25 Steffen R. Hahnel, steffen.hahnel@northwestern.edu, ORCID 0000-0001-8848-0691 26 Stefan Zdraljevic, stefanzdraljevic2018@u.northwestern.edu, ORCID 0000-0003-2883-4616 27 Briana C. Abstract 32 Taken together, our results establish a population-level resource of nematode natural diversity 57 as an important model for the study of mechanisms that give rise to BZ resistance 58 Benzimidazoles (BZ) are essential components of the limited chemotherapeutic arsenal available to 33 control the global burden of parasitic nematodes. The emerging threat of BZ resistance among nearly all 34 nematode species necessitates the development of novel strategies to identify genetic and molecular 35 mechanisms underlying this resistance. All detection of parasitic helminth resistance to BZ is focused on 36 the genotyping of three variant sites in the orthologs of the β-tubulin gene found to confer resistance in the 37 free-living nematode Caenorhabditis elegans. Because of the limitations of laboratory and field 38 experiments in parasitic nematodes, it is difficult to look beyond these three sites, and additional BZ 39 resistance is observed in the field. Here, we took an unbiased genome-wide mapping approach in the 40 free-living nematode species C. elegans to identify the genetic underpinnings of natural resistance to the 41 commonly used BZ, albendazole (ABZ). We found a wide range of natural variation in ABZ resistance in 42 natural C. elegans populations. In agreement with known mechanisms of BZ resistance in parasites, we 43 find that a majority of the variation in ABZ resistance among wild C. elegans strains is caused by variation 44 in the β-tubulin gene ben-1. This result shows empirically that resistance to ABZ naturally exists and 45 segregates within the C. elegans population, suggesting that selection in natural niches could enrich for 46 resistant alleles. We identified 25 distinct ben-1 alleles that are segregating at low frequencies within the 47 C. elegans population, including many novel molecular variants. Population genetic analyses indicate that 48 ben-1 variation arose multiple times during the evolutionary history of C. elegans and provide evidence 49 that these alleles likely occurred recently because of local selective pressures. Additionally, we find 50 purifying selection at all five β-tubulin genes, despite predicted loss-of-function resistants variants in ben-1, 51 indicating that BZ resistance in natural niches is a stronger selective pressure than loss of one β-tubulin 52 gene. Furthermore, we use genome-editing to show that the most common parasitic nematode β-tubulin 53 allele that confers BZ resistance, F200Y, confers resistance in C. elegans. Importantly, we identified a 54 novel genomic region that is correlated with ABZ resistance in the C. . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Rodriguez, briana.rodriguez@northwestern.edu 28 Yuehui Zhao, yzhao349@gatech.edu 29 Patrick T. McGrath, patrick.mcgrath@biology.gatech.edu, ORCID 0000-0002-1598-3746 30 Erik C. Andersen, erik.andersen@northwestern.edu, ORCID 0000-0003-0229-9651 31 . CC-BY 4.0 International license a 1 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Abstract 32 Abstract 32 Benzimidazoles (BZ) are essential components of the limited chemotherapeutic arsenal available to 33 control the global burden of parasitic nematodes. The emerging threat of BZ resistance among nearly all 34 nematode species necessitates the development of novel strategies to identify genetic and molecular 35 mechanisms underlying this resistance. All detection of parasitic helminth resistance to BZ is focused on 36 the genotyping of three variant sites in the orthologs of the β-tubulin gene found to confer resistance in the 37 free-living nematode Caenorhabditis elegans. Because of the limitations of laboratory and field 38 experiments in parasitic nematodes, it is difficult to look beyond these three sites, and additional BZ 39 resistance is observed in the field. Here, we took an unbiased genome-wide mapping approach in the 40 free-living nematode species C. elegans to identify the genetic underpinnings of natural resistance to the 41 commonly used BZ, albendazole (ABZ). We found a wide range of natural variation in ABZ resistance in 42 natural C. elegans populations. In agreement with known mechanisms of BZ resistance in parasites, we 43 find that a majority of the variation in ABZ resistance among wild C. elegans strains is caused by variation 44 in the β-tubulin gene ben-1. This result shows empirically that resistance to ABZ naturally exists and 45 segregates within the C. elegans population, suggesting that selection in natural niches could enrich for 46 resistant alleles. We identified 25 distinct ben-1 alleles that are segregating at low frequencies within the 47 C. elegans population, including many novel molecular variants. Population genetic analyses indicate that 48 ben-1 variation arose multiple times during the evolutionary history of C. elegans and provide evidence 49 that these alleles likely occurred recently because of local selective pressures. Additionally, we find 50 purifying selection at all five β-tubulin genes, despite predicted loss-of-function resistants variants in ben-1, 51 indicating that BZ resistance in natural niches is a stronger selective pressure than loss of one β-tubulin 52 gene. Furthermore, we use genome-editing to show that the most common parasitic nematode β-tubulin 53 allele that confers BZ resistance, F200Y, confers resistance in C. elegans. Importantly, we identified a 54 novel genomic region that is correlated with ABZ resistance in the C. elegans population but independent 55 of ben-1 and the other β-tubulin loci, suggesting that there are multiple mechanisms underlying BZ 56 resistance. Abstract 32 elegans population but independent 55 of ben-1 and the other β-tubulin loci, suggesting that there are multiple mechanisms underlying BZ 56 2 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Author summary 59 Nematode parasites have a tremendous impact on human health with almost two billion people infected 60 worldwide. The control of nematode infections relies mainly on the efficacy of a limited repertoire of 61 anthelmintic compounds, including the benzimidazoles (BZ). Already a significant problem in veterinary 62 medicine, increasing evidence exists for the development of BZ resistance in nematodes that infect 63 humans. Laboratory screens and field surveys identified β-tubulin genes as major determinants of BZ 64 resistance in nematodes but detailed population-wide genetic analyses of resistance mechanisms are only 65 just beginning. Therefore, we took advantage of the free-living model organism Caenorhabditis elegans to 66 study the genetic basis of resistance to the commonly used BZ, albendazole (ABZ) in a natural nematode 67 population. Performing genome-wide association mappings, we were able to identify extreme 68 heterogeneity in the β-tubulin gene ben-1 as a major determinant of ABZ resistance. Moreover, our study 69 provided new insights into the effects of missense and loss-of-function alleles at this locus, and how 70 anthelmintic resistance could have developed within a natural nematode population. 71 3 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Introduction 72 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint [12] have independently identified numerous β-tubulin mutant alleles. These findings were then translated 100 to veterinary-relevant nematodes, where three major single-nucleotide variants (SNVs) in parasitic 101 nematode β-tubulin genes were found to be highly correlated with BZ resistance. The most common 102 variant causes a tyrosine to phenylalanine amino acid change at position 200 (F200Y) [13]. Other SNVs 103 linked to BZ resistance have been described at positions 167 (F167Y) and 198 (E198A) in several 104 nematode species [14,15]. Although all three of these missense variants have been shown to reduce BZ 105 binding affinity to purified β-tubulins in vitro [16,17], the BZ β-tubulin binding site remains experimentally 106 uncharacterized. Furthermore, these missense variants do not explain all of the BZ resistance observed in 107 the field [4,18]. This discrepancy might indicate that additional, non-target related components such as 108 drug efflux pumps and detoxification enzymes might contribute to BZ resistance [4,19,20]. 109 Additional genetic variants associated with natural BZ resistance can be identified using 111 quantitative genetic approaches that consider genotypic and phenotypic variation present within a wild 112 population of parasitic nematodes. However, parasitic nematodes are not easily amenable to quantitative 113 genetic approaches because of their complicated life cycles, their poorly annotated reference genomes, 114 and limited molecular and genetic tools [21–23]. Recent successes in the hookworm parasites found 115 genomic intervals that underlie ivermectin resistance [21,24], but similar approaches have not been 116 applied to BZ resistance yet. By contrast, the free-living nematode species C. elegans has a short life 117 cycle, a well annotated reference genome [25–27], and an abundance of molecular and genetic tools for 118 the characterization of BZ responses [28–30]. A recent example that applied quantitative genetic 119 approaches to investigated BZ responses in C. elegans used a mapping population generated between 120 two genetically divergent strains to identify a quantitative trait loci (QTL) linked to BZ resistance [30]. Introduction 72 Parasitic nematodes have a tremendous impact on global health and socio-economic development, 73 especially in the developing world [1]. They are among the most widespread human pathogens, and 74 almost two billion people are estimated to suffer from infection of one or multiple nematode species [1,2]. 75 The main endemic areas of nematode infections are highly correlated with tropical and subtropical regions 76 worldwide. Because of their detrimental impact on human health, several nematode infections belong to a 77 class of diseases designated by the World Health Organization (WHO) as Neglected Tropical Diseases 78 (NTDs). Altogether, the loss of disability-adjusted life years (DALY) caused by parasitic nematodes is 79 conservatively estimated to be 10 million DALYs per year, which ranks them among the top of all NTDs 80 [1]. Apart from this drastic impact on human health, several nematode species infect a variety of key crops 81 and livestock causing substantial economic losses throughout the world [3]. 82 Global control of nematode infections relies on the efficacy of a limited repertoire of anthelmintic 84 drugs, including benzimidazoles (BZ). BZs are frequently used in mass drug administration (MDA) 85 programs to treat parasitic nematode infections in endemic regions. However, the long-term success of 86 these MDA programs is limited by the persistence and re-emergence of nematode infections in these 87 regions. The most significant of these factors is the high reinfection rate caused by long-term parasitic 88 nematode reservoirs, which necessitates frequent deworming of affected communities [2]. As a 89 consequence of continuous drug-pressure and relaxation cycles, parasite populations have the potential to 90 develop anthelmintic resistance. This resistance is already a significant problem in veterinary medicine [4], 91 and resistance in the parasitic nematodes that infect humans is becoming increasingly evident. For 92 example, reduced cure rates have been observed for soil-transmitted helminthiases, which might be 93 attributable to BZ resistance [5–7]. To face this growing threat, detailed knowledge of anthelmintic 94 resistance mechanisms is required to improve diagnostic tools for field surveys and educate drug- 95 treatment strategies in MDA programs. 96 4 of 52 In vitro studies have shown that BZs inhibit the polymerization of microtubules [8–10]. Mutagenesis 98 screens that selected for BZ resistance in Saccharomyces cerevisiae [11] and Caenorhabditis elegans 99 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. Genetically distinct C. elegans natural isolates respond differently to ABZ 143 We investigated the differences in responses across a panel of C. elegans wild strains to one of the most 144 commonly used BZ compounds (albendazole, ABZ) in human and veterinary medicine [4,27,37]. To test 145 the effects of ABZ treatment on C. elegans, we exposed four genetically divergent C. elegans isolates to 146 various ABZ concentrations and measured the number and length of progeny the animals produced. We 147 used these two traits because they represent endpoint measures for ABZ efficacy against human parasitic 148 nematodes. After four days of exposure to ABZ, we detected a decrease in brood size and progeny length 149 for all four assayed strains (Supplemental Figure 1;; Supplemental data 1 and 2). Additionally, this assay 150 revealed differential ABZ sensitivity among these wild strains, as measured by brood size and progeny 151 length. For subsequent genome-wide association (GWA) mapping experiments, we used an ABZ 152 concentration of 12.5 µM, which is a concentration that induced robust differences in ABZ responses 153 among the four assayed wild strains. 154 Introduction 72 CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Introduction 72 121 Importantly, the QTL identified in this study did not overlap with β-tubulin genes, suggesting that 122 quantitative genetic approaches in C. elegans can be used to discover novel mechanisms associated with 123 BZ resistance. 124 5 of 52 In the present study, we leveraged the power of C. elegans natural diversity to perform genome- 126 wide association (GWA) mappings for BZ resistance. We used a set of 249 wild C. elegans isolates 127 5 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint available through the C. elegans Natural Diversity Resource (CeNDR) [27] to identify genomic regions that 128 contribute to albendazole (ABZ) resistance. ABZ is a broadly administered BZ used to treat parasitic 129 nematode infections in humans and livestock [4,30]. We show that a major source of ABZ resistance in the 130 C. elegans population is driven by putative loss-of-function (LoF) variants in the ben-1 locus. Notably, we 131 found 25 distinct ben-1 alleles with low minor allele frequencies (MAF) that contribute to ABZ resistance. 132 We show that these putative ben-1 LoF variants arose independently during the evolutionary history of the 133 C. elegans species, which suggests that local BZ selective pressures might have contributed to the 134 extreme allelic heterogeneity at this locus. We next made use of the extensive molecular toolkit available 135 in C. elegans to verify that the introduction of the ben-1 LoF alleles do not result in any detectable fitness 136 consequences in standard laboratory conditions, but does confer ABZ resistance. Taken together, these 137 results suggest that C. elegans and likely free-living stages of parasitic species encounter natural 138 compounds that promote BZ resistance through selection for standing or de novo variation in conserved 139 nematode-specific β-tubulin genes. 140 6 of 52 . including the ben-1 locus 157 To identify genomic loci that underlie strain-specific ABZ responses, we performed genome-wide 158 association (GWA) mappings using HTA fitness data obtained from exposing 209 C. elegans to either 159 DMSO (control) or ABZ and DMSO conditions (Supplemental data 3 and 4). We employed two statistical 160 methods to perform GWA mappings using the regressed animal length (q90.TOF) and normalized brood 161 size (norm.n) traits, a single-marker and a gene-burden approach [39,40]. Using the single-marker 162 approach [32,35–37], we identified three distinct quantitative trait loci (QTL) that explained variation in 163 animal length among wild isolates exposed to ABZ, whereas the brood-size trait did not map to any 164 significant genomic loci (Supplemental data 5 and 6). Two of the animal-length QTL we identified are 165 located on chromosome II, and the third is located on chromosome V (Figure 1A). The QTL on the left arm 166 of chromosome II spans from 25 kb to 3.9 Mb and has a peak-marker position at 458 kb. The second QTL 167 7 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint on chromosome II has a peak-marker position at 11 Mb and spans from 10.17 Mb to 11.6 Mb. The QTL on 168 the right arm of chromosome V spans from from 18.04 Mb to 18.68 Mb, with the peak marker located at 169 18.35 Mb (Supplemental Figure 2;; Supplemental data 5). Remarkably, the β-tubulin gene ben-1, shown to 170 be the major determinant for ABZ resistance in the C. elegans laboratory strain N2 [12,13], did not map 171 using this single marker-based approach (Figure 1A). 172 To complement the single-marker mapping described above, we performed a gene-burden 174 approach to map BZ resistance [39,40]. including the ben-1 locus 157 Compared to the single-marker approach that includes SNVs with 175 a minimum 5% minor allele frequency among all strains, the gene-burden approach incorporates all rare 176 strain-specific variation within a gene for association testing [39,56]. With this alternative mapping 177 strategy, we identified the same two QTL on chromosome II that we identified with the single-marker 178 approach. In addition to the chromosome II QTL, we found a significant association between variation at 179 the ben-1 locus on chromosome III and animal-length variation in response to ABZ (Figure 1B). These 180 results suggest that no variants are shared above 5% minor allele frequency among the wild isolates at 181 the ben-1 locus (chromosome III, 3,537,688-3,541,628 bp). An additional significant gene was detected on 182 the right arm of chromosome I (Supplemental data 7 and 8). 183 C. elegans ABZ resistance correlates with extreme allelic heterogeneity at the ben-1 locus 185 8 of 52 To explain the differences between the single-marker and gene-burden based GWA mapping results with 186 respect to ben-1, we investigated the natural variation found at this genomic locus in more detail (Figure 2;; 187 Supplemental data 9). We first used the snpeff function of the cegwas package [27,57] to look for SNVs in 188 ben-1 with predicted moderate-to-high effects on gene function [27,57]. These moderate-to-high impact 189 variants include missense variants, splice donor and acceptor variants, and alternative start and stop 190 codons that might disrupt the open reading frame of ben-1. 191 192 In the set of 209 strains used for the GWA mapping, we identified 19 strains with moderate-to-high 193 impact variants in the ben-1 locus, as predicted by snpEff [57], including 13 strains with amino-acid 194 substitutions. Amino-acid substitutions in β-tubulin genes are important markers for BZ resistance in 195 To explain the differences between the single-marker and gene-burden based GWA mapping results with 186 respect to ben-1, we investigated the natural variation found at this genomic locus in more detail (Figure 2;; 187 Supplemental data 9). We first used the snpeff function of the cegwas package [27,57] to look for SNVs in 188 ben-1 with predicted moderate-to-high effects on gene function [27,57]. These moderate-to-high impact 189 variants include missense variants, splice donor and acceptor variants, and alternative start and stop 190 codons that might disrupt the open reading frame of ben-1. 191 8 of 52 In the set of 209 strains used for the GWA mapping, we identified 19 strains with moderate-to-high 193 impact variants in the ben-1 locus, as predicted by snpEff [57], including 13 strains with amino-acid 194 substitutions. Amino-acid substitutions in β-tubulin genes are important markers for BZ resistance in 195 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. C. elegans ABZ resistance correlates with extreme allelic heterogeneity at the ben-1 locus 185 ; https://doi.org/10.1101/372623 doi: bioRxiv preprint parasitic nematodes and are hypothesized to reduce binding affinity of BZs to β-tubulin [16,17]. In the 196 C. elegans panel of wild strains, we identified six strains that have the F200Y mutation in BEN-1, which is 197 known to be the most common BZ resistance marker in livestock parasites [13]. Additionally, we detected 198 novel amino-acid substitutions in BEN-1, which have not previously been associated with ABZ resistance 199 in nematodes. These missense variants include an A185P substitution variant, found in two strains, as 200 well as the E69G, Q131L, S145F, M257I, and D404N substitution variants, all of which were unique to 201 single strains. In contrast to the amino-acid variants F200Y, E198A, and F167Y found in parasitic 202 nematodes, which cluster at the putative ABZ binding site [58–60], these novel variants are more 203 distributed throughout the protein structure (Supplemental Figure 3). We classified a strain as resistant to 204 ABZ if its animal-length phenotype after ABZ exposure was greater than 50% of the difference between 205 the most and least ABZ-resistant strains (Figure 2A). Eleven of the twelve strains with amino-acid 206 substitutions are more resistant to ABZ than strains with no variation at the ben-1 locus. Similarly, strains 207 with predicted high-impact variants in ben-1 are resistant to ABZ treatment. The strains with high-impact 208 variants include five strains with unique premature stop codons and one strain with a predicted splice- 209 donor variant at the end of exon 1. However, many ABZ-resistant strains in the C. elegans population do 210 not contain any of these rare and common variants. Therefore, these strains either have other types of 211 deleterious variants at the ben-1 locus or are resistant to ABZ through a distinct mechanism. To 212 differentiate these two possibilities, we manually curated all of the strains raw sequence read alignment 213 files (BAM files), publicly available through the CeNDR website [27]. This in-depth investigation revealed 214 extreme allelic heterogeneity at the ben-1 locus in the natural C. elegans population. Twenty-seven strains 215 had either rare deletions or insertions in ben-1 (22 deletions, 5 insertions;; Supplemental data 9). 216 Interestingly, a subset of individuals with missense variants described above were found to be in perfect 217 linkage disequilibrium (LD) with nearby deletion alleles. C. elegans ABZ resistance correlates with extreme allelic heterogeneity at the ben-1 locus 185 All six strains with the F200Y allele share the 218 same deletion of approximately 160 bp that partially removes exons 3 and 4. The close proximity of the 219 parasitic nematodes and are hypothesized to reduce binding affinity of BZs to β-tubulin [16,17]. In the 196 C. elegans panel of wild strains, we identified six strains that have the F200Y mutation in BEN-1, which is 197 known to be the most common BZ resistance marker in livestock parasites [13]. Additionally, we detected 198 novel amino-acid substitutions in BEN-1, which have not previously been associated with ABZ resistance 199 in nematodes. These missense variants include an A185P substitution variant, found in two strains, as 200 well as the E69G, Q131L, S145F, M257I, and D404N substitution variants, all of which were unique to 201 single strains. In contrast to the amino-acid variants F200Y, E198A, and F167Y found in parasitic 202 nematodes, which cluster at the putative ABZ binding site [58–60], these novel variants are more 203 distributed throughout the protein structure (Supplemental Figure 3). We classified a strain as resistant to 204 ABZ if its animal-length phenotype after ABZ exposure was greater than 50% of the difference between 205 the most and least ABZ-resistant strains (Figure 2A). Eleven of the twelve strains with amino-acid 206 substitutions are more resistant to ABZ than strains with no variation at the ben-1 locus. Similarly, strains 207 with predicted high-impact variants in ben-1 are resistant to ABZ treatment. The strains with high-impact 208 variants include five strains with unique premature stop codons and one strain with a predicted splice- 209 donor variant at the end of exon 1. However, many ABZ-resistant strains in the C. elegans population do 210 not contain any of these rare and common variants. Therefore, these strains either have other types of 211 deleterious variants at the ben-1 locus or are resistant to ABZ through a distinct mechanism. To 212 differentiate these two possibilities, we manually curated all of the strains raw sequence read alignment 213 files (BAM files), publicly available through the CeNDR website [27]. This in-depth investigation revealed 214 extreme allelic heterogeneity at the ben-1 locus in the natural C. elegans population. Twenty-seven strains 215 had either rare deletions or insertions in ben-1 (22 deletions, 5 insertions;; Supplemental data 9). C. elegans ABZ resistance correlates with extreme allelic heterogeneity at the ben-1 locus 185 239 240 Among the remaining ABZ-resistant strains, we identified a putative transposon insertion in exon 5 241 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint to the locus of the β-tubulin gene tbb-2, which is in relative close genomic distance (chromosome III, 224 4,015,769 - 4,017,643 bp). Additionally, the amino acid substitution (E69G), which is only observed in a 225 single isotype (KR314), co-occurs with a deletion in exon 2 that is predicted to cause a frameshift in the 226 ben-1 open reading frame. 227 Of all the strains with ben-1 SNVs and indels, only three individuals were not classified as resistant 229 to ABZ treatment. The least resistant of these three strains was the CB4856 strain, which contains a nine 230 base pair deletion near the end of the ben-1 coding sequence. Because this in-frame deletion variant is at 231 the end of the coding sequence, we hypothesized that CB4856 still contains a functional copy of the ben-1 232 gene and this variant likely does not confer resistance to ABZ. The next least resistant strain is QG2075, 233 which shares a four base pair deletion in exon 1 of ben-1 with the ABZ-resistant isotype WN2033. During 234 the growth phase of our HTA assay, we noted that the QG2075 strain has a slow-growth phenotype in 235 normal growth conditions, which is likely confounding our phenotypic measurements in ABZ. Finally, the 236 JU2862 strain has a unique four base pair deletion in exon 4 of ben-1, that is predicted to cause a 237 frameshift in the ben-1 open reading frame. We note that JU2862 is right at our arbitrary ABZ-resistance 238 threshold and upon re-phenotyping with higher replication might be classified as ABZ resistant. 239 Among the remaining ABZ-resistant strains, we identified a putative transposon insertion in exon 5 241 of ben-1 in the JU3125 strain. The genomic origin of this putative transposon insertion is from position 242 17.07 Mb on chromosome X and corresponds to a cut and paste DNA transposon Tc5B, which is part of 243 the TcMar-Tc4 transposon superfamily [61,62]. C. elegans ABZ resistance correlates with extreme allelic heterogeneity at the ben-1 locus 185 216 Interestingly, a subset of individuals with missense variants described above were found to be in perfect 217 linkage disequilibrium (LD) with nearby deletion alleles. All six strains with the F200Y allele share the 218 same deletion of approximately 160 bp that partially removes exons 3 and 4. The close proximity of the 219 annotated F200Y to this 160 bp deletion suggested that the F200Y variant is actually an error in read 220 9 of 52 to the locus of the β-tubulin gene tbb-2, which is in relative close genomic distance (chromosome III, 224 4,015,769 - 4,017,643 bp). Additionally, the amino acid substitution (E69G), which is only observed in a 225 single isotype (KR314), co-occurs with a deletion in exon 2 that is predicted to cause a frameshift in the 226 ben-1 open reading frame. 227 228 Of all the strains with ben-1 SNVs and indels, only three individuals were not classified as resistant 229 to ABZ treatment. The least resistant of these three strains was the CB4856 strain, which contains a nine 230 base pair deletion near the end of the ben-1 coding sequence. Because this in-frame deletion variant is at 231 the end of the coding sequence, we hypothesized that CB4856 still contains a functional copy of the ben-1 232 gene and this variant likely does not confer resistance to ABZ. The next least resistant strain is QG2075, 233 which shares a four base pair deletion in exon 1 of ben-1 with the ABZ-resistant isotype WN2033. During 234 the growth phase of our HTA assay, we noted that the QG2075 strain has a slow-growth phenotype in 235 normal growth conditions, which is likely confounding our phenotypic measurements in ABZ. Finally, the 236 JU2862 strain has a unique four base pair deletion in exon 4 of ben-1, that is predicted to cause a 237 frameshift in the ben-1 open reading frame. We note that JU2862 is right at our arbitrary ABZ-resistance 238 threshold and upon re-phenotyping with higher replication might be classified as ABZ resistant. 239 240 Among the remaining ABZ-resistant strains, we identified a putative transposon insertion in exon 5 241 of ben-1 in the JU3125 strain. The genomic origin of this putative transposon insertion is from position 242 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. C. elegans ABZ resistance correlates with extreme allelic heterogeneity at the ben-1 locus 185 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint to the locus of the β-tubulin gene tbb-2, which is in relative close genomic distance (chromosome III, 224 4,015,769 - 4,017,643 bp). Additionally, the amino acid substitution (E69G), which is only observed in a 225 single isotype (KR314), co-occurs with a deletion in exon 2 that is predicted to cause a frameshift in the 226 ben-1 open reading frame. 227 228 Of all the strains with ben-1 SNVs and indels, only three individuals were not classified as resistant 229 to ABZ treatment. The least resistant of these three strains was the CB4856 strain, which contains a nine 230 base pair deletion near the end of the ben-1 coding sequence. Because this in-frame deletion variant is at 231 the end of the coding sequence, we hypothesized that CB4856 still contains a functional copy of the ben-1 232 gene and this variant likely does not confer resistance to ABZ. The next least resistant strain is QG2075, 233 which shares a four base pair deletion in exon 1 of ben-1 with the ABZ-resistant isotype WN2033. During 234 the growth phase of our HTA assay, we noted that the QG2075 strain has a slow-growth phenotype in 235 normal growth conditions, which is likely confounding our phenotypic measurements in ABZ. Finally, the 236 JU2862 strain has a unique four base pair deletion in exon 4 of ben-1, that is predicted to cause a 237 frameshift in the ben-1 open reading frame. We note that JU2862 is right at our arbitrary ABZ-resistance 238 threshold and upon re-phenotyping with higher replication might be classified as ABZ resistant. C. elegans ABZ resistance correlates with extreme allelic heterogeneity at the ben-1 locus 185 Finally, the MY518 strain is resistant to ABZ but did not 244 contain any of the above classes of variation. However, we did identify a 1 kb inversion that spans exon 1 245 and the promoter region of ben-1. Remarkably, all structural variants present in ben-1 are only present in 246 one or few wild strains and are mostly predicted to cause loss of ben-1 function (Figure 2). Altogether, the 247 putative loss-of-function variants described above explain 73.8% of the phenotypic variation present in the 248 natural C. elegans population. 249 10 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 252 Within-species selective pressures at the ben-1 locus 253 276 However, when we consider ben-1 coding variation and non-coding variation separately, we found the 277 Hcoding statistic to be 0.34 and the Hnoncoding statistic to be -5.5, which indicate that when considering high 278 frequency derived alleles, the coding sequence of this locus is evolving neutrally and that there are many 279 The presence of multiple low-frequency ben-1 alleles that confer resistance to ABZ treatment suggests 254 that selection might have acted on this locus within the C. elegans population. To determine if this 255 hypothesis is possible, we calculated the Ka/Ks ratio between ben-1 genes from C. elegans and from two 256 diverged nematode species C. remanei and C. briggsae [63]. The Ka/Ks ratio between the ben-1 coding 257 sequences of C. elegans and these two diverged species is ~0.008, which indicates that the evolution of 258 this locus is constrained across species. However, among the 249 wild isolates within the C. elegans 259 species [27], we identified 10 synonymous and 22 nonsynonymous, stop-gained, or splice-site variants in 260 the ben-1 locus when compared to the N2 reference genome [25]. We note that the synonymous variants 261 identified in ben-1 are primarily found in wild strains that contain putative loss-of-function alleles. These 262 results indicate that, despite the evolutionary constraint we observe at the ben-1 locus across nematode 263 species, an excess of potentially adaptive mutations (~2.2X nonsynonymous variants) have arisen within 264 the C. elegans species. These results are consistent with our estimates of Tajima’s D [64] at the ben-1 265 locus. When we considered all variant types across coding and non-coding regions of ben-1, we found 266 Tajima’s D to be -2.02. When we only consider putative loss-of-function variants in the ben-1 locus, the 267 Tajima’s D estimate is -2.59. This large negative value of Tajima’s D likely reflects the high number of rare 268 ben-1 alleles present in the C. elegans population. We next calculated Tajima’s D for the genomic region 269 surrounding ben-1 (Figure 3A;; Supplemental data 10). These results show a strong negative dip in 270 Tajima’s D (< -1.5) between 3.50 and 3.55 Mb, which is the region directly surrounding ben-1 (3.537 - 271 3.541 Mb). This observation indicates that the C. elegans population has undergone a population 272 expansion or that the region surrounding ben-1 may have been subject to selective pressures. Within-species selective pressures at the ben-1 locus 253 Within-species selective pressures at the ben-1 locus 253 The presence of multiple low-frequency ben-1 alleles that confer resistance to ABZ treatment suggests 254 that selection might have acted on this locus within the C. elegans population. To determine if this 255 hypothesis is possible, we calculated the Ka/Ks ratio between ben-1 genes from C. elegans and from two 256 diverged nematode species C. remanei and C. briggsae [63]. The Ka/Ks ratio between the ben-1 coding 257 sequences of C. elegans and these two diverged species is ~0.008, which indicates that the evolution of 258 this locus is constrained across species. However, among the 249 wild isolates within the C. elegans 259 species [27], we identified 10 synonymous and 22 nonsynonymous, stop-gained, or splice-site variants in 260 the ben-1 locus when compared to the N2 reference genome [25]. We note that the synonymous variants 261 identified in ben-1 are primarily found in wild strains that contain putative loss-of-function alleles. These 262 results indicate that, despite the evolutionary constraint we observe at the ben-1 locus across nematode 263 species, an excess of potentially adaptive mutations (~2.2X nonsynonymous variants) have arisen within 264 the C. elegans species. These results are consistent with our estimates of Tajima’s D [64] at the ben-1 265 locus. When we considered all variant types across coding and non-coding regions of ben-1, we found 266 Tajima’s D to be -2.02. When we only consider putative loss-of-function variants in the ben-1 locus, the 267 Tajima’s D estimate is -2.59. This large negative value of Tajima’s D likely reflects the high number of rare 268 ben-1 alleles present in the C. elegans population. We next calculated Tajima’s D for the genomic region 269 surrounding ben-1 (Figure 3A;; Supplemental data 10). These results show a strong negative dip in 270 Tajima’s D (< -1.5) between 3.50 and 3.55 Mb, which is the region directly surrounding ben-1 (3.537 - 271 3.541 Mb). This observation indicates that the C. elegans population has undergone a population 272 expansion or that the region surrounding ben-1 may have been subject to selective pressures. To 273 differentiate between these two possibilities, we calculated Fay and Wu’s H [65] and Zeng’s E [66] for ben- 274 1 (Figure 3A;; Supplemental data 11). We noticed that these two statistics showed peaks around the ben-1 275 locus, which do not correspond to our observations of multiple low minor allele frequency alleles. Within-species selective pressures at the ben-1 locus 253 To 273 differentiate between these two possibilities, we calculated Fay and Wu’s H [65] and Zeng’s E [66] for ben- 274 1 (Figure 3A;; Supplemental data 11). We noticed that these two statistics showed peaks around the ben-1 275 locus, which do not correspond to our observations of multiple low minor allele frequency alleles. 276 However, when we consider ben-1 coding variation and non-coding variation separately, we found the 277 Hcoding statistic to be 0.34 and the Hnoncoding statistic to be -5.5, which indicate that when considering high 278 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint high-frequency derived alleles in the intronic and UTR regions. By contrast, we found the Ecoding statistic, which considers low and high frequency alleles, to be -1.7 and the Enoncoding to be 3.4. Taken together, these results indicate that recent selective pressures have acted on the ben-1 locus. . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint high-frequency derived alleles in the intronic and UTR regions. By contrast, we found the Ecoding statistic, 280 which considers low and high frequency alleles, to be -1.7 and the Enoncoding to be 3.4. Taken together, 281 these results indicate that recent selective pressures have acted on the ben-1 locus. 282 If localized selective pressures, possibly caused by increased levels of environmental BZ, drove 284 the loss of ben-1 function in a subset of wild strains, then we could expect to see geographic clustering of 285 resistant strains. When we looked at the distribution of strains with putative ben-1 loss-of-function variants, 286 we observed no trend in the sampling locations of these individuals (Figure 3B). Within-species selective pressures at the ben-1 locus 253 However, we do note that 287 highly diverse strains sampled from various locations on the Pacific Rim [27,37,38] do not harbor any 288 putative ben-1 loss-of-function alleles (Figure 3B). This diverse set of strains is hypothesized to represent 289 the ancestral state of C. elegans, because they do not show signs of chromosome-scale selective sweeps 290 throughout their genomes [38]. The lack of ben-1 loss-of-function variants in these ancestral strains 291 suggests that variation in ben-1 arose after individuals in the species spread throughout the world. We 292 next considered that strains with putative ben-1 loss-of-function alleles might have been isolated from 293 similar local environments and substrates. Strains that are predicted to have a functional ben-1 gene have 294 been isolated from a greater diversity of environmental sampling locations (Supplemental Figure 4A) and 295 sampling substrates (Supplemental Figure 4B) than strains with predicted loss-of-function variants in ben- 296 1. However, a hypergeometric test for enrichment within specific locations and substrates showed no 297 signs of significant enrichment. This lack of enrichment might be caused by biases in global coverage of 298 C. elegans sampling and low sampling density. Nevertheless, the observation that the diversity of ben-1 299 alleles arose independently on various branches of the C. elegans phylogeny lends support to the 300 hypothesis that local selective pressures have acted on the ben-1 locus (Figure 3C). 301 ben-1 natural variants confer BZ resistance to sensitive C. elegans strains 303 12 of 52 The majority of the variants we observe at the ben-1 locus are predicted to result in the loss of gene 304 function. Our findings in C. elegans are in contrast to findings in parasitic nematode populations that 305 describe the F200Y and other missense variants as the major variants contributing to BZ resistance [13]. 306 To test whether the putative loss-of-function variants in ben-1 that we observe in the C. elegans population 307 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint confer the same level of BZ resistance as the known F200Y allele, we introduced a ben-1 deletion in the 308 N2 strain using CRISPR/Cas9 (referred to as Del). In addition to the Del strain, we introduced the F200Y 309 ben-1 allele into the N2 strain (referred to as F200Y) to test whether this allele confers BZ resistance in 310 C. elegans and to compare the levels of BZ resistance between the loss-of-function and missense alleles. 311 We exposed the N2 parental, Del, and F200Y strains to 12.5 µM ABZ using the HTA described above. 312 Both the Del and F200Y variants conferred ABZ resistance to the otherwise sensitive N2 parental strain 313 (Figure 4A;; Supplemental Figure 5;; Supplemental Data 12 and 13). Interestingly, we found no significant 314 difference in BZ resistance between the Del and F200Y strains (p-value = 0.99, TukeyHSD). These results 315 show that putative loss-of-function variants of ben-1 and known parasitic BZ resistance alleles confer BZ 316 resistance in otherwise isogenic C. elegans genetic backgrounds. 317 CC 0 te at o a ce se a To complement the results from the liquid-based HTA experiments, we performed a plate-based 319 competition assay. In the competition experiments, we individually competed the F200Y, Del, and parental 320 N2 strains against an N2 strain that contains a barcode sequence (PTM229). ben-1 natural variants confer BZ resistance to sensitive C. elegans strains 303 We quantified the relative 321 allele frequencies of the barcoded strain for the first, third, fifth, and seventh generations of the competition 322 assay (Figure 4B;; Supplemental Data 14 and 15). Throughout the competition assay, and for all strains 323 tested, the relative frequencies of the barcoded strain did not significantly deviate from the initial frequency 324 when grown on DMSO plates. These results suggest that in standard laboratory conditions, the BEN-1 325 F200Y and ben-1 deletion alleles do not have fitness consequences. We observed the same trend when 326 we competed the N2 and the barcoded N2 strains on ABZ plates. However, the allele frequencies of the 327 barcoded strain dropped to ~20% when competed against strains that contain either of the two ben-1 328 alleles on ABZ plates (relative fitness w = ~ 1.3, both). These two independent assays show that the two 329 ben-1 alleles confer BZ resistance with no negative fitness consequence under standard laboratory growth 330 conditions. 331 Discussion 351 ABZ is a broadly administered BZ used to treat parasitic nematode infections in humans and livestock 352 [4,67]. Already a significant problem in veterinary medicine, the heavy reliance of ABZ and other BZ 353 compounds in MDA programs, the small repertoire of other anthelmintic compounds, and high rates of re- 354 infection of parasitic nematodes in endemic regions around the world have raised the fear of the 355 emergence of BZ resistance among parasitic nematode populations that infect humans [4,68]. Though the 356 BZ-resistance alleles found in model organisms through classical genetic screens have had little direct 357 translatability to parasitic nematodes, these classic alleles have been instrumental toward the elucidation 358 of the mechanism of action of BZs [4,69]. Recent advances in sequencing technologies have enabled 359 researchers to take a quantitative genetics approach to search for novel mechanisms of anthelmintic 360 resistance in natural parasitic and non-parasitic nematodes [27,29,30,70–72]. In the present study, we 361 leveraged genetic diversity within the C. elegans population to study the genetic basis of ABZ resistance 362 within this species. 363 15 of 52 To identify the genetic basis of ABZ resistance in C. elegans, we used two unbiased genome-wide 365 association approaches, single-marker [35,36] and gene-burden [40] based mappings. We found that the 366 C. elegans population harbors extreme allelic heterogeneity at the ben-1 locus, and that this variation 367 contributes to differential ABZ resistance among wild isolates (Figure 1B, Figure 2A). The variants in the 368 ben-1 locus that contribute to ABZ resistance are present at low allele frequencies (<0.05 MAF) within the 369 global C. elegans population and arose independently during the evolutionary history of the species 370 (Figure 3). As a result of this complex demographic history, common variants (>0.05 MAF) used as 371 genetic markers in the mappings near the ben-1 locus are not in LD with the ben-1 alleles that confer ABZ 372 resistance. Therefore, we were unable to detect an association between the ben-1 locus and ABZ 373 response with the single-marker based mapping approach. We note that burden-based mapping 374 approaches are only possible in species with well annotated genomes and genome-wide genetic variation, 375 such as C. elegans and other model organisms. To extend this type of analysis to parasitic species, more 376 effort will need to be made toward improving parasitic nematode reference genomes and the accumulation 377 of genome-wide variation resources [73,74]. Additional genomic intervals contribute to ABZ resistance in the C. elegans population 333 As a result of this complex demographic history, common variants (>0.05 MAF) used 371 genetic markers in the mappings near the ben-1 locus are not in LD with the ben-1 alleles that confer A 372 resistance. Therefore, we were unable to detect an association between the ben-1 locus and A 373 response with the single-marker based mapping approach. We note that burden-based map 374 approaches are only possible in species with well annotated genomes and genome-wide genetic varia 375 such as C. elegans and other model organisms. To extend this type of analysis to parasitic species, m 376 effort will need to be made toward improving parasitic nematode reference genomes and the accumula 377 of genome-wide variation resources [73,74]. In addition to the strong association between ABZ respon 378 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Additional genomic intervals contribute to ABZ resistance in the C. elegans population 333 13 of 52 Above, we showed that extreme allelic heterogeneity at the ben-1 locus explains 73.8% of the phenotypic 334 variation in response to ABZ treatment. To identify potential genomic loci that contribute to the remaining 335 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 26.2% of the observed BZ response variation, we statistically corrected the animal length phenotype in 336 response to ABZ treatment by using the presence of a putative loss-of-function variant at the ben-1 locus 337 in a strain as a covariate for linear regression. After correcting for variation at the ben-1 locus, we 338 performed GWA mapping using both the single-marker (Figure 5A;; Supplemental Data 16, 17, and 18) 339 and gene-burden (Supplemental Figure 6;; Supplemental Data 19, 20, and 21) approaches. Both of these 340 approaches resulted in the disappearance of the two QTL on chromosome II and the QTL on chromosome 341 V (Figure 5A, Supplemental Figure 7). This result suggests that the cumulative variation at the ben-1 locus 342 is in complex interchromosomal linkage disequilibrium (LD) with these three loci (Supplemental Figure 7), 343 despite the loci on chromosomes II and V not being in strong LD with each other (Supplemental Figure 344 2B). Interestingly, using the single-marker GWA-mapping approach, we found an additional genomic locus 345 on chromosome X that was significantly associated with ABZ resistance (Figure 5A;; Supplemental Figure 346 8). However, the gene-burden based approach did not result in significant associations between ABZ 347 resistance and rare variation within any genes. These results suggest that ABZ resistance in the 348 C. elegans population is caused by at least two loci because additional genetic variation present in the 349 C. elegans population contributes to phenotypic variation in response to ABZ treatment. 350 14 of 52 . Additional genomic intervals contribute to ABZ resistance in the C. elegans population 333 CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Discussion 351 ABZ is a broadly administered BZ used to treat parasitic nematode infections in humans and lives 352 [4,67]. Already a significant problem in veterinary medicine, the heavy reliance of ABZ and other 353 compounds in MDA programs, the small repertoire of other anthelmintic compounds, and high rates o 354 infection of parasitic nematodes in endemic regions around the world have raised the fear of 355 emergence of BZ resistance among parasitic nematode populations that infect humans [4,68]. Though 356 BZ-resistance alleles found in model organisms through classical genetic screens have had little d 357 translatability to parasitic nematodes, these classic alleles have been instrumental toward the elucida 358 of the mechanism of action of BZs [4,69]. Recent advances in sequencing technologies have ena 359 researchers to take a quantitative genetics approach to search for novel mechanisms of anthelm 360 resistance in natural parasitic and non-parasitic nematodes [27,29,30,70–72]. In the present study 361 leveraged genetic diversity within the C. elegans population to study the genetic basis of ABZ resista 362 within this species. 363 364 To identify the genetic basis of ABZ resistance in C. elegans, we used two unbiased genome-w 365 association approaches, single-marker [35,36] and gene-burden [40] based mappings. We found that 366 C. elegans population harbors extreme allelic heterogeneity at the ben-1 locus, and that this varia 367 contributes to differential ABZ resistance among wild isolates (Figure 1B, Figure 2A). The variants in 368 ben-1 locus that contribute to ABZ resistance are present at low allele frequencies (<0.05 MAF) within 369 global C. elegans population and arose independently during the evolutionary history of the spe 370 (Figure 3). Discussion 351 In addition to the strong association between ABZ responses 378 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint and variants in the ben-1 locus, we identified a novel ABZ-response QTL on chromosome X after 379 accounting for variants in the ben-1 locus (Figure 5). The identification of this QTL suggests that 380 mechanisms independent of β-tubulin could account for BZ resistance in the C. elegans and in parasitic 381 nematode populations. 382 16 of 52 383 The majority (22/25) of ben-1 alleles that correlate with ABZ resistance are predicted to result in 384 the loss of ben-1 function (Figure 2). These results suggest that the function of the ben-1 gene is not 385 essential to the survival of C. elegans in natural habitats. This observation is in close agreement with 386 previous work using the laboratory-derived N2 strain, which showed that putative ben-1 loss-of-function 387 alleles do not confer observable defects in standard laboratory growth conditions [12]. However, our 388 analysis of Ka/Ks between the C. elegans ben-1 and the distantly related C. briggsae and C. remanei β- 389 tubulin coding sequences suggest that evolution of this gene is highly constrained. A possible explanation 390 for this discrepancy is that the ben-1 gene has a highly specialized function in natural settings and that 391 survival in the presence of a strong selective pressure, such as the presence of increased concentrations 392 of environmental BZs, outweighs the necessity of this specialization. A possible source of environmental 393 BZs are microbes that C. elegans might contact in their natural habitat. Discussion 351 For example, 5,6- 394 dimethylbenzimidazole is a BZ derivative that is produced naturally by prokaryotes [75]. However, whether 395 these BZ derivatives bind to and inhibit β-tubulins remains unclear. Alternatively, natural selection on 396 C. elegans ben-1 might occur by contamination of nematode niches by synthetic BZs produced by 397 humans. Initially developed as fungicides in the 1960s, several BZs, including thiabendazole, benomyl, 398 and carbendazim are extensively used to treat crops, fruits, and grains [76]. A third source of 399 environmental BZs might be runoff from livestock farms because they are used extensively to prevent 400 parasite infections in these animals [77–80]. The observation that putative ben-1 loss-of-function alleles 401 are only found at low frequencies in the C. elegans population might suggest that individuals with these 402 alleles are eventually removed from the population, though we see no evidence for a decrease in fitness in 403 laboratory competitions experiments, which might not recapitulate natural settings (Figure 4). Taken 404 together, our observations indicate that the independent putative ben-1 loss-of-function alleles might have 405 arose recently within the C. elegans population. Considering these findings in retrospect, it is extremely 406 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint fortunate that N2 was used for initial studies of BZ sensitivity. Rapid progress on the mechanism of action 407 for BZ compounds from the free-living C. elegans model to parasites might have not occurred otherwise, 408 strongly illustrating the pitfalls associated with the study of a single genetic background to elucidate 409 anthelmintic resistance. 410 fortunate that N2 was used for initial studies of BZ sensitivity. Rapid progress on the mechanism of action 407 for BZ compounds from the free-living C. elegans model to parasites might have not occurred otherwise, 408 strongly illustrating the pitfalls associated with the study of a single genetic background to elucidate 409 anthelmintic resistance. Discussion 351 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint genes that encode for β-tubulins in C. elegans are mec-7 and tbb-4. Both genes are only expressed in a 435 subset of neuronal cells involved in chemo- and mechanosensation, and both have a phenylalanine at 436 amino acid position 200 [88,90–92]. tbb-6 encodes for a highly diverged C. elegans β-tubulin 437 (Supplemental Figure 10) that could be a stress-resistant tubulin because it is highly expressed during the 438 unfolded protein response [93]. Despite having phenylalanine at amino acid position 200, these tubulins 439 (TBB-4, MEC-7, and TBB-6) are likely not involved in BZ response because we see no difference in ABZ- 440 resistance between control and ABZ conditions for BZ-resistant strains (Supplemental Data 12). Of the six 441 β-tubulin genes described above, we only observe variation in ben-1 and tbb-6, both of which have low- 442 frequency variants with high predicted functional effects. These observations are also shown by estimates 443 of Tajima’s D at these loci (Supplemental Figure 11). 444 18 of 52 The β-tubulin repertoire among parasitic nematode species is highly diverse likely because of 446 multiple gene duplication events [94]. For example, H. contortus has four known β-tubulin genes, of which 447 Hco-tbb-iso-1 and Hco-tbb-iso-2 are thought to be the targets of BZ, and Hco-tbb-iso-3 and Hco-tbb-iso-4 448 are closely related in sequence and expression pattern to C. elegans tbb-4 and mec-7 [94]. Despite the 449 presence of a phenylalanine at amino acid position 200 in all four of the β-tubulins, the focus of the 450 parasitology community is on missense variants present in Hco-tbb-iso-1. However, our data argue that it 451 is important to consider the high level of sequence similarity among the β-tubulin genes. We see that 452 nearby structural variants, like deletions, can alter sequence read alignments and misannotate 453 orthologous β-tubulin gene sequences to artifactually create these variant sites. Discussion 351 410 The remaining (3/25) ben-1 alleles we found to be correlated with ABZ resistance in the C. elegans 412 population result in missense variants. These missense variants are S145F, A185P, and M257I and are 413 found in one, two, and one C. elegans strain, respectively. Remarkably, substitutions of the amino acid 414 residues A185 (A185S) and M257 (M257L) were previously described to confer BZ resistances in Tapesia 415 yallundae and Aspergillus nidulans, respectively [81,82]. Of these two previously identified residues, M257 416 is postulated to directly interact with BZs, because it is in close three-dimensional proximity to the known 417 BZ interacting F200 residue [60]. The S145F allele is in close proximity to the highly conserved GGGTGS 418 motif of the GTP binding and hydrolysis site [83]. A fourth missense mutation Q131L, which is present in a 419 strain that was not phenotyped in our mapping study because it grows slowly under normal laboratory 420 conditions, is located near the β-tubulin/ɑ-tubulin interaction interface [84,85]. Upon re-phenotyping this 421 isotype with the Q131L variant, we found that it is resistant to ABZ treatment (Supplemental Figure 9). A 422 final missense variant (D404N) present in the C. elegans population was not correlated with ABZ 423 resistance (Figure 2). Though theoretical structure-based evidence suggests that most of the missense 424 variants in the C. elegans population associated with ABZ resistance cause a non-functional β-tubulin, 425 further experiments are necessary to confirm these results. 426 17 of 52 The high prevalence of putative ben-1 loss-of-function variants in the natural C. elegans population 428 stands in stark contrast to the relative paucity of allelic diversity found in orthologous β-tubulin genes in 429 parasitic nematodes. Perhaps parasitic nematode species do not have similarly high levels of functional 430 redundancy of β-tubulins as has been observed in C. elegans. The main C. elegans β-tubulins are tbb-1 431 and tbb-2, which are both ubiquitously highly expressed, functionally redundant in laboratory conditions, 432 and contain a tyrosine at position 200 [86–88]. The co-expression of tbb-1, tbb-2, and ben-1 in the nervous 433 system [89] suggests that only one β-tubulin with F200 is required to confer sensitivity to BZs. Two other 434 17 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. Discussion 351 Careful read alignments 454 and high read depth will be necessary to interpret variation across highly divergent parasites with 455 orthologous β-tubulin genes. Additionally, similar to our observations of ben-1 in C. elegans, deletion 456 alleles of H. contortus Hco-tbb-iso-2 have been observed in few field studies, but their importance for BZ 457 resistance remains unclear [95,96]. This observation led Kwa and colleagues [95,97] to hypothesize that 458 BZ resistance requires two steps. First, mutation of F200 to tyrosine in one β-tubulin isoform, and second, 459 deletion of the second β-tubulin that contains F200. This hypothesis aligns well with our observation that 460 the single F200Y change in C. elegans BEN-1 confers BZ resistance in the presence of TBB-1 (Y200) and 461 TBB-2 (Y200). An additional layer of complexity comes from recent evidence in Trichuris trichiura that non- 462 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint BEN-1 β-tubulins might also play a role in BZ resistance in diverse parasitic species [98]. Our results 463 showing that BZ resistance is a complex trait within the C. elegans population (Figure 5) and others’ 464 similar observations within parasitic nematode populations necessitates further investigation into the 465 genetic and molecular mechanisms that underlie this trait. 466 467 19 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. Discussion 351 ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Animals were cultured at 20°C on modified nematode growth medium (NGM), containing 1% agar and 473 0.7% agarose [31] and seeded with OP50 E. coli. Prior to each assay, strains were passaged for at least 474 four generations without entering starvation or encountering dauer-inducing conditions [31]. For the 475 genome-wide association (GWA) studies, 249 wild isolates from CeNDR (version 20170531) were used as 476 described previously [27,32]. Construction of ben-1 allele-replacement and ben-1 deletion strains are 477 described in the corresponding section. All strain information can be found in Supplemental table 1. 478 479 High-throughput fitness assay 480 The high-throughput fitness assays (HTA) were performed as described previously [32] with the exception 481 that each strain was assayed in four technical replicates that consisted of four independent bleach 482 synchronization steps across two days with two independent drug and control preparations. In short, 483 strains were propagated for four generations on agar plates, followed by bleach synchronization. The 484 embryos were titered to 96-well microtiter plates at a final concentration of approximately one embryo per 485 microliter of K medium [33] with modified salt concentrations (10.2 mM NaCl, 32 mM KCl, 3 mM CaCl2, 3 486 mM MgSO4). After overnight incubation, hatched L1 larvae were fed with 5 mg/mL HB101 bacterial lysate 487 (Pennsylvania State University Shared Fermentation Facility, State College, PA) and cultured for two days 488 until they reached the L4 larval stage. Using the large particle flow cytometer COPAS BIOSORT (Union 489 Biometrica, Holliston MA), three L4 larvae per well were sorted into new microtiter plates containing 490 modified K medium, 10 mg/mL HB101 lysate, 50 μM kanamycin, and either 12.5 µM albendazole (ABZ) 491 dissolved in 1% DMSO or 1% DMSO alone as a control. During the following four-day incubation, animals 492 were allowed to mature and to produce offspring. Fitness parameters, including traits for animal length and 493 brood size, were measured for each population under drug and control conditions using the COPAS 494 BIOSORT platform after 96 hours. To facilitate body straightening for more accurate length 495 determinations, animals were treated with sodium azide (50 mM) immediately before measurement. 496 497 20 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 498 Dose-response experiments 526 Albendazole (Sigma Aldrich Product #A4673) (ABZ) dose-response experiments were performed on a set 527 of four genetically divergent C. elegans strains, including the laboratory strain N2 and three different wild 528 isolates (CB4856, JU775, DL238), to determine suitable drug concentrations for subsequent GWA 529 experiments. To this end, strain-specific drug responses were measured in four technical replicates using 530 the HTA as described above at concentrations of 3.125 μM, 6.25 μM, 12.5 μM, and 25 μM ABZ. A suitable 531 drug concentration for subsequent GWA experiments was selected based on the lowest concentration in 532 which we observed a significant difference in ABZ response among strains for brood size and animal 533 length (Supplemental data 1 and 2). 534 Processing of fitness traits for genetic mapping 499 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Processing of fitness traits for genetic mapping 499 Fitness parameters measured by the COPAS BIOSORT platform were processed using the R package 500 easysorter [34] with some modifications to incorporate technical replicates into the analysis. The 501 easysorter package is specifically developed for this type of data and includes reading and modifying of 502 the raw data, pruning of anomalous data points, and regression of control and experimental phenotypes. 503 Measured parameters include time-of-flight (animal length), extinction (optical density), and total object 504 count (brood size) for each well. In short, the function read_data reads in raw phenotype data and runs a 505 support vector machine to identify and eliminate air bubbles, which can be confused with nematodes. In 506 the next step, the function remove_contamination removes data obtained from microtiter wells that were 507 manually identified to contain bacterial or fungal contamination. To generate summarized statistics for 508 each well, the function sum_plate calculates the 10th, 25th, 50th, 75th, and 90th quantiles for all fitness 509 parameters obtained. In this process, brood size is normalized by the number of animals sorted originally 510 in each well. Next, biologically impossible data points passing certain cut-offs (n > 1000, n < 5, norm.n > 511 350) are eliminated (function bio_prune). We removed outlier replicates if they were outside 1.8 times the 512 standard deviation of that strain’s median phenotype of a particular trait. To adjust differences among 513 replicates of each assay, a linear model was applied using the formula (phenotype ~ experiment + assay), 514 which replaced the easysorter function regress (assay = TRUE). The experiment component of the linear 515 model corresponds to two independent drug preparations of the same strains and the assay component 516 corresponds to blocks of different strains performed across multiple weeks. In addition, the bamf_prune 517 function of the easysorter package was skipped, because previous outlier removal based on technical 518 replicates made this step unnecessary. Finally, drug-specific phenotype data is calculated using the 519 regress (assay = FALSE) function of the easysorter package. This function fits a linear model with the 520 formula (phenotype ~ control phenotype) to account for any differences in population parameters present 521 in the control DMSO-only conditions. 522 21 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. Burden Testing 555 Burden test analyses were performed using RVtests [39] and the variable-threshold method [40]. We 556 called SNV using bcftools [41] with settings previously described [27,37,39]. We next performed 557 imputation using BEAGLE v4.1 [42] with window set to 8000, overlap set to 3000, and ne set to 17500. 558 Within RVtests, we set the minor allele frequency range from 0.003 to 0.05 for burden testing. 559 560 Genome-wide association (GWA) mappings 536 22 of 52 The GWA mappings were performed on the processed ABZ HTA phenotype data of 209 C. elegans wild 537 isolates. In short, the easysorter processed phenotype data was analysed using the cegwas R package for 538 association mapping [27]. This package uses the EMMA algorithm for performing association mapping 539 and correcting for population structure [35], which is implemented by the GWAS function in the rrBLUP 540 package [36]. In detail, the GWAS function was used with the following command: rrBLUP::GWAS 541 (min.MAF = 0.05, P3D = FALSE). The kinship matrix used for association mapping was generated using 542 whole-genome high-quality single-nucleotide variants (SNVs) [37] and the A.mat function from the rrBLUP 543 package. All SNVs included in the marker set for GWA mapping had a minimum 5% minor allele 544 frequency in the 240 isotype set [38]. Quantitative trait loci (QTL) were defined by at least one SNV that 545 passed the Bonferroni-corrected threshold and were processed further using fine mapping, as described 546 previously [32]. Computational fine mapping of the genomic regions of interest was performed as 547 described previously [32]. We used the following linear model to correct for the presence of a putative ben- 548 1 LoF variant: 𝑙𝑚(𝑎𝑛𝑖𝑚𝑎𝑙 𝑙𝑒𝑛𝑔𝑡ℎ ̴ (𝑏𝑒𝑛−1 𝐿𝑜𝐹)). The list of strains that were considered to have putative 549 LoF variants and the ben-1-corrected phenotype data are presented (Supplemental data 16). We did not 550 include CB4856 as a strain with a putative LoF variant because it has a 9 bp in-frame deletion near the 551 end of the gene. We performed single-marker mappings as described above and gene-burden mappings 552 as described below using these regressed phenotypes (Supplemental data 17-21). 553 22 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 554 Generation of ben-1 allele replacement and deletion strains 561 All ben-1 edited strains assayed in this study were generated in the N2 background by CRISPR/Cas9- 562 mediated genome editing, using a co-CRISPR approach [43] and Cas9 ribonucleoprotein (RNP) delivery 563 [44]. For the ben-1 F200Y allele replacement strains, sgRNAs for ben-1 and dpy-10 were ordered from 564 Synthego (Redwood City, CA) and injected at final concentrations of 5 μM and 1 μM, respectively. Single- 565 stranded oligodeoxynucleotides (ssODN) templates for homology-directed repair (HDR) of ben-1 and dpy- 566 10 (IDT, Skokie IL) were used at final concentrations of 6 μM and 0.5 μM, respectively. Cas9 protein (IDT, 567 Product #1074182) was added to the injection mixture at a concentration of 5 μM and incubated with all 568 other components for ten minutes at room temperature prior to injection. All concentrations used for the 569 sgRNA-mediated allele replacement were adapted from the work of Prior and colleagues [45]. N2 ben-1 570 deletion strains were generated using two ben-1 specific crRNAs synthesized by IDT (Skokie, IL), which 571 targeted exon 2 and exon 4. For the injection mixture, ben-1 crRNAs were used at final concentration of 572 8.3 μM each, mixed with dpy-10 crRNA and tracrRNA (IDT, Product #1072532) at final concentrations of 573 1.2 μM and 17.6 μM, respectively, and incubated at 95°C for five minutes. After cooling to room 574 temperature, Cas9 protein was added at a final concentration of 15.25 μM (IDT Product #1074181) and 575 incubated for five minutes before dpy-10 ssODN was added to a concentration of 5 μM. 576 23 of 52 RNP injection mixtures were microinjected into the germline of young adult hermaphrodites (P0) 578 and injected animals were singled to fresh 6 cm NGM plates 18 hours after injection. Two days later, F1 579 progeny were screened, and animals expressing a Rol phenotype were transferred to new plates and 580 allowed to generate progeny (F2). Afterwards, F1 animals were genotyped by PCR. For the ben-1(F200Y) 581 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. Generation of ben-1 allele replacement and deletion strains 561 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint allele replacement, ssODN repair templates contained the desired edit and a conservative change of the 582 PAM site to prevent sgRNA:Cas9 cleavage. In addition, the PAM change introduced a new restriction site. 583 PCRs were performed using the primers oECA1297 and oECA1298, and PCR products were incubated 584 with BTsCI restriction enzyme (R0647S, New England Biolabs, Ipswich, MA) to identify successfully edited 585 animals by differential band patterns. For genotyping of ben-1 deletion strains, the primers oECA1301 and 586 oECA1302 were used and successful deletions were identified by shorter PCR products. Non-Rol progeny 587 (F2) of F1 animals positive for the desired edits were propagated on separate plates to generate 588 homozygous progeny. F2 animals were genotyped afterwards, and PCR products were Sanger 589 sequenced for verification. Generated ben-1(F200Y) replacement strains and ben-1 deletion strains were 590 phenotyped for ABZ response using the HTA as described above. All oligonucleotide sequences are listed 591 in the supplement (Supplemental table 2). 592 . CC BY 4.0 International license a Competition assays 594 CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint and seventh weeks into a one-locus genic selection model [46]. All oligonucleotide sequences are listed in 610 the supplement (Supplemental table 2). 611 and seventh weeks into a one-locus genic selection model [46]. All oligonucleotide sequences are listed in 610 the supplement (Supplemental table 2). 611 Statistical analyses 619 Phenotype data are shown as Tukey box plots and p-values were used to assess significant differences of 620 strain phenotypes in allele-replacement experiments. All calculations were performed in R using the 621 TukeyHSD function on an ANOVA model with the formula (phenotype ~ strain). P-values less than 0.05 622 after Bonferroni correction for multiple testing were considered to be significant. 623 Competition assays 594 24 of 52 Pairwise multi-generation competition assays were performed between a ben-1 wild-type strain PTM229 595 and the ben-1 edited strains, containing either the F200Y allele replacement or a ben-1 deletion. All strains 596 were generated in the N2 background. For the assay, strains were bleach-synchronized and embryos 597 were transferred to 10 cm NGM plates. 48 hours later, seven L4 larvae per strain were transferred to 50 598 fresh 6 cm NGM plates containing either 1.25 µM ABZ for drug selection or DMSO as control. The ABZ 599 concentration was determined by dose-response assays beforehand as the lowest concentration that 600 caused a developmental delay in PTM229 and N2 when added to NGM. For each strain combination, 50 601 plates were grown, representing 10 technical replicates of five independent populations. Plates were 602 grown for one week until starvation. Animals were transferred to fresh plates by cutting out a 0.5 cm³ agar 603 chunk. After every culture transfer, starved animals were washed off the plates with M9, and DNA was 604 collected using the Qiagen DNeasy Kit (Catalog #69506). Allele frequencies of PTM229 compared to wild- 605 type N2 or the ben-1 edit strains in each replicate populations were measured using Taqman analysis in a 606 Bio-Rad QX200 digital droplet PCR system. Digital PCR was performed following the standard protocol 607 provided by Bio-Rad with the absolute quantification method. To calculate the relative fitness w of the 608 competitive strains, we used linear regression to fit the relative allele frequencies at the first, third, fifth, 609 24 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint and seventh weeks into a one-locus genic selection model [46]. All oligonucleotide sequences are listed in the supplement (Supplemental table 2). . Computational modelling of ben-1 variants 613 We obtained the predicted BEN-1 peptide sequence from WormBase [25]. We used the online utility 614 PHYRE2 to generate a homology model of the predicted BEN-1 peptide [47]. We specified the intensive 615 homology search option for running PHYRE2. Visualization of the BEN-1 homology model and highlighting 616 of the variants of interest was performed using PyMol [48]. 617 Population Genetics 625 Sliding window analysis of population genetic statistics was performed using the PopGenome package in 626 R [49,50]. All sliding window analyses were performed using the imputed SNV VCF available on the 627 CeNDR website with the most diverged isotype XZ1516 set to the outgroup [27,42,51]. Window size was 628 set to 100 SNVs with a slide distance of one SNV. The Tajima’s D calculation, using all of the manually 629 curated variants in the ben-1 locus, was performed using modified code from the developers of the LDhat 630 package [52]. Estimates of Ka/Ks were performed using an online service 631 (http://services.cbu.uib.no/tools/kaks). Multiple sequence alignments were performed using MUSCLE [53] 632 through the online service (https://www.ebi.ac.uk/Tools/msa/muscle/). The β-tubulin neighbor-joining 633 phylogeny construction was performed using Jalview 2 [54]. Linkage disequilibrium (LD) of QTL markers 634 was calculated using the genetics [55] package in R. LD calculations are reported as 𝑟 = −𝐷 / 𝑠𝑞𝑟𝑡( 𝑝(𝐴) ∗ 635 𝑝(𝑎) ∗ 𝑝(𝐵) ∗ 𝑝(𝑏) )where 𝐷 = 𝑝(𝐴𝐵) − 𝑝(𝐴) ∗𝑝(𝐵).r = -D / sqrt( p(A) * p(a) * p(B) * p(b) ) 636 638 25 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint References 648 648 1. Hotez PJ, Alvarado M, Basáñez M-G, Bolliger I, Bourne R, Boussinesq M, et al. The global burden of 649 disease study 2010: interpretation and implications for the neglected tropical diseases. PLoS Negl 650 Trop Dis. 2014;;8: e2865. 651 2. Lustigman S, Prichard RK, Gazzinelli A, Grant WN, Boatin BA, McCarthy JS, et al. A research agenda 652 for helminth diseases of humans: the problem of helminthiases. PLoS Negl Trop Dis. 2012;;6: e1582. 653 3. Charlier J, van der Voort M, Kenyon F, Skuce P, Vercruysse J. Chasing helminths and their economic 654 impact on farmed ruminants. Trends Parasitol. 2014;;30: 361–367. 655 4. Kotze AC, Hunt PW, Skuce P, von Samson-Himmelstjerna G, Martin RJ, Sager H, et al. Recent 656 advances in candidate-gene and whole-genome approaches to the discovery of anthelmintic 657 resistance markers and the description of drug/receptor interactions. Int J Parasitol Drugs Drug 658 Resist. 2014;;4: 164–184. 659 5. De Clercq D, Sacko M, Behnke J, Gilbert F, Dorny P, Vercruysse J. Failure of mebendazole in 660 treatment of human hookworm infections in the southern region of Mali. Am J Trop Med Hyg. 661 1997;;57: 25–30. 662 6. Albonico M, Bickle Q, Ramsan M, Montresor A, Savioli L, Taylor M. Efficacy of mebendazole and 663 levamisole alone or in combination against intestinal nematode infections after repeated targeted 664 mebendazole treatment in Zanzibar. Bull World Health Organ. 2003;;81: 343–352. 665 7. Humphries D, Simms BT, Davey D, Otchere J, Quagraine J, Terryah S, et al. Hookworm infection 666 among school age children in Kintampo north municipality, Ghana: nutritional risk factors and 667 response to albendazole treatment. Am J Trop Med Hyg. 2013;;89: 540–548. 668 8. Neff NF, Thomas JH, Grisafi P, Botstein D. Isolation of the beta-tubulin gene from yeast and 669 demonstration of its essential function in vivo. Cell. 1983;;33: 211–219. 670 9. Laclette JP, Guerra G, Zetina C. Inhibition of tubulin polymerization by mebendazole. Biochem 671 Biophys Res Commun. 1980;;92: 417–423. 672 10. Ireland CM, Gull K, Gutteridge WE, Pogson CI. The interaction of benzimidazole carbamates with 673 mammalian microtobule protein. Biochem Pharmacol. 1979;;28: 2680–2682. 674 11. Thomas JH, Neff NF, Botstein D. Isolation and characterization of mutations in the beta-tubulin gene 675 of Saccharomyces cerevisiae. Genetics. 1985;;111: 715–734. 676 12. Driscoll M, Dean E, Reilly E, Bergholz E, Chalfie M. Genetic and molecular analysis of a 677 Caenorhabditis elegans beta-tubulin that conveys benzimidazole sensitivity. Acknowledgements 639 We thank Shannon C. Brady, Kyle Siegel, and members of the Andersen lab for editing the manuscript for 640 flow and content. We thank members of the Andersen lab for making reagents used in the experiments 641 presented in the manuscript. 642 We thank Shannon C. Brady, Kyle Siegel, and members of the Andersen lab for editing the manuscript fo 640 flow and content. We thank members of the Andersen lab for making reagents used in the experiments 641 presented in the manuscript. 642 26 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint References 648 J Cell Biol. 1989;;109: 678 2993–3003. 679 13. Kwa MS, Veenstra JG, Roos MH. Benzimidazole resistance in Haemonchus contortus is correlated 680 with a conserved mutation at amino acid 200 in beta-tubulin isotype 1. Mol Biochem Parasitol. 681 1994;;63: 299–303. 682 14. Silvestre A, Cabaret J. Mutation in position 167 of isotype 1 beta-tubulin gene of Trichostrongylid 683 nematodes: role in benzimidazole resistance? Mol Biochem Parasitol. 2002;;120: 297–300. 684 15. Ghisi M, Kaminsky R, Mäser P. Phenotyping and genotyping of Haemonchus contortus isolates 685 reveals a new putative candidate mutation for benzimidazole resistance in nematodes. Vet Parasitol. 686 2007;;144: 313–320. 687 27 f 52 16. Lacey E, Gill JH. Biochemistry of benzimidazole resistance. Acta Trop. 1994;;56: 245–262. 688 27 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 17. Lubega GW, Prichard RK. Specific interaction of benzimidazole anthelmintics with tubulin: high-affinity 689 binding and benzimidazole resistance in Haemonchus contortus. Mol Biochem Parasitol. 1990;;38: 690 221–232. 691 18. Krücken J, Fraundorfer K, Mugisha JC, Ramünke S, Sifft KC, Geus D, et al. Reduced efficacy of 692 albendazole against Ascaris lumbricoides in Rwandan schoolchildren. Int J Parasitol Drugs Drug 693 Resist. 2017;;7: 262–271. 694 19. Yilmaz E, Ramünke S, Demeler J, Krücken J. Comparison of constitutive and thiabendazole-induced 695 expression of five cytochrome P450 genes in fourth-stage larvae of Haemonchus contortus isolates 696 with different drug susceptibility identifies one gene with high constitutive expression in a multi- 697 resistant isolate. Int J Parasitol Drugs Drug Resist. 2017;;7: 362–369. 698 20. Jones LM, Flemming AJ, Urwin PE. NHR-176 regulates cyp-35d1 to control hydroxylation-dependent 699 metabolism of thiabendazole in Caenorhabditis elegans. Biochem J. 2015;;466: 37–44. 700 21. Doyle SR, Bourguinat C, Nana-Djeunga HC, Kengne-Ouafo JA, Pion SDS, Bopda J, et al. 32. Zdraljevic S, Strand C, Seidel HS, Cook DE, Doench JG, Andersen EC. Natural variation in a single 726 amino acid substitution underlies physiological responses to topoisomerase II poisons. PLoS Genet. 727 2017;;13: e1006891. 728 33. Boyd WA, Smith MV, Freedman JH. Caenorhabditis elegans as a model in developmental toxicology. 729 References 648 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Methods Mol Biol. 2012;;889: 15–24. 730 730 34. Shimko TC, Andersen EC. COPASutils: An R Package for Reading, Processing, and Visualizing Data 731 from COPAS Large-Particle Flow Cytometers. PLoS One. 2014;;9: e111090. 732 35. Kang HM, Zaitlen NA, Wade CM, Kirby A, Heckerman D, Daly MJ, et al. Efficient control of population 733 structure in model organism association mapping. Genetics. 2008;;178: 1709–1723. 734 36. Endelman JB. Ridge Regression and Other Kernels for Genomic Selection with R Package rrBLUP. 735 The Plant Genome Journal. 2011;;4: 250. 736 37. Cook DE, Zdraljevic S, Tanny RE, Seo B, Riccardi DD, Noble LM, et al. The Genetic Basis of Natural 737 Variation in Caenorhabditis elegans Telomere Length. Genetics. 2016;;204: 371–383. 738 38. Andersen EC, Gerke JP, Shapiro JA, Crissman JR, Ghosh R, Bloom JS, et al. Chromosome-scale 739 selective sweeps shape Caenorhabditis elegans genomic diversity. Nat Genet. 2012;;44: 285–290. 740 39. Zhan X, Hu Y, Li B, Abecasis GR, Liu DJ. RVTESTS: an efficient and comprehensive tool for rare 741 variant association analysis using sequence data. Bioinformatics. 2016;;32: 1423–1426. 742 40. Price AL, Kryukov GV, de Bakker PIW, Purcell SM, Staples J, Wei L-J, et al. Pooled association tests 743 for rare variants in exon-resequencing studies. Am J Hum Genet. 2010;;86: 832–838. 744 41. Li H. A statistical framework for SNP calling, mutation discovery, association mapping and population 745 genetical parameter estimation from sequencing data. Bioinformatics. 2011;;27: 2987–2993. 746 42. Browning BL, Browning SR. Genotype Imputation with Millions of Reference Samples. Am J Hum 747 Genet. 2016;;98: 116–126. 748 43. Kim H, Ishidate T, Ghanta KS, Seth M, Conte D Jr, Shirayama M, et al. A co-CRISPR strategy for 749 efficient genome editing in Caenorhabditis elegans. Genetics. 2014;;197: 1069–1080. 750 44. Paix A, Folkmann A, Rasoloson D, Seydoux G. High Efficiency, Homology-Directed Genome Editing 751 in Caenorhabditis elegans Using CRISPR-Cas9 Ribonucleoprotein Complexes. Genetics. 2015;;201: 752 47–54. 753 45. Prior H, Jawad AK, MacConnachie L, Beg AA. Highly Efficient, Rapid and Co-CRISPR Independent 754 Genome Editing in Caenorhabditis elegans. G3: Genes\|Genomes\|Genetics. 2017;; g3.300216.2017. 755 46. Zhao Y, Long L, Xu W, Campbell RF, Large EL, Greene JS, et al. Laboratory evolution from social to 756 solitary behavior in the N2 reference strain is unnecessary for its fitness advantages [Internet]. References 648 Genome- 701 wide analysis of ivermectin response by Onchocerca volvulus reveals that genetic drift and soft 702 selective sweeps contribute to loss of drug sensitivity. PLoS Negl Trop Dis. 2017;;11: e0005816. 703 22. Holden-Dye L, Walker RJ. Anthelmintic drugs and nematicides: studies in Caenorhabditis elegans. 704 WormBook. 2014;; 1–29. 705 23. Hunt PW, Kotze AC, Knox MR, Anderson LJ, McNally J, LE Jambre LF. The use of DNA markers to 706 map anthelmintic resistance loci in an intraspecific cross of Haemonchus contortus. Parasitology. 707 2010;;137: 705–717. 708 24. Stear MJ, Boag B, Cattadori I, Murphy L. Genetic variation in resistance to mixed, predominantly 709 Teladorsagia circumcincta nematode infections of sheep: from heritabilities to gene identification. 710 Parasite Immunol. 2009;;31: 274–282. 711 25. Stein L, Sternberg P, Durbin R, Thierry-Mieg J, Spieth J. WormBase: network access to the genome 712 and biology of Caenorhabditis elegans. Nucleic Acids Res. 2001;;29: 82–86. 713 26. C. elegans Sequencing Consortium. Genome sequence of the nematode C. elegans: a platform for 714 investigating biology. Science. 1998;;282: 2012–2018. 715 27. Cook DE, Zdraljevic S, Roberts JP, Andersen EC. CeNDR, the Caenorhabditis elegans natural 716 diversity resource. Nucleic Acids Res. 2016;;45: D650–D657. 717 28. Strange K. Drug Discovery in Fish, Flies, and Worms. ILAR J. 2016;;57: 133–143. 718 29. Ghosh R, Andersen EC, Shapiro JA, Gerke JP, Kruglyak L. Natural variation in a chloride channel 719 subunit confers avermectin resistance in C. elegans. Science. 2012;;335: 574–578. 720 30. Zamanian M, Cook DE, Zdraljevic S, Brady SC, Lee D, Lee J, et al. Discovery of unique loci that 721 underlie nematode responses to benzimidazoles [Internet]. 2017. doi:10.1101/116970 722 31. Andersen EC, Bloom JS, Gerke JP, Kruglyak L. A variant in the neuropeptide receptor npr-1 is a 723 major determinant of Caenorhabditis elegans growth and physiology. PLoS Genet. 2014;;10: 724 e1004156. 725 32. Zdraljevic S, Strand C, Seidel HS, Cook DE, Doench JG, Andersen EC. Natural variation in a single 726 amino acid substitution underlies physiological responses to topoisomerase II poisons. PLoS Genet 727 2017;;13: e1006891. 728 33. Boyd WA, Smith MV, Freedman JH. Caenorhabditis elegans as a model in developmental toxicology. 729 33. Boyd WA, Smith MV, Freedman JH. Caenorhabditis elegans as a model in developmental toxicology 729 28 of 52 28 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. References 648 757 bioRxiv. 2018. p. 309997. doi:10.1101/309997 758 47. Kelley LA, Mezulis S, Yates CM, Wass MN, Sternberg MJE. The Phyre2 web portal for protein 759 modeling, prediction and analysis. Nat Protoc. 2015;;10: 845–858. 760 48. De Lano WL, California U. PyMOL: An Open-Source Molecular Graphics Tool. Available: 761 http://www.ccp4.ac.uk/newsletters/newsletter40/11_pymol.pdf 762 49. Pfeifer B, Wittelsbürger U, Ramos-Onsins SE, Lercher MJ. PopGenome: An Efficient Swiss Army 763 Knife for Population Genomic Analyses in R. Mol Biol Evol. Oxford University Press;; 2014;;31: 1929– 764 1936. 765 50. Team RC. R: A language and environment for statistical computing. R Foundation for Statistical 766 Computing, Vienna, Austria. 2014. 2016. 767 29 of 52 51. Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. The variant call format and 768 VCFtools. Bioinformatics. 2011;;27: 2156–2158. 769 29 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 52. McVean G, Awadalla P, Fearnhead P. A Coalescent-Based Method for Detecting and Estimating Recombination From Gene Sequences. Genetics. Genetics;; 2002;;160: 1231–1241. 53. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;;32: 1792–1797. 54. Waterhouse AM, Procter JB, Martin DMA, Clamp M, Barton GJ. Jalview Version 2--a multiple sequence alignment editor and analysis workbench. Bioinformatics. 2009;;25: 1189–1191. 55. Warnes G, Gorjanc G, Leisch F, Man M. Genetics: Population genetics. R package version 1.3. 6. 2012. 56. Kishore SP, Blank E, Heller DJ, Patel A, Peters A, Price M, et al. Modernizing the World Health 778 Organization List of Essential Medicines for Preventing and Controlling Cardiovascular Diseases. J 779 Am Coll Cardiol. 2018;;71: 564–574. 780 57. Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, et al. A program for annotating and 781 predicting the effects of single nucleotide polymorphisms, SnpEff: SNPs in the genome of Drosophila 782 melanogaster strain w1118;; iso-2;; iso-3. Fly . 2012;;6: 80–92. 783 58. Aguayo-Ortiz R, Méndez-Lucio O, Medina-Franco JL, Castillo R, Yépez-Mulia L, Hernández-Luis F, et 784 al. Towards the identification of the binding site of benzimidazoles to β-tubulin of Trichinella spiralis: 785 insights from computational and experimental data. J Mol Graph Model. 2013;;41: 12–19. 786 59. 68. Clarke NE, Doi SAR, Wangdi K, Chen Y, Clements ACA, Nery SV. Efficacy of anthelminthic drugs 807 and drug combinations against soil-transmitted helminths: a systematic review and network meta- 808 analysis. Clin Infect Dis. 2018;; doi:10.1093/cid/ciy423 809 770 771 References 648 Aguayo-Ortiz R, Méndez-Lucio O, Romo-Mancillas A, Castillo R, Yépez-Mulia L, Medina-Franco JL, 787 et al. Molecular basis for benzimidazole resistance from a novel β-tubulin binding site model. J Mol 788 Graph Model. 2013;;45: 26–37. 789 60. Robinson MW, McFerran N, Trudgett A, Hoey L, Fairweather I. A possible model of benzimidazole 790 binding to β-tubulin disclosed by invoking an inter-domain movement. J Mol Graph Model. 2004;;23: 791 275–284. 792 61. Hubley R, Finn RD, Clements J, Eddy SR, Jones TA, Bao W, et al. The Dfam database of repetitive 793 DNA families. Nucleic Acids Res. 2016;;44: D81–9. 794 62. Oosumi T, Garlick B, Belknap WR. Identification of putative nonautonomous transposable elements 795 associated with several transposon families in Caenorhabditis elegans. J Mol Evol. 1996;;43: 11–18. 796 63. Li WH, Wu CI, Luo CC. A new method for estimating synonymous and nonsynonymous rates of 797 nucleotide substitution considering the relative likelihood of nucleotide and codon changes. Mol Biol 798 Evol. 1985;;2: 150–174. 799 64. Tajima F. Statistical method for testing the neutral mutation hypothesis by DNA polymorphism. 800 Genetics. 1989;;123: 585–595. 801 65. Fay JC, Wu CI. Hitchhiking under positive Darwinian selection. Genetics. 2000;;155: 1405–1413. 802 65. Fay JC, Wu CI. Hitchhiking under positive Darwinian selection. Genetics. 2000 66. Zeng K, Fu Y-X, Shi S, Wu C-I. Statistical tests for detecting positive selection by utilizing high- 803 frequency variants. Genetics. 2006;;174: 1431–1439. 804 67. Furtado LFV, de Paiva Bello ACP, Rabelo ÉML. Benzimidazole resistance in helminths: From 805 problem to diagnosis. Acta Trop. 2016;;162: 95–102. 806 67. Furtado LFV, de Paiva Bello ACP, Rabelo ÉML. Benzimidazole resistance in helminths: From 805 problem to diagnosis. Acta Trop. 2016;;162: 95–102. 806 68. Clarke NE, Doi SAR, Wangdi K, Chen Y, Clements ACA, Nery SV. Efficacy of anthelminthic drugs 807 and drug combinations against soil-transmitted helminths: a systematic review and network meta- 808 analysis. Clin Infect Dis. 2018;; doi:10.1093/cid/ciy423 809 30 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 69. Gilleard JS. Understanding anthelmintic resistance: the need for genomics and genetics. Int J 810 Parasitol. 2006;;36: 1227–1239. 811 70. References 648 Large EE, Xu W, Zhao Y, Brady SC, Long L, Butcher RA, et al. Selection on a Subunit of the NURF 812 Chromatin Remodeler Modifies Life History Traits in a Domesticated Strain of Caenorhabditis 813 elegans. PLoS Genet. Public Library of Science;; 2016;;12: e1006219. 814 71. Redman E, Sargison N, Whitelaw F, Jackson F, Morrison A, Bartley DJ, et al. Introgression of 815 ivermectin resistance genes into a susceptible Haemonchus contortus strain by multiple 816 backcrossing. PLoS Pathog. 2012;;8: e1002534. 817 72. Valentim CLL, Cioli D, Chevalier FD, Cao X, Taylor AB, Holloway SP, et al. Genetic and Molecular 818 Basis of Drug Resistance and Species-Specific Drug Action in Schistosome Parasites. Science. 819 American Association for the Advancement of Science;; 2013;;342: 1385–1389. 820 73. Howe KL, Bolt BJ, Cain S, Chan J, Chen WJ, Davis P, et al. WormBase 2016: expanding to enable 821 helminth genomic research. Nucleic Acids Res. 2016;;44: D774–80. 822 74. Howe KL, Bolt BJ, Shafie M, Kersey P, Berriman M. WormBase ParaSite - a comprehensive resource 823 for helminth genomics. Mol Biochem Parasitol. 2017;;215: 2–10. 824 75. Crofts TS, Men Y, Alvarez-Cohen L, Taga ME. A bioassay for the detection of benzimidazoles reveals 825 their presence in a range of environmental samples. Front Microbiol. 2014;;5. 826 doi:10.3389/fmicb.2014.00592 827 76. Oliver RP, Geoff Hewitt H. Fungicides in Crop Protection, 2nd Edition. CABI;; 2014. 828 77. Prchal L, Podlipná R, Lamka J, Dědková T, Skálová L, Vokřál I, et al. Albendazole in environment: 829 faecal concentrations in lambs and impact on lower development stages of helminths and seed 830 germination. Environ Sci Pollut Res Int. 2016;;23: 13015–13022. 831 78. Perruchon C, Pantoleon A, Veroutis D, Gallego-Blanco S, Martin-Laurent F, Liadaki K, et al. 832 Characterization of the biodegradation, bioremediation and detoxification capacity of a bacterial 833 consortium able to degrade the fungicide thiabendazole. Biodegradation. 2017;;28: 383–394. 834 79. Cycoń M, Mrozik A, Piotrowska-Seget Z. Bioaugmentation as a strategy for the remediation of 835 pesticide-polluted soil: A review. Chemosphere. 2017;;172: 52–71. 836 80. de Oliveira Neto OF, Arenas AY, Fostier AH. Sorption of thiabendazole in sub-tropical Brazilian soils. 837 Environ Sci Pollut Res Int. 2017;;24: 16503–16512. 838 81. Hawkins NJ, Fraaije BA. Predicting Resistance by Mutagenesis: Lessons from 45 Years of MBC 839 Resistance. Front Microbiol. 2016;;7: 1814. 840 82. Oakley BR. Tubulins in Aspergillus nidulans. Fungal Genet Biol. 2004;;41: 420–427. 841 82. Oakley BR. Tubulins in Aspergillus nidulans. References 648 Fungal Genet Biol. 2004;;41: 420– 83. Hesse J, Thierauf M, Ponstingl H. Tubulin sequence region beta 155-174 is involved in binding 842 exchangeable guanosine triphosphate. J Biol Chem. 1987;;262: 15472–15475. 843 84. McKean PG, Vaughan S, Gull K. The extended tubulin superfamily. J Cell Sci. 2001;;114: 2723–2733. 844 85. Löwe J, Li H, Downing KH, Nogales E. Refined structure of alpha beta-tubulin at 3.5 A resolution. J 845 Mol Biol. 2001;;313: 1045–1057. 846 86. Lu C, Srayko M, Mains PE. The Caenorhabditis elegans microtubule-severing complex MEI-1/MEI-2 847 katanin interacts differently with two superficially redundant beta-tubulin isotypes. Mol Biol Cell. 848 2004;;15: 142–150. 849 31 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 87. Ellis GC, Phillips JB, O’Rourke S, Lyczak R, Bowerman B. Maternally expressed and partially 850 redundant beta-tubulins in Caenorhabditis elegans are autoregulated. J Cell Sci. 2004;;117: 457–464 851 88. Hutter H, Suh J. GExplore 1.4: An expanded web interface for queries on Caenorhabditis elegans 852 protein and gene function. Worm. 2016;;5: e1234659. 853 89. Hurd DD. Tubulins in C. elegans. WormBook. 2018;; 1–34. 854 89. Hurd DD. Tubulins in C. elegans. WormBook. 2018;; 1–34. 854 90. Hao L, Thein M, Brust-Mascher I, Civelekoglu-Scholey G, Lu Y, Acar S, et al. Intraflagellar transport 855 delivers tubulin isotypes to sensory cilium middle and distal segments. Nat Cell Biol. 2011;;13: 790– 856 798. 857 91. Bounoutas A, O’Hagan R, Chalfie M. The multipurpose 15-protofilament microtubules in C. elegans 858 have specific roles in mechanosensation. Curr Biol. 2009;;19: 1362–1367. 859 92. Savage C, Hamelin M, Culotti JG, Coulson A, Albertson DG, Chalfie M. mec-7 is a beta-tubulin gene 860 required for the production of 15-protofilament microtubules in Caenorhabditis elegans. Genes Dev. 861 1989;;3: 870–881. 862 93. Munkácsy E, Khan MH, Lane RK, Borror MB, Park JH, Bokov AF, et al. DLK-1, SEK-3 and PMK-3 863 Are Required for the Life Extension Induced by Mitochondrial Bioenergetic Disruption in C. elegans. 864 PLoS Genet. 2016;;12: e1006133. 865 94. Saunders GI, Wasmuth JD, Beech R, Laing R, Hunt M, Naghra H, et al. Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 Figure 1: GWA mappings of 209 C elegans wild isolates in response to AB 881 Supplemental Figure 2: A) Wild isolate phenotypes split by the the presence of the ALT or REF allele at 882 the QTL identified using the single-marker mapping approach. B) LD of peak markers identified with the 883 single-marker mapping approach. 884 Supplemental Figure 2: A) Wild isolate phenotypes split by the the presence of the ALT or REF allele at 882 the QTL identified using the single-marker mapping approach. B) LD of peak markers identified with the 883 single-marker mapping approach. 884 Figure 2: A) ABZ resistance phenotypes colored by the presence of variation in the ben-1 locus. B) 885 Detailed figure showing natural variation found in ben-1 among C. elegans wild isolates. 886 Supplemental Figure 3: 3D structure of C. elegans ben-1 homology model, highlighting the observed 887 missense variants present in the C. elegans population. 888 Figure 2: A) ABZ resistance phenotypes colored by the presence of variation in the ben-1 locus. B) 885 Detailed figure showing natural variation found in ben-1 among C. elegans wild isolates. 886 Supplemental Figure 3: 3D structure of C. elegans ben-1 homology model, highlighting the observed 887 missense variants present in the C. elegans population. 888 Figure 3: A) Neutrality statistics of genomic regions surrounding the ben-1 locus. B) Global sampling 889 locations of C. elegans wild isolates. C) C. elegans phylogeny. 890 Figure 3: A) Neutrality statistics of genomic regions surrounding the ben-1 locus. B) Global sampling 889 locations of C. elegans wild isolates. C) C. elegans phylogeny. 890 Supplemental Figure 4: Distribution of sampling A) environment and B) substrate where C. elegans 891 isolates with putative LoF (ALT) or REF alleles at the ben-1 locus. 892 Supplemental Figure 4: Distribution of sampling A) environment and B) substrate where C. elegans 891 isolates with putative LoF (ALT) or REF alleles at the ben-1 locus. 892 Figure 4: Fitness of F200Y and deletion alleles of the ben-1 gene using the A) high-throughput and B) 893 competitive growth assays. 894 Figure 4: Fitness of F200Y and deletion alleles of the ben-1 gene using the A) high-throughput and B) 893 competitive growth assays. References 648 Characterization and 866 comparative analysis of the complete Haemonchus contortus β-tubulin gene family and implications 867 for benzimidazole resistance in strongylid nematodes. Int J Parasitol. 2013;;43: 465–475. 868 95. Kwa MS, Kooyman FN, Boersema JH, Roos MH. Effect of selection for benzimidazole resistance in 869 Haemonchus contortus on beta-tubulin isotype 1 and isotype 2 genes. Biochem Biophys Res 870 Commun. 1993;;191: 413–419. 871 96. Beech RN, Prichard RK, Scott ME. Genetic variability of the beta-tubulin genes in benzimidazole- 872 susceptible and -resistant strains of Haemonchus contortus. Genetics. 1994;;138: 103–110. 873 97. Prichard R. Genetic variability following selection of Haemonchus contortus with anthelmintics. Trends 874 Parasitol. 2001;;17: 445–453. 875 98. Demeler J, Krüger N, Krücken J, von der Heyden VC, Ramünke S, Küttler U, et al. Phylogenetic 876 characterization of β-tubulins and development of pyrosequencing assays for benzimidazole 877 resistance in cattle nematodes. PLoS One. 2013;;8: e70212. 878 32 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 894 Figure 4: Fitness of F200Y and deletion alleles of the ben-1 gene using the A) high-throughput and B) 893 competitive growth assays. 894 Supplemental Figure 5: HTA results for all ben-1 edit strains in the presence of ABZ 895 Figure 5: A) Genome-wide single-marker mapping of ABZ length phenotype after correcting isotype 896 phenotypes based on the presence of putative ben-1 LoF variation. B) Fine-mapping of chromosome X 897 QTL. 898 Figure 5: A) Genome-wide single-marker mapping of ABZ length phenotype after correcting isotype 896 phenotypes based on the presence of putative ben-1 LoF variation. B) Fine-mapping of chromosome X 897 QTL. 898 phenotypes based on the presence of putative ben-1 LoF variation. B) Fine-mapping of chromosome X 897 QTL. 898 Supplemental Figure 6: Genome-wide gene-burden mapping of ABZ length phenotype after correcting 899 isotype phenotypes based on the presence of putative ben-1 LoF variation. 900 Supplemental Figure 6: Genome-wide gene-burden mapping of ABZ length phenotype after correcting 899 isotype phenotypes based on the presence of putative ben-1 LoF variation. 900 Supplemental Figure 6: Genome-wide gene-burden mapping of ABZ length phenotype after correcting 899 isotype phenotypes based on the presence of putative ben-1 LoF variation. 900 Supplemental Figure 7: Wild isolate phenotypes split by the presence of the ALT or REF allele at the 901 QTL identified using the single-marker mapping approach. We highlight the presence of putative ben-1 902 LoF variants. 903 Supplemental Figure 7: Wild isolate phenotypes split by the presence of the ALT or REF allele at the 901 QTL identified using the single-marker mapping approach. We highlight the presence of putative ben-1 902 LoF variants. 903 Supplemental Figure 7: Wild isolate phenotypes split by the presence of the ALT or REF allele at the 901 QTL identified using the single-marker mapping approach. We highlight the presence of putative ben-1 902 LoF variants. 903 Supplemental Figure 8: Wild isolate phenotypes after ben-1 LoF regression split by the isotype genotype 904 at the chromosome X QTL peak marker. 905 Supplemental Figure 8: Wild isolate phenotypes after ben-1 LoF regression split by the isotype genotype 904 at the chromosome X QTL peak marker. 905 33 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Supplemental Figure 9: HTA re-phenotyping results for selected C elegans wild strains in the presence 906 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Supplemental Figure 9: HTA re-phenotyping results for selected C. elegans wild strains in the presence of ABZ . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Supplemental Figure 9: HTA re-phenotyping results for selected C. elegans wild strains in the presence 906 of ABZ 907 Supplemental Figure 10: Neighbor joining phylogenetic tree of the six C. elegans β-tubulins, the C. 908 remanei and C. briggsae ben-1 orthologs, the Haemonchus contortus ben-1 ortholog, and the Necator 909 americanus ben-1 ortholog. 910 Supplemental Figure 11: Tajima’s D for the genomic regions surrounding ben-1 orthologs in the 911 C. elegans genome. 912 Supplemental Figure 9: HTA re-phenotyping results for selected C. elegans wild strains in the presence 906 of ABZ 907 Supplemental Figure 10: Neighbor joining phylogenetic tree of the six C. elegans β-tubulins, the C. 908 remanei and C. briggsae ben-1 orthologs, the Haemonchus contortus ben-1 ortholog, and the Necator 909 americanus ben-1 ortholog. 910 Supplemental Figure 10: Neighbor joining phylogenetic tree of the six C. elegans β-tubulins, the C. 908 remanei and C. briggsae ben-1 orthologs, the Haemonchus contortus ben-1 ortholog, and the Necator 909 americanus ben-1 ortholog. 910 Supplemental Figure 11: Tajima’s D for the genomic regions surrounding ben-1 orthologs in the 911 C. elegans genome. 912 34 of 52 . Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Supplemental Data 913 914 TS1: HTA ABZ dose response data_raw(.Rda) 915 TS2: HTA ABZ dose response data_processed(.tsv) 916 TS3: HTA phenotype data of C. elegans wild strains_raw(.Rda) 917 TS4: HTA phenotype data of C. elegans wild strains_processed(.tsv) 918 TS5: GWA single marker mapping data_processed(.tsv) 919 TS6: GWA single marker fine mapping data(.tsv) 920 TS7: GWA gene-burden mapping phenotype data(.ped) 921 TS8: GWA gene-burden mapping data(.tsv) 922 TS9: Positions and effects for all ben-1 variants(.tsv) 923 TS10: Summary of population genetics data(.Rda) 924 TS11: Tajima’s D, Fay’s Wu and Zeng’s E for ben-1 (.tsv) 925 TS12: HTA phenotype data of C. elegans ben-1 mutants_summarized(.tsv) 926 TS 13: HTA phenotype data of C. elegans ben-1 mutants_processed(.tsv) 927 TS 14: competition assay data_raw(.tsv) 928 TS 15: relative fitness calculations(.xml) 929 TS 16: ben-1 regressed with ben-1 covariate(.tsv) 930 TS 17: GWA single-marker mapping data after ben-1 regression(.tsv) 931 TS 18: GWA single-marker fine mapping data after ben-1 regression(.tsv) 932 TS 19: GWA gene-burden mapping phenotype data after ben-1 regression(.ped 933 TS 20: GWA gene-burden mapping data after ben-1 regression(.assoc) 934 TS 21: GWA gene-burden mapping data after ben-1 regression(.tsv) 935 TS 22: Tajima’s D for other C. elegans beta-tubulins (.tsv) 936 TS 23: HTA re-phenotyping of C. elegans wild strains(.Rda) 937 Supplemental Table 1: strains(.csv) 938 Supplemental Table 2: Oligonucleotide sequences(.csv) 939 940 940 35 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 941 942 943 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 944 942 943 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 944 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 944 A) The single-marker based Manhattan plot for animal length (q90.TOF) in the presence of ABZ is shown. 945 Each dot represents an SNV that is present in at least 5% of the assayed population. The genomic 946 location of each SNV is plotted on the x-axis, and the statistical significance of the correlation between 947 genotype and phenotype is plotted on the y-axis. SNVs are colored red if they pass the genome-wide 948 Bonferroni-corrected significance threshold, which is shown by a gray horizontal line. Genomic regions of 949 interest are represented by cyan rectangles surrounding each QTL. In total, three QTL were identified, 950 located on chr II and chr V. B) A genome-wide manhattan plot of animal length (q90.TOF) after ABZ 951 exposure based on the gene-burden mapping approach is shown. Each dot represents a single 952 C. elegans gene with its genomic location plotted on the x-axis, and the statistical significance of the 953 correlation between genotype and phenotype plotted on the y-axis. Genes passing the burden test 954 statistical significance threshold are colored in red. The burden test analysis identified genes significantly 955 correlated with ABZ resistance on chr I, chr II, and chr III. 956 A) The single-marker based Manhattan plot for animal length (q90.TOF) in the presence of ABZ is s 945 36 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available unde The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a , who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; /372623 . CC-BY 4.0 International license a under, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available unde The copyright holder for this preprint (which was not this version posted July 19, 2018. ; .1101/372623 Variants in ben-1 are highly correlated with ABZ resistance lot for the phenot pic response of C elegans strains nder ABZ e pos re is sho n Each ba . CC-BY 4.0 International license a review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: nt 957 958 Figure 2: Variants in ben-1 are highly correlated with ABZ resistance 959 Figure 2: Variants in ben-1 are highly correlated with ABZ resistance 959 A) A bar plot for the phenotypic response of C. elegans strains under ABZ exposure is shown. Each bar 960 represents a single wild strain included in the HTA. Strains are sorted by their relative resistance to ABZ 961 based on the mean animal length (q90.TOF) from four replicate measures. All strains that were found to 962 have variants with predicted moderate-to-high impact and/or structural variation are colored by their 963 specific type of variant. Strains similar to the N2 reference genome with respect to the ben-1 locus are 964 shown in grey. The black dotted line marks the point halfway point between the most and least resistant 965 ABZ strains. B) An overview of variants found in the ben-1 locus among C. elegans wild strains is shown. 966 The genomic position of each variant is shown on the x-axis (S: splice donor variant;; M: missense variant 967 leading to amino acid substitution;; crossed circle (⦻): alternative stop codon;; triangle: deletion;; I: insertion;; 968 37 of 52 . Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint percent sign: inversion;; T: transposon insertion. Colors correspond to colors in (A). All strains with ben-1 969 variants are listed on the y-axis in increasing order by their length in ABZ. For variants shared by two or 970 more strains, the mean phenotype value of all corresponding strains determines their placement on the y- 971 axis. A representation of the ben-1 gene model is shown below the variant summary panel, including 5’ 972 and 3’ UTRs (grey rectangle and bar), five exons (blue bars), and four introns (thin lines). Additionally, 973 missense variants leading to amino acid substitutions are highlighted with their exact position and the 974 corresponding amino acid exchange in the single letter code. Missense variants are colored orange if the 975 corresponding strains do not contain any other ben-1 variation. Missense variants that are always 976 associated with deletion variants are colored blue. 977 percent sign: inversion;; T: transposon insertion. Colors correspond to colors in (A). All strains with ben-1 969 variants are listed on the y-axis in increasing order by their length in ABZ. For variants shared by two or 970 more strains, the mean phenotype value of all corresponding strains determines their placement on the y- 971 axis. A representation of the ben-1 gene model is shown below the variant summary panel, including 5’ 972 and 3’ UTRs (grey rectangle and bar), five exons (blue bars), and four introns (thin lines). Additionally, 973 missense variants leading to amino acid substitutions are highlighted with their exact position and the 974 corresponding amino acid exchange in the single letter code. Missense variants are colored orange if the 975 corresponding strains do not contain any other ben-1 variation. Missense variants that are always 976 associated with deletion variants are colored blue. 977 percent sign: inversion;; T: transposon insertion. Supplemental Figure 1: Dose response of four genetically divergent C. elegans strains to ABZ exposure 880 Figure 1: GWA mappings of 209 C elegans wild isolates in response to ABZ 881 Colors correspond to colors in (A). All strains with ben-1 969 variants are listed on the y-axis in increasing order by their length in ABZ. For variants shared by two or 970 more strains, the mean phenotype value of all corresponding strains determines their placement on the y- 971 axis. A representation of the ben-1 gene model is shown below the variant summary panel, including 5’ 972 and 3’ UTRs (grey rectangle and bar), five exons (blue bars), and four introns (thin lines). Additionally, 973 missense variants leading to amino acid substitutions are highlighted with their exact position and the 974 corresponding amino acid exchange in the single letter code. Missense variants are colored orange if the 975 corresponding strains do not contain any other ben-1 variation. Missense variants that are always 976 associated with deletion variants are colored blue. 977 978 38 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 979 Figure 3: Population-level summary of ben-1 variants A) F d W ’ H ( i k) T ji ’ D (bl k) d Z ’ E (bl ) f th i i di b Figure 3: Population level summary of ben 1 variants 980 Figure 3: Population-level summary of ben-1 variants 981 A) Fay and Wu’s H (pink), Tajima’s D (black), and Zeng’s E (blue) for the genomic region surrounding ben- 982 1. Genomic position is shown on the x-axis, and the value for each displayed statistic is shown on the y- 983 axis. The neutrality statistics were calculated using a sliding window approach (100 SNVs window size 984 and 1 SNV slide distance) using only SNV data. B) The global distribution of strains that contain moderate- 985 to-high predicted variation in ben-1. Each dot corresponds to the sampling location of an individual strain 986 and is colored by the type of variant discovered in the ben-1 locus. C) The genome-wide phylogeny of 249 987 C. elegans strains showing that variation in the ben-1 locus occurred independently multiple times during 988 the evolutionary history of the species. The dots on individual branch nodes correspond to strains with 989 variation in ben-1 and have the same color code as in panel B. 990 39 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; /372623 991 992 Figure 4: F200Y allele replacement and ben-1 deletion confer resistance to the ABZ sensitive N2 993 t i 994 Figure 4: F200Y allele replacement and ben-1 deletion confer resistance to the ABZ sensitive N2 993 strain 994 A) Tukey box plots of the animal length phenotypes for at least 100 replicates of the generated ben-1 995 allele-replacement strains after ABZ treatment is shown. The y-axis represents the animal length 996 phenotype after correcting for growth in DMSO conditions. 980 Figure 3: Population-level summary of ben-1 variants 981 Both ben-1 allele strains are significantly more 997 resistant to ABZ treatment than the N2 strain (p-value < 1E-10, TukeyHSD). B) The results from a multi- 998 generation competition experiment between a barcoded N2 strain and the ben-1 allele-replacement strains 999 are shown. The generation number is shown on the x-axis, and the allele frequency of the ben-1 allele- 1000 replacement strains is shown on the y-axis. 1001 Figure 4: F200Y allele replacement and ben-1 deletion confer resistance to the ABZ sensitive N2 993 Figure 4: F200Y allele replacement and ben-1 deletion confer resistance to the ABZ sensitive N2 993 strain 994 Figure 4: F200Y allele replacement and ben-1 deletion confer resistance to the ABZ sensitive N2 993 strain 994 A) Tukey box plots of the animal length phenotypes for at least 100 replicates of the generated ben-1 995 allele-replacement strains after ABZ treatment is shown. The y-axis represents the animal length 996 phenotype after correcting for growth in DMSO conditions Both ben 1 allele strains are significantly more 997 Figure 4: F200Y allele replacement and ben-1 deletion confer resistance to the ABZ sensitive N2 993 strain 994 A) Tukey box plots of the animal length phenotypes for at least 100 replicates of the generated ben-1 995 allele-replacement strains after ABZ treatment is shown. The y-axis represents the animal length 996 phenotype after correcting for growth in DMSO conditions. Both ben-1 allele strains are significantly more 997 resistant to ABZ treatment than the N2 strain (p-value < 1E-10, TukeyHSD). B) The results from a multi- 998 generation competition experiment between a barcoded N2 strain and the ben-1 allele-replacement strains 999 are shown. The generation number is shown on the x-axis, and the allele frequency of the ben-1 allele- 1000 replacement strains is shown on the y-axis. 1001 40 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. 980 Figure 3: Population-level summary of ben-1 variants 981 ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1003 003 1004 1005 1004 1005 Figure 5: Regression of the putative ben-1 LoF variants identifies a novel QTL on chromosome X 1006 Figure 5: Regression of the putative ben-1 LoF variants identifies a novel QTL on chromosome X 1006 of the putative ben-1 LoF variants identifies a novel QTL on chromosome X Figure 5: Regression of the putative ben-1 LoF variants identifies a no 006 Figure 5: Regression of the putative ben-1 LoF variants identifies a novel QTL on chromosome X 1006 A) Manhattan plot of the single-marker based GWA mapping showing that a nominally significant QTL on 1007 chromosome X is associated with residual phenotypic variation in response to ABZ treatment. Each dot 1008 represents an SNV that is present in at least 5% of the assayed population. The genomic location of each 1009 SNV is plotted on the x-axis, and the statistical significance is plotted on the y-axis. SNVs are colored red 1010 if they pass the Bonferroni-corrected significance threshold, which is shown by a gray horizontal line. The 1011 genomic region of interest is represented by cyan rectangle surrounding the chromosome X QTL. B) Fine- 1012 mapping of the chromosome X QTL showing all variants with moderate-to-high predicted effects within the 1013 chromosomes X QTL region of interest. The genomic position in Mb is shown on the x-axis, and the 1014 −𝑙𝑜𝑔10(𝑝) value, which represents the strength of the association between residual ABZ resistance and 1015 the plotted variant, is shown on the y-axis. Dots are colored by the type of variant. 1016 A) Manhattan plot of the single-marker based GWA mapping showing that a nominally significant QTL on 1007 chromosome X is associated with residual phenotypic variation in response to ABZ treatment. Each dot 1008 represents an SNV that is present in at least 5% of the assayed population. The genomic location of each 1009 SNV is plotted on the x-axis, and the statistical significance is plotted on the y-axis. SNVs are colored red 1010 if they pass the Bonferroni-corrected significance threshold, which is shown by a gray horizontal line. The 1011 genomic region of interest is represented by cyan rectangle surrounding the chromosome X QTL. B) Fine- 1012 mapping of the chromosome X QTL showing all variants with moderate-to-high predicted effects within the 1013 chromosomes X QTL region of interest. 980 Figure 3: Population-level summary of ben-1 variants 981 The genomic position in Mb is shown on the x-axis, and the 1014 −𝑙𝑜𝑔10(𝑝) value, which represents the strength of the association between residual ABZ resistance and 1015 the plotted variant, is shown on the y-axis. Dots are colored by the type of variant. 1016 41 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1019 1020 Supplemental Figure 1: 1021 Dose response of four genetically divergent C. elegans strains after ABZ exposure 1022 The box plots show a representative dose-response experiment on four C. elegans strains for A) animal 1023 length (q90.TOF) and B) brood size (norm.n). The ABZ concentration is plotted on the x-axis, and 1024 individual replicate trait values subtracted from the mean in DMSO control conditions is plotted on the y- 1025 axis. Each box represents four technical replicates. 1026 1020 Supplemental Figure 1: 1021 Dose response of four genetically divergent C. elegans strains after ABZ exposure 1022 Dose response of four genetically divergent C. elegans strains after ABZ exposure 1022 response of four genetically divergent C. elegans strains after ABZ exposure Dose response of four genetically divergent C. elegans strains afte 1022 The box plots show a representative dose-response experiment on four C. elegans strains for A) animal 1023 length (q90.TOF) and B) brood size (norm.n). The ABZ concentration is plotted on the x-axis, and 1024 individual replicate trait values subtracted from the mean in DMSO control conditions is plotted on the y- 1025 axis. Each box represents four technical replicates. 1026 The box plots show a representative dose-response experiment on four C. elegans strains for A) animal 1023 length (q90.TOF) and B) brood size (norm.n). The ABZ concentration is plotted on the x-axis, and 1024 individual replicate trait values subtracted from the mean in DMSO control conditions is plotted on the y- 1025 axis. Each box represents four technical replicates. 1026 42 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1027 1028 Supplemental Figure 2: 1029 Single-marker mapping summary 1030 A) Regressed animal length (q90.TOF) phenotypes in the presence of ABZ. Each dot represents t 1031 regressed animal length of four replicates. Strains are grouped by the presence of the REF 1032 genotype at the peak QTL marker identified in the single-marker GWA mapping approach. B) 1033 disequilibrium (LD) as measured by the correlation coefficient between peak QTL markers in 1034 formula for the correlation coefficient r = -D / sqrt( p(A) * p(a) * p(B) * p(b) ). 1035 . CC-BY 4.0 International license a 7 8 43 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint Dose response of four genetically divergent C. elegans strains after ABZ exposure 1022 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1036 1037 Supplemental Figure 3: 1038 Amino acid substitutions found in C. elegans wild strains are distributed throughout the BEN-1 1039 structure 1040 An in silico model of the BEN-1 structure (green), binding to an alpha-tubulin (purple) is shown. Nov 1041 identified amino acid substitutions among C. elegans wild strains are highlighted in color. Most allele 1042 were correlated with ABZ resistance (orange). The F200Y mutation, known as a major BZ resistanc 1043 marker in parasitic nematodes is shown in pink. 1044 1036 1036 1037 Supplemental Figure 3: 1038 Amino acid substitutions found in C. elegans wild strains are distributed throughout the BEN-1 1039 1037 Supplemental Figure 3: 1038 Amino acid substitutions found in C. elegans wild strains are distributed throughout the BEN-1 1039 structure 1040 An in silico model of the BEN-1 structure (green), binding to an alpha-tubulin (purple) is shown. Novel 1041 identified amino acid substitutions among C. elegans wild strains are highlighted in color. Most alleles 1042 were correlated with ABZ resistance (orange). The F200Y mutation, known as a major BZ resistance 1043 marker in parasitic nematodes is shown in pink. 1044 1045 Amino acid substitutions found in C. elegans wild strains are distributed throughout the BEN-1 1039 44 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; 72623 . CC-BY 4.0 International license a who has granted bioRxiv a license to display the preprint in perpetuity. Dose response of four genetically divergent C. elegans strains after ABZ exposure 1022 It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; 372623 1046 Supplemental Figure 4: 1047 Sampling environment and substrate of wild C. elegans isolates 1048 The fractions of wild C. elegans isolates sampled in a given A) location and on a given B) substrate are 1049 shown. Colors for the stacked bar plots correspond to different A) sampling locations and B) substrates. 1050 046 Supplemental Figure 4: 1047 Sampling environment and substrate of wild C. elegans isolates 048 Sampling environment and substrate of wild C. elegans isolat 1048 The fractions of wild C. elegans isolates sampled in a given A) location and on a given B) substrate are 1049 shown. Colors for the stacked bar plots correspond to different A) sampling locations and B) substrates. 1050 The fractions of wild C. elegans isolates sampled in a given A) location and on a given B) substrate are 1049 shown. Colors for the stacked bar plots correspond to different A) sampling locations and B) substrates. 1050 45 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1052 1053 1054 Supplemental Figure 5: 1055 Complete ben-1 CRISPR allele HTA assay 1056 Tukey box plots of the animal length phenotypes of the generated ben-1 allele-replacement strains after 1057 ABZ treatment is shown. Blue boxes correspond to independent F200Y allele strains and pink boxes 1058 correspond to independent Del strains. The y-axis represents the animal-length phenotype after correcting 1059 for growth in DMSO conditions. 1060 1053 1054 Supplemental Figure 5: 1055 Complete ben-1 CRISPR allele HTA assay 1056 Tukey box plots of the animal length phenotypes of the generated ben-1 allele-replacement strains after 1057 ABZ treatment is shown. Blue boxes correspond to independent F200Y allele strains and pink boxes 1058 correspond to independent Del strains. The y-axis represents the animal-length phenotype after correcting 1059 for growth in DMSO conditions. 1060 46 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1062 1063 1064 S 1063 Supplemental Figure 6: 1065 Gene-burden mapping of putative ben-1 LoF regressed phenotype 1066 Gene-burden manhattan plot of the ben-1 corrected animal length (q90.TOF) after ABZ exposure. Animal- 1067 length phenotypes in the presence of ABZ were adjusted based on the presence of a putative loss-of- 1068 function variant in ben-1. Each dot represents a single gene of the C. elegans genome with its genomic 1069 location plotted on the x-axis, and the test statistic plotted on the y-axis. Genes passing the burden test 1070 statistical significance threshold are colored in red. 1071 47 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a ho has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; 2623 1073 1074 Supplemental Figure 7: 1075 Phenotypes of isolates split by genotypes at peak QTL markers identified using the single-marker 1076 073 Supplemental Figure 7: 1075 Phenotypes of isolates split by genotypes at peak QTL markers identified using the single-marker 1076 Phenotypes of isolates split by genotypes at peak QTL markers identified using the single-marker 1076 Regressed animal length (q90.TOF) phenotypes in the presence of ABZ. Each dot represents the mean 1078 regressed animal length of four replicates per strain. Strains are grouped by the presence of the REF or 1079 ALT genotype at the peak QTL marker identified in the single-marker GWA mapping approach. Dot colors 1080 correspond to the identified variant class in the ben-1 locus. 1081 48 of 52 . CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . Gene-burden mapping of putative ben-1 LoF regressed phenotype 1066 CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1082 1083 1084 1085 Supplemental Figure 8: 1086 Animal length phenotypes after correcting for the presence of putative ben-1 loss-of-function 1087 1082 1083 1083 1084 1084 1085 Supplemental Figure 8: 1086 Animal length phenotypes after correcting for the presence of putative ben-1 loss-of-function 1087 variants 1088 Animal length phenotypes in the presence of ABZ after correction for the presence of a putative loss-of- 1089 function variant in ben-1. Each dot represents the mean regressed animal length of four replicates per 1090 strain. Strains are split into groups based on the presence of the REF or ALT allele at the chromosome X 1091 QTL peak marker. Dots are colored by the presence of a putative loss-of-function variant in ben-1, where 1092 pink corresponds to strains with a putative loss-of-function variant, and blue corresponds to strains with no 1093 putative loss-of-function variants. 1094 1095 Animal length phenotypes after correcting for the presence of putative ben-1 loss-of-function 1087 Animal length phenotypes in the presence of ABZ after correction for the presence of a putative loss-of- 1089 function variant in ben-1. Each dot represents the mean regressed animal length of four replicates per 1090 strain. Strains are split into groups based on the presence of the REF or ALT allele at the chromosome X 1091 QTL peak marker. Dots are colored by the presence of a putative loss-of-function variant in ben-1, where 1092 pink corresponds to strains with a putative loss-of-function variant, and blue corresponds to strains with no 1093 putative loss-of-function variants. 1094 49 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1096 1097 1098 Supplemental Figure 9: 1099 Re-phenotyping of a slow-growing ben-1 variant strain 1100 Tukey box plots of the animal length phenotypes of N2 and three wild isolates in the presence of ABZ. Gene-burden mapping of putative ben-1 LoF regressed phenotype 1066 The 1101 y-axis represents the animal length phenotype after correcting for growth in DMSO conditions. 1102 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under 1096 1097 1098 . CC-BY 4.0 International license a y p ) , g p y p p p p y Re-phenotyping of a slow-growing ben-1 variant strain 1100 Tukey box plots of the animal length phenotypes of N2 and three wild isolates in the presence of ABZ. The 1101 y-axis represents the animal length phenotype after correcting for growth in DMSO conditions. 1102 50 of 52 . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint 1103 Supplemental Figure 10: Phylogenetic analysis of β-tubulins 1104 Neighbor-joining tree of the amino acid sequences of the six known C. elegans β-tubulins and paralogs in 1105 C. remanei (Cre), C. briggsae (Cbr), H. contortus (Hcont), and Necator americanus. 1106 1103 Supplemental Figure 10: Phylogenetic analysis of β-tubulins 1104 Neighbor-joining tree of the amino acid sequences of the six known C. elegans β-tubulins and paralogs in 1105 C. remanei (Cre), C. briggsae (Cbr), H. contortus (Hcont), and Necator americanus. 1106 1107 Supplemental Figure 10: Phylogenetic analysis of β-tubulins 1104 Neighbor-joining tree of the amino acid sequences of the six known C. elegans β-tubulins and paralogs in 05 C. remanei (Cre), C. briggsae (Cbr), H. contortus (Hcont), and Necator americanus. 06 Neighbor-joining tree of the amino acid sequences of the six known C. elegans β-tubulins and paralogs in 1105 C. remanei (Cre), C. briggsae (Cbr), H. contortus (Hcont), and Necator americanus. 1106 1107 Neighbor-joining tree of the amino acid sequences of the six known C. elegans β-tubulins and paralogs in 1105 C. remanei (Cre), C. briggsae (Cbr), H. contortus (Hcont), and Necator americanus. 1106 51 of 52 . Gene-burden mapping of putative ben-1 LoF regressed phenotype 1066 CC-BY 4.0 International license a ertified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under The copyright holder for this preprint (which was not this version posted July 19, 2018. ; https://doi.org/10.1101/372623 doi: bioRxiv preprint . CC-BY 4.0 International license a d by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under 1108 plemental Figure 11: ma’s D at the ben-1 locus and other β-tubulin genes ma’s D calculated from SNV data of genomic regions surrounding A) ben-1, B) tbb-1, C) tbb-2, D) Tajima’s D at the ben-1 locus and other β-tubulin genes Tajima’s D calculated from SNV data of genomic regions surrounding A) ben-1, B) tbb-1, C) tbb-2, D) tbb- 1113 4, E) tbb-6, and F) mec-7. Genomic position in Mb is plotted on the x-axis, and Tajima’s D is plotted on the 1114 y-axis. Tajima’s D was calculated using a sliding window with a 100 SNV window and a one-SNV step 1115 size. Two red bars for each panel correspond to the start and end positions for the corresponding gene. 1116 Tajima’s D calculated from SNV data of genomic regions surrounding A) ben-1, B) tbb-1, C) tbb-2, D) tbb- 1113 4, E) tbb-6, and F) mec-7. Genomic position in Mb is plotted on the x-axis, and Tajima’s D is plotted on the 1114 y-axis. Tajima’s D was calculated using a sliding window with a 100 SNV window and a one-SNV step 1115 size. Two red bars for each panel correspond to the start and end positions for the corresponding gene. 1116 52 of 52
https://openalex.org/W3041519188	https://europepmc.org/articles/pmc7350753?pdf=render	English	null	Acceptability of a hypothetical preventative HIV vaccine among people who use drugs in Vancouver, Canada	BMC public health	2,020	cc-by	7,914	Fleming et al. BMC Public Health (2020) 20:1081 https://doi.org/10.1186/s12889-020-09202-6 Fleming et al. BMC Public Health (2020) 20:1081 https://doi.org/10.1186/s12889-020-09202-6 Open Access © 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. Abstract Background: As research on HIV vaccines continues to advance, studies exploring the feasibility of this intervention are necessary to inform uptake and dissemination strategies with key populations, including people who use drugs (PWUD). Methods: We conducted 25 in-depth qualitative interviews examining HIV vaccine acceptability among PWUD in Vancouver, Canada. Participants were recruited from an ongoing prospective cohort of HIV-negative PWUD. Data were coded using NVivo, and analyzed thematically. Results: Acceptability was framed by practical considerations such as cost and side effects, and was influenced by broader trust of government bodies and health care professionals. While an HIV vaccine was perceived as an important prevention tool, willingness to be vaccinated was low. Results suggest that future vaccine implementation must consider how to minimize the burden an HIV vaccine may place on PWUD. Centering the role of health care providers in information dissemination and delivery may assist with uptake. Conclusions: Our findings suggest improvements in care and improved patient-provider relationships would increase the acceptability of a potential HIV vaccine among this population. Keywords: Vaccines, HIV/AIDS, HIV prevention, Qualitative research Keywords: Vaccines, HIV/AIDS, HIV prevention, Qualitative research people who use or inject drugs [8, 9], sex workers [10, 11] and men who have sex with men (MSM) [12, 13]. Uptake among these key populations presents an important chal- lenge for the dissemination of new biomedical prevention efforts, including people who use drugs (PWUD). Further, how PWUD perceive and engage with prevention interven- tions can be influenced by their experiences of co-morbid conditions (e.g., hepatitis C, mental illness) [14, 15], as well as by less visible social, structural, and economic barriers [16–19]. Previous research has found willingness to receive an HIV vaccine among PWUD was generally low, and shaped by perceptions of self-efficacy [8] and stigmatization of HIV [20]. This is of particular importance when con- trasted with research examining acceptability and uptake of a hepatitis B (HBV) and C (HCV) vaccines, for which Acceptability of a hypothetical preventative HIV vaccine among people who use drugs in Vancouver, Canada Taylor Fleming1,2, Jenna Valleriani1,3, Cara Ng1, Lisa Maher4,5, Will Small1,6 and Ryan McNeil1,7,8* Background As candidate prophylactic HIV vaccines progress through clinical trials [1–3], it is critical to assess acceptability among potential recipients to inform implementation strat- egies. Past studies have found that the acceptability of HIV vaccines depends on a variety of important factors, includ- ing perceptions of safety, potential side effects, distrust of health care systems and providers, and considerations of social risks (e.g., discrimination) [4, 5]. These works have generally focused on populations believed to be at high risk of HIV acquisition, such as sexually active youth [6, 7], * Correspondence: ryan.mcneil@yale.edu 1British Columbia Centre on Substance Use, Vancouver, BC, Canada 7Department of Medicine, Yale School of Medicine, New Haven, CT, USA Full list of author information is available at the end of the article * Correspondence: ryan.mcneil@yale.edu 1British Columbia Centre on Substance Use, Vancouver, BC, Canada 7Department of Medicine, Yale School of Medicine, New Haven, CT, USA Full list of author information is available at the end of the article Fleming et al. BMC Public Health (2020) 20:1081 Page 2 of 9 Page 2 of 9 reports of hypothetical and real-world uptake are attributed to lack of knowledge of HBV and HCV [9, 21, 22], rather than stigma. eligible for this study if they had completed a follow-up interview with V-DUS in the previous six months, were current regular users of illicit drugs, and were HIV- negative. Participant characteristics are reported in Table 1. We aimed to oversample women and Indigenous persons relative to their representation among the local drug-using population in recognition of the ways they are uniquely impacted by HIV risk. The study was approved by the research ethics board of Providence Health Care and the University of British Columbia. g Much previous work on the acceptability of HIV vaccines among PWUD has relied predominantly on quantitative research [23–25] and has failed to consider the complex interplay of social, structural and environ- mental contexts and their impact on health outcomes in this population. The broader literature increasingly iden- tifies the need to shift away from a focus on individual choices and behaviours to understanding of underlying conditions and contexts, in order to address HIV risk [19, 26]. For example, gendered power relations may play a role in women’s HIV-related vulnerability [27, 28], and considerations of contextual factors such as vulner- ability to sexual and physical violence, distinctively shape women’s HIV-risk profiles [29, 30]. Semi-structured interviews with 25 participants took place in a private room, and were conducted by JV and CN using an interview guide to facilitate discussion. Both interviewers are extensively trained in qualitative research methods, including interviewing, and have prior experience conducting interviews with PWUD. Before beginning the interview, JV and CN explained the study, reviewed the consent form with participants, and re- ceived written informed consent. Interviews focused on general immunization history, perceptions of HIV risk, current risk behaviours, knowledge of public health As research on candidate HIV vaccines continues to advance, centering the voices of PWUD as a key popula- tion for biomedical prevention interventions will allow for a more nuanced examination of uptake and feasibility currently lacking in available studies. In this qualitative study, we explore the perceptions of PWUD towards a hypothetical HIV vaccine in Vancouver, Canada, and their implications for the acceptability and uptake of an effica- cious vaccine. Table 1 Participant characteristics Participants (n = 25) Age Mean 50.2 Range 34–73 Gender Men 14 (56%) Women 11 (44%) Ethnicity White 10 (40%) Indigenous 13 (52%) Other 2 (8%) Drugs use in past 30 daysa Heroin 15 (60%) Cocaine (powder) 7 (28%) Crack cocaine 4 (16%) Crystal methamphetamine 8 (32%) Fentanyl 10 (40%) Frequency of drug use Daily 15 (60%) 3–4 times per week 3 (12%) One or fewer times per week 7 (28%) Would receive HIV vaccineb Yes 12 (48%) No 10 (40%) Undecided 2 (8%) aNote participants could select multiple substances bHypothetical approved preventative HIV vaccine Table 1 Participant characteristics Gender This ethno-epidemiological study draws on semi- structured interviews conducted with 25 participants of the Vancouver Drug Users Study (V-DUS). V-DUS is a prospective cohort study of HIV-negative people who use drugs, operating in Vancouver, Canada since 1996. This cohort has been described in detail elsewhere [31]. In short, V-DUS participants are recruited through snowball sampling and street outreach, and are eligible to participate in the cohort study if they are 18 years or older, HIV negative, and used illicit drugs in the past 30 days at baseline. Participants complete a standardized interviewer-administered questionnaire and provide blood samples at baseline and bi-annual follow-up visits. Ethno-epidemiology aims to merge both epidemiological and qualitative methods to increase understanding of how social-structural contexts impact patterns of health and social harms [32]. V-DUS participants have well-established relationships with our research program through their involvement in the longitudinal cohort study. V-DUS participants who were eligible for the present study were contacted by a V-DUS team member, and invited to participate in a one-time interview at a storefront research office located in Vancouver’s Downtown Eastside neighbourhood—a low-income inner city neighbourhood with a high preva- lence of HIV and illicit drug use. Participants were Fleming et al. BMC Public Health (2020) 20:1081 Page 3 of 9 new public health initiative for a vaccine shown to be efficacious: efforts and campaigns around HIV, and willingness to receive a hypothetical preventative HIV vaccine. For the purposes of this study, this vaccine was assumed to be an intramuscular injection [33, 34]. The interview guide developed for this study (see Additional file 1) focused on a hypothetical proven HIV vaccine, however, lines of questioning were flexible enough to allow for more general discussion of HIV vaccines, including clinical trial participation. Interviews were audio-recorded, and lasted between 45 and 75 min each. Participants received $30 cash honorarium for their time. Upon reaching saturation, interviews were then transcribed verbatim, coded, and analyzed thematically using NVivo 12. Partic- ipants were assigned pseudonyms using a random name generator. I don’t want to be one of the guinea pigs though … I don’t want to be, like, you know, what I was saying, they have to do trials and stuff? I don’t want to be in that group of people that do the trials. I don’t care how much money they give me. Gender Similar to others, Charlotte invoked the image of “guinea pigs” to describe clinical trial participants, and would under no circumstances participate as a test subject to demonstrate efficacy of a candidate vaccine. Conversely, other participants believed a preventative HIV vaccine would be useful, but would be unwilling to receive such a vaccine in any capacity, regardless of demonstrated efficacy. These participants framed their unwillingness to participate through their perceived risk to HIV exposure. Similar to other participants, ‘Andy’ (62-year-old white man) characterized his level of risk as low, by contrasting himself to other PWUD who may participate in conventionally high HIV-risk behaviours (e.g., sex work, MSM), or who participate in the same behaviours as himself but in ‘riskier’ ways (e.g., higher frequency of injecting): We drew on both deductive and inductive approaches [35] during analysis to focus on the uptake and acceptabil- ity of a hypothetical preventative HIV vaccine, potential concerns around a hypothetical vaccine, relationships with health care service providers, and practical considerations that shape acceptability (e.g., cost, dosing regimens, side effects). We developed an initial coding framework in- formed by a priori categories extracted from the interview guide, and our research team (JV, CN and RM) met after reviewing 3–5 transcripts to further refine the coding framework based on emerging themes. Once final themes were established, data were recoded to ensure trustworthi- ness of final themes. Two team members (JV and CN) in- dependently coded each of the remaining transcripts to establish inter-coder reliability, and the research team met regularly to discuss findings as a group. Well if you’re MSM or you’re an injection drug user, I think it would be a smart thing to take if it was proven that it was working. If you were sure, you know 99 percent that it was going to prevent it, then you should, especially if you are engaging in things that we know, there are two ways of getting HIV, let’s face it – sex and injection drugs – that’s it. So, if you are engaging in those risky behaviours, then why not. You would be stupid not to because the alternative is really shitty … I don’t have any, you know, sex no, there is no need [for a vaccine], and I don’t share rigs. The one time when I do my occasional shot now I use a brand-new rig out of a package. Willingness to receive an HIV vaccine g Participants were generally knowledgeable about and accepting of vaccines other than the influenza vaccine, with which many reported they had prior negative experiences. Participants viewed vaccines as an important public and individual health tool, and most had received them as necessary (e.g., childhood vaccinations). Few participants were unaware of what a vaccine is or how it works, although these individuals were more skeptical about a hypothetical preventative HIV vaccine. Further, partici- pants acknowledged the important role that vaccines played in both public and individual health, although they were unwilling to receive any vaccine that was not demon- strated to be effective and accepted among the general public. For example, while most participants, including ‘Charlotte’ (49-year-old white woman) considered a future HIV vaccine useful, many reported not wanting to be among the first to be vaccinated, whether as a participant in a clinical trial of a candidate vaccine or as part of Within these narratives, participants sought to distance themselves from stigma associated with drug use, while simultaneously reinforcing stigmatization of PWUD who may be among the most socially and economically marginalized. Were a vaccine proven to be efficacious through clinical trials and offered as part of a new public health initiative, participants still viewed the vaccine to be somewhat experimental, with Charlotte continuing “I would want to see other people get the vaccine first before trying it,” even though, as members of a key population PWUD may be among the first offered a hypothetical preventative HIV vaccine. Many felt it was important to wait before getting a newly approved vaccine as part of a public health Fleming et al. BMC Public Health (2020) 20:1081 Page 4 of 9 from others within their peer networks. ‘Harvey’ (60- year-old white man) explained: initiative because of any potential unknown consequences, and multiple participants expressed concern about being one of the first to get a new vaccine. The idea of a new preventative HIV vaccine concerned participants, who were unsure if that would come with unknown and poten- tially long-term side effects, and whether safety of the vac- cine had been established; this is despite the implication that such a vaccine would have already been approved for public use, suggesting that public acceptance of a vaccine’s efficacy shaped willingness to receive a vaccine just as much, if not more than clinical acceptance (e.g., clinical trial results): You can’t always trust the government. Willingness to receive an HIV vaccine I think they think they’re making the best decision in your interest. I don’t know, at my age, I know things about the government that they’ve done to friends of mine and at the time they were doing these things they weren’t so forthcoming with what was really going on. Some participants spoke about rejecting future vaccin- ation campaigns because of a long-standing suspicion of government initiatives. For example, ‘Laura’ (39-year-old Indigenous woman) explained, “The government’s looking out for themselves. The people are nothing” when discussing her faith in government-sponsored vaccination initiatives. When not relating this mistrust to past negative experi- ences with governmental initiatives and institutions, partic- ipants cited conspiracy theories, such as population control or concerns about the “chemical composition” of vaccines, as why they would be skeptical of any future preventative HIV vaccines, regardless of proven efficacy. For ‘Luke’ (62-year-old white man), targeting a high-risk group, such as PWUD, under the guise of HIV vaccination was viewed as a potential strategy to eliminate socially and morally stigmatized groups: I mean, it would have to be like really, really tested, and really show that it works, you know? Like what proof do you know? The only proof is would you take it and then you go and have sex with an HIV-positive person and see if it works or not. (‘Simon,’ 44-year-old white man) Moreover, for Simon and others, the evidence support- ing an HIV vaccine still had to be considered with the risk of it not working as intended when deciding whether or not to vaccinate. For example, ‘Nick’ (58-year-old Black man) questioned the role of public support in relation to the seasonal influenza vaccine, noting that despite public health campaigns purporting the benefits of vaccination, many who were vaccinated still became ill with the flu each year: “I’m taking it [flu vaccine] because they [public health messaging] tell me it’s good for me. I’m hoping they’re telling me the truth, but I don’t know.” Who are they gonna get rid of first? They’re gonna get rid of the drug addicts, the prisoners … you know, because we’re overpopulated … it’s common sense. The planet can only hold so many people, and there’s only so much food, there’s only so much water, there’s only so much of everything. And we keep going the way we’re going, what’s gonna happen, right? I mean, who goes first, right? Willingness to receive an HIV vaccine It’s you or I? All participants expressed that they would want more information about the effectiveness of an HIV vaccine and its side effects before they exposed themselves to it. For most participants, mild to moderate side effects would be acceptable, such as a sore arm or fever, and would not impede uptake, as Nick explained: Participants who subscribed to conspiracy theories surrounding vaccinations had no personal experiences to support these claims. Rather, belief in these theories origi- nated from reports through peer networks and personal research: “I just read that [conspiracy] on the internet and I got scared” (‘Marco,’ 38-year-old white man). I’d have to hear them [the side effects] … I don’t want to, as I said, I don’t want to go blind … I’d have to hear what the side effects are, and do I really want to go through it. Like, if the side effects are just headaches, I could take that. While participants were generally distrustful of government and pharmaceutical companies, the majority spoke extensively about the trust they had in their indi- vidual health care providers with regards to medical care and information about vaccinations, suggesting that participants trusted clinicians to scrutinize public health policy and potentially harmful government initiatives for them. Participants unanimously said they would want to receive information about the vaccination from a health care professional, such as a doctor or nurse, because they considered them knowledgeable and trusted them. The vast majority of participants had a regular physician, Immunization schedule and cost Immunization schedule and cost Immunization scheduling was also an important consid- eration in uptake and feasibility. While in general fewer doses were more acceptable, most participants indicated a preference for a single injection with no follow-up ‘booster shot.’ Approximately half reported that they would accept more doses on a schedule if required. How- ever, many, including ‘Andy’ (62-year-old white man) cited the challenges of adhering to an immunization schedule, particularly a longer immunization schedule, while actively using drugs: Luke echoed this trust in his individual physician as a source of information about an HIV vaccine. If my next-door neighbours tell it to me, well, you know what I mean? I might believe it, I might not, and then I don’t know if I, you know. But if my doctor told me to, I’d take it as gospel, basically, right? Well that’s an issue with drug users, especially with drug users, because getting them to come in for follow ups are hard, and if the vaccine is based on the fact that you have to have three of them, say you had to have it three times two months apart, that could be an issue … but you never know with a drug user where, they could be in jail, they could be in a fucking different province, they could be anywhere. And so the follow up would be difficult. Some participants indicated that they would defer to their doctor’s authority or medical expertise, and were open to future vaccinations even though they did not believe they were at high risk for HIV. For example, ‘Donald’ (34-year-old Indigenous man) said, “If they [my doctor] really thought I needed it, I probably would do it.” Here we see that doctors carry authority with these participants in ways which could increase the acceptabil- ity and uptake of a preventative HIV vaccine. Even those participants who stated they would not be vaccinated (see Table 1) said that they would consider it if their doctor felt it was very important. More broadly, Andy underscores the barriers to engaging in preventative HIV care for criminalized populations within the context of drug prohibition, as the act of meeting daily survival needs and the risk of law enforcement lends a level of unpredictability to participants’ lives, which may prohibit regular follow-up care. Institutional trust and mistrust Skepticism towards a hypothetical HIV vaccine was predominantly framed by a mistrust in government and pharmaceutical companies, as well as by past negative experiences within health care systems. Many partici- pants felt that these agencies and institutions cared little about the public, generally, and PWUD, in particular. These perceptions were shaped by negative personal experiences with government agencies (e.g., criminal justice system, child welfare system), as well as reports Page 5 of 9 Fleming et al. BMC Public Health (2020) 20:1081 Page 5 of 9 and the relationships these participants had developed with their health care providers helped shape their atti- tudes towards a potential HIV vaccine, as Nick described: sexually active. Some of this was related to concerns about the stigma of accessing an HIV vaccine, which participants believed would be eliminated if it were given to everybody uniformly at a young age as part of a childhood immunization schedule, in contrast to being targeted at high-risk individuals and populations. I’d need to sit down and talk to my doctor and hear what they got to tell me about it. I just can’t jump in to it because someone says we’ve got a vaccine. I want to know about the vaccine. I want to know if it can help me. I want to know if it can hurt me. I want to know about it before I take it … I’ve got a really good doctor. y Compulsory vaccination In general, participants reported that they would want to receive the vaccination at a fixed site clinic, and very few in- dividuals preferred receiving vaccinations through outreach services (e.g., ‘door to door’ or mobile campaigns). Such services may have implications for privacy, confidentiality, and social stigmatization, as Nick acknowledged: As an extension of this more general distrust for govern- ment, most participants rejected the idea of a compulsory HIV vaccination policy for adults. Some participants raised concerns that targeting certain key populations, such as PWUD, would be discriminatory or further stig- matizing, while most asserted their individual autonomy to make health care decisions for themselves or in conjunction with their physician, as ‘Michelle’ (50-year- old Indigenous woman) highlighted: “It’s a little invasive and intrusive … I just think that people should be able to make their own decisions.” It should be noted that while health care providers seem to play a key role in partici- pants’ decision to participate in new HIV prevention tech- nologies, participants were strongly opposed to any type of coercive or compulsory approach to vaccination. I wouldn’t want to do it. I don’t want everybody knowing my business... if you’re going to this place, then everybody thinks you got AIDS … I go to my own doctor, no one knows what I got. Participants identified cost as a major barrier or deter- rent to vaccine uptake. Participants unanimously agreed that a preventative HIV vaccine should be free for vulnerable populations and those on limited incomes. Within this, some believed that a tested and approved HIV vaccine should be free of charge because they viewed it as a critical public health issue. Michelle was Half of our participants were, however, open to the idea of a tested and approved HIV vaccine being deliv- ered to children in schools, even as a mandatory vaccin- ation, and particularly before young people become Page 6 of 9 Page 6 of 9 Fleming et al. BMC Public Health (2020) 20:1081 increase the likelihood of uptake and completion of a vaccine schedule [22, 38], implementation of an HIV vaccine will require the development of strategies to specifically target initial vaccine uptake, particularly given the stigmatization of HIV and reluctance to en- gage with any HIV-associated services. y Compulsory vaccination Moreover, it is important to consider how traditional public health campaigns for HIV prevention have targeted particu- lar high-risk groups [39], which may contribute to the stigmatization of those who are already marginalized. New approaches to HIV prevention must consider how stigma can affect or impede the uptake of new interventions. among the few participants who would be potentially willing to pay for a preventative HIV vaccine depending on the cost ($20–$100): Because we are fucking Canadians and we’re not from the USA… There should be no reason why we shouldn’t have free anything … especially vaccines in regard to health and stuff. No matter what risk group we’re in, we deserve it …Doesn’t matter what do we do in our life, what we do on the fucking sidelines in our home and it’s our business. The above quotation is notable in light of the focus on behavioural risk factors and the transmission of HIV. Michelle highlights that individuals hold autonomy over their lives, even if these choices place them at risk for HIV acquisition. Michelle also draws on the right to access (free) health care within the context of Canada’s publicly funded health care system, including new vaccinations, asserting that cost should not be a barrier to uptake. This also demonstrates the importance of limiting the financial burden vaccinations may place on individuals. Our findings also suggest that the patient-provider rela- tionship is a critical component to reaching PWUD and that improvements in care, as well as patient-provider rela- tionships, would increase the acceptability and uptake of a potentially new HIV vaccine among PWUD. This reported trust in health care providers was somewhat surprising considering the consistent lack of trust in government agencies and institutions shaped by negative past experi- ences within the health care system reported by our partic- ipants. Further, past research with PWUD has described prevailing attitudes of mistrust of health care providers and authorities as a major barrier to optimal engagement in preventative care, including vaccine uptake [21, 40]. Our participants’ narratives demonstrate that doctors with established positive relationships with PWUD patients exert considerable influence on individuals’ decision to consider being vaccinated against HIV, and that this group is critical to engagement in preventative care for such key populations. This is consistent with other studies that re- port that positive relationships between PWUD and health care providers lead to better health outcomes [41, 42]. y Compulsory vaccination This is also consistent with previous childhood vaccination research demonstrating that trust in health care providers is a key factor in parents’ decisions to vaccinate their children, and that this relationship is an important site of intervention in increasing vaccination uptake [43, 44]. While doctors may influence decisions by PWUD to be vaccinated, most of our participants described a conversa- tion with their doctor as part of their decision making around uptake of a licensed HIV vaccine. Strategies to encourage uptake of a potential HIV vaccine by PWUD must also consider social and structural barriers which shape access to, and experiences with, health services and government bodies. In particular, the mistrust in govern- ment reported by our participants has significant implica- tions for vaccination campaigns, including skepticism around candidate and licensed hypothetical vaccines, resulting in lower uptake and feasibility of these products. Broad institutional mistrust (e.g., government conspiracies, “Big Pharma”) has also been implicated in vaccine hesitancy within the contexts of childhood vaccinations Discussion HIV vaccine acceptability among PWUD in our study was framed by practical considerations such as concerns about cost, vaccine schedules, and side effects. Accept- ability was also influenced by perceptions and trust of government bodies, for-profit pharmaceutical industries, and health care professionals. In our study, a HIV vac- cine was generally considered a useful preventative tool by PWUD, although willingness to personally receive such a vaccination was low. While roughly half of our participants were willing to re- ceive a hypothetical preventative HIV vaccine, willingness to participate amongst these individuals was near-unanimously conditional on widespread uptake and acceptance of this vaccine as safe and effective by the general public, and most rejected compulsory HIV immunization for adults. Despite this, many suggested a mandatory HIV vaccination policy for children, framing it as a way to protect future genera- tions and achieve high coverage because of its potential to eliminate the stigma associated with being identified in a high-risk group later in life. Previous studies, however, have suggested a mandatory policy would be unlikely, as it requires endorsement and acceptance by the general popu- lation [36, 37]. The development of strategies for HIV vaccination uptake specifically tailored to PWUD, especially those considered at high risk or hard-to-reach, would help to ensure broad HIV vaccine coverage, particularly consider- ing those who are at higher risk of transmission report being less likely to receive a preventative HIV vaccine [25]. While past hepatitis B research has found that, for PWUD willing to receive a preventative vaccination, cash incentives can Fleming et al. BMC Public Health (2020) 20:1081 Page 7 of 9 Page 7 of 9 [45] and communicable disease pandemics such as the 2009 H1N1 event [46]. Decoupling public health policy and practice from pharmaceutical companies and political powers will be an important step in promoting uptake of a potential HIV vaccine, particularly for PWUD, whose lived realities are criminalized by these very institutions. and as such, a majority of our sample had a primary health care physician. Considering the importance of the patient- provider relationship in acceptability and uptake of an emerging HIV vaccine, populations of PWUD less engaged with health care services may be more difficult to reach in dissemination. To encourage real-world uptake of a future preventa- tive HIV vaccine among PWUD, barriers such as cost, dosing regimens and mode of dissemination will need to be addressed. Conclusion h l While PWUD in the current study demonstrate a general acceptance of vaccines, they were less willing to receive a hypothetical preventative HIV vaccine when considering the potential for unknown long-term risks, were one to be proven effective and made publicly available. The lack of willingness by our participants to be vaccinated was also related to a distrust of government bodies and pharma- ceutical companies, often juxtaposed against a reported trust in their health care providers. This highlights the importance of consolidating engagement of PWUD with health care providers as new prevention technology, such as HIV vaccines, emerge. However, even despite willing- ness to discuss this with a health care provider, PWUD did underscore the importance of agency in health care decision-making, rather than a compulsory policy for adults. Centralizing the role of health care providers in the dissemination of information and delivery of the HIV vaccine may assist with uptake among PWUD. Practically, vaccine schedule, cost, and where the vaccine would be administered (due to privacy concerns), can also shape up- take and dissemination. Thus, in the future, strategies should be tailored to meet the needs of PWUD and other high-risk groups to promote broad HIV vaccine coverage. While this study adds to the small body of qualitative research on HIV vaccine acceptability, and highlights how social and structural barriers may shape access to this intervention for PWUD, this study has some important limitations. The V-DUS cohort is made up of both former and current PWUD, and is generally an older cohort of individuals [49] which may limit the generalizability of the findings to younger PWUD. Further, despite being over- represented within our sample, Indigenous participants’ perspectives on HIV vaccine acceptability did not speak to racialized experiences, meaning that when asked directly by the interviewer, they did not feel being indigenous impacted their risk of HIV transmission, or access to vaccines, including information and conversations from their health care provider. Additionally, participants gen- erally reported that that being Indigenous did not impact their access to other support and health services, nor did they bring up the impact of being Indigenous on health care access or vaccine access independently. Given that Indigenous peoples in Canada experiences inequitable access to health care, and a disproportionate burden of risk of HIV, future vaccine acceptability research should examine how race and histories of colonization impact these views. Acknowledgements The authors thank the study participants for their contribution to this research, as well as current and past researchers and staff. Abbreviations bb e at o s HIV: Human immunodeficiency virus; MSM: Men who have sex with men; PWUD: People who use drugs; V-DUS: Vancouver Drug Users Study HIV: Human immunodeficiency virus; MSM: Men who have sex with men; PWUD: People who use drugs; V-DUS: Vancouver Drug Users Study Authors’ contributions TF and RM developed the study protocol. JV and CN contributed to data collection. JV, CN, and RM contributed to the data analysis. TF, JV, and RM wrote the original manuscript. CN, LM, WS, and RM provided input on the final version of the manuscript. All authors have read and approved the manuscript. Discussion However, the impact of stigma as a barrier to uptake continues to underpin HIV prevention efforts, as is consistent with past research on intervention imple- mentation among PWUD [47, 48]. For example, many of our participants did not want to discuss HIV-related topics (e.g., health services, risk behaviours) in detail, and when speaking to vaccination administration, indi- cated that were they to receive it, they would want to receive this vaccination in a clinic because of concerns around privacy, and not wanting to utilize HIV-specific agencies or services for fear of being identified as some- one with HIV. This suggests that public health interven- tions should consider encouraging more positive HIV vaccine attitudes, and the creation of partnerships with health care providers can help offer accurate and destig- matizing information around HIV risk and risks of a potential vaccine as important piece of disseminating an HIV vaccine. Conclusion h l Lastly, the current context, the Downtown Eastside of Vancouver, Canada, is unique in that it repre- sents a greater concentration of harm reduction and health services and PWUD than likely found elsewhere, Supplementary information pp y Supplementary information accompanies this paper at https://doi.org/10. 1186/s12889-020-09202-6. Additional file 1. Interview guide Author details 18. Milloy MJ, Kerr T, Buxton J, Rhodes T, Guillemi S, Hogg R, et al. Dose- response effect of incarceration events on nonadherence to HIV antriretroviral therapy among injection drug users. J Infect Dis. 2011;203(9) 1215–21. 1British Columbia Centre on Substance Use, Vancouver, BC, Canada. 2Interdisciplinary Graduate Studies Program, University of British Columbia, Vancouver, BC, Canada. 3National Institute for Cannabis Health and Education, Toronto, ON, Canada. 4The Kirby Institute for Infection and Immunity, Faculty of Medicine, University of New South Wales, Sydney, NSW, Australia. 5The Burnet Institute, Melbourne, VIC, Australia. 6Faculty of Health Sciences, Simon Fraser University, Burnaby, BC, Canada. 7Department of Medicine, Yale School of Medicine, New Haven, CT, USA. 8Program in Addiction Medicine, Yale School of Medicine, Yale University, 367 Cedar Street, New Haven, CT, USA. 19. Milloy MJ, Marshall BD, Kerr T, Buxton J, Rhodes T, Montaner J, et al. Social and structural factors associated with HIV disease progression among illicit drug users: a systematic review. AIDS. 2012;26(9):1049–63. 20. Barrington C, Moreno L, Kerrigan D. Local understanding of an HIV vaccine and its relationship with HIV-related stigma in the Dominican Republic. AIDS Care. 2007;19(7):871–7. Addiction Medicine, Yale School of Medicine, Yale University, 367 Cedar Street, New Haven, CT, USA. 21. Park JN, White B, Bates A, Enriquez J, Liao L, Maher L. Motivators and barriers influencing willingness to participate in candidate HCV vaccine trials: perspectives of people who inject drugs. Drug Alcohol Depend. 2012; 123(1–3):35–40. Received: 20 November 2019 Accepted: 2 July 2020 Funding This work was supported by the US National Institutes of Health (R01DA043408). RM and WS are supported by Michael Smith Foundation for Health Research Scholar Awards. RM is also supported by a Canadian Institute of Health Research New Investigator Award. TF is supported by a Frederick Banting and Charles Best Canada Graduate Scholarships from CIHR. Page 8 of 9 Page 8 of 9 Page 8 of 9 Fleming et al. BMC Public Health (2020) 20:1081 LM is supported by the award of an Australian National Health and Medical Research Council Senior Research Fellowship. The sponsors had no role in the design and conduct of this research study or preparation of this manuscript. workers, injection and non-injection drug users, and men who have sex with men in Spain. AIDS Behav. 2010;14(3):607–17. workers, injection and non-injection drug users, and men who have sex with men in Spain. AIDS Behav. 2010;14(3):607–17. 11. Etcheverry MF, Evans JL, Sanchez E, Mendez-Arancibia E, Merono M, Gatell JM, et al. Enhanced retention strategies and willingness to participate among hard-to-reach female sex workers in Barcelona for HIV prevention and vaccine trials. Hum Vaccin Immunother. 2013;9(2):420–9. Received: 20 November 2019 Accepted: 2 July 2020 22. Topp L, Day CA, Wand H, Deacon RM, van Beek I, Haber PS, et al. A randomised controlled trial of financial incentives to increase hepatitis B vaccination completion among people who inject drugs in Australia. Prev Med. 2013;57(4):297–303. References 1. Barouch DH, Tomaka FL, Wegmann F, Stieh DJ, Alter G, Robb ML, et al. Evaluation of a mosaic HIV-1 vaccine in a multicentre, randomised, double- blind, placebo-controlled, phase 1/2a clinical trial (APPROACH) and in rhesus monkeys (NHP 13-19). Lancet. 2018;392(10143):232–43. 1. Barouch DH, Tomaka FL, Wegmann F, Stieh DJ, Alter G, Robb ML, et al. Evaluation of a mosaic HIV-1 vaccine in a multicentre, randomised, double- blind, placebo-controlled, phase 1/2a clinical trial (APPROACH) and in rhesus monkeys (NHP 13-19). Lancet. 2018;392(10143):232–43. 1. Barouch DH, Tomaka FL, Wegmann F, Stieh DJ, Alter G, Robb ML, et al. Evaluation of a mosaic HIV-1 vaccine in a multicentre, randomised, double- blind, placebo-controlled, phase 1/2a clinical trial (APPROACH) and in rhesus monkeys (NHP 13-19). Lancet. 2018;392(10143):232–43. 23. Young AM, DiClemente RJ, Halgin DS, Sterk CE, Havens JR. HIV vaccine acceptability among high-risk drug users in Appalachia: a cross-sectional study. BMC Public Health. 2014;14:537. 2. Hammer SM, Sobieszczyk ME, Janes H, Karuna ST, Mulligan MJ, Grove D, et al. Efficacy trial of a DNA/rAd5 HIV-1 preventive vaccine. N Engl J Med. 2013;369(22):2083–92. 2. Hammer SM, Sobieszczyk ME, Janes H, Karuna ST, Mulligan MJ, Grove D, et al. Efficacy trial of a DNA/rAd5 HIV-1 preventive vaccine. N Engl J Med. 2013;369(22):2083–92. 24. Newman PA, Duan N, Lee SJ, Rudy ET, Seiden DS, Kakinami L, et al. HIV vaccine acceptability among communities at risk: the impact of vaccine characteristics. Vaccine. 2006;24(12):2094–101. 3. Safrit JT, Fast PE, Gieber L, Kuipers H, Dean HJ, Koff WC. Status of vaccine research and development of vaccines for HIV-1. Vaccine. 2016; 34(26):2921–5. 3. Safrit JT, Fast PE, Gieber L, Kuipers H, Dean HJ, Koff WC. Status of vaccine research and development of vaccines for HIV-1. Vaccine. 2016; 34(26):2921–5. 25. Newman PA, Lee SJ, Rudy ET, Diamant A, Duan N, Nakazono T, et al. Endorsement of compulsory HIV vaccination policy among populations at high risk of HIV exposure (LA VOICES). Prev Sci. 2014;15(3):428–35. 4. Newman PA, Woodford MR, Logie C. HIV vaccine acceptability and culturally appropriate dissemination among sexually diverse Aboriginal peoples in Canada. Glob Public Health. 2012;7(1):87–100. 26. Coupland H, Page K, Stein E, Carrico A, Evans J, Dixon T, et al. Structural interventions and social suffering: responding to amphetamine-type stimulant use among female entertainment and sex workers in Cambodia. Int J Drug Policy. 2019;64:70–8. 5. Rudy ET, Newman PA, Duan N, Kelly EM, Roberts KJ, Seidan DS. Consent for publication Not applicable. 16. Meyer JP, Althoff AL, Altice FL. Optimizaing care for HIV-infected people who use drugs: evidence-based approaches to overcoming healthcare disparities. Clin Infect Dis. 2013;57(9):1309–17. Not applicable. Availability of data and materials 12. Knight R, Small W, Carson A, Shoveller J. Complex and conflicting social norms: implications for implementation of future HIV pre-exposure prophylaxis (PrEP) interventions in Vancouver. Canada PLoS One. 2016;11(1): e0146513. The data underlying these findings is presented in this manuscript in support of presented themes and subthemes as anonymized participant quotations. Further data cannot be made publicly available as per study approval by the research ethics board of Providence Health Care and the University of British Columbia. Interested researchers may direct data access requests to the corresponding author. 13. Charlebois ED, Das M, Porco TC, Havlir DV. The effect of expanded antiretroviral treatment strategies on the HIV epidemic among men who have sex with men in San Francisco. Clin Infect Dis. 2011;52(8):1046–9. The authors declare that they have no competing interests. The authors declare that they have no competing interests. Competing interests 17. Wolfe D. Paradoxes in antiretroviral treatment for injecting drug users: access, adherence and structural barriers in Asia and the former Soviet Union. Int J Drug Policy. 2007;18(4):246–54. Ethics approval and consent to participate 14. Page K, Tsui J, Maher L, Choopanya K, Vanichseni S, Mock PA, et al. Biomedical HIV prevention including pre-exposure prophylaxis and opiate agonist therapy for women who inject drugs. JAIDS. 2015;69(S2):S19–S175. This study was approved by the research ethics board of Providence Health Care and the University of British Columbia. All participants provided written informed consent. 15. Altice FL, Kamarulzaman A, Soriano VV, Schecter M, Friedland GH. Treatment of medical, psychiatric, and substance-use comorbidities in people infected with HIV who use drugs. Lancet. 2010;376:367–87. References Vaccine rejecting Parents' engagement with expert systems that inform vaccination programs. J Bioeth Inq. 2017;14(1):65–76. 46. Yaqub O, Castle-Clarke S, Sevdalis N, Chataway J. Attitudes to vaccination: a critical review. Soc Sci Med. 2014;112:1–11. 47. Ibragimov U, Cooper HL, Haardorfer R, Dunkle KL, Zule WA, Wong FY. Stigmatization of people who inject drugs (PWID) by pharmacists in Tajikistan: sociocultural context and implications for a pharmacy-based prevention approach. Harm Reduct J. 2017;14(1):64. 48. Pilgrim N, Musheke M, Raymond HF, Keating R, Mwale J, Banda L, et al. Quality of care and HIV service utilization among key populations in Zambia: a qualitative comparative analysis among female sex workers, men who have sex with men and people who use drugs. AIDS Care. 2019;31(4):460–4. 49. Cheng T, Small W, Dong H, Nosova E, Hayashi K, DeBeck K. An age-based analysis of nonmedical prescription opioid use among people who use illegal drugs in Vancouver, Canada. Substance Abuse Treatment, Prevention, and Policy. 2018;13(1):41. 49. Cheng T, Small W, Dong H, Nosova E, Hayashi K, DeBeck K. An age-based analysis of nonmedical prescription opioid use among people who use illegal drugs in Vancouver, Canada. Substance Abuse Treatment, Prevention, and Policy. 2018;13(1):41. References HIV vaccine acceptability among women at risk: perceived barriers and facilitators to futre HIV vaccine uptake. AIDS Educ Prev. 2005;17(3):253–67. 27. Iverson J, Page K, Madden A, Maher L. HIV, HCV and health-related harms among women who inject drugs- implications for prevention and treatment. JAIDS. 2015;69(S2):S176–S81. p 6. Mbunda T, Tarimo EA, Chalamilla G, Bakari M, Sandstrom E, Kulane A. The influence of community members on participation by youth in an HIV vaccine trial in Tanzania. PLoS One. 2016;11(12):e0168660. p 6. Mbunda T, Tarimo EA, Chalamilla G, Bakari M, Sandstrom E, Kulane A. The influence of community members on participation by youth in an HIV vaccine trial in Tanzania. PLoS One. 2016;11(12):e0168660. 28. Higgins JA, Hoffman S, Dworkin SL. Rethinking gender, heterosexual men, and women's vulnerability to HIV/AIDS. Am J Public Health. 2010;100(3):435–45. 7. Jaspan HB, Berwick JR, Myer L, Mathews C, Flisher AJ, Wood R, et al. Adolescent HIV prevalence, sexual risk, and willingness to participate in HIV vaccine trials. J Adolesc Health. 2006;39(5):642–8. 29. Boyd J, Collins AB, Mayer S, Maher L, Kerr T, McNeil R. Gendered violence and overdose prevention sites: a rapid ethnographic study during an overdose epidemic in Vancouver. Canada Addiction. 2018;113(12):2261–70. 8. Dhalla S, Poole G, Singer J, Patrick DM, Wood E, Kerr T. Cognitive factors and willingness to participate in an HIV vaccine trial among HIV-negative injection drug users. Vaccine. 2010;28(7):1663–7. 30. Blankenship KM, Reinhard E, Sherman SG, El-Bassel N. Structural interventions for HIV preventions among women who use drugs: a global perspective. JAIDS. 2015;69(S2):S140–S5. 9. Lally MA, Montstream-Quas SA, Tanaka S, Tedeschi SK, Morrow KM. A qualitative study among injection drug using women in Rhode Island: attitudes toward testing, treatment, and vaccination for hepatitis and HIV. AIDS Patient Care STDs. 2008;22(1):53–64. 31. Wood E, Hogg RS, Lima V, et al. Highly active antiretroviral therapy and survival in hiv-infected injection drug users. JAMA. 2008;300(5):550–4. 10. Etcheverry MF, de Lazzari E, Fuchs JD, Merono M, Sierra E, Del Romero J, et al. Pilot study assessing HIV vaccine trial readiness among female sex 32. Lopez AM, Bourgois P, Wenger LD, Lorvick J, Martinez AN, Kral AH. Interdisciplinary mixed methods research with structurally vulnerable Page 9 of 9 Fleming et al. BMC Public Health (2020) 20:1081 Fleming et al. BMC Public Health (2020) 20:1081 populations: case studies of injection drug users in San Francisco. Int J Drug Policy. 2013;24(2):101–9. References populations: case studies of injection drug users in San Francisco. Int J Drug Policy. 2013;24(2):101–9. 33. Jin X, Morgan C, Yu X, DeRosa S, Tomaras GD, Montefiori DC, et al. Multiple factors affect immunogenicity of DNA plasmid HIV vaccines in human clinical trials. Vaccine. 2015;33(20):2347–53. 34. Baden LR, Liu J, Li H, Johnson JA, Walsh SR, Kleinjan JA, et al. Induction of HIV-1-specific mucosal immune responses following intramuscular recombinant adenovirus serotype 26 HIV-1 vaccination of humans. J Infect Dis. 2015;211(4):518–28. 35. Bradley EH, Curry LA, Devers KJ. Qualitative data analysis for health services research: developing taxonomy, themes, and theory. Health Serv Res. 2007; 42(4):1758–72. 36. Liau A, Zimet GD. Undergraduates’s perception of HIV immunization: attitudes and behaviours as determining factors. International Journal of AIDS. 2000;11:445–50. 37. Newman PA, Logie C. HIV vaccine acceptability: a systematic review and meta-analysis. Aids. 2010;24(11):1749–56. 38. World Health Organization. Guidance on prevention of viral hepatitis B and C among people who inject drugs. Geneva: World Health Organization; 2012. 39. Gibson DR, Zhang G, Cassady D, Pappas L, Mitchell J, Kegeles SM. Effectiveness of HIV prevention social marketing with injecting drug users. Am J Public Health. 2010;100(10):1828–30. 40. Maher L, White B, Hellard M, Madden A, Prins M, Kerr T, et al. Candidate hepatitis C vaccine trials and people who inject drugs: challenges and opportunities. Vaccine. 2010;28(45):7273–8. 41. Knowlton AR, Arnsten JH, Eldred LJ, Wilkinson JD, Shade SB, Bohnert AS, et al. Antiretroviral use among active injection-drug users: the role of patient-provider engagement and structural factors. AIDS Patient Care STDs 2010;24(7):421–8. 42. Salvalaggio G, McKim R, Taylor M, Wild TC. Patient–provider rapport in the health Care of People who Inject Drugs. SAGE Open. 2013;3(4): 2158244013509252. 43. Eller NM, Henrikson NB, Opel DJ. Vaccine information sources and parental trust in their child’s health care provider. Health Educ Behav. 2019;46(3):445–53. 43. Eller NM, Henrikson NB, Opel DJ. Vaccine information sources and parental trust in their child’s health care provider. Health Educ Behav. 2019;46(3):445–53. 44. Brown KF, Kroll JS, Hudson MJ, Ramsay M, Green J, Long SJ, et al. Factors underlying parental decisions about combination childhood vaccinations including MMR: a systematic review. Vaccine. 2010;28(26):4235–48. 45. Attwell K, Leask J, Meyer SB, Rokkas P, Ward P. Vaccine rejecting Parents' engagement with expert systems that inform vaccination programs. J Bioeth Inq. 2017;14(1):65–76. 45. Attwell K, Leask J, Meyer SB, Rokkas P, Ward P. 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/W4285772783	https://zenodo.org/records/6855662/files/10.pdf	English	null	Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Literature Review	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	2,358	INTRODUCTION patient underwent several complementary studies, including skin biopsy without presenting a definitive diagnosis. Laboratory with persistence of hypereosinophilia that guided us towards the diagnosis of Idiopathic Hypereosinophilic Syndrome. Eosinophilia is a common finding in clinical practice, but when elevated values of eosinophils are found, it constitutes a diagnostic challenge. Hypereosinophilic syndrome is a group of rare disorders defined by persistent blood hypereosinophilia >1.5 × 109/l and associated to organ damage that result in a wide variety of clinical manifestations: fatigue with nonspecific skin lesions, to endomyocardial fibrosis, neurological compromise and life-threatening evolution. ABSTRACT Eosinophilia is a common finding in clinical practice, but when elevated values of eosinophils are found, it constitutes a diagnostic challenge. Hypereosinophilic syndrome is a group of rare disorders defined by persistent blood hypereosinophilia >1.5 × 109/l and associated to organ damage that result in a wide variety of clinical manifestations: fatigue with nonspecific skin lesions, to endomyocardial fibrosis, neurological compromise and life-threatening evolution. The prognosis of the disease is variable and depends on the variant and the availability of specific treatment. 1- 2 We present the clinical case of a patient with a history of B symptoms and the presence of persistent erythroderma. The patient underwent several complementary studies, including skin biopsy without presenting a definitive diagnosis. Laboratory with persistence of hypereosinophilia that guided us towards the diagnosis of Idiopathic Hypereosinophilic Syndrome. Available on: https://ijmscr.org/ KEYWORDS: Eosinophil, Dermatitis, Erythroderma KEYWORDS: Eosinophil, Dermatitis, Erythroderma Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Literature Review Polanco Llanes Alondra S1, Salazar Quijano Christian A2, Can Pérez Eric E3, Arjona Bojorquez Rashid I4, Luna Garcia Javier5, García Catalán Grisell6, Alonzo Burgos Marcos A7 1,5Internal medicine department, Hospital General de Especialidades Dr. Javier Buenfil Osorio 2General surgery department, Hospital General de Especialidades Dr. Javier Buenfil Osorio 3Internal medicine department. Private practice. 4General surgery department.,Hospital Regional #1 Lic. Ignacio García Téllez. T-1 IMSS. 6General surgery department ,Unidad Médica de Alta Especialidad IMSS #14 7Internal medicine department, Hospital General de Zona #3 IMSS “Jesús María” Aguascalientes. Polanco Llanes Alondra S1, Salazar Quijano Christian A2, Can Pérez Eric E3, Arjona Bojorquez Rashid I4, Luna Garcia Javier5, García Catalán Grisell6, Alonzo Burgos Marcos A7 1,5Internal medicine department, Hospital General de Especialidades Dr. Javier Buenfil Osorio 2General surgery department, Hospital General de Especialidades Dr. Javier Buenfil Osorio 3Internal medicine department. Private practice. 4General surgery department Hospital Regional #1 Lic Ignacio García Téllez T 1 IMSS J , , g 1,5Internal medicine department, Hospital General de Especialidades Dr. Javier Buenfil Osorio 2General surgery department, Hospital General de Especialidades Dr. Javier Buenfil Osorio 3Internal medicine department. Private practice. ARTICLE DETAILS Published On: 18 July 2022 International Journal of Medical Science and Clinical Research Studies ISSN(print): 2767-8326, ISSN(online): 2767-8342 Volume 02 Issue 07 July 2022 Page No: 650-653 International Journal of Medical Science and Clinical Research Studies ISSN(print): 2767-8326, ISSN(online): 2767-8342 Volume 02 Issue 07 July 2022 Page No: 650-653 International Journal of Medical Science and Clinical Research Studies ISSN(print): 2767-8326, ISSN(online): 2767-8342 Volume 02 Issue 07 July 2022 Page No: 650-653 DOI: https://doi.org/10.47191/ijmscrs/v2-i7-10, Impact Factor: 5.365 International Journal of Medical Science and Clinical Research Studie ISSN(print): 2767-8326, ISSN(online): 2767-8342 International Journal of Medical Scienc ISSN(print): 2767-8326, ISSN(online): 2767-8342 Volume 02 Issue 07 July 2022 Page No: 650-653 DOI: https://doi.org/10.47191/ijmscrs/v2-i7-10, Impact Factor: 5.365 Corresponding Author: Polanco Llanes Alondra S CLINICAL CASE A 34-year-old male with no significant chronic-degenerative history. He started symptoms 7 months earlier, with fever of 39 °C predominantly at night, diaphoresis, asthenia, adynamia, and weight loss of 8 kg in one month. Later with itching and erythema in the extremities, spreading to the chest and abdomen. (Figure 1-2) The prognosis of the disease is variable and depends on the variant and the availability of specific treatment. 1- 2 We present the clinical case of a patient with a history of B symptoms and the presence of persistent erythroderma. The 650 Volume 02 Issue 07 July 2022 Corresponding Author: Polanco Llanes Alondra S Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Literature Review Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Fig. 1. Skaly skin and lichenification áreas. Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Literature Review Fig. 1. Skaly skin and lichenification áreas. Fig. 2. Erythroderma of abdomen. Fig. 2. Erythroderma of abdomen. Fig. 1. Skaly skin and lichenification áreas. Fig. 2. Erythroderma of abdomen. Fig. 2. Erythroderma of abdomen. Streptolysins were reported in more than 1600, for which he was treated with amoxicillin / clavulanate, benzathine penicillin and betamethasone. Without clinical improvement, a skin biopsy was taken, finding pityriasis rubra pilaris; being treated with deflazacort, hydroxyzine, cetirizine, methotrexate and prednisone 0.5/mg/kg/hr, which he voluntarily discontinued due to edema of the extremities. Despite treatment he developed cervical, inguinal and axillary lymph node growths. Suspecting mycosis fungoides, a new skin biopsy was taken, which documented psoriasiform and spongiotic dermatitis with lymphohistiocytic infiltrate and eosinophilia. Laboratories with 20,000 leukocytes, 10,000 eosinophils, 8,000 neutrophils, 1000 lymphocytes, 359,000 platelets. Due to the presence of eosinophilia, evaluation by hematology was requested, where a new Biometry is performed with persistence of eosinophilia at levels of 9,500 to 120,000. Biopsy of bone marrow showed: hypercellular bone marrow with accentuated eosinophilia, presence of immature forms in 10 %. (Figure 3-4). Fig. 3. Hypercelular bone marrow. Fig.4. Eosinophilic promielocytes and mature forms. Fig.4. Eosinophilic promielocytes and mature forms. Fig. 3. Hypercelular bone marrow. Hypereosinophilic syndrome was diagnosed, TORCH and hepatitis panel were requested, both being negative. He started treatment with dexamethasone 16mg every 24 hours and imatinib, showing a decrease in eosinophilia and significant clinical improvement. CLINICAL CASE 2011, the International Working Group on Eosinophilic Disorders (ICOG-EO), maintained the criteria regarding the level of eosinophilia in the blood, but modified the duration of hypereosinophilia to 1 month, adding tissue eosinophilia and forms of asymptomatic, associated and eosinophilia. superimposed. 651 Volume 02 Issue 07 July 2022 Corresponding Author: Polanco Llanes Alondra S REFERENCES  Primary (myeloid), Molecular and cytogenetic studies are performed. I. Grzegorz Helbig; Amy D. Klion; (2021). Hypereosinophilic syndromes – An enigmatic group of disorders with an intriguing clinical spectrum and challenging treatment. Blood Reviews, (), –. doi:10.1016/j.blre.2021.100809  Overlay.  Overlay.  Overlay.  Idiopathic or of undetermined significance.5 The clinical presentations of HES are variable and can affect any organ. In the present case, the patient presented severe peripheral eosinophilia with erythroderma, proving dermal involvement by histopathological examination, without presenting compromise at any other systemic level. 6-7 II. Liliana María Tamayo-Quijano, Lina María Aguirre- Hernández. (2018-06-09). Recurrent erythroderma leading to the diagnosis of hypereosinophilic syndrome. Allergy Mexico, 65, supplement 1. II. Liliana María Tamayo-Quijano, Lina María Aguirre- Hernández. (2018-06-09). Recurrent erythroderma leading to the diagnosis of hypereosinophilic syndrome. Allergy Mexico, 65, supplement 1. The most frequent initial manifestations are dermatological, which are characterized by being nonspecific and variable, therefore, clinical suspicion or knowledge of the entity is deficient. III. Leru, PM Eosinophilic disorders: evaluation of the current classification and diagnostic criteria, proposal of a practical diagnostic algorithm. Clin Transl Allergy 9, 36 (2019).https://doi.org/10.1186/s13601- 019-0277-4 Expressions in the skin are generally estimated at 69%, in a series of 44 patients with positive PDGFRA-MHES they were observed in 57% and up to 80% in 21 patients with lymphocytic variant CD3-CD4+ HES (LHES). IV. Gao, Su-jun MD, PhDa; Wei, Wei MD, PhDb; Chen, Jiang-tao MD a ; Tan, Ye-hui MD, PhD; Yu, Cheng- bao MDd ; Litzow, Mark Robert MDc ; Liu, Qiu-ju MD, PhD a,c,* Hypereosinophilic Syndrome Presenting With Multiorgan Infiltration and Deep Vein Thrombosis, Medicine: August 2016 - Volume 95 - Number 35 - p e4658 doi: 10.1097/MD.0000000000004658 Cutaneous exteriorization requires a differential diagnosis with urticaria, mycosis fungoides, adverse drug reactions, contact dermatitis, and atopic dermatitis. Attention should be paid to pruritic erythematous papules, urticaria, angioedema, dermographism, oral and genital ulcers; erythema annulare centrifugum, acral blisters, and erythroderma. Histopathologic examination of the skin lesion is usually nonspecific, with viable eosinophilic infiltration. 8-9 V. V. Valent P, Klion AD, Horny HP, Roufosse F, Gotlib J, Weller PF, et al. Contemporary consensus proposal on criteria and classification of eosinophilic disorders and related conditions. J Allergy Clin Immunol 2012;130:607–12. On the other hand, pulmonary and gastrointestinal manifestations are the second manifestation observed, followed by cardiac and neurological involvement. The most common symptoms of HES are fatigue, itching, and shortness of breath. 10-11 VI. CONCLUSION The condition occurs in men with an estimated 9:1 ratio, especially in FIP1L1-PDGFR α, who are almost exclusively male. However, in the lymphoid and idiopathic variant, the sex ratio seems closer to 1:1.The most frequently affected organs are skin, lungs, intestine, heart, kidneys, eyes and the peripheral nervous system. It is more common in men with a mean age of onset of 50 years. The most serious complication is heart disease. 4 Hypereosinophilic syndrome is a rare disease, sometimes fatal, and one of the organs that can be affected is the skin. Its diagnosis most of the time is given by ruling out other pathologies, and it is possible that only one organ is affected, as in the case of our patient. When the debut is with dermal involvement, the diagnosis is often delayed due to its pleomorphic manifestations and its insidious evolution. Therefore, it is crucial that in all cases of erythroderma, hypereosinophilic syndrome be considered as a differential diagnosis, to initiate timely treatment and improve the patient's clinical prognosis. Hypereosinophilic syndrome is classified into: Hypereosinophilic syndrome is classified into:  Family or hereditary.  Secondary due to causes of reactive eosinophilia,such as helminth infection or drug hypersensitivity. Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Literature Review Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Literature Review DISCUSSION HypereosinophiliaIt is a rare, multisystemic and heterogeneous syndrome with significant mortality. It is characterized by an absolute concentration of eosinophils, greater than 1500/mm3 on two consecutive occasions, persistent at least for 1 month. It is considered moderate with concentrations between 1,500 and 5,000/mm3, and severe, when it is higher than 5,000/mm3. The 2016 World Health Organization recognizes a category of myeloid/lymphoid neoplasms with prominent eosinophilia (M/Leo) and genetic rearrangements of PDGFRA/B, FGFR1, or JAK2. In a patient with myeloid characteristics, tests for myeloid/lymphoid neoplasms with prominent eosinophilia should be performed; if this pathology is ruled out, a diagnosis of chronic eosinophilic leukemia will be considered. If secondary causes are excluded, the diagnosis of idiopathic HES is possible.1 Regarding the pathophysiology,the cytokines that stimulate eosinophil production in the bone marrow are IL-3, IL-5, and granulocyte-macrophage colony-stimulating factor (GM- CSF); produced by CD4 and CD8 T cells in peripheral blood and inflamed tissues. Of these 3 cytokines, IL-5 causes terminal differentiation of eosinophils and is the target of medical treatment. Another pathogenic mechanism is an intrinsic defect of eosinophil-committed neoplastic progenitor cells, caused by mutations involving PDGFR or FGFR1.3 The initial concept of hypereosinophilic syndrome (HES) was introduced by Hardy and Anderson in 1968, later in 1975 Chusid et al established the first diagnostic criteria: persistent absolute blood eosinophil count greater than 1500/mm3 for more than 6 months, with evidence of tissue and organ damage, without any identifiable cause of eosinophilia. In Corresponding Author: Polanco Llanes Alondra S The authors declare no conflict of int The authors declare no conflict of int The authors declare no conflict of int Corresponding Author: Polanco Llanes Alondra S CONFLICT OF INTEREST.  Secondary (lymphocytic variant) with cytokine production from clonal and/or phenotypically aberrant T cells. REFERENCES Klion AD. How I treat hypereosinophilic syndromes. Blood 2015;126:1069–77. 652 Volume 02 Issue 07 July 2022 VII. VII. Ogbogu PU, Bochner BS, Butterfield JH, Gleich GJ, Huss-Marp J, Kahn JE, et al. Hypereosinophilic syndromes: a multicenter, retrospective analysis of clinical characteristics and response to therapy. J Allergy Clin Immunol 2009;124: 1319–25. Treatment should not be delayed. Patients with life- threatening complications such as the current case, which manifests as severe erythroderma, marked eosinophilia, should be treated with high doses of corticosteroids, pending a definitive diagnosis. The response to Imatinib, a tyrosine kinase inhibitor in FIP1L1-PDGFRα positive patients, reaches almost 100%. The low dose of 100 mg/day offers hematologic and molecular remission in most patients. 12 VIII. VIII. Legrand F, Renneville A, MacIntyre E, Mastrilli S, Ackermann F, Cayuela JM, et al. The spectrum of FIP1L1-PDGFRA-associated chronic eosinophilic Corresponding Author: Polanco Llanes Alondra S 652 Volume 02 Issue 07 July 2022 Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Literature Review leukemia. New insights based on a survey of 44 cases. Medicine. 2013;92:e1–9. IX. Lefevre G, Copin MC, Staumont-Salle D, Avenel- Audran M, Aubert H, Taieb H, et al. The lymphoid variant of hypereosinophilic syndrome. Study of 21 patients with CD3-CD4+ aberrant T-cell phenotype. Medicine. 2014;93:255–66. X. Ogbogu PU, Bochner BS, Butterfield JH, Gleich GJ, Huss-Marp J, Kahn JE, et al. Hypereosinophilic syndromes: a multicenter, retrospective analysis of clinical characteristics and response to therapy. J Allergy Clin Immunol 2009;124: 1319–25. XI. Kovacs N, Benjamin Moshkovich O, Nels Symptom assessmen syndrome: toward deve tool. J Allergy Clin Im https://doi. org/10.101 2198(20)30472-4. Onli XII. Arefi M, García JL, Briz JN, Martín-Núñez G, mesylate in patient syndrome. Int J Hemato Hypereosinophilic Syndrome, Erythroderma as Clinical Debut. Case Report and Literature Review leukemia. New insights based on a survey of 44 cases. Medicine. 2013;92:e1–9. XI. Kovacs N, Benjamin Moshkovich O, Nelse XI. Kovacs N, Benjamin K, Holland-Thomas N, Moshkovich O, Nelsen LM, Ortega H, et al. Symptom assessment in hypereosinophilic syndrome: toward development of a patient-reported tool. J Allergy Clin Immunol Pract 2020;May 15. https://doi. org/10.1016/j.jaip.2020.04.069. S2213- 2198(20)30472-4. Online ahead of print. IX. Lefevre G, Copin MC, Staumont-Salle D, Avenel- Audran M, Aubert H, Taieb H, et al. The lymphoid variant of hypereosinophilic syndrome. Study of 21 patients with CD3-CD4+ aberrant T-cell phenotype. Medicine. 2014;93:255–66. XII. X. X. Ogbogu PU, Bochner BS, Butterfield JH, Gleich GJ, Huss-Marp J, Kahn JE, et al. Hypereosinophilic syndromes: a multicenter, retrospective analysis of clinical characteristics and response to therapy. J Allergy Clin Immunol 2009;124: 1319–25. Corresponding Author: Polanco Llanes Alondra S 653 Volume 02 Issue 07 July 2022 VII. XII. Arefi M, García JL, Briz MM, de Arriba F, Rodríguez JN, Martín-Núñez G, et al. Response to imatinib mesylate in patients with hypereosinophilic syndrome. Int J Hematol. 2012; 96:320–26.
https://openalex.org/W4233444944	https://www.researchsquare.com/article/rs-11554/v1.pdf?c=1579144131000	English	null	Molecular Characterization and Antibiotic Resistance of Acinetobacter baumannii in Cerebrospinal Fluid and Blood	Research Square (Research Square)	2,020	cc-by	6,480	Xiaohong Shi Department of Clinical Laboratory Shandong Provincial Qianfoshan Hospital, the First Hospital Affiliated with Shandong First Medical University, Jinan, Shandong, China Huaiqi Jing State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Ran Duan State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Ran Duan State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Ran Duan Yufeng Fan State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Zhenzhou Huang State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Zhenzhou Huang Hong Wang Department of Clinical Laboratory Shandong Provincial Qianfoshan Hospital, the First Hospital Affiliated With Shandong First Medical University, Jinan, Shandong, China Xin Wang Xin Wang State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Huaiqi Jing State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Huaiqi Jing Ran Duan State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Shuai Qin State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Shuai Qin State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Chanping, Beijing, 102206, People's Republic of China Dongyue Lv State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People's Republic of China Yufeng Fan Zhenzhou Huang Page 1/17 Page 1/17 Jiazheng Wang (  wangjiazheng518@163.com ) First Hospital Affiliated with Shandong First Medical University https://orcid.org/0000-0003-0733-169X Conclusion A high level of carbapenem resistance was detected in this study. The CC92 and bla OXA-23 gene were predominant. Five novel STs were detected, and these new STs require further investigation to understand the nature of and to prevent outbreaks caused by A. baumannii . Our study provides additional observations and epidemiological data of CSF and blood A. baumannii strains, which may improve future infection control measures and aid in potential clinical treatment in hospitals and other clinical settings. Results We observed that eighty-eight of the 94 isolates collected were resistant to imipenem or meropenem. Among them, the bla OXA-23 gene was the most prevalent carbapenemase gene with a 91.5% (86/94) detection rate, followed by the bla OXA-24 gene that showed a 2.1% (2/94) detection rate isolates. Among all CRAB observations in this study, isolates with the bla OXA-23 gene were resistant to both imipenem and meropenem. However, isolates positive for the bla OXA-24 gene but negative for the bla OXA-23 gene showed an imipenem-sensitive but meropenem-resistant phenotype. The outcome of multilocus sequence typing analysis showed 21 different STs were distinguished, of which ST195 (25.5%), ST540 (12.8%) and ST208 (11.7%) were most frequently observed. Eighty of the 94 isolates (85.1%) were clustered into CC92, and all CC92 isolates showed a carbapenem resistance phenotype (except AB13). Five novel STs were detected, and most of them were CRAB, some of which belonged to CC92. Research Keywords: Acinetobacter baumannii, cerebrospinal fluid, CSF, blood, blaOXA-23, CC92, carbapenem resistance, infection Posted Date: January 15th, 2020 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. d ll License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Version of Record: A version of this preprint was published at PLOS ONE on February 22nd, 2021. See the published version at https://doi.org/10.1371/journal.pone.0247418. Page 2/17 Methods A total of 50 nonrepetitive CSF A. baumannii isolates and 44 blood isolates were collected. The resistance phenotypes were determined according to the Clinical and Laboratory Standards Institute guidelines. We performed Polymerase Chain Reaction (PCR) experiments to detect the carbapenem resistance mechanism. Finally, we conducted Multilocus Sequence Typing (MLST) to depict the genetic relatedness of these isolates. Background The increasing rates of carbapenem-resistant Acinetobacter baumannii (CRAB) caused nosocomial infections generate significant comorbidity and sometime cause death among patients. Current treatment options are limited. These infections pose great difficulties for infection control and clinical treatment. This study identifies the antimicrobial resistance, carbapenemases and genetic relatedness of A. baumannii isolates from cerebrospinal fluid (CSF) and blood in a hospital in Shandong, China. Background Acinetobacter baumannii is a nonfermentative, gram-negative opportunistic pathogen that often causes disease among immunocompromised patients [1]. A. baumannii is often found in hospitals causing a Page 3/17 Page 3/17 variety of nosocomial infectious diseases, including bloodstream infections, urinary tract infections, meningitis and wound infections [2]. In recent years, A. baumannii has become an important bacterium to identify when treating and controlling infectious diseases because of its remarkable ability to evolve extensive drug resistance to many antibiotics [3]. variety of nosocomial infectious diseases, including bloodstream infections, urinary tract infections, meningitis and wound infections [2]. In recent years, A. baumannii has become an important bacterium to identify when treating and controlling infectious diseases because of its remarkable ability to evolve extensive drug resistance to many antibiotics [3]. Some observations are referred to as Critical Values because these laboratory values may indicate an urgent and a life-threatening situation for patients where treatment protocols indicate immediate therapy must be initiated. This includes microbiology results identifying the bacteria in cerebrospinal fluid (CSF) and blood [4, 5]. Therefore, the presence of A. baumannii strains in CSF and blood is a Critical Value to identify for practitioners as it can poses great difficulties for clinical treatment options. Carbapenems are considered to be the most effective antibiotics against many multidrug-resistant bacteria [6]. However, the increase in carbapenem-resistant Acinetobacter baumannii (CRAB) isolates has recently become a global concern. CHINET surveillance data showed that from 2005 to 2018, the resistance of A. baumannii to imipenem and meropenem increased approximately twofold [7, 8]. The production of carbapenemase is one of the most common and important mechanisms for A. baumannii resistance to carbapenems. Among all carbapenemases, OXA-type (mainly OXA-23, OXA-24, OXA-48, OXA-51 and OXA-58) was the most prevalent [9-11]. In addition, because the coding genes are located in transferable genetic elements and can spread among A. baumannii and even into other bacteria [12, 13], New Delhi metallo-β-lactamase (NDM) and Klebsiella pneumoniae carbapenemase (KPC) producers have also shown significant importance for worldwide prevalence [14, 15]. The multilocus sequence typing (MLST) method has been widely used to depict genetic relatedness and for molecular epidemiological studies of A. baumannii [16]. Previous molecular epidemiological studies have shown that CC92 is highly prevalent throughout China, and it is the clonal complex (CC) with the widest global distribution [17, 18]. In this work, we observe, investigate and characterized 94 A. Background baumannii isolates obtained from CSF and blood we acquired from patients in university hospital in Shandong, China between 2014 and 2019. Distribution of carbapenem resistance genes All isolates were detected for the presence of carbapenem resistance genes (Table 1). In the 94 A. baumannii isolates, all of them harbored the intrinsic blaOXA-51 gene. In contrast, none of them had blaOXA-48, blaOXA-58, NDM-1 or KPC genes. The blaOXA-23 gene was the most prevalent carbapenemase gene with a 91.5% (86/94) detection rate, followed by the blaOXA-24 gene in only 2 (2.1%) isolates. In this study, all of the isolates with blaOXA-23 or blaOXA-24 genes were carbapenem resistant. Among them, isolates with the blaOXA-23 gene were resistant to both imipenem and meropenem. However, isolates positive for the blaOXA-24 gene but negative for the blaOXA-23 gene showed an imipenem-sensitive but meropenem-resistant phenotype. Table 1 Positive rates of carbapenem resistance genes No. of strains Rate of gene (%) OXA-23 OXA-24 OXA-48 OXA-51 OXA-58 NDM1 KPC 94 91.5 2.1 0 100 0 0 0 Distribution of bacterial isolates A. baumannii isolates were collected from two different locations, CSF (n=50) and blood (n=44), from the First Hospital Affiliated with Shandong First Medical University. For the CSF isolates, one was collected in 2014, four in 2015, seventeen in 2016, fifteen in 2017, five in 2018 and eight in 2019. Forty-one of these isolates were obtained from neurosurgery, six from the ICU and three from other wards. Regarding the A. baumannii isolates samples obtained from blood; twelve were collected in 2016, thirteen in 2017, eleven in 2018 and eight in 2019. Twenty-eight blood isolates were obtained from the ICU, six from neurosurgery, four from hematopathology and six from other wards. Page 4/17 Antibiotic susceptibility The antimicrobial susceptibility profiles of 94 A. baumannii strains were shown in Fig.1 and Table S1 in Supplemental Material. More than 90% of the A. baumannii isolates were resistant to ticarcillin/clavulanic acid, piperacillin/tazobactam, ceftazidime and ciprofloxacin. A total of 77.7% of th isolates were resistant to tobramycin, 71.3% were resistant to levofloxacin, 59.6% to trimethoprim/sulfamethoxazole and 52.1% to cefepime. The resistance rate for minocycline and cefoperazone/sulbactam, which are commonly used for CRAB infection, were 19.1% and 45.7% respectively. In contrast, only seven isolates (7.4%) were resistant to tigecycline and all isolates were sensitive to colistin. Among all 94 A. baumannii isolates, 86 (93.6%) isolates were resistant to both imipenem and meropenem. Two isolates showed a meropenem-resistant but imipenem-sensitive phenotype. MLST profile The MLST analysis revealed a total of 21 different STs, including 16 existing STs and 5 novel STs we identified in this study (the new STs were submitted for ST assignment which were ST1967, ST1968, ST1969, ST1970 and ST1971). The profiles of the newly identified ST types are listed in Table 2. Among Page 5/17 Page 5/17 them, 89.4% (84/94) were represented by 11 main STs (having ≥2 isolates), and the prevalent STs were ST195, ST540 and ST208, accounting for 25.5% (24/94), 12.8% (12/94), and 11.7% (11/94), respectively (Fig.2 and Table S2 in the supplemental material). them, 89.4% (84/94) were represented by 11 main STs (having ≥2 isolates), and the prevalent STs were ST195, ST540 and ST208, accounting for 25.5% (24/94), 12.8% (12/94), and 11.7% (11/94), respectively (Fig.2 and Table S2 in the supplemental material). Twelve STs representing 85.1% (80/94) of the isolates were clustered into CC92, with up to 18 different allelic profiles and 56 different isolates being represented. In addition, 9 individual STs accounted for 14 isolates (The detailed MLST profiles can be seen in Table S2 in Supplemental Material). Table 2 Allelic profiles of the new STs found in this study STs gltA gyrB gdhB recA cpn60 gpi rpoD 1967 1 34 3 2 2 178 3 1968 1 3 3 2 2 113 3 1969 1 17 135 12 23 98 6 1970 1 3 3 2 2 160 4 1971 36 34 59 28 4 279 3 MLST, antibiotic susceptibility and carbapenem resistance genes Discussion CSF and blood infection of A. baumannii (especially for CRAB) may be life-threatening and bring great obstacles for clinical treatment [4, 5]. This study offers insight into the molecular characterization and antibiotic resistance of A. baumannii from CSF and blood. CSF and blood infection of A. baumannii (especially for CRAB) may be life-threatening and bring great obstacles for clinical treatment [4, 5]. This study offers insight into the molecular characterization and antibiotic resistance of A. baumannii from CSF and blood. In China, from 2005 to 2018, the resistance rate of A. baumannii for imipenem and meropenem increased from 32.9% to 71.7% and from 41.3% to 78.1%, respectively [7, 8]. Compared to the CHINET surveillance data, the resistance rates of A. baumannii for imipenem and meropenem in our experiment were 91.5% and 93.6%, respectively, higher than the surveillance data. For CSF and blood infection, the recommended dose of antibiotics was usually higher and needed a longer course of treatment than the case of superficial infection. However the CHINET surveillance data contained antibiotic resistance data for bacteria isolated from various sites, including some superficial infections. This might be the cause of the higher drug resistance rate in our study. In addition, the CHINET surveillance data [7, 8, 19, 20] demonstrated that the resistance rate of A. baumannii to meropenem is slightly higher than that to imipenem. This is similar to our observations and results. In fact, our results also support the view that imipenem is more bactericidal [21] and has a higher T>MIC value [22] than meropenem against A. baumannii. In contrast, the resistance rate to tigecycline was only 7.4%, and all isolates showed colistin sensitive phenotype. The resistance rates of these two drugs were far below carbapenem and the other antibiotics tested in this study, suggesting that these two drugs might serve as therapeutic agents to control infections. To investigate the mechanism of carbapenem resistance, carbapenemase-encoding genes were tested in this study. Our results showed that the blaOXA-23 gene existed in most CRAB isolates but was absent for most CSAB isolates, which indicated that blaOXA-23 was the major mechanism for carbapenem resistance of CRAB isolated from CSF and blood. The blaOXA-23 gene was also the most important mechanism for carbapenem resistance in CRAB in China [18, 23] and some other countries [24-26]. MLST, antibiotic susceptibility and carbapenem resistance genes MLST, antibiotic susceptibility and carbapenem resistance genes All of the isolates grouped into CC92 were carbapenem-resistant A. baumannii except for one isolate (AB13). These isolates were also not sensitive to ticarcillin/clavulanic acid, piperacillin/tazobactam, ceftazidime, cefepime, ciprofloxacin and levofloxacin but had variable susceptibilities to cefoperazone/sulbactam, tobramycin, minocycline, tigecycline and trimethoprim/sulfamethoxazole. In contrast, all of the CSAB isolates belonged to individual STs except for one isolate (AB13). These isolates were also sensitive to the other 12 antibiotics tested in this study. Five novel STs were identified in this study. All 4 ST1967 were isolated from CSF and were non-sensitive to all β-lactam antibiotics and quinolones. For ST 1968 isolates, 5 were non-sensitive to 13 antibiotics (except colistin) tested in this study, but the other one isolate (AB73) was sensitive to tobramycin, trimethoprim/sulfamethoxazole and colistin. ST1970 isolate was non-sensitive to 12 antibiotics (except tigecycline and colistin) tested in this study, whereas the ST1971 isolate was sensitive to all antibiotics. ST1969 contained two isolates (AB29 and AB70). Both of these isolates possessed the blaOXA-24 gene and were resistant to meropenem but sensitive to imipenem. Page 6/17 Discussion On the other hand, the blaOXA-24 gene was another mechanism for carbapenem resistance, as blaOXA-24-positive but blaOXA-23- negative isolates in this study showed meropenem-resistance and imipenem-sensitive characteristics. BlaOXA-24-positive A. baumannii strains have been reported in many countries [9, 27, 28], especially in Spain, where blaOXA-24 was the most prevalent gene [9]. Interestingly, even though most blaOXA-24-positive isolates have been reported to be resistant to both imipenem and meropenem, in our experiment, blaOXA- 24-positive strains showed an imipenem-sensitive but meropenem-resistant phenotype. Some molecular biological mechanisms have been reported in many gram-negative bacteria to explain the imipenem- sensitive but meropenem-resistant phenomenon. For example, the transmission of the blaIMP-6 and blaCTX-M-2 plasmids [29] as well as the absence of OmpK35 and the frame shift mutation in OmpK36 [30] have been shown to be important mechanisms for imipenem-sensitive but meropenem-resistant Klebsiella pneumoniae (ISMRKP) strains. Substrate specificities of efflux pumps lead to different drug resistance characteristics for Pseudomonas aeruginosa. As a specific substrate, meropenem could be extruded by many efflux pumps, but imipenem was not affected by these efflux systems [31]. As a result, Page 7/17 Page 7/17 some imipenem-sensitive but meropenem-resistant Pseudomonas aeruginosa strains were detected. However, studies examining this mechanism against A. baumannii are scarce. Both of these strains belonged to a novel ST (ST1969). This molecular mechanism for A. baumannii strains will require further investigation. some imipenem-sensitive but meropenem-resistant Pseudomonas aeruginosa strains were detected. However, studies examining this mechanism against A. baumannii are scarce. Both of these strains belonged to a novel ST (ST1969). This molecular mechanism for A. baumannii strains will require further investigation. ST540, ST195 and ST208 were three major STs for A. baumannii isolated from CSF and blood. Among them, ST195 and ST208 are two dominant STs currently found in China [32-34]. Although ST540 was not the main ST observed in China, it was shown that ST540 was not only one of the three common STs but also the predicted founder of the CC for A. baumannii isolated from blood and CSF. On a global scale, CC92 was the largest and most geographically diverse CC, which was widespread in many countries [34], including China [23]. Although we did not find ST92 isolates (the predicted founder of CC92) in this study, it was shown that most A. baumannii isolated from CSF and blood belonged to CC92 (CC92 was the unique CC clonal group tested). Discussion In addition, 79 of 80 CC92 isolates were CRAB (except for one isolate, AB13), whereas only approximately 50% of the CC92 isolates were resistant to imipenem or meropenem, which suggested that CC92 isolates tend to acquire carbapenem resistance determinants. A previous study reported that CC92 is a widespread variant that has advantages in acquiring resistance determinants and surviving in the nosocomial environment, which renders it preferentially selected under antibiotic pressure [18]. This is a possible reason for the high detection rate of CC92 A. baumannii isolates in CSF and blood and the high correlation between CC92 and carbapenem resistance characteristics. We also observed a total of five new STs. Among them, two novel STs were classified into CC92 and others were individual STs, which suggested that A. baumannii isolates were diverse and still in clonal expansion. As 13 of the 14 isolates in the five new STs were identified as CRAB, close attention should be paid toward these new STs to identify and further limit both transmission and outbreaks. Conclusion In summary, with our study, we described the molecular characterization and antibiotic resistance of A. baumannii from CSF and blood in a hospital in Shandong, China. A high level of carbapenem resistance was detected. The CC92 and blaOXA-23 gene were predominant in this hospital. Five novel STs were detected, and most of them were CRAB, some of which belonged to CC92. This study offers new epidemiological data of CSF and blood A. baumannii strains, which may help to improve infection control measures and clinical treatment in hospitals. Antimicrobial susceptibility test All A. baumannii strains were tested for susceptibility to 14 antibiotics, including ticarcillin/clavulanic acid, piperacillin/tazobactam, ceftazidime, cefoperazone/sulbactam, cefepime, imipenem, meropenem, tobramycin, ciprofloxacin, levofloxacin, minocycline, tigecycline, colistin and trimethoprim/sulfamethoxazole by using a Vitek 2 compact system (bioMérieux, Marcy, France) with AST-N-335 cards. The results were evaluated according to the Clinical and Laboratory Standards Institute (CLSI) criteria except for tigecycline were adapted from the United States Food and Drug Administration breakpoints. Bacterial isolates A total of 94 nonrepetitive A. baumannii isolates were obtained from CSF or blood samples of patients from different departments (neurosurgery, ICU, hematopathology and other wards) at the First Hospital Page 8/17 Page 8/17 Affiliated with Shandong First Medical University (Shandong, China). These samples were obtained from 2014 to 2019. All isolates were identified using MALDI-TOF MS (Bruker) and further verified by PCR products of 16S rDNA sequencing [35]. PCR products were sequenced by TsingkeBioTech Co., Ltd., followed by sequence alignment on the NCBI database. Affiliated with Shandong First Medical University (Shandong, China). These samples were obtained from 2014 to 2019. All isolates were identified using MALDI-TOF MS (Bruker) and further verified by PCR products of 16S rDNA sequencing [35]. PCR products were sequenced by TsingkeBioTech Co., Ltd., followed by sequence alignment on the NCBI database. Table 3 Table 3 Primers used with their respective annealing temperatures. Primer Sequence Amplicon length (bp) Annealing temp (°C) Ref OXA-51- F 5’-ATGAACATTAAAGCACTC-3’ 353 bp 46 °C [36] OXA-51- R 5’-CTATAAAATACCTAATTGTTC-3’ OXA-23- F 5’-GATCGGATTGGAGAACCAGA-3’ 501 bp 53 °C [37] OXA-23- R 5’-ATTTCTGACCGCATTTCCAT-3’ OXA-24- F 5’-GGTTAGTTGGCCCCCTTAAA-3’ 246 bp 53 °C [37] OXA-24- R 5’-AGTTGAGCGAAAAGGGGATT-3’ OXA-58- F 5’-AAGTATTGGGGCTTGTGCTG-3’ 599 bp 53 °C [37] OXA-58- R 5’-CCCCTCTGCGCTCTACATAC-3’ KPC-F 5’-GCTCAGGCGCAACTGTAAGT-3’ 823 bp 55 °C [38] KPC-R 5’-GTCCAGACGGAACGTGGTAT-3’ NDM-1-F 5’-TCTCGACATGCCGGGTTTCGG- 3’ 475 bp 55 °C [38] NDM-1-R 5’-ACCGAGATTGCCGAGCGACTT-3 OXA-48- F 5’-GCGTGGTTAAGGATGAACAC-3’ 438 bp 52 °C [39] OXA-48- R 5’-CATCAAGTTCAACCCAACCG-3’ Multilocus Sequence Typing (MLST) Multilocus sequence typing (MLST) analyses were performed using the Oxford scheme [16], publicly available from the https://pubmlst.org/abaumannii/info/primers_Oxford.shtml. This MLST scheme is based on the sequencing of fragments of seven housekeeping genes: citrate synthase (gltA) DNA gyrase Primers used with their respective annealing temperatures. [37] PCR Experiments PCR assays were carried out using conventional PCR amplification. The target genes included the blaOXA- 51, blaOXA-23, blaOXA-24, blaOXA-58, blaOXA-48, blaNDM-1, and blaKPC genes. Table 3 shows the sequences used for primer design and the annealing temperatures. The reaction conditions of the PCR programs consisted of an initial elongation at 94°C for 5 minutes; followed by 30 cycles of 94°C for 30 seconds, the respective annealing temperatures (Table 3) for 30 seconds, and 72°C for 1 minute; and a final extension step at 72°C for 10 minutes. Page 9/17 Page 9/17 Page 9/17 Availability of data and materials The datasets during and analyze during the current study available from the corresponding author on reasonable request. Acknowledgements Not applicable. Multilocus Sequence Typing (MLST) Multilocus sequence typing (MLST) analyses were performed using the Oxford scheme [16], publicly available from the https://pubmlst.org/abaumannii/info/primers_Oxford.shtml. This MLST scheme is based on the sequencing of fragments of seven housekeeping genes: citrate synthase (gltA), DNA gyrase subunit B (gyrB), glucose dehydrogenase B (gdhB), homologous recombination factor (recA), 60-kDa chaperonin (cpn60), glucose-6-phosphate isomerase (gpi), and RNA polymerase sigma factor (rpoD). Page 10/17 Page 10/17 Amplification reactions were carried out as described previously [18, 23], and sequencing was performed in both directions. Amplification reactions were carried out as described previously [18, 23], and sequencing was performed in both directions. The sequences were aligned with the reference sequence from the MLST database (https://pubmlst.org/abaumannii/) using MEGA (version 4.0). BioEdit (version 7.0.1) was used to determine the allele assignments of the housekeeping genes before composing a profile of each strain. The newly identified STs were submitted to the MLST database curator for approval, and a number was assigned. A minimum-spanning tree using the allelic difference between isolates of the seven housekeeping genes was constructed using BioNumerics (Applied Math). Abbreviations CRAB: carbapenem resisitant Acinetobacter baumannii; A. baumannii: Acinetobacter baumannii; CSF: cerebrospinal fluid; CLSI: Clinical and Laboratory Standards Institute; PCR: polymerase chain reaction; MLST: multilocus sequence typing; STs: sequence types; CC: clonal complex; CHINET: China antimicrobial surveillance network; NDM: New Delhi metallo-β-lactamase; KPC: Klebsiella pneumoniae carbapenemase; ICU: intensive care unit; T: time; MIC: minimum inhibitory concentration; CSAB: carbapenem sensitive Acinetobacter baumannii; ISMRKP: imipenem sensitive but meropenem resistance Klebsiella pneumonia; MALDI-TOF MS: matrix assisted laser desorption ionization time of flight mass spectrometry; NCBI: National Center for Biotechnology Information; T/C: ticarcillin/clavulanic acid; P/T: piperacillin/tazobactam; CAZ: ceftazidime; CSL: cefoperazone/sulbactam; FEP: cefepime; IPM: imipenem; MEM: meropenem; TOB: tobramycin; CIP: ciprofloxacin; LEV: levofloxacin; MNO: minocycline; TGC: tigecycline; COL: colistin; SXT: sulfamethoxazole/trimethoprim. imipenem; MEM: meropenem; TOB: tobramycin; CIP: ciprofloxacin; LEV: levofloxacin; MNO: minocycline; TGC: tigecycline; COL: colistin; SXT: sulfamethoxazole/trimethoprim. Authors’ contributions JZW conceived the idea and designed the experiment. JZW, XHS and HW analyzed the results. JZW, XHS and HW drafted the manuscript. RD, SQ, DYL, YFF and ZZH performed the experiment. XW, HQJ, LZ and BKS participated in manuscript revision. All authors read and approved the final manuscript. XHS and HW contributed equally to this work. XHS and HW contributed equally to this work. Page 11/17 Funding This work was supported by the National Sci-Tech Key Project (2018ZX10713-003-002, 2018ZX10713- 001-002). This work was supported by the National Sci-Tech Key Project (2018ZX10713-003-002, 2018ZX10713- 001-002). Consent for publication Not applicable. Author details 1 Department of Clinical Laboratory Shandong Provincial Qianfoshan Hospital, the First Hospital Affiliated with Shandong First Medical University, Jinan, Shandong, China. 2 State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People’s Republic of China. 3 Beijing Advanced Innovation Center for Big Data-Based Precision Medicine, Bei hang University, Beijing, China. 2 State Key Laboratory for Infectious Disease Prevention and Control, National Institute for Communicable Disease Control and Prevention, Changping, Beijing, 102206, People’s Republic of China. 3 Beijing Advanced Innovation Center for Big Data-Based Precision Medicine, Bei hang University, Beijing, China. 4 Shandong Institute of Industrial Technology for Health Sciences and Precision Medicine, Jinan, China. 5 Luddy School of Informatics, Computing and Engineering, Indiana University, Bloomington, IN, USA Ethics approval and consent to participate 4 Shandong Institute of Industrial Technology for Health Sciences and Precision Medicine, Jinan, China. 5 Luddy School of Informatics, Computing and Engineering, Indiana University, Bloomington, IN, USA Ethics approval and consent to participate Not applicable. Competing interests Authors declare that they have no competing interests. Authors declare that they have no competing interests. References Dahdouh E, Gomez-Gil R, Pacho S, Mingorance J, Daoud Z, Suarez M: Clonality, virulence determinants, and profiles of resistance of clinical Acinetobacter baumannii isolates obtained from a Spanish hospital. PloS one 2017, 12(4):e0176824. 10. Nowak P, Paluchowska P: Acinetobacter baumannii: biology and drug resistance-role of carbapenemases. Folia histochemica et cytobiologica 2016, 54(2):61-74. 10. Nowak P, Paluchowska P: Acinetobacter baumannii: biology and drug resistance-role of carbapenemases. Folia histochemica et cytobiologica 2016, 54(2):61-74. 11. Evans BA, Amyes SG: OXA beta-lactamases. Clinical microbiology reviews 2014, 27(2):241-263. 11. Evans BA, Amyes SG: OXA beta-lactamases. Clinical microbiology reviews 2014, 27(2):241-263. 12. Pagano M, Martins AF, Barth AL: Mobile genetic elements related to carbapenem resistance in Acinetobacter baumannii. Brazilian journal of microbiology : [publication of the Brazilian Society for Microbiology] 2016, 47(4):785-792. 12. Pagano M, Martins AF, Barth AL: Mobile genetic elements related to carbapenem resistance in Acinetobacter baumannii. Brazilian journal of microbiology : [publication of the Brazilian Society for Microbiology] 2016, 47(4):785-792. 13. Partridge SR, Kwong SM, Firth N, Jensen SO: Mobile Genetic Elements Associated with Antimicrobial Resistance. Clinical microbiology reviews 2018, 31(4). 13. Partridge SR, Kwong SM, Firth N, Jensen SO: Mobile Genetic Elements Associated with Antimicrobial Resistance. Clinical microbiology reviews 2018, 31(4). 14. Azimi L, Talebi M, Pourshafie MR, Owlia P, Rastegar Lari A: Characterization of Carbapenemases in Extensively Drug Resistance Acinetobacter baumannii in a Burn Care Center in Iran. International journal of molecular and cellular medicine 2015, 4(1):46-53. 14. Azimi L, Talebi M, Pourshafie MR, Owlia P, Rastegar Lari A: Characterization of Carbapenemases in Extensively Drug Resistance Acinetobacter baumannii in a Burn Care Center in Iran. International journal of molecular and cellular medicine 2015, 4(1):46-53. 15. Chen Y, Zhou Z, Jiang Y, Yu Y: Emergence of NDM-1-producing Acinetobacter baumannii in China. The Journal of antimicrobial chemotherapy 2011, 66(6):1255-1259. 15. Chen Y, Zhou Z, Jiang Y, Yu Y: Emergence of NDM-1-producing Acinetobacter baumannii in China. The Journal of antimicrobial chemotherapy 2011, 66(6):1255-1259. 16. Bartual SG, Seifert H, Hippler C, Luzon MA, Wisplinghoff H, Rodriguez-Valera F: Development of a multilocus sequence typing scheme for characterization of clinical isolates of Acinetobacter baumannii. Journal of clinical microbiology 2005, 43(9):4382-4390. 16. Bartual SG, Seifert H, Hippler C, Luzon MA, Wisplinghoff H, Rodriguez-Valera F: Development of a multilocus sequence typing scheme for characterization of clinical isolates of Acinetobacter baumannii. Journal of clinical microbiology 2005, 43(9):4382-4390. 17. References 1. Hu Y, He L, Tao X, Meng F, Zhang J: Biofilm may not be Necessary for the Epidemic Spread of Acinetobacter baumannii. Scientific reports 2016, 6:32066. 2. Peleg AY, Seifert H, Paterson DL: Acinetobacter baumannii: emergence of a successful pathogen. Clinical microbiology reviews 2008, 21(3):538-582. 3. Fournier PE, Vallenet D, Barbe V, Audic S, Ogata H, Poirel L, Richet H, Robert C, Mangenot S, Abergel C et al: Comparative genomics of multidrug resistance in Acinetobacter baumannii. PLoS genetics 2006, 2(1):e7. 4. Chapman CN, Otis CN: From critical values to critical diagnoses: a review with an emphasis on cytopathology. Cancer cytopathology 2011, 119(3):148-157. 4. Chapman CN, Otis CN: From critical values to critical diagnoses: a review with an emphasis on cytopathology. Cancer cytopathology 2011, 119(3):148-157. Page 12/17 Page 12/17 5. Howanitz PJ, Steindel SJ, Heard NV: Laboratory critical values policies and procedures: a college of American Pathologists Q-Probes Study in 623 institutions. Archives of pathology & laboratory medicine 2002, 126(6):663-669. 6. Biglari S, Alfizah H, Ramliza R, Rahman MM: Molecular characterization of carbapenemase and cephalosporinase genes among clinical isolates of Acinetobacter baumannii in a tertiary medical centre in Malaysia. Journal of medical microbiology 2015, 64(Pt 1):53-58. 6. Biglari S, Alfizah H, Ramliza R, Rahman MM: Molecular characterization of carbapenemase and cephalosporinase genes among clinical isolates of Acinetobacter baumannii in a tertiary medical centre in Malaysia. Journal of medical microbiology 2015, 64(Pt 1):53-58. 7. Hu FP, Guo Y, Zhu DM, Wang F, Jiang XF, Xu YC, Zhang XJ, Zhang CX, Ji P, Xie Y et al: Resistance trends among clinical isolates in China reported from CHINET surveillance of bacterial resistance, 2005-2014. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 2016, 22 Suppl 1:S9-14. 8. Hu F, Guo Y, Yang Y, Zheng Y, Wu S, Jiang X, Zhu D, Wang F: Resistance reported from China antimicrobial surveillance network (CHINET) in 2018. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 2019. 8. Hu F, Guo Y, Yang Y, Zheng Y, Wu S, Jiang X, Zhu D, Wang F: Resistance reported from China antimicrobial surveillance network (CHINET) in 2018. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 2019. 9. References Runnegar N, Sidjabat H, Goh HM, Nimmo GR, Schembri MA, Paterson DL: Molecular epidemiology of multidrug-resistant Acinetobacter baumannii in a single institution over a 10-year period. Journal of clinical microbiology 2010, 48(11):4051-4056. 17. Runnegar N, Sidjabat H, Goh HM, Nimmo GR, Schembri MA, Paterson DL: Molecular epidemiology of multidrug-resistant Acinetobacter baumannii in a single institution over a 10-year period. Journal of clinical microbiology 2010, 48(11):4051-4056. 18. Ruan Z, Chen Y, Jiang Y, Zhou H, Zhou Z, Fu Y, Wang H, Wang Y, Yu Y: Wide distribution of CC92 carbapenem-resistant and OXA-23-producing Acinetobacter baumannii in multiple provinces of China. International journal of antimicrobial agents 2013, 42(4):322-328. 18. Ruan Z, Chen Y, Jiang Y, Zhou H, Zhou Z, Fu Y, Wang H, Wang Y, Yu Y: Wide distribution of CC92 carbapenem-resistant and OXA-23-producing Acinetobacter baumannii in multiple provinces of China. International journal of antimicrobial agents 2013, 42(4):322-328. Page 13/17 Page 13/17 19. Xiao YH, Giske CG, Wei ZQ, Shen P, Heddini A, Li LJ: Epidemiology and characteristics of antimicrobial resistance in China. Drug resistance updates : reviews and commentaries in antimicrobial and anticancer chemotherapy 2011, 14(4-5):236-250. 20. Hu F, Zhu D, Wang F, Wang M: Current Status and Trends of Antibacterial Resistance in China. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2018, 67(suppl_2):S128-S134. 21. Novelli A, Adembri C, Livi P, Fallani S, Mazzei T, De Gaudio AR: Pharmacokinetic evaluation of meropenem and imipenem in critically ill patients with sepsis. Clinical pharmacokinetics 2005, 44(5):539-549. 22. Kuti JL, Florea NR, Nightingale CH, Nicolau DP: Pharmacodynamics of meropenem and imipenem against Enterobacteriaceae, Acinetobacter baumannii, and Pseudomonas aeruginosa. Pharmacotherapy 2004, 24(1):8-15. 23. Fu Y, Zhou J, Zhou H, Yang Q, Wei Z, Yu Y, Li L: Wide dissemination of OXA-23-producing carbapenem-resistant Acinetobacter baumannii clonal complex 22 in multiple cities of China. The Journal of antimicrobial chemotherapy 2010, 65(4):644-650. 24. Merino M, Poza M, Roca I, Barba MJ, Sousa MD, Vila J, Bou G: Nosocomial outbreak of a multiresistant Acinetobacter baumannii expressing OXA-23 carbapenemase in Spain. Microbial drug resistance (Larchmont, NY) 2014, 20(4):259-263. 24. Merino M, Poza M, Roca I, Barba MJ, Sousa MD, Vila J, Bou G: Nosocomial outbreak of a multiresistant Acinetobacter baumannii expressing OXA-23 carbapenemase in Spain. Microbial drug resistance (Larchmont, NY) 2014, 20(4):259-263. 25. References Qu J, Du Y, Yu R, Lu X: The First Outbreak Caused by Acinetobacter baumannii ST208 and ST195 in China. BioMed research international 2016, 2016:9254907. 32. Qu J, Du Y, Yu R, Lu X: The First Outbreak Caused by Acinetobacter baumannii ST208 and ST195 in China. BioMed research international 2016, 2016:9254907. 33. Huang G, Yin S, Gong Y, Zhao X, Zou L, Jiang B, Dong Z, Chen Y, Chen J, Jin S et al: Multilocus Sequence Typing Analysis of Carbapenem-Resistant Acinetobacter baumannii in a Chinese Burns Institute. Frontiers in microbiology 2016, 7:1717. 33. Huang G, Yin S, Gong Y, Zhao X, Zou L, Jiang B, Dong Z, Chen Y, Chen J, Jin S et al: Multilocus Sequence Typing Analysis of Carbapenem-Resistant Acinetobacter baumannii in a Chinese Burns Institute. Frontiers in microbiology 2016, 7:1717. 34. Karah N, Sundsfjord A, Towner K, Samuelsen O: Insights into the global molecular epidemiology of carbapenem non-susceptible clones of Acinetobacter baumannii. Drug resistance updates : reviews and commentaries in antimicrobial and anticancer chemotherapy 2012, 15(4):237-247. 35. Hao H, Liang J, Duan R, Chen Y, Liu C, Xiao Y, Li X, Su M, Jing H, Wang X: Yersinia spp. Identification Using Copy Diversity in the Chromosomal 16S rRNA Gene Sequence. PloS one 2016, 11(1):e0147639. 36. Turton JF, Woodford N, Glover J, Yarde S, Kaufmann ME, Pitt TL: Identification of Acinetobacter baumannii by detection of the blaOXA-51-like carbapenemase gene intrinsic to this species. Journal of clinical microbiology 2006, 44(8):2974-2976. 36. Turton JF, Woodford N, Glover J, Yarde S, Kaufmann ME, Pitt TL: Identification of Acinetobacter baumannii by detection of the blaOXA-51-like carbapenemase gene intrinsic to this species. Journal of clinical microbiology 2006, 44(8):2974-2976. 37. Mostachio AK, van der Heidjen I, Rossi F, Levin AS, Costa SF: Multiplex PCR for rapid detection of genes encoding oxacillinases and metallo-beta-lactamases in carbapenem-resistant Acinetobacter spp. Journal of medical microbiology 2009, 58(Pt 11):1522-1524. 37. Mostachio AK, van der Heidjen I, Rossi F, Levin AS, Costa SF: Multiplex PCR for rapid detection of genes encoding oxacillinases and metallo-beta-lactamases in carbapenem-resistant Acinetobacter spp. Journal of medical microbiology 2009, 58(Pt 11):1522-1524. 38. Chang Y, Luan G, Xu Y, Wang Y, Shen M, Zhang C, Zheng W, Huang J, Yang J, Jia X et al: Characterization of carbapenem-resistant Acinetobacter baumannii isolates in a Chinese teaching hospital. Frontiers in microbiology 2015, 6:910. 38. References Chang Y, Luan G, Xu Y, Wang Y, Shen M, Zhang C, Zheng W, Huang J, Yang J, Jia X et al: Characterization of carbapenem-resistant Acinetobacter baumannii isolates in a Chinese teaching hospital. Frontiers in microbiology 2015, 6:910. 39. Poirel L, Walsh TR, Cuvillier V, Nordmann P: Multiplex PCR for detection of acquired carbapenemase genes. Diagnostic microbiology and infectious disease 2011, 70(1):119-123. 39. Poirel L, Walsh TR, Cuvillier V, Nordmann P: Multiplex PCR for detection of acquired carbapenemase genes. Diagnostic microbiology and infectious disease 2011, 70(1):119-123. References El Bannah AMS, Nawar NN, Hassan RMM, Salem STB: Molecular Epidemiology of Carbapenem- Resistant Acinetobacter baumannii in a Tertiary Care Hospital in Egypt: Clonal Spread of blaOXA-23. Microbial drug resistance (Larchmont, NY) 2018, 24(3):269-277. 25. El Bannah AMS, Nawar NN, Hassan RMM, Salem STB: Molecular Epidemiology of Carbapenem- Resistant Acinetobacter baumannii in a Tertiary Care Hospital in Egypt: Clonal Spread of blaOXA-23. Microbial drug resistance (Larchmont, NY) 2018, 24(3):269-277. 26. Zowawi HM, Sartor AL, Sidjabat HE, Balkhy HH, Walsh TR, Al Johani SM, AlJindan RY, Alfaresi M, Ibrahim E, Al-Jardani A et al: Molecular epidemiology of carbapenem-resistant Acinetobacter baumannii isolates in the Gulf Cooperation Council States: dominance of OXA-23-type producers. Journal of clinical microbiology 2015, 53(3):896-903. 26. Zowawi HM, Sartor AL, Sidjabat HE, Balkhy HH, Walsh TR, Al Johani SM, AlJindan RY, Alfaresi M, Ibrahim E, Al-Jardani A et al: Molecular epidemiology of carbapenem-resistant Acinetobacter baumannii isolates in the Gulf Cooperation Council States: dominance of OXA-23-type producers. Journal of clinical microbiology 2015, 53(3):896-903. 27. Castanheira M, Wanger A, Kruzel M, Deshpande LM, Jones RN: Emergence and clonal dissemination of OXA-24- and OXA-58-producing Acinetobacter baumannii strains in Houston, Texas: report from the SENTRY Antimicrobial Surveillance Program. Journal of clinical microbiology 2008, 46(9):3179- 3180. 27. Castanheira M, Wanger A, Kruzel M, Deshpande LM, Jones RN: Emergence and clonal dissemination of OXA-24- and OXA-58-producing Acinetobacter baumannii strains in Houston, Texas: report from the SENTRY Antimicrobial Surveillance Program. Journal of clinical microbiology 2008, 46(9):3179- 3180. 28. Todorova B, Velinov T, Ivanov I, Dobreva E, Kantardjiev T: First detection of OXA-24 carbapenemase- producing Acinetobacter baumannii isolates in Bulgaria. World journal of microbiology & biotechnology 2014, 30(4):1427-1430. 29. Kayama S, Shigemoto N, Kuwahara R, Ishino T, Imon K, Onodera M, Yokozaki M, Ohge H, Sugai M: The first case of septicemia caused by imipenem-susceptible, meropenem-resistant Klebsiella pneumoniae. Annals of laboratory medicine 2013, 33(5):383-385. 30. Kayama S, Koba Y, Shigemoto N, Kuwahara R, Kakuhama T, Kimura K, Hisatsune J, Onodera M, Yokozaki M, Ohge H et al: Imipenem-susceptible, meropenem-resistant Klebsiella pneumoniae producing OXA-181 in Japan. Antimicrobial agents and chemotherapy 2015, 59(2):1379-1380. Page 14/17 Page 14/17 31. Masuda N, Sakagawa E, Ohya S, Gotoh N, Tsujimoto H, Nishino T: Substrate specificities of MexAB- OprM, MexCD-OprJ, and MexXY-oprM efflux pumps in Pseudomonas aeruginosa. Antimicrobial agents and chemotherapy 2000, 44(12):3322-3327. 32. Figures Page 15/17 Figure 2 Antibiotic resistance rates T/C: ticarcillin/clavulanic acid; P/T: piperacillin/tazobactam; CAZ: ceftazidim CSL: cefoperazone/sulbactam; FEP:cefepime; IPM: imipenem; MEM: meropenem; TOB: tobramycin; CIP: i fl i LEV l fl i MNO i li TGC i li COL li i SXT Figure 4 Minimum spanning tree of 94 Acinetobacter baumannii isolates from cerebrospinal fluid and blood based on MLST. Each ST is represented by a circle sized in proportion to the number of isolates represented by that ST, the colors of the halo surrounding the STs denote types that belong to the same clonal complex, the number of allelic difference between STs is indicated on the branches. The detailed MLST profiles can be seen in Table S2 in Supplementary Material. Minimum spanning tree of 94 Acinetobacter baumannii isolates from cerebrospinal fluid and blood based on MLST. Each ST is represented by a circle sized in proportion to the number of isolates represented by that ST, the colors of the halo surrounding the STs denote types that belong to the same clonal complex, the number of allelic difference between STs is indicated on the branches. The detailed MLST profiles can be seen in Table S2 in Supplementary Material. Figure 2 Antibiotic resistance rates T/C: ticarcillin/clavulanic acid; P/T: piperacillin/tazobactam; CAZ: ceftazidime; CSL: cefoperazone/sulbactam; FEP:cefepime; IPM: imipenem; MEM: meropenem; TOB: tobramycin; CIP: ciprofloxacin; LEV: levofloxacin; MNO: minocycline; TGC: tigecycline; COL: colistin; SXT: sulfamethoxazole/trimethoprim. Page 16/17 Figure 4 Minimum spanning tree of 94 Acinetobacter baumannii isolates from cerebrospinal fluid and blood based on MLST. Each ST is represented by a circle sized in proportion to the number of isolates represented by that ST, the colors of the halo surrounding the STs denote types that belong to the sam clonal complex, the number of allelic difference between STs is indicated on the branches. The detaile MLST profiles can be seen in Table S2 in Supplementary Material. Figure 4 Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. supplementarymaterials.docx supplementarymaterials.docx Page 17/17
https://openalex.org/W2277125992	https://hal.archives-ouvertes.fr/hal-03277615/file/OPERAS_Future_of_Scholarly_Communication_06.2021.pdf	English	null	The future of scholarly communication	Cambridge University Press eBooks	2,013	cc-by	17,875	Future of Scholarly Communication Maciej Maryl, Marta Blaszczyńska, Agnieszka Zalotyńska, Laurence Taylor, Karla Avanço, Ana Balula, Anna Buchner, Lorena Caliman, Claire Clivaz, Carlos Costa, et al. To cite this version: Maciej Maryl, Marta Blaszczyńska, Agnieszka Zalotyńska, Laurence Taylor, Karla Avanço, et al.. Future of Scholarly Communication. [Research Report] OPERAS. 2021. hal-03277615 Distributed under a Creative Commons Attribution 4.0 International License HAL Id: hal-03277615 https://hal.science/hal-03277615v1 Submitted on 4 Jul 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 Future of Scholarly Communication Forging an inclusive and innovative research infrastructure for scholarly communication in the social sciences and humanities Future of Scholarly Communication Forging an inclusive and innovative research infrastructure for scholarly communication in the social sciences and humanities Forging an inclusive and innovative research infrastructure for scholarly communication in the social sciences and humanities Future of Scholarly Communication Forging an inclusive and innovative research infrastructure for scholarly communication in the social sciences and humanities Digital Object Identifier (DOI): 10.5281/zenodo.5017705 License: Creative Commons Attribution 4.0 International (CC BY 4.0) This report collects together the findings of Work Packa- ge 6 (Innovation) of the OPERAS-P project. It provides an overview of the phenomena relevant to the future de- velopment and functioning of OPERAS. The report is ro- bust (covers a wide range of issues relevant to OPERAS and scholarly communication), empirically-tested (with inputs from 655 participants from 33 countries), and sta- keholder-validated (consulted with over three hundred stakeholders). (including consensus procedures and accountability when growing). This chapter recommends keeping values at the core of the community, adopting a reflexive attitude towards governance, and allo- wing digital participation and decision-making, while favouring spaces for open innovation. Keeping com- munication channels open to relevant EU initiatives has also been suggested. Chapter 2 Innovative business models develops, collates, and shares information about alternative funding models for open access (OA) books, based on a thorough exploration of the standpoints of two cru- cial stakeholders in the OA book publishing ecosystem: libraries and publishers. The work revealed the pola- risation of the academic library landscape and the fragmentation and diversity of medium-sized acade- mic book publishing as well as an interest in open access book-related initiatives and non-BPC models. The main challenges include the scarcity of human resources and specific funding, which coincides with an overabundance of OA projects. A lack of technical expertise and existing evaluation systems are also seen as obstacles. This chapter recommends recogni- sing regional differences and the diversity of busi- ness models, as well as rethinking evaluation sys- tems, uniting regional hubs, and developing skills in digital publishing. This report focuses on the future of scholarly com- munication, trying to define the main challenges it faces and proposing recommendations for systemic changes. OPERAS is the main addressee of this report, as it can undertake various activities (in the form of services, tools, training, and advocacy) relevant to other groups. However, in order to achieve real and long lasting chan- ge, this report appeals to other stakeholders as well, including e-infrastructures, publishers, SSH resear- chers, research performing organisations, policy ma- kers, and funders. Each chapter provides an overview of the main fin- dings and challenges with an emphasis on recommenda- tions for OPERAS and other stakeholders. Index of content Abstract Executive summary Introduction and methodology Stakeholders CHAPTER 1 OPERAS governance: community values, scalability and future challenges Introduction Main Findings Challenges Recommendations Data and further reading CHAPTER 2 Innovative business models Introduction Main Findings Challenges Recommendations Data and further reading CHAPTER 3 The road to FAIR Social Sciences and Humanities Introduction Main Findings Challenges Recommendations Data and further reading CHAPTER 4 Innovative models of bibliodiversity in scholarly publications Introduction Main Findings Challenges Recommendations Data and further reading CHAPTER 5 Future of scholarly writing in SSH Introduction Main Findings Challenges Recommendations Data and further reading CHAPTER 6 Quality assessment of SSH research: innovations and challenges Introduction Main Findings Challenges Recommendations Data and further reading The future (instead of conclusions) 4 4 6 7 10 10 11 12 13 16 18 18 19 19 22 22 23 24 24 28 30 30 31 31 32 8 14 20 26 45 34 34 35 36 37 40 40 41 42 43 38 This report discusses the scholarly communication issu- es in Social Sciences and Humanities that are relevant to the future development and functioning of OPERAS. The outcomes collected here can be divided into two groups of innovations regarding 1) the operation of OPE- RAS, and 2) its activities. The “operational” issues include the ways in which an innovative research infrastructure should be governed (Chapter 1) as well as the business mo- dels for open access publications in Social Sciences and Humanities (Chapter 2). The other group of issues is dedi- cated to strategic areas where OPERAS and its services may play an instrumental role in providing, enabling, or unlocking innovation: FAIR data (Chapter 3), bibliodi- versity and multilingualism in scholarly communication (Chapter 4), the future of scholarly writing (Chapter 5), and quality assessment (Chapter 6). Each chapter provi- des an overview of the main findings and challenges with emphasis on recommendations for OPERAS and other stakeholders like e-infrastructures, publishers, SSH rese- archers, research performing organisations, policy ma- kers, and funders. Links to data and further publications stemming from work concerning particular tasks are lo- cated at the end of each chapter. The chapters presented in this report collect together the main findings of the more elaborated task reports that can be found in the OPERAS Innovation Lab community on Zenodo, with associated data stored in a dedicated OPERAS-P collection at Nakala. Chapter 3 The road to FAIR Social Sciences and Humanities provides the knowledge base toge- ther with the views of the community in order to find the most suitable way of achieving the maximum uptake of FAIR principles (making data findable, accessible, interoperable and reusable) by the SSH community. The research highlighted a broad understanding of data and the need for the coordination of activities as well as providing success stories and examples. The main challenges include a low awareness of FAIR and a lack of incentives and rewards, together with a diversity of data-related practices in SSH. This chapter recom- mends close consideration of research communities’ needs and the preservation of domain specificities, while also providing coordination, training, acknowled- gement, and reward with regard to FAIR data practices. Chapter 1 OPERAS governance: community values, scalability, and future challenges. This chapter analyses both the current models of governance imple- mented by research infrastructures, and the innovati- ve forms of governance in other types of organisations to derive strategies and suggestions on how OPERAS may further develop its shared culture, common identity, and values. The research has shown the need for a strong community-based organisation, a satisfaction with the current model, and a need for scalability, flexibility, and reflexivity in the organisation. The main challenges include the practical assessment of the current mo- del, and keeping decision-making agile and scalable 4 Chapter 4 Innovative models of bibliodiversity in scholarly publications provides a theoretical and empirical exploration of multilingualism in scholarly communication together with a conceptual design for a platform prototype for shared translation services at the scholarly communication level. The research showed the research relevance of multilingualism, which enables global interaction with multinational and multidisciplinary research. The main challenge is to boost balanced multilingualism and to make national production internationally relevant. This chapter recommends developing a community-ba- sed platform to support scholarly translation as well as the amplification, recognition, and incentivisation of multilingual practices. Chapter 5 Future of scholarly writing in SSH explores current writing practices in SSH to inform fu- ture OPERAS activities on researchers’ needs regar- ding publishing technologies and both ongoing and up- coming transformations of scholarly communication. The main findings include differentiation between digita- lly-enabled and digital writing (and between the scho- larly needs associated with both practices). Innovation is seen as a chance to improve the sharing of ideas with audiences. Novel formats and genres are consi- dered more appropriate for certain content for several reasons: they are liberating, communicative, inte- ractive, and collaborative; and they enable versio- ning and updating. The main challenges include the lack of recognition of novel practices, as innovation is impeded by such factors as quality assessment, prestige, and competencies. This chapter recom- mends developing publishing guidelines with re- gard to innovative genres, supporting novel com- munication practices, and providing training and targeted services for innovative genres in SSH. g g Chapter 6 Quality assessment of SSH research: innovations and challenges explores the following areas: how excellence and other peer review proxies are constructed and (re)negotiated in everyday practices across SSH disciplines, who is involved in the processes and who remains outside them, what the boundaries of peer review are in terms of inclusiveness, and how the processes are aligned or misaligned with research realities. The main findings include the observation that peer review is embedded in the broader systems of academic power structures and has a crucial role in shaping disciplinary identities in SSH. Chapter 6 identifies such challenges as the shortage of evaluati- ve labour, gaining recognition for reviewing records, and the reinforcement of existing power structures through peer-review. This chapter recommends deve- loping responsible research metrics at the EU level as well as coordinated advocacy and training reports by OPERAS and DARIAH. In terms of the conceptu- al prototypes of new services, it posits a Book Review Certification Service extension as well as the administra- tion of peer-review records. The work collected in this report will serve as the basis for OPERAS’ activities and its future servi- ces. This research will be continued by OPERAS Inno- vation Lab, led by the Institute of Literary Research of the Polish Academy of Sciences. 5 Introduction and methodology Each chapter provides an overview of the main findings and challenges with an emphasis on recommendations for OPERAS and other stakeholders (see the next sec- tion). Links to data and further publications stemming from the work on particular tasks are located at the end of each chapter. The chapters presented in this report collect together the main findings of the more elabora- ted task reports that can be found in the OPERAS Inno- vation Lab community on Zenodo, with associated data stored in a dedicated OPERAS-P collection at Nakala. The work on this report proved how crucial user re- search and stakeholder engagement is for our under- standing of needs in regard to scholarly communication and ways they may be implemented. This work will be continued as the OPERAS Lab, coordinated by IBL PAN, which will focus on defining OPERAS’ responses to ac- tual user needs as well as on envisioning and prototyping new services. Some prototypes have already been de- veloped and conceptualised within the framework of the OPERAS-P project, namely the OPERAS living book, as well as a concept for integrated services for digital scholarly editions. This report focuses on the future of scholarly commu- nication, trying to define the main challenges it faces and proposing recommendations for systemic changes. Each chapter provides an overview of the main findings and challenges with an emphasis on recommendations for OPERAS and other stakeholders (see the next sec- tion). Links to data and further publications stemming from the work on particular tasks are located at the end of each chapter. The chapters presented in this report collect together the main findings of the more elabora- ted task reports that can be found in the OPERAS Inno- vation Lab community on Zenodo, with associated data stored in a dedicated OPERAS-P collection at Nakala. Finally, the outputs are stakeholder-validated, which means that the key stakeholder groups (research infrastructures, researchers, institutions, policy makers, funders) were consulted throughout the process. i Preliminary findings were presented during the Future of Scholarly Communication online conference (24–26 February). Each task presented its outputs in one of six dedicated sessions, receiving feedback from three invi- ted experts and members of the audience. The event at- tracted 342 participants from 46 countries all over the world. The feedback gathered informed the final stage of the drafting of the reports and recommendations. A detailed programme is available in this blog post. Introduction and methodology Maciej Maryl and Marta Błaszczyńska This report collects together the findings of Work Pac- kage 6 (Innovation) of the OPERAS-P project, the main objective of which was to produce a robust, empirical- ly tested, and stakeholder-validated foundational body of knowledge regarding phenomena relevant to the future development and functioning of OPERAS. Let us unpack this statement. 1 person Survey Interview Workshop WP6 conference Breakdown of participants in WP6 activities in Europe The work is robust because it covers a wide range of issues relevant to OPERAS, and scholarly communication in general, as identified by previous OPERAS work (esp. white papers). WP6 concentrated on innovation within the various aspects of scholar- ly communication and the role of research infrastruc- tures (RIs) in this process in order to achieve a better understanding of the rapidly changing scholarly com- munication environment in which OPERAS operates. 1 person Survey Interview Workshop WP6 conference The outcomes collected here can be divided into two groups of innovations, on the one hand, OPERAS’ ope- rations, and on the other, its activities. The “operational” issues include ways in which an innovative research in- frastructure should be governed (Chapter 1) as well as business models for open access publications in Social Sciences and Humanities (Chapter 2). The other group of issues is dedicated to strategic areas where OPE- RAS and its services might play an instrumental role in providing, enabling, or unlocking innovation: FAIR data (Chapter 3), bibliodiversity and multilingualism in scho- larly communication (Chapter 4), the future of scholarly writing (Chapter 5), and quality assessment (Chapter 6). Breakdown of participants in WP6 activities in Europe The work has been empirically tested because it is crucial to address the real needs of the community. Thus, the tasks within the Work Package shared similar workflows, which consisted of a state-of-the art review and soliciting input from stakeholders through surveys, interviews, and workshops. Altogether the research involved 655 participants from 33 countries on four con- tinents (with a focus on Europe, given the aim of the pro- ject). There were 57 people interviewed, 134 filled out the surveys, and 464 participated in dedicated workshops. This report focuses on the future of scholarly commu- nication, trying to define the main challenges it faces and proposing recommendations for systemic changes. ¹ The difference between a research infrastructure and an e-infrastructure lies in their disci- plinary scope: while OPERAS, as a research infrastructure, focuses on the needs of the SSH community, e-infrastructures such as OpenAIRE or European Open Science Cloud, addresses researchers in different fields. Introduction and methodology The work on this report proved how crucial user re- search and stakeholder engagement is for our under- standing of needs in regard to scholarly communication and ways they may be implemented. This work will be continued as the OPERAS Lab, coordinated by IBL PAN, which will focus on defining OPERAS’ responses to ac- tual user needs as well as on envisioning and prototyping new services. Some prototypes have already been de- veloped and conceptualised within the framework of the OPERAS-P project, namely the OPERAS living book, as well as a concept for integrated services for digital scholarly editions. 6 Countries represented in WP6 activities Countries represented in WP6 activities Stakeholders Although the main aim of this report is to provide guidan- ce for OPERAS, a European research infrastructure for scholarly communication in SSH, the recommen- dations stemming from our work have a wider reach. This is because scholarly communication faces systemic problems and one could hardly expect that changes made just by one stakeholder group (a research infrastructu- re in this case) would sustainably transform the whole ecosystem. A good example of a systemic issue is the perceived lack of prestige accorded to innovative forms of writing. In order to change this, policy makers would need to provide adequate assessment mechanisms, funders ought to allow such publications to be funded, research infrastructures should provide services and tools for easy publishing, publishers would need to recogni- se the value of such work and, finally, researchers need to accept them as valid outputs. That is why we address our recommendations to different stakeholder groups, all of whom play important roles in OPERAS’ work. Publishers, especially scholarly-led initiatives, belong to one of the key stakeholder groups as they focus on the publication part of the scholarly communication workflow. Researchers in SSH are the main stakeholders in OPE- RAS, especially digital humanities scholars who may have more advanced needs with regard to publishing their outputs (e.g. linking text with data and code). Research performing organisations like universities or research institutes are important because institutional regulations and practices serve as an immediate sphere of reference for scholars. This report also provides some guidance for policy makers, i.e., relevant bodies at the national or European level who shape research policies and may provide top- -down support for certain activities. Likewise, funders may adjust their schemes in order to embrace and enco- urage innovation in certain activities. E-infrastructures Publishers Researchers in SSH Research performing organisations Policy makers Funders OPERAS The current model, which seems satisfactory The current model, which seems satisfactory e c e o e , c see s sa s ac o y The current model, mixing, for example, an executive assembly and an assembly of the commons, seems satisfactory to members as it allows for poly-centric governance. The latter is based on institutions’ represen- tatives, but also on expertise and distributed decision- -making. However, at this moment in time it remains a merely theoretical model whose efficiency needs to be assessed in practice during the coming months. OPERAS OPERAS is the main addressee of this report, as it can undertake various activities (in the form of services, tools, training, and advocacy) relevant to other groups. However, e-infrastructures1 like publishing plat- forms, repositories, and digital libraries, are also an im- portant target for these recommendations in regard to ad- dressing the particular disciplinary needs of SSH scholars. 7 01 CHAPTER Authors: Valérie Schafer and Lars Wieneke (C2DH, University of Luxembourg) Contributors: Pierre Mounier (EHESS) and Suzanne Dumouchel (CNRS Partners involved: University of Luxembourg and CNRS Analysing failures and keeping flexibility and reflexi- vity Analysing failures and keeping flexibility and reflexi- vity Analysing failures and keeping flexibility and reflexi- vity The second workshop, in February 2021, was part of the OPERAS WG6 workshop, which was also organised online, with this second one being more specifically focused on OPERAS governance and its perspectives. Adjustments are needed within the whole life cycle of a project, while there is also the need for some analy- sis of the successes and failures. These are the constant needs at the crossroad of governance and management and should also be taken into account in the governance model. Finally, an internal workshop lasting half a day in May 2021 gathered together several OPERAS governan- ce stakeholders (members of the executive assembly, scientific committee) to address the most urgent issues and challenges related to the future of OPERAS and its governance. Temporalities are key Time is as much a question for the individual participants involved in the project (in particular, the time spent wor- king for OPERAS, vs. their own general workload) as it is for the longer temporalities of the project (and in particu- lar, the transition to an association faced with the need for institutionalisation, reliability, and results, but which also wishes to pursue a flexible and creative process). The research on stakeholders was based on a survey and several workshops. The survey was conducted mainly during March 2020 and was disseminated thro- ugh internal OPERAS-P channels, Twitter, and the OPE- RAS-P blog. The survey’s 25 questions took approxima- tely 15–20 minutes to answer and were framed around the values of OPERAS-P. We received 26 answered surveys overall (24 from OPERAS members, 2 from external respondents). Scalability and multi-layered architecture Linked to the previous point, there is a key need for scalability as well as a need for a multi-layered model that may allow, at the same time, innovation and main- tenance, and centrality and decentralisation. Hybrid mo- dels can exist and cohabit within the same organisation in order to avoid both the risk of amateurism and esta- blishment, and to combine the strengths and virtues of both.ll We co-organised three workshops on the topic of OPERAS governance and, more generally, the gover- nance of research infrastructures. The first was entirely dedicated to digital governance and research infrastructu- res, and was organised by our team in collaboration with a scientific committee. It took place remotely, due to the COVID-19 crisis, in September 2020. Introduction Task 6.1 of Work Package 6 aimed to analyse both current models of governance implemented by rese- arch infrastructures, and innovative forms of governance in other types of organisations, to derive strategies and suggestions on how OPERAS may further develop its shared culture, common identity, and values, and balance the needs of the stakeholders to ensure efficiency and reliability along with the capacity to be open to new models of governance that emerge from the digital environment. The activities within this task were divided into three phases: 1) mapping the state of the art, 2) stakeholder research, and 3) recommen- dations. A strong community-based organisation Participants in OPERAS-P show great motivation in con- tributing to the commons (i.e. cultural resources acces- sible and sharable by all members of the community), as well as “learning while contributing,” and they give importance to the content of discussions. Respect, enga- gement, trust, and the common good are requirements for their participation as well as common and clear go- als and deadlines. There is a need for both community- -based values, such as trust and transparency, but also for efficiency and clarity (see, e.g., the results of the survey and last workshop in May 2021). Accountability is also underlined. Regarding the state of the art, the diversity of possi- ble approaches to governance explains the abundance of literature, and the impossibility of being exhaustive in a field that is as much a matter of management stu- dies as it is of studies in STS, or the sociology of inno- vation, or digital humanities, or the information and communication sciences. Therefore, we selected three key topics that mostly gathered together the issues we wanted to explore within this work package: gover- nance and values; research infrastructures, trading zo- nes, and interdisciplinarity; and finally, knowledge com- mons and P2P productions. OPERAS in a complex environmenti As underlined in our final workshop, OPERAS has to go along with other research infrastructures at the EU level (i.e. with Triple, COPIM, EOSC, etc.) and others. OPERAS will become an important player in this field, and so these adherent responsibilities have to be taken into account. Within our task we had the chance to meet and to collaborate with several actors in the field of digital governance, notably the co-founders of Meoh, a think- -and-do-tank based in Brussels that studies how social trust can inform new models of cooperation and gover- nance in a networked society; and with COPIM (Com- munity-led Open Publication Infrastructures for Mono- graphs). 10 Keeping decision making agile and scalable p g g g With the growth of OPERAS and its institutionalisa- tion, there is a need to combine several cultures, roles, and visions, and, notably, to combine the place of profes- sionals with that of members who are representatives of their institutions. Empowerment OPERAS will have to ensure that it keeps its structure open and inclusive. It will have to deal with various de- grees of availability, levels of involvement, and should ensure that hidden power and excessive bureaucracy can’t take hold. Incentives for participants and recogni- sing participation should also be constantly evaluated. At the same time, new members need to find roles and functions within the network that match their abi- lities and needs. Consensus procedures Consensus procedures How can OPERAS keep consensus procedures whi- le growing? When is voting needed? Does OPERAS’ growth also lead to a growing number of members and decision makers? Assessing the designed model The current model seems satisfactory from a theoreti- cal point of view. It is balanced, multi-layered, has clear responsibilities, and includes several bodies (executi- ve assembly, general assembly, assembly of the com- mons, scientific advisory committee, coordination team, etc.); but it needs to be assessed over the coming months in order to ensure there is coordination between the layers. Consolidation and professionalisation as the main challenges The professionalisation and consolidation of OPERAS is needed, but should not diminish the requirement for constant innovation, horizontality, and creativity. Naming and explaining Some of the terms used in the governance scheme don’t match their actual function. For example, the Executive assembly is less of an assembly and much more of an execu- tive team that drives actions within the network. Other functions need clear explanations of their roles and actions to develop a shared understanding and to allow newcomers to quickly understand how OPERAS is organised. Favouring spaces for open innovation (task forces, working groups) The governance model hybridises both formal groups and more flexible and temporary groups (for example task forces and working groups) that may be more P2P and short- -term, and may target a precise interest or goal. Flexible task forces can help maintain innovation while also allowing for a more structured arrangement for daily operations. The success of this model also relies on clear communication and information, with few, but efficient, channels and the regular renewal of roles and mandates. Coming to terms with the goals of OPERAS Coming to terms with the goals of OPERAS OPERAS understands its main mission in managing and fostering the actions and in- terests of its constituents, therefore, the governance scheme should clearly highlight the relevance of this task. In reality, the role of the Coordination Team seems to be much more relevant to the overall mission and more coordination between the Coordination Team and the Executive Assembly is needed to distribute management tasks. Supporting scholars involved in OPERAS Scholars involved in OPERAS need to be recognised by their institutions and receive some form of incentive for their involvement. Accountability Transparency and trust are key for the development of OPERAS. Its values must also reflect the openness that is at the core of the project. This has to be clear for OPE- RAS’ internal members as well as those who are exter- nal, whether they be funders, partners, stakeholders, etc. 11 Recommendation Keeping values at the core of the community Regular updating and evaluation of the code of conduct and guidelines, and the reinforce- ment of shared values. Interacting with research infrastructure Beyond the need for transparency, trust, and accountability, a communication channel is needed to allow feedback. Scholars should be encouraged to participate in, and to enter into, OPERAS’ services in an (inter)active mode that favours their input and requires their commitment. They should accept OPERAS’ values, terms of use, etc. Answering the need for maintenance, accountability, etc. Exchange with the external EU environment As a key player in the European field, and notably in OA, OPERAS has to work with EU partners such as Triple EOSC, and should, therefore, identify representatives and de- cision makers that may create bridges. Supporting scholars involved in OPERAS Answering the need for maintenance, accountability, etc. There is a need for a clear governance model that allows policy makers and funders to identify the decision-making processes and roles in OPERAS while accepting a hybrid model of governance between institutionalisation and flexibility. Community and the commons At the heart of OPERAS’ reflection, the notion of the commons also poses major challen- ges and needs to find its place in the governance model while respecting the requirement to comply with legislative and economic constraints, and decision-making processes that must be efficient. As a central value of OPERAS, it needs to be refined in order to become a lever for OPERAS’ growth, which must be supported by shared governance. The assem- bly of the commons needs to be engaged – an engagement plan is needed. Having a reflexive attitude towards governance Recommendations Target audience OPERAS/ Research infrastructure Recommendation Having a reflexive attitude towards governance The creation of a WG entirely dedicated to OPERAS’ governance to assess progress, strengths, failures, etc. Allowing digital participation and decision-making Allowing digital participation and decision-making In order to allow all members to participate, a hybrid format (f2f and remote) should be developed that will also permit remote consensus or voting. Favouring spaces for open innovation (task forces, working groups) Key players to be heard and representedi The field of publishing is rapidly changing and adapting. OPERAS is a place to discuss and co-shape these changes. 12 Data and further reading • The full task report is available OPERAS Innovation Lab community on Zenodo • Survey data are available at OPERAS-P collection at Nakala. • The Knowledge Infrastructures and Digital Governance report from a two-day workshop held on 7 and 8 Sep- tember 2020, with the aim of combining theoretical and practical perspectives on issues that are constantly developing as a result of the wide-ranging forms of re- search infrastructure and challenges facing digital gover- nance. • Recording of a DGO workshop with Meoh (with 10 participants from 8 nationalities, 8 countries, and 3 conti- nents) on multi stakeholder cooperation during the 21st Annual International Conference on Digital Government Research in June 2020 trust and transparency efficiency and clarity involvement workload reliability and results a flexible and creative process centrality decentralisation Balancing the needs of different stakeholders for better governance Balancing the needs of different stakeholders for better governance 13 02 CHAPTER Introduction Following the investigation of the academic library land- scape, we focused on publishers. We analysed nine Europe-based OA book publishers who used business models that either departed from relying on Book Pro- cessing Charges (BPCs) completely or used mixed models in which BPCs were one of several revenue streams. In order to better understand how they wor- ked, we interviewed representatives of these nine pu- blishing houses. We looked at several crucial aspects that would help us both identify common threads and pinpoint the particularities of the applied models. We examined each case according to the following areas of interest: 1) the publisher’s general profile, 2) workflows, 3) the business model, 4) sustainability, and 5) challen- ges. The interviews will be hosted online by the Open Ac- cess Books Network and will form the basis of a broader community collated collection of publishers’ profiles. New presses will be encouraged to submit their own responses to the template used for these interviews in order to create an open database of business models for OA book case studies. The main objective of the OPERAS-P Task 6.2 (Innovati- ve business models) was to develop, collate, and share information about alternative funding models for open access (OA) books. In order to fulfil this general goal, we wanted to better understand the standpoints of two crucial stakeholders in the OA book publishing ecosystem: libraries on the one hand and publishers on the other. We first investigated the academic library systems in 14 European countries to examine how they were set up and how they dealt with OA books. Second, we had a closer look at publishers and the intricacies of chosen publishing models for OA books as they are applied across the European landscape. In both cases we identified the most important challenges that the examined cases were facing, in the realms of administra- tion, legal issues, infrastructure, funding, among others. Seeking to understand the academic library landsca- pe in Europe, we conducted interviews with librarians who represented 14 European countries: OPERAS core members (Croatia, France, Germany, Greece, Italy, Poland, Portugal, Slovenia, and the Netherlands), along with the addition of Spain and the Nordic countries (Denmark, Finland, Norway, and Sweden). Introduction Represen- tatives of each of the analysed countries were asked general questions in the following areas of interest: 1) the general characteristics of library systems con- cerning e-content and OA publications, 2) the libra- ry community and open access, 3) OA book policies, 4) OA book funding, 5) library/scholar-led OA book pu- blishing initiatives, and 6) the integration of OA books with library systems. Three workshops organised for small publishers inte- rested in non-BPC models helped us grasp what challen- ges they encountered and where they could use help. The first of these workshops gathered representatives of six presses who presented their models, with this workshop attracting a large audience of over 200 regi- strants. The following two workshops were focused on the three specific business models that the audience of the first workshop was most interested in: the busi- ness models of the publishers Open Book Publishers and Punctum Books, and the “Opening the Future” bu- siness model. Four workshops – including two regional events for the Nordic countries (Denmark, Finland, Norway, Sweden) and for Southern Europe (Croatia, Greece, Slovenia), as well as two country-based ones for Ger- many and Poland – gave us the chance to ask more specific questions about roadblocks, selection crite- ria, and budget allocations concerning OA book-related projects for the wider group of library representatives. A survey, which participants were asked to complete prior to these workshops, revealed the level of familiarity lib- rarians had with existing OA book publishing initiatives. A set of short reports discussing the main take-aways from each of the workshops was published in the form of blog posts on the COPIM website. The work undertaken for Task 6.2 relied heavily on the OA book community’s engagement. If it weren’t for the willingness and enthusiasm of our partners – the interviewees, the workshop participants, resear- chers, and librarians – we would not have been able to fulfil our aims. Over the course of the project, we have had the pleasure of working with representatives of over 50 organisations, and we would like to express our grati- tude to all of them. Introduction Based on a systematic literature review, desk-based research, and, perhaps even more importantly, on what we heard from the European library community in inter- views and workshops, we compiled a report on acade- mic libraries in Europe and OA books, which we have made available to the community as a living document on the COPIM website – it is open to comments (see Data and further reading section for links). It is our inten- tion to add more country-based cases to it in the future. In trying to gain an overview of existing OA book po- licies in Europe, we examined over 60 policies, and cre- ated a summary of 27 cases that mentioned OA books specifically. We have made this file available to the com- munity as a living document through the Open Access Books Network, to which new cases can be added. Based on a systematic literature review, desk-based research, and, perhaps even more importantly, on what we heard from the European library community in inter- views and workshops, we compiled a report on acade- mic libraries in Europe and OA books, which we have made available to the community as a living document on the COPIM website – it is open to comments (see Data and further reading section for links). It is our inten- tion to add more country-based cases to it in the future. In trying to gain an overview of existing OA book po- licies in Europe, we examined over 60 policies, and cre- ated a summary of 27 cases that mentioned OA books specifically. We have made this file available to the com- munity as a living document through the Open Access Books Network, to which new cases can be added. 16 ² Other countries, e.g., France, also support large scale OA book publishing initiatives, but not necessarily through university libraries. The United Kingdom, which was not included in this report, is a particularly fertile ground for small-scale OA book initiatives (see, for exam- ple, Open Book Publishers). ” It is nice to have a stable source of revenue. At the same time, I guess it is a problem... I. Melinščak Zlodi, interview, 16.02.2021 The BPC model treats researchers in a very unequal way, and so that’s why we didn’t want to adopt it. L. Kaakinen, interview, 18.02.2021 The main fear was that when we started to publish books open access, we wouldn’t sell so many printed books, but this has not really happened. M. Rudolf, interview, 18.02.2021 ” It is nice to have a stable source of revenue. At the same time, I guess it is a problem... The BPC model treats researchers in a very unequal way, and so that’s why we didn’t want to adopt it. The main fear was that when we started to publish books open access, we wouldn’t sell so many printed books, but this has not really happened. M. Rudolf, interview, 18.02.2021 Excerpts from interviews on innovative business models Excerpts from interviews on innovative business models 17 Local relevance and impact are crucial Several publishers expressed concern about relying on a single source of revenue when publishing OA bo- oks. This challenge has been raised both by publishers who depend on national subsidies and those relying on the BPC model. Academic libraries stress the importance of providing metrics that show the impact of OA books at a local le- vel. Local relevance has also been stressed in the con- text of multilingualism and the importance of recognizing publications in local languages, especially in SSH. One size will not fit all Discrepancies between library systems across Europe as well as those among publishers operating in different regional circumstances show that it is impossible to find a single EU-wide model for OA book publishing. It is ho- wever possible to identify several regional trends and similarities between the examined countries. The academic library landscape is polarised Publishers are actively interested in non-BPC models There is a number of small publishers exploring alternative business models for open access books. This interest is seen as coming both from OA born pres- ses and those thinking about either switching to OA completely or combining OA and non-OA publishing. One size will not fit all e acade c b a y a dscape s po a sed There are deep discrepancies between the members of the examined European countries when it comes to dealing with open access issues. In the Nordic coun- tries, Germany, and the Netherlands it has become one of the pivotal aspects of scholarly communication; insti- tutions are supportive, and there are funding schemes that allow libraries to invest in OA book publishing initia- tives. Other regions still struggle with the full integra- tion of OA publications in their library ecosystems: there is insufficient funding, not enough human resources, and little decision-making autonomy at the institutional level and hence little room for experimentation. Fragmentation and diversity The considerable longtail of smaller and medium-sized academic book publishers differs in terms of how they are set up, with their structures varying when it comes to areas such as revenue, costs, legal affairs, production, and distribution for academic book publishing. Individual- ly, these presses are uniquely positioned and deeply-ro- oted within their communities in order to best serve the- ir particular scholarly community. Collectively, they play a vital and key role in realising a transition to open access for books as part of the broader spectrum that is scholar- ly book publishing. Lack of technical expertise Library and scholar-led OA book publishing initiatives are rare There is a considerable lack of technical skills within both the publishing and library communities when it co- mes to producing digital books. Both groups reported challenges when dealing with issues of the production and distribution of digital files. This lack of expertise usu- ally results in these parts of the publishing process being outsourced or neglected. These initiatives have not (yet) gained momentum in continental Europe. While there are several emer- ging projects involving libraries, in most cases they are not large scale. Among the pioneers of innovative OA book publishing models are Germany, Denmark, Finland, and Sweden. There are also a few examples of projects partially subsidised by national funders. In the majority of investigated countries, however, such initiatives do not exist2. OA book-related projects overdose Libraries reported a certain frustration concerning the multiplicity of existing OA book initiatives, which often operated different business models and offered dif- ferent services. This plethora of options makes it difficult to make a sound decision as to which project to support and which to reject. The scarcity of human resources The scarcity of human resources The scarcity of human resources Both librarians and publishers reported a scarcity of human resources as one of the main road blocks in developing and implementing OA book publishing initiatives. Scarcity of OA book-specific fundingi i We identified only four countries (Germany, the Nether- lands, Finland, and Norway) that have OA book-dedicated funds, some at the national level, others at the institutio- nal or funder’s level. Of the remaining examined coun- tries such funds do not exist. Interest in open access, book-related initiatives Interest in open access, book related initiatives There is an incontestable interest in OA book pu- blishing initiatives coming from libraries and publishers. The abundance of numerous library associations, which treat open access as one of the critical points of discus- sion, show the scale and importance of library enga- gement in open access publishing practices. There is a number of small publishers exploring alternative busi- ness models for open access books. This interest is seen as coming both from OA born presses and those thin- king about either switching to OA completely or combi- ning OA and non-OA publishing. Main Findings The academic library landscape is polarised Data and further reading • Interview transcripts and regional workshop survey responses are available at OPERAS-P collection at Na- kala. • Interview transcripts and regional workshop survey responses are available at OPERAS-P collection at Na- kala. Support knowledge exchange Facilitate best-practice exchanges in the form of workshops, open databases with case studies, and toolkits in order to create a dialogue within the OA books community and allow stakeholders to learn from each other, especially in particularly challenging are- as (e.g. production and distribution of digital books). Unite in regional hubs Find organisations interested in OA book publishing that operate under similar regional circumstances as your own, and create local hubs to exchange best practices in OA book publishing. Stay informed about innovative business models in the OA book landscape, evaluate which of them could work for you, combine different approaches, and create hybrids to find the one that will best suit your particular needs. Learn to speak fluent digital Seek opportunities to develop skills in digital publishing and look for partners who might assist you in this process. Seek opportunities to develop skills in digital publishing and look for partners who might assist you in this process. Recognise the diversity of business models Recognise the diversity of business models When creating policies pertaining to OA books, acknowledge the existence of diverse business models, and provide funding to facilitate innovation. Recognise regional differences When creating policies pertaining to OA books, take into consideration specific regional circumstances rather than trying to impose a unified policy, which might work for some, but excludes others. Rethink evaluation systems Create evaluation systems that recognise OA books as a valid and valuable publishing output, and reflect on the diversity of publishers to encourage researchers to publish their research in OA and expand their publishing choices. Existing evaluation systems Existing evaluation systems applied to measure scholars’ performance, often do not recognise OA books as a va- luable publishing output and discriminate against small- -scale OA book publishers. 18 Recommendations Target audience Recommendation Policy makers Policy makers and funders Research institutions and policy makers Librarians and publishers OPERAS Recognise regional differences When creating policies pertaining to OA books, take into consideration specific regional circumstances rather than trying to impose a unified policy, which might work for some, but excludes others. Recognise the diversity of business models When creating policies pertaining to OA books, acknowledge the existence of diverse business models, and provide funding to facilitate innovation. Rethink evaluation systems Create evaluation systems that recognise OA books as a valid and valuable publishing output, and reflect on the diversity of publishers to encourage researchers to publish their research in OA and expand their publishing choices. Unite in regional hubs Find organisations interested in OA book publishing that operate under similar regional circumstances as your own, and create local hubs to exchange best practices in OA book publishing. Pick, choose, and experiment Stay informed about innovative business models in the OA book landscape, evaluate which of them could work for you, combine different approaches, and create hybrids to find the one that will best suit your particular needs. Learn to speak fluent digital Seek opportunities to develop skills in digital publishing and look for partners who might assist you in this process. Support knowledge exchange Facilitate best-practice exchanges in the form of workshops, open databases with case studies, and toolkits in order to create a dialogue within the OA books community and allow stakeholders to learn from each other, especially in particularly challenging are- as (e.g. production and distribution of digital books). Recommendation Recognise regional differences 4. Library Support for OA Books Workshop: the Southern European perspective • Review of existing OA books policies • Review of existing OA books policies A document collating existing funder OA book policies in Europe. The document is open so that the community can add new entries and collaboratively work on upda- ting this database. • Full report on academic libraries and open access bo- oks in Europe is available as a static file, and as an open document that the community can comment on. • The full 6.2. task report is available OPERAS Innova- tion Lab community on Zenodo • A blog post on a workshop for publishers on innovati- ve business models for books. • Series of blog post on regional workshops for librarians: 1. Library Support for OA Books Workshop: the German perspective • A blog post on a workshop for publishers on innovati- ve business models for books. • Publishers’ case studies database 3. Library Support for OA Books Workshop: the Scandi- navian perspective 19 03 CHAPTER pi Everything is data, or could be Within the FAIR framework, the concept of data is in- tended to be as universal as possible, including data- sets, publications, software, etc. Although accepted in various SSH fields – such as social sciences, histo- ry, linguistics, and digital humanities – the notion of data continues to be regularly discussed in the SSH con- text. The literature review and the research community workshops have also shown ways of integrating SSH’s various data types and scientific methods into a cohe- rent digital ecosystem. For instance, it seems possible to make a broad distinction between source and result data, or to consider the well-established processes of resources’ curation and the management of the social sciences and humanities as FAIR-enabling practices. Therefore, OPERAS-P Task 6.3 was aimed at provi- ding the knowledge base together with the views of the community to find the most suitable way to make FAIRi- fication of SSH data possible, using a threefold approach: • Speak the same language g g The first step was to focus on the different kinds of data in SSH and the issues arising from implementing FAIR principles via a thorough review of the rich and growing literature about data in SSH. • Work with the community • Work with the community Through focus groups and workshops, the task engaged stakeholders in unveiling perceptions about FAIR data, the needs of the different communities, and the challen- ges facing FAIRification regarding various disciplines. Advocating for FAIR adoption means explaining its benefits i Advocating for FAIR adoption means explaining its benefits i From previous findings, it appears that advocacy for FAIR principles remains the first step when consi- dering the SSH landscape. In order to expand the ad- option of FAIR principles within the SSH, it is the prin- ciples’ final purpose that should be outlined, showing how they increase the quality of research and can inte- grate even convergent, although pre-existing, practices. • Showing the road ahead Based on the review and the workshops, the task suggested some directions for OPERAS concerning the FAIRification of data, including FAIRification tool pro- totypes and measures to further engage the community both in discussions and implementation. Main Findings The FAIR principles (Findability, Accessibility, Interopera- bility, and Reusability) are a set of foundational guidelines aimed at improving the management of digital scholarly resources for both humans and machines. In conside- ring digital objects as a whole, focusing on data manage- ment and reuse, and allowing for cross-disciplinary rese- arch, the FAIR principles provide an innovative approach for Social Sciences and Humanities’ (SSH) practices. More fragmented, less data-centric, and in some fields, dealing with physical objects, the SSH environment still needs some guidance in exploring the full potential of FAIR principles. As a major component for integrating the European Open Science Cloud (EOSC), FAIR prin- ciples are also, more broadly, an important tool for open science. Capturing the SSH in transition FAIR principles, which are essentially digital, mostly da- ta-centric, and oriented towards automated processes, are to a certain point an adequate device for capturing current SSH research practices. The levels of awareness, skills, and engagement concerning FAIR principles cha- racterise, beyond disciplines, distinct communities and reveal the transitional period of the SSH global landsca- pe. To avoid falling into mere opacity, the agnosticism of generic principles like FAIR requires the various com- munities’ specificities to be taken into account. Coordination is key for FAIRification Task 6.3 worked in synergy with parallel activities carried out by the CO-OPERAS Implementation Network (IN) within the GOFAIR initiative, which was aimed at the FAIRification of SSH data and publications. CO-OPERAS IN is coordinated by OpenEdition and UniTo, with the re- gular contributions from Huma-Num. The various workshops and communications made it obvious that the FAIRification of SSH implied a wide diversity of actors. All the actors involved in the data ge- neration process should also be actors of the “FAIRifica- tion-chain”: researchers, data stewards, repository ma- nagers, librarians, and publishers. Regarding the direct actors of data generation, there is, more specifically, a need to converge on metadata standards, potential- ly by sharing a minimal metadata set. Coordination at a broader level is also required to ensure a con- sistent FAIR ecosystem for the SSH. With that prospect, OPERAS intends to collaborate with SSH ERICs Cess- da, CLARIN, and DARIAH as well as with projects like SSHOC and EOSC-Pillar. ries, examples, and tools need more visibilit FAIR stories, examples, and tools need more visibility SSH researchers feel that there is a lack of recognition when their research relies concretely on implemen- tations of FAIR principles. This poor reward system for researchers making their data actually findable, acces- sible, interoperable, and reusable obviously hinders the research community’s engagement with the adoption of FAIR principles. Moreover, this deprives the community 22 22 of examples and stories that they could take inspira- tion from. Through CO-OPERAS IN, OPERAS carried out initiatives to address this issue. First, a blog is cur- rently being established that will offer a common space for discussion and experience sharing. Second, in order to provide guidance for the FAIRification of publications, a major object of SSH research, on-going work will pro- vide a FAIRification toolkit dedicated to publishing plat- forms. FAIR awareness is still low FAIR awareness is still low General knowledge regarding specific but major aspects of FAIR, such as persistent identifiers, metadata inte- roperability standards, and open licenses, is very uneven in the SSH environment, if not simply lacking. Given that the acronym is most certainly becoming more widely known, it is obvious that FAIRification requires more than such a superficial understanding. Advocacy efforts and de- dicated training are therefore required to increase the le- vel and the quality of FAIR awareness. However, this may also rely on an effort of “translation,” adapting the FAIR analytical grid to well-established and functional practices that are convergent with, if not identical to, FAIR principles. Diversity and complexity of the SSH As mentioned above, the SSH landscape does not offer a coherent or uniform landscape. However, rather than “fragmentation,” we should simply speak, in this case, of the diversity and complexity of the SSH research environment. When it comes to data, the diversity incre- ases immediately because of the various typologies and methods involved. It implies a slight adjustment of the ob- jective: instead of bringing all the SSH communities into a single (and impossible) model, we should look for simila- rities that could work as hooks – able to connect together all the different parts of a rich and lively environment. FAIR is not open, but it supports open science This is a twofold challenge. FAIR is regularly descri- bed as distinct from openness; just as FAIR has regu- larly become the companion of open science policies. The statement “open as possible, closed as necessary,” although handy for general presentations, offers unfortu- nately poor guidance for concrete FAIR implementations. In fact, open licensing allows for reuse in degrees that are not entirely described by the pair “open/closed.” This represents the first challenge in terms of com- munication. Moreover, dealing with humans’ creations and phenomena, SSH sources and outputs often include personal information, property rights, or even sensitive data. Reusability, as outlined by SSH researchers, is thus characterized by legal challenges that require specific guidance and expertise. Acknowledge and reward FAIRification of data and publications, making them part of the research assessment. Use the FAIR principles as a grid to assess the FAIRness of publishing systems, to enhance both content visibility and the quality of publishers’ information systems. Expand and improve advocacy by offering both explanations of FAIR principles and examples of FAIR tools and services. Provide training on FAIR data and metadata that takes into account the disciplines, data types, and the existing standards’ specificities. Coordinate with other infrastructures and projects involved in FAIRification to offer consistent guidance to their respective communities. Recommendation Consider the research communities’ needs in order to align implementation policies with research practices and purposes. Preserve the domain specificities regarding data, digital objects, methods, etc., and adapt FAIR implementations accordingly. Preserve the multilingualism and bibliodiversity of the SSH environment. Address the sustainability of FAIR services to ensure data reuse in the longer term. Collaborate on common minimal metadata sets, allowing for cross-disciplinary research and the building of FAIR digital objects. Incentives and rewards Another challenge already mentioned concerns the glo- bal reward system of contemporary research. Incentives to adopt FAIR principles and rewards for FAIR imple- mentations can only partially rely on research networks and infrastructures. The FAIRness level and quality of research should be part of funders and policymakers’ assessment processes. 23 Recommendations Target audience Recommendation Research performing organisations Scholars Scholars and research infrastructures Funders and Policy makers Publishers OPERAS Consider the research communities’ needs in order to align implementation policies with research practices and purposes. Preserve the domain specificities regarding data, digital objects, methods, etc., and adapt FAIR implementations accordingly. Preserve the multilingualism and bibliodiversity of the SSH environment. Address the sustainability of FAIR services to ensure data reuse in the longer term. Collaborate on common minimal metadata sets, allowing for cross-disciplinary research and the building of FAIR digital objects. Produce an inventory of existing FAIRification tools, enriched with new tools dedica- ted to SSH specific objects, such as publications or cultural heritage materials. Acknowledge and reward FAIRification of data and publications, making them part of the research assessment. Use the FAIR principles as a grid to assess the FAIRness of publishing systems, to enhance both content visibility and the quality of publishers’ information systems. Expand and improve advocacy by offering both explanations of FAIR principles and examples of FAIR tools and services. Provide training on FAIR data and metadata that takes into account the disciplines, data types, and the existing standards’ specificities. Coordinate with other infrastructures and projects involved in FAIRification to offer consistent guidance to their respective communities. Produce an inventory of existing FAIRification tools, enriched with new tools dedica- ted to SSH specific objects, such as publications or cultural heritage materials. Acknowledge and reward FAIRification of data and publications, making them part of the research assessment. Data and further reading • The full report from this task is available at OPERAS Innovation Lab community on Zenodo. • Reports from the national workshops on Definition of Data for FAIR SSH (in chronological order): Porto (in- ternational), Turin, Coimbra, Göttingen, Paris, Brussels (international). • T. Biro, E. Giglia, “Humanities and Data: for a communi- ty-driven path towards FAIRness”, March 2020, recording from the Berlin OpenScience conference. 24 24 Key issues for FAIR data in SSH FAIR data in the SSH • A snapshot of the SSH transitional period • Everything is data, or could be • Advocating for FAIR adoption means explaining its benefits • For FAIRification, coordination is key • FAIR stories, examples and tools need more visibility Main findings Challenges • FAIR awareness is still low • Diversity and complexity of the SSH • FAIR is not open but supports open science • Incentives and rewards Recommendations RESEARCH PERFORMING ORGANISATIONS • Consider the research communities’ needs to align implementation policies with research practices • Preserve the domain specificities and adopt FAIR • Preserve multilingualism and bibliodiversity • Address the sustainability of their FAIR services • Produce an inventory of existing SSH FAIRification tools SCHOLARS • Collaborate on common minimal metadata sets allowing for cross-disciplinary research and the building of FAIR digital objects FUNDERS • Acknowledge and reward FAIRification practices PUBLISHERS • Use the FAIR principles as a grid to assess publication FAIRness OPERAS • Improve advocacy and enhance training for FAIR adapted to SSH • Coordinate with other infrastructures and projects involved in FAIRification to offer consistent guidance FAIR data in the SSH • A snapshot of the SSH transitional period • Everything is data, or could be • Advocating for FAIR adoption means explaining its benefits • For FAIRification, coordination is key • FAIR stories, examples and tools need more visibility SCHOLARS 25 04 CHAPTER Introduction In respect to the problem, the study reflects a gap in the recent literature, namely, identifying factors that influence the dynamics underlying language selection and the use of multilingualism within scholarly commu- nication. The database selected was Google Scholar, and the search terms used were “scholarly communica- tion,” “language,” and “multilingualism,” combined with the Boolean operator “AND.” The search, undertaken on 6 April 2020, yielded 152 works. These results were re- viewed to exclude duplicates, PhD and Master’s disser- tations, and works that did not meet this literature re- view’s goals. To be within the selection criteria, the works had to 1) have a DOI (Digital Object Identifier) code, 2) be published in open access between 2019 and 2020, and 3) be written in English, French, German, Portugu- ese, Italian, or Spanish. This resulted in 12 documents being selected that were then analysed by resorting to qualitative content analysis of the abstract and conc- lusion sections. Subsequently, the final category fra- mework reflected the corpus codification structure that emerged from the analysis. Given the growing need to strengthen the bonds be- tween stakeholders involved in scholarly communication and multilingualism, this task was directed by a three- -fold purpose: 1) synthesise evidence in the literature as to the innovative dynamics of knowledge-sharing and scholarly communication within linguistically diver- se scholarly contexts and research networks; 2) have a better understanding of the role of multilingualism within bibliodiversity in scholarly communication thro- ugh the lens of publishers, researchers, and translators; and 3) present the conceptual design of a future OPE- RAS Translation Platform aimed at supporting translation services at the scholarly communication level. Aim of the study • to prepare a theoretical background to discuss the use of multilingualism in scholarly communication; • to identify, analyse, and understand the innovative dynamics of working practices and knowledge-sharing within linguistically diverse scholarly contexts and rese- arch networks; Survey Multilingualism in Scholarly Communication: This report presents the main results and an in-depth analysis of the survey Multilingualism in Social Sciences and Humanities, which was conducted during the sum- mer of 2020 (from 19 June to 20 August), in the form of an online survey distributed among researchers, translators, and publishers within the OPERAS network and other channels. A total of 359 participants respon- ded to the survey in which they were given a common initial set of questions, followed by their own contribu- tion according to three different perspectives (resear- chers, translators, and publishers), separate or combi- ned depending on the respondents’ profiles. Following the first step of the literature review, the empirical survey led to two main objectives: to collect evidence as to the role of multilingualism within bibliodiversity in scholar- ly communication, and to contribute to the conceptual design of a platform prototype for community-owned translation services at the scholarly communication level, both involving the needs of publishers, translators, and researchers. • to identify and analyse the motivations behind these practices (questionnaires/focus groups – how tools may answer to needs); • to formulate recommendations/guidelines for OPERAS and other stakeholders regarding the future implemen- tation of a service aimed at enhancing multilingualism; • to prepare the conceptual design of a platform pro- totype for a shared translation service at the scholarly communication level (involving publishers, translators, and researchers). Literature review. The literature review was a qualitative study of an explo- ratory nature; the method used is in the scope of an in- tegrative literature review, summarising prior research to clarify research trends based on in vivo content analy- sis of the selected corpus. This method follows several stages, starting with problem formulation, which frames the data collection; it is then followed by selection, tre- atment, analysis, and the final presentation of results. 28 OPERAS-P SIG for Multi- lingualism Survey Scholarly debate Endowment of national languages Final Report OPERAS-P Multilingualism in Scholarly Communication (19th June to 20th August 2020) Future OPERAS translation service monitoring practices and stimulating discussion Multilingualism within Scholarly Communication in SSH – a literature review (May 2021) Ana Balula and Delfim Leão Multilingualism in the SSH (3rd October 2020) OPERAS Research Infrastructure in 90’: Mission, Vision, Action at the OPERAS 2020 Conference Publishers and Publishing Stances: a Contribution to Multilingualism and Bibliodiversity (24th April 2021) Joint workshop OPERAS SIG on Multilingualism and Advocacy, with the OPERAS-P and the TRIPLE projects, and APEES – the Portuguese Association of Higher Education Presses Multilingualism in Scholarly Communication: a Preliminary Report (SIG + OPERAS-P 6.4) (21st September 2020) Innovative Models of Bibliodiversity in Scholarly Publications promoting workshops 2020 Workshop Multilingualism - satellite event of the OASPA Conference 2020 Summary of activities undertaken during the work on the report on Innovative models of bibliodiversity in scholarly publications Survey Multilingualism in Scholarly Communication (19th June to 20th August 2020) Future OPERAS translation service Endowment of national languages Multilingualism in the SSH (3rd October 2020) OPERAS Research Infrastructure in 90’: Mission, Vision, Action at the OPERAS 2020 Conference promoting workshops Multilingualism in Scholarly Communication: a Preliminary Report (SIG + OPERAS-P 6.4) (21st September 2020) 2020 Workshop Multilingualism - satellite event of the OASPA Conference 2020 Summary of activities undertaken during the work on the report on Innovative models of bibliodiversity in scholarly publications Summary of activities undertaken during the work on the report on Innovative models of bibliodiversity in scholarly publications 29 Research relevance “Englishisation” does not seem to fully address the in- tended main goals of research concerning information sharing and discussion, or the co-construction of know- ledge, for which multilingualism can be an important asset, while promoting inclusiveness and equity among researchers. Boosting balanced multilingualism In the literature review, the classification of the cor- pus regarding the dynamics between multilingualism and scholarly communication in SSH was identified in vivo and structured as four categories: A scenario that has become increasingly clear during the development of the different phases of the task is that multilingualism must be perceived as a strong manifestation of bibliodiversity, which is particularly im- portant in SSH. This does not preclude the use of English as a communication language, as long as the advanta- ges of using a lingua franca do not risk turning it into the lingua unica of scientific and scholarly communica- tion. Instead, innovative solutions must be implemented so they have the ability to enhance balanced multilingu- alism in scholarly communication, in information-sharing, and in collaborative knowledge construction. Conceiving that platform as a social infrastructure Conceiving that platform as a social infrastructure By federating technical knowledge and scholarly experti- se, the social infrastructure will stimulate the sharing of tools, methodologies, and practices so that a broad user community can test and scale what is being developed separately by individual partners. Challenges Boosting balanced multilingualism Content curation The possibility of having reliable multilingual research information available definitely contributes to an effi- cient dissemination of the research (and research data) produced in national languages, as well as contributing to communication among publishers and researchers – thus promoting the development of intercultural, com- parative, and/or complementary studies in SSH. In this context, multilingual content curation is crucial. Developing a community based translation plaform The lack of a platform to support translation to different languages is a limitation for the federative nature of OPE- RAS’ consortium. Perceive and value multilingualism as a strong manifestation of bibliodiversity, which is particularly important in the area of Social Sciences and Humanities. Develop a platform to support scholarly translation, that is community based by boosting the collaborative work of researchers, translators, and publishers, by creating conditions for cooperation and providing information that will enable each scholarly work to identify an appropriate publisher profile, a suitable scientific milieu, and the right partnership in order to disseminate specialised or local scientific production to a wider environment. Get directly involved in studying and promoting the development of a translation platform as one of OPERAS’ future services. Recommendations Target audience Scholars Publishers Recommendation View the amplification of multilingualism as an advantage for fostering international collaborative works and for promoting interculturality, inclusion, and equity. Improve the scholarly communication landscape at the international scale; helping what usually tends to be considered “national” (the use of local languages) to become more clearly “international” (by putting them on the radar of wider networks and within the scope of collaborative interest groups worldwide). Enhance expertise, particularly when it is combined with reciprocity, in order to stimu- late networking and improve bibliodiversity through multilingualism. Advocate for the implementation of innovative solutions that have the ability to enhan- ce balanced multilingualism in scholarly communication, information-sharing, collabo- rative knowledge construction, and careers recognition and credit. Perceive and value multilingualism as a strong manifestation of bibliodiversity, which is particularly important in the area of Social Sciences and Humanities. Reputation Given that a considerable amount of SSH research is pu- blished as monographs and/or in local languages, the use of these databases to evaluate research and establish the researchers or institution’s reputations is necessarily fallacious and limited. Making national production internationally relevant The literature review demonstrated that the notion of international publishing is closely linked to the idea of publishing in English in large international publishing houses; however, by putting a broad universe of small publishers and their authors in contact with each other, it will be possible to find an alternative way to internatio- nalise scholarly production, enhance specific catalogues, and relaunch multilingualism as an expression of bibliodi- versity, inclusion, and scientific maturity. Balanced multilingualism in scholarly communication This is considered to be a golden breakthrough, which embraces information-sharing, collaborative knowledge construction, and equity by enabling global interaction with multinational and multidisciplinary research (and re- searchers), thus mitigating the hurdles underlying static, poor translations while connecting research worldwide. In regard to the survey, the results showed that there was a strong openness among researchers, translators, and publishers in viewing the amplification of multilin- gualism as an advantage both for fostering international collaborative works and for promoting interculturality, inclusion, and equity; highlighting that, • a collaborative system that uses expertise in specific areas to support and facilitate translations could reduce time consumption, and mitigate the risks of flaws and high prices that may be attached to the translation pro- cess; • and, the exchange of experiences and specificities, among researchers who speak different languages but share the same areas of study, has the potential to make a relevant contribution to enriching international collabo- ration and to make an impact on works. 30 Data and further reading • Data related to the survey Multilingualism in Scholarly Communication are available at OPERAS-P collection at Nakala. • Full report from this task is available at OPERAS Inno- vation Lab community on Zenodo. • The issue concerning the literature review and balan- ced multilingualism is further elaborated in: Ana Balula and Delfim Leão, “Multilingualism within Scholarly Com- munication in SSH – a literature review,” JLIS.it 12, 2 (May 2021): 88-98. DOI: dx.doi.org/10.4403/jlis.it-12672 31 05 CHAPTER Digitally-enabled vs. digital writing Writing is a deeply social and technologically supported activity; with the discovery, storing, curating, and inter- preting of research resources being part of the writing process, as each of these activities influences the out- come. We distinguish between digitally-enabled writing and digital writing. Both refer to writing as a textual prac- tice supported by various digital tools, but differ in the degree to which writing harnesses the full potential of digital technology by establishing different kinds of ma- terials, such as data, visualisation, or pictures, in a single output; and in the degree to which the final output differs from the traditional codex format and linear narrative. • 56 case studies complement these insights with an analysis of selected innovative tools and services. Some of the insights gained in this study were imple- mented in two conceptual prototypes for innovative services. The study touches on various dimensions of scholar- ly communication. First, in order to open-up the study to various materials, we agreed to treat the “scholarly text” broadly, not only as a standard, written articulation, but rather as an expression that can employ different media. Digitally-enabled vs. digital writing Second, we prepared working definitions for the main concepts pertinent to the task: communicating (the act of sharing a text through various formal or infor- mal channels); specificity of SSH (scholarly communi- cation practices in Social Sciences and Humanities that are different from other fields); writing (the act of gene- rating a text, understood as the expression of an argu- ment that may use various media, formats, and genres); collaboration (collective activities that are undertaken in writing, communicating, publishing, and peer-review); tools (the services and software used in the process of writing, communicating, and publishing at various sta- ges of the researchers’ workflow); publishing (the act of disseminating a text through a formal process, inc- luding intermediaries); innovative forms and genres (text used by scholars to transmit their argument that are beyond the traditional formats of the journal article, book, report, etc.); audiences (the public who enga- ge with scholarly texts and their authors); evaluating (the critical assessment of the products of all types of scholarly communication, i.e., writing, communicating, publishing); innovative forms of peer-review (peer-re- view practices that go beyond the commonly accepted forms to address the perceived deficiencies of the sys- tem); academic prestige (widespread respect attached to certain practices by scholarly communities); and po- wer structures (dynamic systems of hierarchy and influ- ence in scholarly communication). The tools supporting writing are chosen according to individual preferences, disciplinary needs, and competences. Adapting the Levi-Strauss approach3, we distinguish between two types of tool users: engi- neers and bricoleurs. The engineers are experts in many specialist tools and fluidly switch between them. Brico- leurs, on the other hand, still combine digital practices with analogue, offline ones. Simpler tools are used at the ideation stage of a project, while more advanced tools support more mature stages. The choice of tool in a collaborative setting is a tra- de-off between needs and functionalities, and is often a matter of a common denominator between the competences of the team members. It would not be an exaggeration to say that all of our interviewees who write collaboratively have used Google docs for this pro- cess. Innovation is seen as a chance to improve the sha- ring of ideas with audiences, thanks to technologi- cal affordances. Innovation is, then, understood either in terms of form, i.e. novel means of communicating ideas, allowing for expression in other media and linking data with text (e.g. ³ Lévi-Strauss, Claude. The savage mind, Chicago 1966, pp. 17-18. Cf. Antonijevic, Smiljana and Ellysa Stern Cahoy:“Researcher as Bricoleur: Contextualizing humanists’ digital work- flows”. DHQ. 12(3) (2018). Future of scholarly writing in SSH Authors: Maciej Maryl, Marta Błaszczyńska, Agnieszka Szulińska, Anna Buchner, Piotr Wciślik, Iva Melinščak Zlodi, Jadranka Stojanovski, Elisa Nury, Claire Clivaz, Bartłomiej Szleszyński, Kajetan Mojsak, Mateusz Franczak Contributors: Interviewing and coding: Erzsébet Tóth-Czifra Interviewing: Michael Kaiser, Valérie Schaefer, Vera Chiquet, Mate Juric, and Krešimir Zauder Case study analysis: Paweł Rams, Katarzyna Jarzyńska, Renata Rokicka, Izabella Ratyńska, Hanna Kwaśny, Marcin Wilkowski and Zofia Leśnik Internal review: Pierre Mounier, Jennifer Edmond Other: Agata Morka, Magdalena Wnuk Partners involved: Institute of Literary Research of the Polish Academy of Sciences (IBL PAN), DARIAH-ERIC, Swiss Institute of Bioinformatics, Max Weber Foundation, Open Book Publishers, University of Luxembourg and University of Zadar. Specificity of SSHf The work undertaken in Task 6.5 was aimed at exploring current writing practices in SSH and, thus, will inform future OPERAS activities on researchers’ needs regar- ding publishing technologies, and both ongoing and up- coming transformations of scholarly communication. There are differences in scholarly communication be- tween SSH and other disciplines as well as within the disciplines of SSH itself. These concern issues ranging from output genres and the aims of peer-review, to col- laboration strategies and funding. The main communi- cation genre reflects the features that are valued most by particular disciplines: in the case of the sciences, this is the timely reporting of facts through journal articles, while the humanities value the depth and breadth of the interpretation conveyed by a monograph. In order to address these aims the research team ad- opted a methodology combining three approaches: In order to address these aims the research team ad- opted a methodology combining three approaches: In order to address these aims the research team ad- opted a methodology combining three approaches: The literature review, presenting our interpretation of the current trends in scholarly communication. The literature review, presenting our interpretation of the current trends in scholarly communication. • 41 interviews with scholars and publishers provided insights into the practices and needs of the actors within scholarly communication with regard to the innovative aspects of scholarly communication. Thirty-two have been transcribed and analysed, and nine have been sum- marised for further use in research. Challenges prefer publishing in publications with good quality me- tadata, in high impact international journals that are in- dexed in international citation indexes, in the English language, and in reputable monograph series that will attract many book reviews. Articles are considered more practical for communication due to their conformity to metrics and their speed of publication, even though, in the humanities, the monograph continues to confer scholarly reputation. Innovative forms of writing do not yet have an esta- blished position in academia. Some respondents had already expected novel solutions from their colleagues and referred to digital outputs (such as blogs or tweets) in their own work, whereas others saw them as underva- lued and difficult to cite. Lack of digital competences impedes the transition to digital forms of publication. Many traditional forms of humanities’ outputs have their counterparts in the digi- tal space (e.g. scholarly digital editions and monographs). While these expand the possibilities of a given form and their use in further research, they also require a high degree of digital competence. The competence barrier and associated learning curve often prevents the experts in a field engaging with novel forms. Novel formats and genres are considered more ap- propriate for certain content for several reasons: they are liberating, communicative, interactive, and colla- borative; and they enable versioning and updating. Open access venues are favoured due to institutional mandates, personal principles or ideologies, speed of publication, and the possibility of reaching a wider au- dience. Due to problems with visibility for scholarly as- sessment, early career researchers are more constra- ined in their choice of innovative forms than established scholars. Many scholars consider the APC/BPC model of open access exploitative, and prohibitive without pro- ject funding. Innovation is impeded by such factors as quali- ty assessment, prestige, competencies, and a lack of established standards for referencing novel forms. The issue of how to use novel sources in a scholarly text is one of the challenges of 21st-century scholarly writing. These challenges push scholars toward practices of do- uble referencing and double publication, whereby the tra- ditional publication provides prestige for the novel form. Scholars tend to associate new formats of scholar- ly communication with the possibility for wider so- cietal outreach, however, they are aware that it often does not correlate with bibliometric impact, and that for many reasons other avenues with high societal outreach are considered inferior by academics. Digitally-enabled vs. digital writing computational essays, web books, living books, video essays), or access, i.e., providing access to more traditional types of outputs, including grey literature. Innovation is also considered helpful in reaching wider, often non academic, audiences (e.g. blo- gs, podcasts, videos). Data and software are considered valid outputs of SSH research. The choice of traditional types of publication is affected by discoverability and prestige. Authors try to choose publication venues based on their expec- ted future discoverability and visibility. They often prefer The choice of traditional types of publication is affected by discoverability and prestige. Authors try to choose publication venues based on their expec- ted future discoverability and visibility. They often prefer 34 prefer publishing in publications with good quality me- tadata, in high impact international journals that are in- dexed in international citation indexes, in the English language, and in reputable monograph series that will attract many book reviews. Articles are considered more practical for communication due to their conformity to metrics and their speed of publication, even though, in the humanities, the monograph continues to confer scholarly reputation. Recommendation Target audience Publishers, researches Develop publishing guidelines with regard to innovative genres. This should be done in cooperation with scholars to allow for the publishing of supplemental material, data, and code from a study alongside the text. Publishers should not discourage innovation. Support novel communication practices, going beyond traditional formats, inclu- ding data preparation, and publication. Extend the definition of approved project out- puts and provide direct funding for open science practices, e.g., support project members who report data preparation activities, and fund open access publications. Funders, policy makers, Institutions Recognise novel communication forms as valid scholarly outputs and incentivise their use. The lack of rewards halt innovation, as scholars turn to more “recognised” types of outputs. Encourage novel types of outputs and recognise the value of innovations in scho- larly communication. Treat digital, innovative formats as “equal” to established genres (books, ebooks, articles) in teaching and referencing. If you have used such a resource (e.g. video clip, blog post or website), you should quote it directly instead of looking for a “traditional” publication where the author may have said something similar. Offset competence deficits by engaging in team work. Scholars lacking digital skills should partner with colleagues who are able to take care of technical issues. Provide services that respond to the innovative needs of scholars, allowing the con- nections between writing and data to be established (both at a publishing and a discovery level). Recognise the variety of roles scholars take up in scholarly communication: as authors, lead authors in collaborative writing, journal or book editors, reviewers, data managers, or software developers. In each of these roles they perform scholarly activities that often go unnoticed or unrewarded, which, in effect, may discourage researchers from undertaking them. Make your research available using new channels and try to integrate novel com- munication skills into your work and teaching curricula. Teaching new communication tools should be widespread as it concerns a fundamental scholarly activity. Provide targeted training in innovative publishing tools and services, tailored to different levels of users. Lowering the threshold for using novel genres is key to their wider uptake. Provide integrated, sustainable, and modular services for innovative genres in SSH (e.g. digital scholarly editions) that can be used by scholars or publishers. Integration may help in the standardisation of novel formats. Create policies for open access and data practices, providing clear guidance for scholars. Challenges Power structures block innovation. Researchers them- selves, and the community more broadly, are recognised as important actors in the SSH scholarly communication landscape. Depending on their approach, they can play the role of guardians of the status quo, or innovation facilitators. Innovative forms of writing could challenge traditional structures, giving more gatekeeping power to the wider readership community.i Editors of journals and commissioning editors for books are considered power brokers in academia, however, their position might now be challenged with the emergence of new platforms for scholarly commu- nication (in particular, based on open, collaborative peer review), which gives more gatekeeping power to the community. There is a large gap between the apparent benefits of open access and the present criteria for academic career advancement, and scholars fear that publishing in open access could impair their chances of employ- ment, diminish the value of their CV, or reduce their ca- reer prospects. Lack of digital competence keeps many researchers who are experts in their fields from using digital forms of scholarly publication. This is a group of re- searchers who would be willing to use digital tools and forms of research publication in some form, but do not do so because the “entry threshold” is too high. Popularisation of research is not encouraged by pre- stige structures despite its high societal impact. Scho- lars lack clear incentives to engage with wider audiences. The digital environment is open to scholars without technical skills. Many tools and services are currently prepared for non-tech-savvy users from academic circ- les. Workshops, training, and support services become crucial in making these services accessible and achie- ving impact. Competence gaps discourage the uptake of tools. Scholars often fail to use tools or services that could be beneficial for their services because they lack the com- petences to take full advantage of them. Tools and service providers aim to create commu- nities around particular projects. They let scholars define, design, evaluate, comment, test, and perform other kinds of activities, which allows them to become co-authors rather than passive end-users. 35 Recommendation Do not discourage novel publication practices but rather be vocal about the need for policy makers to recognise them as valid scholarly outputs. 36 Data and further reading • Interview transcripts together with the interview scenario and coding scheme are available at OPERAS-P collection at Nakala. • Elisa Nury, “New models of scholarly writing. DH+ SIB contribution to OPERAS-P”, Digital Humanities+ blog, 202137. • Elisa Nury, “New models of scholarly writing. DH+ SIB contribution to OPERAS-P”, Digital Humanities+ blog, 202137. • The full report from this task is available at OPERAS Innovation Lab community on Zenodo. • Maciej Maryl, Marta Błaszczyńska, Agnieszka Szuliń- ska, Paweł Rams, “The Case for an Inclusive Scholarly Communication Infrastructure for Social Sciences and Humanities”, F1000Research, 2020, volume 9. DOI: doi. org/10.12688/f1000research.26545.1. • The issue of evaluation and peer review is further ela- borated in: Erzsébet Tóth-Czifra, “Rethinking text, techné and tenure: evaluation and peer review challenges aro- und Virtual Research Environments in the Arts and Hu- manities”. In “Ancient Manuscripts and Virtual Research Environments”, ed. Claire Clivaz and Garrick V. Allen, spe- cial issue, Classics@ 18 (2021). [N.p.] (forthcoming)36. • Maciej Maryl, Marta Błaszczyńska, Agnieszka Szuliń- ska, Paweł Rams, “Transforming the publishing landsca- pe to create an inclusive SSH scholarly infrastructure”, F1000 Blogs38, 2021. • Elisa Nury and Claire Clivaz, with Marta Błaszczyńska, Michael Kaiser, Agata Morka, Valérie Schaefer, Jadranka Stojanovski and Erzsébet Tóth-Czifra, “Open Research Data and Innovative Scholarly Writing: OPERAS highli- ghts”, Proceedings of the Swiss Data Research Day 2020, Makhlouf Shabou Basma et al. (eds.), RESSI 2021, forthcoming. • Magdalena Wnuk, Maciej Maryl, Marta Błaszczyń- ska, “Report on the OPERAS-P workshop >>The Future of Scholarly Communication<<”, OPERAS Blog39, 2021. • Magdalena Wnuk, Maciej Maryl, Marta Błaszczyń- ska, “Report on the OPERAS-P workshop >>The Future of Scholarly Communication<<”, OPERAS Blog39, 2021. • “>>OPERAS-PL – communication in humanities and social sciences<<: report on the Polish Node meeting” OPERAS Blog40, 2021. • Maciej Maryl, Marta Błaszczyńska, Bartłomiej Szleszyń- ski, Tomasz Umerle, “Dane badawcze w literaturoznaw- stwie” [Research data in literary studies], Teksty Drugie, 2021, issue 2, pp. 13-44. DOI: 10.18318/td.2021.2.2. Data and further reading Use new channels to make your research available Recognise and incentivise novel communication forms Provide policies and guidelines on open access and research data Encourage innovative publications Recognise the variety of roles in scholarly communication Recognise and reference innovative formats Support the use of innovative communication practices and formats in grant projects Develop publishing guidelines with regard to innovative genres Provide innovative services respon- ding to the needs of scholars Provide innovative services that could be employed by publishers OPERAS Scholars Funders Publishers Research institutions Policy makers How to support innovative communication practices among researchers How to support innovative communication practices among researchers How to support innovative communication practices among researchers Research institutions Provide innovative services respon- ding to the needs of scholars Recognise and reference innovative formats 37 06 CHAPTER Main Findings This chapter aims to present the results of the work con- ducted in Task 6.6 (Quality Assessment of SSH Research: Innovations and Challenges) of the OPERAS-P (Open Scholarly Communication in the European Research Area for Social Sciences and Humanities – Preparation) project. The task aimed to better understand the ways in which peer review works in actual SSH practices. In the present report, we analyse key aspects of peer review that normally remain hidden from analysis. This work supports the development of relevant OPE- RAS activities and services by informing them about current trends, gaps, and community needs in research evaluation. This entails 1) teasing out the underlying reasons behind the persistence of certain proxies in the system (such as the “impact factors of the mind” that continue to assign tacit prestige to certain pu- blishers and forms of scholarship), and 2) the analysis of emerging trends and future innovation in peer re- view activities within the SSH domain. This latter com- prises two areas: innovation in peer review workflows (the different flavours of openness, novel practices, and tools), and the peer review of digital scholarly ob- jects (such as digital critical editions, data, software etc.). Peer review is embedded in the broader systems of academic power structures, commonly referred to as the prestige economy. There is tension between bibliometrics and disciplinary community norms of excellence. The three most frequently discussed and most contro- versial functions of peer review have been identified as: constructive improvement of scholarly works, gateke- eping, and constructing/shaping disciplinary identities and boundaries. Gatekeeping and improvement mechanisms are some- times seen as opposing processes, as gatekeeping often gives rise to the strengthening of established power positions. The shortage of reviewers opens the door for young scholars to establish themselves as reviewers. This, of course, does not automatically mean that young scholars have equal opportunities to gain experience in reviewing or enter a gatekeeping position. One’s ne- tworks and institutional prestige can be a game-chan- ger here. Besides, our respondents repeatedly voiced the need to support PhD students and early career rese- archers in becoming thoughtful reviewers. Main Findings The goal of our study was to gain an in-depth under- standing of how the notion of excellence and other peer review proxies are constructed and (re)negotia- ted in everyday practices across SSH disciplines, who is involved in the processes and who remains outsi- de them, what are the boundaries of peer review in terms of inclusiveness of content types, and how are the processes aligned or misaligned with research realities. To achieve this, we undertook and analysed 32 in-depth interviews with scholars about their motivations, challen- ges, and experiences with novel practices in scholarly writing and in peer-review. This input and the encoded and pseudonymised interview transcripts will be sha- red as open data in a certified data repository (NAKA- LA) together with a rich documentation of the process so that our interpretations, conclusions, and the resulting recommendations are clearly delineable from the rich input we were working with and that are, thus, openly reusable for other purposes. The special “flavours” of peer review in SSH, as reflec- ted in the interviews, include: • Peer review has a crucial role in shaping discipli- nary identities. y • Editorial curation is central to research evaluation, and editors are in an especially powerful gatekeeping position. • Publication forums are strongly associated with scholarly networks. • Peer review in SSH deviates from its positivist tra- ditions: quality judgements are situated deeply in smaller epistemic cultures and, therefore, in many cases, resist the pass/fail approach. y p pp • There is a diversity of scholarly content types, often involving multimedia, that remain outside the sco- pe of formal peer review. Publishing review texts anonymously alongside publi- cations turned out to be the flavour of openness that en- joyed the most support by our respondents, with some even endorsing it. The main incentives reported by our respondents were purely scholarly in nature (e.g. advancing one’s field, curiosity, chances to contribute to the knowledge commons; the prestige of invitation), rather than moneta- ry or other in-kind rewards (APC discounts, vouchers etc.). Assessing the quality of scholarship and continuing the discussion around them is a much more abundant and prevalent activity than is channelled in formal peer review discourses. It occurs naturally in conference di- scussions, on social media, in book reviews, including their new media equivalents such as podcasts and lite- rature reviews. Authors: Erzsébet Tóth-Czifra, Marta Błaszczyńska Contributors: Methodology, interviewing and coding: Anna Buchner, Maciej Maryl Interviewing and coding: Jadranka Stojanovski, Iva Melinščak Zlodi Interviewing: Michael Kaiser, Valérie Schaefer, Claire Clivaz, Elisa Nury, Vera Chiquet, Mate Juric, and Krešimir Zauder Internal review: Jennifer Edmond, Maciej Maryl Partners involved: DARIAH-EU (Jennifer Edmond, Erzsébet Tóth-Czifra); Institute of Literary Research of the Polish Academy of Sciences (Maciej Maryl, Marta Błaszczyńska, Anna Buchner, Agnieszka Szulińska, Paweł Rams, Mateusz Franczak); University of Zadar (Jadranka Stojanovski, Iva Melinščak Zlodi, Krešimir Zauder) Authors: Erzsébet Tóth-Czifra, Marta Błaszczyńska Contributors: Methodology, interviewing and coding: Anna Buchner, Maciej Maryl Interviewing and coding: Jadranka Stojanovski, Iva Melinščak Zlodi Interviewing: Michael Kaiser, Valérie Schaefer, Claire Clivaz, Elisa Nury, Vera Chiquet, Mate Juric, and Krešimir Zauder Internal review: Jennifer Edmond, Maciej Maryl Partners involved: DARIAH-EU (Jennifer Edmond, Erzsébet Tóth-Czifra); Institute of Literary Research of the Polish Academy of Sciences (Maciej Maryl, Marta Błaszczyńska, Anna Buchner, Agnieszka Szulińska, Paweł Rams, Mateusz Franczak); University of Zadar (Jadranka Stojanovski, Iva Melinščak Zlodi, Krešimir Zauder) Main Findings These spontaneous evaluation practices are perfor- med with the sole intention of continuing a meaningful scholarly dialogue and advancing one’s field. 40 40 Challenges While the act of reviewing is perceived as an important part of academic work, it is difficult to find ways to ad- minister recognition and likewise gain recognition for one’s review record, especially in the case of tra- ditional, blind peer review. In this drought of reviewers, prestigious, well-esta- blished journals attract more reviewers, not just more authors. As a result, established proxies of excel- lence are easily being reproduced. This poses difficulties for the evaluation of interdisciplinary research, and also challenges the inclusion of (born-)digital outputs in for- mal assessment systems. This goes against the pres- sing need for re-harmonizing reviewing practices and research realities. The shortage of evaluative labour is recognised as the key challenge that the institution of formal peer review needs to overcome. It affects and shapes both the pool of reviewers, publishing workflows (inc- luding, of course, the peer review process itself), and the range of scholarship that is eligible for peer review. The institution of peer review can reinforce existing po- wer structures and make it harder for certain scholars to contribute to the community. While editors find it difficult to find good reviewers for their journals, there are groups who are much less likely to get asked to re- view others’ work. The shortage of evaluative labour is recognised as the key challenge that the institution of formal peer review needs to overcome. It affects and shapes both the pool of reviewers, publishing workflows (inc- luding, of course, the peer review process itself), and the range of scholarship that is eligible for peer review. Opening up the peer review processes turned out to be especially challenging in these research con- texts, with strong and complex, but not univocal, com- munity resistance against them. The institution of peer review can reinforce existing po- wer structures and make it harder for certain scholars to contribute to the community. While editors find it difficult to find good reviewers for their journals, there are groups who are much less likely to get asked to re- view others’ work. Recommendation Information management systems that are publicly owned, inclusive, and have a broad range of content types are absolute infrastructural prerequisites for implementing respon- sible research metrics that are transparent and under the control of research communities and ministries. The current tendency for proprietary, closed systems to gain important positions in delivering research metrics poses a significant threat to transparency and community control. Research Infrastructures, OPERAS OPERAS has already invested in such transparent, public infrastructure by implementing the OPERAS Metrics Service, a service that enables the transparent tracking of OA book usage. As a next step, we recommend that OPERAS launch a working group dedi- cated to responsible research metrics that functions as a European level knowledge hub for experts in charge of implementing research metrics in OPERAS’ member countries. Such a coordinated effort could 1) ensure interoperability across national Cur- rent Research Information Systems (CRIS), and 2) could inform future OPERAS services on a regular basis. We also recommend coordinating with ENRESSH along these lines. Enabling the citability of all the various kinds of research outputs beyond the research pa- per is a first step towards these outputs being taken into account for formal assessment. We recommend OPERAS coordinate with DARIAH on advocacy and training efforts towards a better citation culture in the SSH. As a trust building instrument, the transparent but labour-efficient communication of editorial policies and workflows (including how decisions are made and by whom, what kinds of pre-filtering mechanisms are in place, and the average time frame for publi- cations) is crucial to managing expectations for both authors and reviewers. OPERAS co- uld consider extending the Book Peer Review Certification Service in this direction. Being able to administer one’s reviewing record in a publicly owned information management system is an absolute prerequisite for appropriately rewarding peer review activities. Based on previous experience gained through the Open Access Book Peer Review Certification service, OPERAS should explore the possibilities of buil- ding such an infrastructure, which operates with the minimum possible administrative costs for both the publisher and the author/institution (Maybe in collaboration with the CRIS system and its various implementations in OPERAS’ member countries? Building on previous work on the SSH research assessment within the ENRESSH project could be a good starting point). Challenges Venues of scholarly communication, peer review • Limited peer review capacities cannot scale up to the expansion of scholarly communication • Prestige economy: conservative quality proxies are easier to reproduce • The review of interdisciplinary, digital scholarship remains challenging Scholarly work scholarly outputs • Difficult to build capacity for reviewing • A range of content types remain out of the scope of peer review Formal research assessment, policies • Publisher prestige still plays a major role • Reviewing activities are unrewarded reviewing remains a merciless task Inspired by: Eve (2015). Peer review – challenges to its expected functionality translate Peer review – challenges to its expected functionality Venues of scholarly communication, peer review • Limited peer review capacities cannot scale up to the expansion of scholarly communication • Prestige economy: conservative quality proxies are easier to reproduce • The review of interdisciplinary, digital scholarship remains challenging Formal research assessment, policies • Publisher prestige still plays a major role • Reviewing activities are unrewarded reviewing remains a merciless task 41 Recommendation Research Infrastructures, OPERAS, Policy makers Research Infrastructures, OPERAS, Policy makers Publishers, Research Infrastructures In an increasingly complex research assessment landscape, where the visibility of au- tomated workflows, knowledge graphs, and scholarly outputs in information manage- ment systems play an increasingly important role, publishers need to make sure that their content is findable and accessible not only for humans but for machines too so as to enable citation and usage tracking. Authors cannot be disadvantaged in terms of citations and visibility because they are publishing with smaller publishing houses. The ongoing efforts of OPERAS to provide support for smaller publishers so they can upscale their workflows to digital and become interoperable with bigger scholarly informa- tion systems (e.g. providing help with implementing PID systems, developing tools for converting domain-specific formats to global standards) is of vital importance. We recommend continuing and extending this work, for example, with an HTML me- tadata enhancement toolbox that enables publishers to also increase their HTML meta- data quality. Encouraging benevolence and constructiveness in the evaluation guidelines of pu- blication venues could contribute to a healthier and more effective peer review culture. Encouraging benevolence and constructiveness in the evaluation guidelines of pu- blication venues could contribute to a healthier and more effective peer review culture. Publication venues awarding badges to their top reviewers, not only on quantitative but also qualitative basis, could serve as an incentive for constructive improvement. We recommend publishing venues seek ways to better connect or channel informal evaluation practices into formal peer review systems For instance inviting authors Publication venues awarding badges to their top reviewers, not only on quantitative but also qualitative basis, could serve as an incentive for constructive improvement. Publication venues awarding badges to their top reviewers, not only on quantitative but also qualitative basis, could serve as an incentive for constructive improvement. We recommend publishing venues seek ways to better connect or channel informal evaluation practices into formal peer review systems. For instance, inviting authors of review blog posts to upgrade or turn their text into a formal peer review. We recommend publishing venues seek ways to better connect or channel informal evaluation practices into formal peer review systems. For instance, inviting authors of review blog posts to upgrade or turn their text into a formal peer review. 42 To ease the burden of gatekeeping, publication venues should consider implementing a model of peer review similar to Plos One where the scope of peer review is restricted to checking the integrity of scholarly processes and the soundness of the publication rather than making assumptions about their importance or innovation potential. Research metrics, a) need to be developed in conversation with the commu- nities being measured, b) need to be used for the intention they were designed for, and c) need to be applied after situations where infrastructure is needed to support a metric-based approach A crucial step towards capacity building would be if all European countries followed Dutch formal assessment policies, which reward reviewing activities. Even though we are well aware of the “one size doesn’t fit all” golden rule in EU-level research policies, we cannot see any specific contextual issue that would prevent its implementation in a diversity of national contexts across Europe. We recommend OPERAS further investigate any possible infra- structural or policy obstacles. Policy makers, Scholarly communities Policy makers, Scholarly communities Introducing quantitative measures for research evaluation seems to be, to a certain extent, unavoidable so as to enable scholarly works from very different disciplines and regions to be compared. Encouraging benevolence and constructiveness in the evaluation guidelines of pu- blication venues could contribute to a healthier and more effective peer review culture. However, to resolve the conflict between research metrics and research realities, both geographical peculiarities and disciplinary communities of practice need to be taken into account in a flexible and multi-dimensional system of metrics. We recommend further developing the HuMetricsHSS in this direction. Harmonizing HuMetricsHSS’s efforts with the DARIAH Impact Working Group would facilitate coordination from the doma- in-specific angle across geographical regions in Europe and beyond. Building innovative peer review practices on top of already established, proven instan- ces of informal evaluation practices rather than designing them from scratch could be taken as an assurance for community uptake. • Tóth-Czifra, Erzsébet. (2021, March). Quality assess- ment of SSH research: innovations and challenges – workshop slides. Zenodo. Data and further reading • Interview transcripts together with the interview scenario and coding scheme are available at OPERAS-P collection at Nakala. • Elisa Nury and Claire Clivaz, with Marta Błaszczyńska, Michael Kaiser, Agata Morka, Valérie Schaefer, Jadranka Stojanovski and Erzsébet Tóth-Czifra, “Open Research Data and Innovative Scholarly Writing: OPERAS highli- ghts,” Proceedings of the Swiss Data Research Day 2020, Makhlouf Shabou Basma et al. (eds.), RESSI 2021, (forthcoming). • The full report from this task is available at OPERAS Innovation Lab community on Zenodo. • The issue of evaluation and peer review is further ela- borated in: Erzsébet Tóth-Czifra, “Rethinking text, techné and tenure: evaluation and peer review challenges aro- und Virtual Research Environments in the Arts and Hu- manities.” In “Ancient Manuscripts and Virtual Research Environments,” ed. Claire Clivaz and Garrick V. Allen, spe- cial issue, Classics@ 18 (2021). [N.p.](forthcoming). • Tóth-Czifra, Erzsébet. (2021, March). Quality assess- ment of SSH research: innovations and challenges – workshop slides. Zenodo. 43 Overview of the literature Key areas for the investigation of peer review practices in the Humanities Overview of the literature Key areas for the investigation of peer review practices in the Humanities Transparency • The many shades of openness from black box to fully transparent workflows • Administration, tracking • Book peer review as an even bigger mystery/less standarized practice The prestige economy • The symbolic capital and institutional rewarding criteria; container- -level evaluation • Reconsidering academic incentive and reward mechanisms to recognise quality assurance work (reviewing, editing) Special flavours of peer review in SSH • Quality judgements are situated in small epistemic cultures • Editorial curation is central • Book peer review is a diverse and not always transparent practice • Open peer review remains on the level of experiments • ...or the current lack thereof, and its impact on the pace of publishing • Reconsidering the function of peer review: a systemic gatekeeper vs a (continuous) process of collaborative improvement • Around novel digital scholarly objects • Around nice/multidisciplinary areas • “Unofficial” peer review happening outside of academic publishing • Decoupling peer-review from the journal article • ... The Future (instead of conclusions) The Future (instead of conclusions) Data and further reading towards both the individuals involved and the institution itself • The social components of peer-review: control, governance, power dynamics and biases (gender bias, gatekeeping by senior people, etc.) Trust and rethinking power relations Innovations and new evaluation frameworks Rewards, recognition, incentives and capacity building for peer review Rewards, recognition, incentives and capacity building for peer review Rewards, recognition, incentives and capacity building for peer review Transparency • The many shades of openness from black box to fully transparent workflows • Administration, tracking • Book peer review as an even bigger mystery/less standarized practice The prestige economy • The symbolic capital and institutional rewarding criteria; container- -level evaluation • Reconsidering academic incentive and reward mechanisms to recognise quality assurance work (reviewing, editing) Special flavours of peer review in SSH • Quality judgements are situated in small epistemic cultures • Editorial curation is central • Book peer review is a diverse and not always transparent practice • Open peer review remains on the level of experiments • ...or the current lack ther on the pace o • Reconsider a systemic process o • A • A • “ outs • D the • ... to and th • The soc governance bias, gatekeepi Tr po Inn ev Rewards, recognition, i and capacity building f Transparency • The many shades of openness from black box to fully transparent workflows • Administration, tracking • Book peer review as an even bigger mystery/less standarized practice penness from black box transparent workflows stration, tracking as an even ry/less ctice y d es ntral diverse nt practice • ...or the current lack t on the pa • Recons a syste proce • an • The governa bias, gateke I p y • Around novel digital scholarly objects • Around nice/multidisciplinary areas • “Unofficial” peer review happening outside of academic publishing • Decoupling peer-review from the journal article Innovations and new evaluation frameworks 44 The body of work collected in this report and the associa- ted research outputs from Work Package 6 of the OPE- RAS-P project lay the foundations for the future develop- ment of OPERAS and its services to the SSH community. We are extremely grateful to the almost one thousand people from various countries and stakeholder commu- nities for contributing in various ways to the findings of this report through surveys, interviews, workshop participation, or in other ways. It is thanks to you that the authors of this report were able to keep the findings and recommendations as close to actual community ne- eds as possible. These outcomes will inform the future work of OPE- RAS in three major ways. First, they will provide a ratio- nale for further operation, and a background for strategic decisions. Second, they will serve as blueprints for co- operation with various stakeholders and will help to keep OPERAS’ offering in close alignment with actual scholar- ly needs. Finally, they provide rough sketches of novel services as well as a portfolio of concrete activities to be pursued in the nearest future. This scoping exercise will be continued by OPERAS’ Innovation Lab, an initiative spearheaded by IBL PAN, which will carry out research on users’ needs and fu- ture services. We want the OPERAS Lab to become a space for discussing, envisioning, testing, and prototy- ping new solutions for the community. This report is just the beginning. 45
https://openalex.org/W4393143729	https://www.mdpi.com/2504-3900/97/1/82/pdf?version=1711156229	English	null	Gas Sensing Capabilities of CuInS2/ZnO Core–Shell Quantum Dot	null	2,024	cc-by	1,646	Antonio Orlando 1,2,, Guglielmo Trentini 1,2, Pietro Tosato 2 , Soufiane Krik 1 , Matteo Valt 2 , Andrea Gaiardo 2 and Luisa Petti 1 Antonio Orlando 1,2,, Guglielmo Trentini 1,2, Pietro Tosato 2 , Soufiane Krik 1 , Matteo Valt 2 , Andrea Gaiardo 2 and Luisa Petti 1 1 Sensing Technologies Lab, Faculty of Engineering, Free University of Bolzano-Bozen, Piazza Università 5, 39100 Bolzano, Italy; guglielmo.trentini@student.unibz.it (G.T.); soufiane.krik@unibz.it (S.K.); luisa.petti@unibz.it (L.P.) 1 Sensing Technologies Lab, Faculty of Engineering, Free University of Bolzano-Bozen, Piazza Università 5, 39100 Bolzano, Italy; guglielmo.trentini@student.unibz.it (G.T.); soufiane.krik@unibz.it (S.K.); luisa.petti@unibz.it (L.P.) p 2 Sensors and Devices Center, Bruno Kessler Foundation, Via Sommarive 18, 38123 Trento, Italy; ptosato@fbk.eu (P.T.); mvalt@fbk.eu (M.V.); gaiardo@fbk.eu (A.G.) * Correspondence: antonio.orlando@natec.unibz.it; Tel.: +39-0461-31415033 † Presented at the XXXV EUROSENSORS Conference Lecce Italy 10–13 September 2023 p 2 Sensors and Devices Center, Bruno Kessler Foundation, Via Sommarive 18, 38 ptosato@fbk.eu (P.T.); mvalt@fbk.eu (M.V.); gaiardo@fbk.eu (A.G.) Correspondence: antonio.orlando@natec.unibz.it; Tel.: +39-04 * Correspondence: antonio.orlando@natec.unibz.it; Tel.: +39-0461-31415033 † Presented at the XXXV EUROSENSORS Conference, Lecce, Italy, 10–13 September 2023 † Presented at the XXXV EUROSENSORS Conference, Lecce, Italy, 10–13 Sept Abstract: Chemoresistive gas sensors are surely one of the easiest and most commonly used methods to monitor the presence of different polluting gases. Nevertheless, there are still several challenges to overcome in order for these sensors to be widely used. In particular, the selectivity and sensitivity of chemoresistive gas sensors towards a wide range of analytes need to be improved. This is why new sensing materials capable of detecting different analytes in a sensitive and selective manner are being investigated. In this regard, this work is focused on the development and characterization of a new sensing material based on the quantum dot (QD) core–shell of CuInS2/ZnO (CIS-ZO). Optimized films of the QD core–shell of CIS-ZO were integrated into a micro-electromechanical system (MEMS)-based gas sensor platform, showing excellent sensing performance versus different gases and especially towards ethanol (C2H5OH). Keywords: chemoresistive gas sensor; quantum dots; micro-electromechanical system 1. Introduction In recent years, low dimensional materials (2D, 1D, 0D) have attracted researchers’ attention thanks to their high-surface-to-volume ratio, low cost, and optimal electrical proprieties. In particular, in the field of gas sensing, they have represented a step forward thanks to the high active surface area, allowing a strong interaction with target molecules within rapid response times. Moreover, the possibility to tailor these materials with different properties could also enable the selective detection of specific target gases, allowing the realization of highly selective devices [1]. Copyright: © 2024 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/). proceedings proceedings proceedings proceedings Citation: Orlando, A.; Trentini, G.; Tosato, P.; Krik, S.; Valt, M.; Gaiardo, A.; Petti, L. Gas Sensing Capabilities of CuInS2/ZnO Core–Shell Quantum Dot. Proceedings 2024, 97, 82. https:// doi.org/10.3390/proceedings 2024097082 g y Starting from metal oxide-based semiconductors (especially tin oxide, zinc oxide, and tungsten oxide) that are commonly used as gas sensing materials, many recent studies have focused on the fabrication of gas sensors with metal oxide-based quantum dots (QDs) as a sensing material. Academic Editors: Pietro Siciliano and Luca Francioso In this work, we developed a micro-electromechanical system (MEMS)-based gas sensor that uses the QD core–shell of CuInS2/ZnO (CIS-ZO) as a sensing material for the detection of different gases. In particular, ethanol (C2H5OH) showed very high sensitivity and fast response/recovery. Published: 22 March 2024 2. Materials and Methods In this study, the MEMS-based gas sensor platform presented in [2] was used for the investigation of the sensing proprieties of CIS-ZO. CIS-ZO was obtained after the oxidation at 650 ◦C in air of QDs of CuInS2/ZnS (ZCIS) synthesized with the procedure reported by W. Hu et al. [3]. The powder of the obtained CIS-ZO was characterized using scanning electron microscopy (SEM), as shown in Figure 1a. https://www.mdpi.com/journal/proceedings Proceedings 2024, 97, 82. https://doi.org/10.3390/proceedings2024097082 Proceedings 2024, 97, 82 2 of 3 conce (b) (a) (b) igure 1. (a) SEM image on CIS-ZO powder. (b) Response of CIS-ZO–based gas sensors s 𝑅𝑒𝑠𝑝𝑜𝑛𝑠𝑒 ൌ ሺ 0 R R ሻ െ 1 with diﬀerent concentrations of ethanol (5, 10, 15, and 20 ppm) Figure 1. (a) SEM image on CIS-ZO powder. (b) Response of CIS-ZO–based gas sensors calculated as Response = ( R0 Rg ) −1 with different concentrations of ethanol (5, 10, 15, and 20 ppm) at 0 RH% and 400 ◦C. (a) (a) (b) ure 1. (a) SEM image on CIS-ZO powder. (b) Response of CIS-ZO–based gas sensors 𝑒𝑠𝑝𝑜𝑛𝑠𝑒 ൌ ሺ 0 R R ሻ െ 1 with diﬀerent concentrations of ethanol (5, 10, 15, and 20 ppm Figure 1. (a) SEM image on CIS-ZO powder. (b) Response of CIS-ZO–based gas sensors calculated as Response = ( R0 Rg ) −1 with different concentrations of ethanol (5, 10, 15, and 20 ppm) at 0 RH% and 400 ◦C. g R d 400 °C. Discussion In order to perform an optimal deposition of the CIS-ZO on the MEMS device, the CIS-ZO powder was mixed with an organic matrix in order to obtain a screen-printable paste. The paste was then deposited on the device by means of screen printing, using the Aurel 1520B. The SEM image of the CIS-ZO (Figure 1a) showed an average diamet noparticles of 5 nm. The sensing performances of CIS-ZO were tested using th ricated gas sensor platform. For the sensing measurements, the sensing ma The fabricated device was bounded on a gold support (TO-39) in order to integrate it into a customized chamber for the resistance measurements. The chamber was then connected to a dedicated gas mixing system capable of obtaining different concentrations of different gases. ted up to 400 devices were 3. Discussion Conflicts of Interest: The authors declare no conflicts of interest. 1. Galstyan, V. “Quantum Dots: Perspectives in next-Generation Chemical Gas Sensors”—A Review. Anal. Chim. Acta 2021, 1152, 238192. [CrossRef] [PubMed] ted up to 400 devices were 3. Discussion e devices were tested versus 5, 10, 15, and 20 ppm of ethanol (Figure 1b). Th wed excellent reproducibility and stability in the response. Moreover, they s h response (18.5 vs. 20 ppm of ethanol) and excellent recovery. hor Contributions: Conceptualization, A.O., A.G., M.V. and L.P.; methodolo estigation, A.O.; data curation, A.O. and G.T.; writing—original draft preparat ting—review and editing, A.G., S.K., P.T., M.V. and L.P.; supervision, S.K., M.V., A.G The SEM image of the CIS-ZO (Figure 1a) showed an average diameter of the nanopar- ticles of 5 nm. The sensing performances of CIS-ZO were tested using the MEMS-fabricated gas sensor platform. For the sensing measurements, the sensing material was heated up to 400 ◦C, and the relative humidity in the gas chamber was maintained at 0%. The devices were tested versus 5, 10, 15, and 20 ppm of ethanol (Figure 1b). The results showed excel- lent reproducibility and stability in the response. Moreover, they showed a high response (18.5 vs. 20 ppm of ethanol) and excellent recovery. authors have read and agreed to the published version of the manuscript. ding: This research received no external funding. itutional Review Board Statement: Not applicable. Author Contributions: Conceptualization, A.O., A.G., M.V. and L.P.; methodology, A.O.; investiga- tion, A.O.; data curation, A.O. and G.T.; writing—original draft preparation, A.O.; writing—review and editing, A.G., S.K., P.T., M.V. and L.P.; supervision, S.K., M.V., A.G. and L.P. All authors have read and agreed to the published version of the manuscript. ormed Consent Statement: Not applicable. Funding: This research received no external funding. ormed Consent Statement: Not applicable. Funding: This research received no external funding. pp a Availability Statement: The data that support Institutional Review Board Statement: Not applicable. a Availability Statement: The data that su esponding author, upon reasonable requ Informed Consent Statement: Not applicable. a Availability Statement: The data that su esponding author, upon reasonable requ Informed Consent Statement: Not applicable. p g p q nﬂicts of Interest: The authors declare no conﬂicts of interest. Data Availability Statement: The data that support the findings of this study are available from the corresponding author, upon reasonable request. p g p q nﬂicts of Interest: The authors declare no conﬂicts of interest. Data Availability Statement: The data that support the findings of this study are available from the corresponding author, upon reasonable request. Conflicts of Interest: The authors declare no conflicts of interest. 2. Gaiardo, A.; Novel, D.; Scattolo, E.; Crivellari, M.; Picciotto, A.; Ficorella, F.; Iacob, E.; Bucciarelli, A.; Petti, L.; Lugli, P.; et al. Optimization of a Low-Power Chemoresistive Gas Sensor: Predictive Thermal Modelling and Mechanical Failure Analysis. Sensors 2021, 21, 783. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. [ ] 3. Hu, W.; Yang, S.; Huang, J. Composition Effect on the Carrier Dynamics and Catalytic Performance of CuInS 2 /ZnS Quantum Dots for Light Driven Hydrogen Generation. J. Chem. Phys. 2019, 151, 214705. [CrossRef] [PubMed] References 1. Galstyan, V. “Quantum Dots: Perspectives in next-Generation Chemical Gas Sensors”—A Review. Anal. Chim. Acta 2021, 1152, 238192. [CrossRef] [PubMed] Proceedings 2024, 97, 82 3 of 3 2. Gaiardo, A.; Novel, D.; Scattolo, E.; Crivellari, M.; Picciotto, A.; Ficorella, F.; Iacob, E.; Bucciarelli, A.; Petti, L.; Lugli, P.; et al. Optimization of a Low-Power Chemoresistive Gas Sensor: Predictive Thermal Modelling and Mechanical Failure Analysis. Sensors 2021, 21, 783. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W4229054576	https://link.springer.com/content/pdf/10.1007/s13239-021-00557-4.pdf	English	null	Three-Dimensional Analysis of the In Vivo Motion of Implantable Cardioverter Defibrillator Leads	Cardiovascular engineering and technology	2,021	cc-by	6,329	ABBREVIATIONS ABBREVIATIONS N Number SD Standard deviation Min Minimum Max Maximum Q1 First quartile Q3 Third quartile S Distance to lead tip (along the lead); the lead tip equals the tip of the extended helix max C(s) Maximum curvature of the lead centerline during cardiac cycle at a particular distance to tip min C(s) Minimum curvature of the lead centerline during cardiac cycle at a particular distance to tip Cmax Maximum absolute curvature of the lead centerline during cardiac cycle Camp(s) Curvature amplitude during cardiac cycle at a particular distance to tip; Camp(s) = [max C(s) min C(s)]/2 Camp Maximum curvature amplitude of the lead centerline s(Cmax) Distance of Cmax location to lead tip s(Camp) Distance of Camp location to lead tip Abstract—Better understanding of the lead curvature, move- ment and their spatial distribution may be beneﬁcial in developing lead testing methods, guiding implantations and improving life expectancy of implanted leads. Abstract—Better understanding of the lead curvature, move- ment and their spatial distribution may be beneﬁcial in developing lead testing methods, guiding implantations and improving life expectancy of implanted leads. Objective—The aim of this two-phase study was to develop and test a novel biplane cine-ﬂuoroscopy-based method to evaluate input parameters for bending stress in leads based on their in vivo 3D motion using precisely determined spatial distributions of lead curvatures. Potential tensile, compres- sive or torque forces were not subjects of this study. q j y Methods—A method to measure lead curvature and curva- ture evolution was initially tested in a phantom study. In the second phase using this model 51 patients with implanted ICD leads were included. A biplane cine-ﬂuoroscopy record- ing of the intracardiac region of the lead was performed. The lead centerline and its motion were reconstructed in 3D and used to deﬁne lead curvature and curvature changes. The maximum absolute curvature Cmax during a cardiac cycle, the maximum curvature amplitude Camp and the maximum curvature Cmax@amp at the location of Camp were calculated. These parameters can be used to characterize fatigue stress in a lead under cyclical bending. 1 Results—The medians of Camp and Cmax@amp were 0.18 cm1 and 0.42 cm1, respectively. The median location of Cmax was in the atrium whereas the median location of Camp occurred close to where the transit through the tricuspid valve can be assumed. Increased curvatures were found for higher slack grades. Address correspondence to Tamas Szili-Torok, Department of Cardiology, Erasmus MC, Postbus 2040, 3000 CA Rotterdam, The Netherlands. Electronic mail: t.szilitorok@erasmusmc.nl ABBREVIATIONS Conclusion—Our results suggest that reconstruction of 3D ICD lead motion is feasible using biplane cine-ﬂuoroscopy. Lead curvatures can be computed with high accuracy and the results can be implemented to improve lead design and testing. Three-Dimensional Analysis of the In Vivo Motion of Implantable Cardioverter Deﬁbrillator Leads TAMAS SZILI-TOROK ,1 JENS RUMP,2 TORSTEN LUTHER,2 and SING-CHIEN YAP1 1Department of Cardiology, Erasmus MC, Postbus 2040, 3000 CA Rotterdam, The Netherlands; and 2Biotronik SE & Co. KG, Woermannkehre 1, 12359 Berlin, Germany (Received 20 January 2021; accepted 8 June 2021; published online 29 June 2021) Associate Editor Igor Eﬁmov oversaw the review of this article. Cardiovascular Engineering and Technology, Vol. 13, No. 1, February 2022 ( 2021) pp. 129–138 https://doi.org/10.1007/s13239-021-00557-4 Cardiovascular Engineering and Technology, Vol. 13, No. 1, February 2022 ( 2021) pp. 129–138 https://doi.org/10.1007/s13239-021-00557-4 Cardiovascular Engineering and Technology, Vol. 13, No. 1, February 2022 ( 2021) pp. 129–138 https://doi.org/10.1007/s13239-021-00557-4 BIOMEDICAL ENGINEERING SOCIETY Original Article Original Article INTRODUCTION Keywords—Implantable cardioverter deﬁbrillator, Right ven- tricular, Lead motion, Lead positioning, 3D reconstruction. In vivo lead motion has a signiﬁcant impact on lead performance. Therefore, it recently became a new fo- cus in the regulatory ﬁeld as well as in the ﬁeld of implantable cardiac devices, as lead failure continues to be a crucial cause for malfunctions of the device system. The importance of lead motion analysis becomes especially evident when we look at recently 129 1869-408X/22/0200-0129/0 2021 The Author(s) SZILI-TOROK et al. 130 ously implanted leads were assessed. The Institutional MEC (Medical Ethics Committee) approved the study and all patients gave written informed consent. Pa- tients with an implanted ICD- or cardiac resynchro- nization therapy deﬁbrillator (CRT-D) system were enrolled during a regular patient follow-up at least 3 months after implantation of the device system. It was explicitly favored to enroll patients who received their ICD or CRT-D system from diﬀerent implanting physicians. This was realized and the 51 patients were spread among 13 implanting electrophysiologists. A biplane cine-ﬂuoroscopy procedure was conducted subsequent to the routine follow-up procedure. The cine-ﬂuoroscopy window was limited to the intracar- diac region, which ensured imaging of all electrically active elements of the leads (lead tip, shock coils, ring electrodes—including DX atrial dipole electrodes if applicable) during the full cycle length. The patients were in lying, i.e., horizontal position. No study-re- lated follow-up procedure was required. Completion of the imaging and device follow-up was deﬁned as the end of the study. reported lead-related complications such as lead externalization issues and fractures.1,2,6 Moreover, the development of a new engineering standard for cardiac rhythm management systems is in progress5 to for- mulate requirements on lead fatigue performance based on in vivo lead motion data. Currently, stan- dards are based on established harmonized testing methods, but they do not take lead model speciﬁc use conditions, e.g. by in vivo lead curvatures and cyclic curvature changes, into account. Routine static X-ray images are not able to assess such dynamic lead behavior and therefore cannot be used to analyze lead movements. Hoﬀman et al.10 showed that the determination of the three-dimen- sional in vivo positions of the leads is feasible by two synchronously acquired X-ray images with different view angles to minimize errors due to heart move- ments. Study Design This study was a prospective, exploratory, non- randomized, single-center, feasibility study, and regis- tered at the public national register of the Central Committee on Research Involving Human Subjects. Inclusion criteria were: INTRODUCTION A very limited number of studies are available assessing intracardiac in vivo curvature of leads.4,8,14,18 The aim of our study was to develop and test a novel method to analyze implanted deﬁbrillator lead motion in three dimensions. The bending stress strongly depends on the curvature and the curvature amplitude, whereby the actual strain also depends on the structural properties of the lead and diﬀer depending on lead type and manufacturer. The quan- titative analysis of the speciﬁc stress was beyond the scope of this study. Although diﬀerent biplane imaging systems are clinically approved and in use, a single-center design was chosen for this study to ensure eﬃcient study execution and to avoid possible diﬀerences in imaging systems or system setups. The imaging was done with a Siemens Artis Zee biplane system together with the appropriate software. In addition to the biplane cine-ﬂuoroscopy images, data of the following categories were recorded: demographic data, medical conditions, type of im- planted devices and leads, lead measurements at fol- low-up, X-ray procedural data, and adverse events during the study. Biomechanical Analysis The software tool, developed and successfully tested by Biotronik (Biotronik SE & Co. KG, Berlin, Ger- many), was used to reproduce the spatial (3D) lead movement from the biplane cineﬂuoroscopic imaging data (Fig. 1), and to precisely calculate curvatures and curvature changes of the lead. Version control was done via Lock Modify Write. The latest version was veriﬁed using artiﬁcial datasets based on in vivo data to determine the sensitivity and robustness. The assess- ment of the error, repeatability and reproducibility was done by the phantom study described further below. Patient has provided written informed consent; pa- tient has a Biotronik ICD or CRT-D and at least a Biotronik Linox Smart S DX lead or any other Bio- tronik lead model; patient is able to attend the X-ray procedure following a routine follow-up visit; none of the leads was implanted within the last 3 months; pa- tient had no cardiac intervention within the last 2 months. Exclusion criteria were: Patient age less than 18 years; patient is pregnant or breastfeeding; any complication of the implanted sys- tem at the time of enrollment. The clinical data sets were used to assess the bend- ing stress in leads based on the in vivo lead motion and deformation using associated modeling parameters, e.g. lead redundancy and curvatures. The lead redun- dancy was evaluated according to slack grades devel- The study was conducted in two phases. After developing a method to measure lead curvature and curvature evolution, it was initially tested in a phantom study. In the second phase using this method previ- BIOMEDICAL ENGINEERING SOCIETY BIOMEDICAL ENGINEERING SOCIETY B E S Three-Dimensional Analysis of the In Vivo Motion 131 maximum of max C(s) for all s yielded the maximum absolute curvature Cmax that occurred during a cardiac cycle in the reconstructed lead. Cmax is a measure for the highest static bending stress in the lead. The max- imum curvature amplitude Camp is a measure for the highest dynamic bending stress in the lead. Addition- ally, the maximum curvature Cmax@Camp at the loca- tion of Camp was determined from max C(s). Both Camp and Cmax@Camp can be used to characterize the relevant intracardiac fatigue stress in a lead under cyclical bending. Leaving potential unknown external forces aside, static curvature of the lead has a smaller inﬂuence on the lead’s lifetime than dynamic changes within the observed range of curvatures. Biomechanical Analysis Therefore, a combination of the maximum absolute curvature at any position with the maximum curvature amplitude may overestimate the mechanical in vivo bending stress of the lead. Therefore, pairing the maximum curvature amplitude and the maximum curvature at the location of the maximum curvature amplitude could be a better measure for the dynamic bending stress a lead is ex- posed to. oped by the Ottawa Heart Institute.6 Slack grades for our classiﬁcation are exempliﬁed in Fig. 2. Curvatures were calculated from 3D reconstruction of the lead centerline. The reconstruction was per- formed for all recorded frames of a cardiac cycle (30 frames/s). The intracardiac reconstruction length was at least 15 cm starting from the distal end to capture the locations of the maximum intracardiac curvature and curvature amplitude. The intracardiac lead curvature was analyzed by the following parameters (see nomenclature): Cmax; Camp; Cmax@Camp; s Cmax ð Þ; s Camp Figure 3 illustrates the most important parameters of the curvature analysis. First, at each discrete time step (t) of the cardiac cycle and at each discrete dis- tance to tip (s) the lead centerline curvature was cal- culated. From that, curves of the maximum curvature during the cardiac cycle (max C(s)) and the minimum curvature during the cardiac cycle (min C(s)) could be provided for the reconstructed intracardiac region. The Finally, the locations of Cmax and Camp on the reconstructed lead were analyzed by their distance to tip (s(Cmax) and s(Camp)). Often, Cmax@Camp equals Cmax. This means that the maximum absolute curva- ture occurs at the location of Camp. FIGURE 1. 3D reconstruction of a lead from biplane cine- ﬂuoroscopy. Left: LAO; and right RAO views with a minimum of 45 degrees difference. Prior to the patient analysis, the uncertainty of the curvature estimation of the software tool developed by Biotronik was analyzed regarding its reproducibility and repeatability via a phantom study. Patient Phase: In Vivo Three-Dimensional Lead Imaging and Reconstruction The imaging procedure was scheduled in conjunc- tion with a routine patient and device follow-up visit. During the procedure, the cine-ﬂuoroscopy ﬁeld of view was limited to the intracardiac region to focus on all electrically active elements of the leads, i.e. lead tip, shock coils, and ring electrodes including DX atrial dipole electrodes of Biotronik Linox smart S DX leads. All images had to fulﬁll the following requirements: Biplanar X-ray images of phantom 1 and phantom 2 were acquired at two diﬀerent angulations and diﬀer- ent positioning of the phantoms. The imaging system was the same Siemens Artis Zee Biplane at the Eras- mus Medical Center Rotterdam that was used for image acquisition from patients in the second (patient) phase of the study. Frame rate of 30 frames/s; the recording duration is about 4 s to acquire at least 3 entire cardiac cycles; the angulation between the two biplane views must be greater than 30, at best 90; and the complete intracardiac lead region is visible in both biplane images during the entire cardiac cycles. The user interface for the determination of the imaged lead path was programmed with Matlab R2013b. The optimization to determine the geometry of the imaging system was based on the Nelder-Mead algorithm.13 After identifying associated image pixels of the lead path in the two views, the 3D coordinates of the path were calculated according to epipolar geom- etry.9 The 3D coordinates were smoothed by a cubic spline based on the Fortran function Smooth.15 The curvature was piecewise calculated with a sliding win- dow of 5 mm width. The cine-ﬂuoroscopy was performed by a trained team led by the corresponding author. The following steps were elements of the examination procedure for a particular patient after enrollment: admission to the lab equipped with the biplane ﬂuoroscopy, patient on the table and placing standard ECG leads. Then positioning of the imaging system to ensure appropri- ate biplane views and veriﬁcation of frame rate. It was followed by cine-ﬂuoroscopy while the patient holds his/her breath. Breath hold was preferred to ensure that the entire lead portion remained within the ﬁeld of view during imaging. Before the patient left the room, we veriﬁed the acquired images and proper storage of the data. Phantom Study: Uncertainty Analysis Two 3D lead paths (phantom 1 and phantom 2) based on in vivo lead paths were constructed with four known local maximum curvatures each. The curva- tures were given by planar path segments composed of a fourth-order polynomial of the form y ¼ ax4 þ bx3 þ dx FIGURE 1. 3D reconstruction of a lead from biplane cine- ﬂuoroscopy. Left: LAO; and right RAO views with a minimum of 45 degrees difference. The local maxima of the curvature of phantom 1 were: 0.75, 1.0, 0.25, and 0.5 cm1. Phantom 2 had the Slack grade 0 Slack grade 1 Slack grade 2 Slack grade 3 Slack grade 4 FIGURE 2. Slack classiﬁcation. From left to right: grade 0: no slack, grade 1: minimal, grade 2: normal, grade 3: mildly excessiv and grade 4: very excessive slack. Slack grade 3 Slack grade 2 Slack grade 4 Slack grade 1 Slack grade 0 FIGURE 2. Slack classiﬁcation. From left to right: grade 0: no slack, grade 1: minimal, grade 2: normal, grade 3: mildly excessive, and grade 4: very excessive slack FIGURE 2. Slack classiﬁcation. From left to right: grade 0: no slack, grade 1: minimal, grade 2: normal, grade 3: mildly excessive, and grade 4: very excessive slack. BIOMEDICAL ENGINEERING SOCIETY 132 SZILI-TOROK et al. FIGURE 3. Illustration of estimated curvatures and curvature amplitudes calculated from 3D ICD lead centerline reconstruction. FIGURE 3. Illustration of estimated curvatures and curvature amplitudes calculated from 3D ICD lead centerline reconstruction. URE 3. Illustration of estimated curvatures and curvature amplitudes calculated from 3D ICD lead cente of estimated curvatures and curvature amplitudes calculated from 3D ICD lead centerline reconstruction. local maximal curvatures: 0.2, 0.4, 0.3, and 0.5 cm1 (Fig. 4). The planar segments were rotated in space and combined to form a three-dimensional path with curvatures of zero at the junctions. The coordinates of these synthetic lead paths were used to build a phan- tom body with a 3D printer, with channels in the body to place Linox Smart SD 65/16 leads, manufactured by Biotronik, in a stable position. The bodies of phantom 1 and phantom 2 were designed by Biotronik and printed with acrylic glass powder by an external sup- plier. local maximal curvatures: 0.2, 0.4, 0.3, and 0.5 cm1 (Fig. 4). The planar segments were rotated in space and combined to form a three-dimensional path with curvatures of zero at the junctions. Phantom Study: Uncertainty Analysis The coordinates of these synthetic lead paths were used to build a phan- tom body with a 3D printer, with channels in the body to place Linox Smart SD 65/16 leads, manufactured by Biotronik, in a stable position. The bodies of phantom 1 and phantom 2 were designed by Biotronik and printed with acrylic glass powder by an external sup- plier. reconstruction were done by 2 diﬀerent operators (re- ferred as ‘‘reconstructor’’) to determine the repro- ducibility of the post processing. Three-Dimensional Analysis of the In Vivo Motion 133 FIGURE 4. Results of the phantom study. (a) Spatial distribution of the curvature of phantom 1 (left) and phantom 2 (right). The values were calculated numerically for the theoretical lead path. (b) Whiskers plot of all reconstructed curvatures of phantom 1 (left) and phantom 2 (right). The red marks represent the median of the spatial distribution of the curvature. FIGURE 4. Results of the phantom study. (a) Spatial distribution of the curvature of phantom 1 (left) and phantom 2 (right). The values were calculated numerically for the theoretical lead path. (b) Whiskers plot of all reconstructed curvatures of phantom 1 (left) and phantom 2 (right). The red marks represent the median of the spatial distribution of the curvature. tomies. Finally, clinical data sets have been acquired from 51 patients, and intracardiac lead curvatures and assessment of lead slack were statistically analyzed. Pearson’s correlation coeﬃcient was used to measure the correlation between slack grade and maximal lead curvature amplitude and maximum curvature at the location of maximum curvature amplitude. During the phantom analysis the following statistical methods were applied: since the mechanical stress is mainly due to high curvatures with high changes of amplitude over the heart cycle, the main focus of the analysis was on the local maxima of the curvature. A statistical anal- ysis, including an analysis of variance (ANOVA), was performed. position. Thereafter, ECG and the supportive systems were removed. At the end of the procedure, the patient was released and the data were archived as DICOM RAW data. Patient Phase: In Vivo Three-Dimensional Lead Imaging and Reconstruction If a second cine-ﬂuoroscopy was necessary, we rearranged the imaging system and/or patient To assess the reproducibility of the localization of the lead path in the images, 5 diﬀerent operators (re- ferred as ‘‘identiﬁer’’) manually marked the position of the lead paths. This localization was done with 5 images of each phantom (2) and view (2) to determine the repeatability. Thereby, a total of 100 pairs of bi- plane images were evaluated. The optimization and BIOMEDICAL ENGINEERING SOCIETY BIOMEDICAL ENGINEERING SOCIETY B E S Three-Dimensional Analysis of the In Vivo Motion Phantom Study The local maxima of the reconstructed curvature were in good agreement with the theoretical curvature distribution (Fig. 4). The mean error (bias) over all curvature maxima was 3.6 ± 11.3%. The estimated curvature showed a tendency to overestimate the the- oretical curvature, especially for the curvature 1.0 cm1 (bias 4%). The maximal relative variance occurred at a curvature of 0.25 cm1 with 18.5%. The most relevant results of the ANOVA are summarized in Table 1. The main inﬂuence on the variance was the repetition of the curvature estimation (> 90%). The different identiﬁers, reconstructors and amplitudes of the local curvature maxima had only a minor effect on the variance. Patient Study The slack grade distribution was analyzed for all ICD leads included in our study. All leads had an apical tip location. The mean slack grade was TABLE 1. Results of the ANOVA on phantoms 1 and 2 with the reconstructor (Operator) and local maxima of curvature. TABLE 1. Results of the ANOVA on phantoms 1 and 2 with the reconstructor (Operator) and local maxima of curvature. Source Standard deviation % of total variation Repeatability 0.109 92.03 Operator 0.000 0.00 Interaction (OP) 0.018 2.36 Reproducibility 0.018 2.36 Total gage R&R 0.111 94.39 Part-to-part 0.027 5.61 Total variation 0.114 100.00 Part also referred as ‘‘arch’’ as parameter of variance. BIOMEDICAL ENGINEERING SOCIETY Source Standard deviation % of total variation Patient Population The patient population consisted of 51 patients in routine medical care with an implanted Biotronik ICD or CRT-D device and Biotronik leads. None of the patients had any cardiac intervention within the last two months prior to the study. All patients had ICD leads with active ﬁxation in the right ventricular apical region. The mean age of the patients was 56.4 ± 13.6 years (ranging from 18 to 83 years). Thirty-four (67%) male and 17 (33%) female patients were included. Presence of heart failure was reported in 19 (37%) patients, 6/19 (32%) were in NYHA class I, 12/19 (63%) in NYHA class II, and 1/19 (5%) in NYHA class III. Recent measurements on left ven- tricular ejection fraction were available for 31 patients, showing a mean value of 39% (ranging from 20% to 71%). Considering all 51 right ventricular ICD leads the median of the maximum curvature amplitude Camp was 0.18 cm1. The frequency distribution of this parameter can be seen in Fig. 6a. The median Camp was 0.18 cm1 (Q1–Q3 0.13–0.25, range 0.04–0.48). The median Cmax@amp was 0.42 cm1 (Q1–Q3 0.33– 0.62, range 0.12–1.24). The curvature 0.42 cm1 was also the curvature with the highest count in the patient data (Fig. 6b). As expected, this median value was considerably lower than the median of the overall maximum curvature of 0.53 cm1. The observed locations of maximum absolute cur- vature Cmax and maximum curvature amplitude Camp were spread over the whole intracardiac region and were equal in about 45% of analyzed ICD leads. The median location of Cmax was in the proximal region of the intracardiac lead segment (median 10.7 cm, Q1–Q3 6.6–12.3, range 1.7–17.9), which is most likely in the atrium, whereas the median location of Camp occurred at a more central region of the intracardiac lead seg- ment (median 6.0 cm, Q1–Q3 4.8–8.0, range 1.3–18.5) close to where the transit through the tricuspid valve can be assumed. The frequency distributions of s(Cmax) and s(Camp) are shown in Fig. 6c. RESULTS 2.3 ± 0.9. While all slack grades were present in our study, grades 3 (45%) and 2 (29%) were most fre- quently observed. Maximum lead curvature ampli- tudes (correlation r = 0.64, p value = 4.e-7) and maximum curvatures at the location of maximum curvature amplitude (correlation r = 0.63, p value = 8.e-7) were found to signiﬁcantly increase with lead slack. These correlations are illustrated in Fig. 5. 2.3 ± 0.9. While all slack grades were present in our study, grades 3 (45%) and 2 (29%) were most fre- quently observed. Maximum lead curvature ampli- tudes (correlation r = 0.64, p value = 4.e-7) and maximum curvatures at the location of maximum curvature amplitude (correlation r = 0.63, p value = 8.e-7) were found to signiﬁcantly increase with lead slack. These correlations are illustrated in Fig. 5. Statistical Analysis Exploratory data analyses e.g. age, gender and heart condition, were used to describe the patient popula- tion. The sample size was deﬁned to be at least 45 eligible imaging data sets from enrolled patients. This sample size was deﬁned based on an estimation to acquire a feasible number of imaging data sets within a normal range of lead motion patterns and heart ana- 134 SZILI-TOROK et al. Device Interrogation After the Patient Study During ICD interrogation two lead-related prob- lems were detected, one at 2 months and the other at 26 months after the end of the study. One patient experienced lead dislodgement requiring repositioning. The ICD lead had minimal slack (Ottawa slack grade 0) and a relatively low Cmax = 0.6 9 median of Cmax (median of all 51 patients). The other patient had lead dysfunction (oversensing). The ICD lead had excessive intracardiac slack and a high Cmax, 2.4 x median of Cmax (Fig. 7). The location of highest Cmax was in the right atrium, which also demonstrated a relatively high Camp. The patient later underwent a full system extraction. Patient Study Patient Study DISCUSSION We developed a novel method to analyze implanted deﬁbrillator lead motion in three dimensions. The major ﬁnding of this initial evaluation is that this Part also referred as ‘‘arch’’ as parameter of variance. BIOMEDICAL ENGINEERING SOCIETY BIOMEDICAL ENGINEERING SOCIETY Three-Dimensional Analysis of the In Vivo Motion 135 FIGURE 5. (a) correlation between slack grade and maximum curvature amplitude. (b) correlation between slack grade and maximum curvature at location of maximum curvature amplitude. FIGURE 5. (a) correlation between slack grade and maximum curvature amplitude. (b) correlation between slack grade and maximum curvature at location of maximum curvature amplitude. FIGURE 6. (a) frequency distribution of the maximum curvature amplitude. (b) frequency distribution of the maximum curvature at the location of the maximum curvature amplitude. (c) frequency distribution of the location of maximum curvature s(Cmax) and maximum curvature amplitude s(Camp). FIGURE 6. (a) frequency distribution of the maximum curvature amplitude. (b) frequency distribution of the maximum curvature at the location of the maximum curvature amplitude. (c) frequency distribution of the location of maximum curvature s(Cmax) and maximum curvature amplitude s(Camp). Patient Study Reveals Potential Lead Positioning Issues Patient Study Reveals Potential Lead Positioning Issues below 10% and errors of 30–50% are not unusual.11,16,17 The challenges increase even more in 3D space.7 In order to overcome these challenges in this study, the 3D coordinates were calculated using epipolar geometry in combination with an applied smooth cubic spline. As a result, the curvature was determined by solving a linear equation with multiple coordinates in two dimensions. Obviously, due to these multiple steps in the post processing, our analytical approach to estimate the curvature error became very complex. Because of the limited contrast of X-ray images compared to optical images the identiﬁcation of the lead path was performed manually. This manual selection of the lead path is more adaptive to changes in contrast and image noise although it may be prone to spatial errors of the reconstruction calculation. On the other hand, during reconstruction of the 3D points of the lead path, the use of a parametric function describing the run of the path is a crucial factor for the curvature estimation. Therefore, the underlying func- tion used in the software was continuously differen- tiable up to the second derivation to avoid breaks at the junctions of the function segments. This allowed avoiding smoothing in the post processing such as spatial averaging.14 Our data clearly demonstrate that the variance is mainly due to the variance of repeata- bility. The individual identiﬁer/reconstructor or parameters like image geometry and amplitude of curvature had only a negligible inﬂuence. Of note, large out-of-plane lead segments increased the impact of errors made during the manual localization of the imaged lead path. Therefore, the determination of the reconstruction error was assumed to be a worst-case error scenario. Taking this into account, the uncer- tainty of our method relating the curvature estimation was comparatively low (3.6% mean overestimation, 11.3% SD) and hence suitable for the determination of in vivo curvatures. The values published by Baxter et al.3 with maxi- mum intracardiac curvatures from 0.27 to 0.78 cm1 and mean maximum curvature 0.48 cm1 (0.18 cm1 SD) for 20 patients are comparable to the data shown in our study. A curvature of 1.4 cm1 reported by Liu et al.14 was based on a single patient with a dataset taken from a prior study,18 with a position of the tip of the pacing lead in the right ventricular outﬂow tract. Lessons Learned from the Phantom Study assessment based on biplane ﬂuoroscopy recording is feasible, highly accurate and can be implemented for the assessment of the intracardiac in vivo movement of right ventricular implantable cardioverter-deﬁbrillator (ICD) leads. assessment based on biplane ﬂuoroscopy recording is feasible, highly accurate and can be implemented for the assessment of the intracardiac in vivo movement of right ventricular implantable cardioverter-deﬁbrillator (ICD) leads. Several studies regarding the calculation of a digital path’s curvature in computer vision in 2D have shown that it is very challenging to keep the curvature error BIOMEDICAL ENGINEERING SOCIETY BIOMEDICAL ENGINEERING SOCIETY 136 SZILI-TOROK et al. FIGURE 7. ICD lead with very excessive slack and malfunction. Left: 3D reconstruction of leads with highlighted maximum curvatures. Right: envelope of curvatures along the ICD lead. FIGURE 7. ICD lead with very excessive slack and malfunction. Left: 3D reconstruction of leads with highlighted maximum curvatures. Right: envelope of curvatures along the ICD lead. below 10% and errors of 30–50% are not unusual.11,16,17 The challenges increase even more in 3D space.7 In order to overcome these challenges in this study, the 3D coordinates were calculated using epipolar geometry in combination with an applied smooth cubic spline. As a result, the curvature was determined by solving a linear equation with multiple coordinates in two dimensions. Obviously, due to these multiple steps in the post processing, our analytical approach to estimate the curvature error became very complex. Because of the limited contrast of X-ray images compared to optical images the identiﬁcation of the lead path was performed manually. This manual selection of the lead path is more adaptive to changes in contrast and image noise although it may be prone to spatial errors of the reconstruction calculation. On the other hand, during reconstruction of the 3D points of the lead path, the use of a parametric function describing the run of the path is a crucial factor for the curvature estimation. Therefore, the underlying func- tion used in the software was continuously differen- tiable up to the second derivation to avoid breaks at the junctions of the function segments. This allowed avoiding smoothing in the post processing such as spatial averaging.14 Our data clearly demonstrate that the variance is mainly due to the variance of repeata- bility. The individual identiﬁer/reconstructor or parameters like image geometry and amplitude of curvature had only a negligible inﬂuence. Patient Study Reveals Potential Lead Positioning Issues The clinical data demonstrate that increased intracar- diac slack was associated with higher static and dy- namic lead stress as expressed as maximum curvature at the location of maximum curvature amplitude, and maximum curvature amplitude. This may imply that excessive intracardiac slack may negatively impact lead survival. This is highlighted by the patient in our study who required lead extraction after the detection of noise (Fig. 7). Interestingly, a previous case–control study demonstrated that excessive intracardiac slack was associated with lead conductor fracture in Med- tronic Fidelis 6949 leads.12 Previously, the dynamic nature of slack was poorly investigated. The con- structed model provides insight into the location of potential mechanical lead stress, which is in the atrium for maximum curvature and at the level of the tricus- pid valve for the maximum curvature amplitude. This information can be used to facilitate the development of more durable leads. Lessons Learned from the Phantom Study Of note, large out-of-plane lead segments increased the impact of errors made during the manual localization of the imaged lead path. Therefore, the determination of the reconstruction error was assumed to be a worst-case error scenario. Taking this into account, the uncer- tainty of our method relating the curvature estimation was comparatively low (3.6% mean overestimation, 11.3% SD) and hence suitable for the determination of in vivo curvatures. CONFLICT OF INTEREST 3 ata ocess g V sua at o a d a s ss o , 00 . 8Harrigan, T., R. Kirkeeide, S. Jalal, T. Beveridge, and B. D. Montgomery. Assessment of pacing lead curvature and strain with three dimensional reconstruction of biplane cineangiographic images in vivo. J. Am. Coll. Cardiol. 27:345A, 1996. Tamas Szili-Torok declares that he received a research grant from Biotronik, Germany. Dr Torsten Luther and Dr Jens Rump are employees of Biotronik, Germany. Dr Sing Yap has no conﬂict of interest to declare. 9Hoffmann, K. R., A. Sen, L. Lan, K.-G. Chua, J. Esthappan, and M. Mazzucco. A system for determination of 3D vessel tree centerlines from biplane images. Int. J. Cardiac Imaging 16:315–330, 2000. REFERENCES 1Altman, P. A., J. M. Meagher, D. W. Walsh, and D. A. Hoffmann. Rotary bending fatigue of coils and wires used in cardiac lead design. J. Biomed. Mater. Res. 43:21–37, 1998. 2Baxter, W. W., and A. D. McCulloch. In vivo ﬁnite element model-based image analysis of pacemaker lead mechanics. Med. Image. Anal. 5:255–270, 2001. CONCLUSIONS g 3Baxter, W., J. Morissette, D. Roberts, and S. Schwartz. Acutely implanted cardiac lead shape measurement: physician prefrerence only? In: Summer Bioengineering Conference, 2003. In conclusion reconstruction of 3D lead motion is feasible using biplane cine-ﬂuoroscopy and lead cur- vatures can be computed with high accuracy. An in- crease of dynamic stress with increasing slack was conﬁrmed. Our results can be implemented to improve lead design and testing. In the future even lead posi- tioning can be assisted using slack assessment as a guiding tool and it may provide useful information before lead extraction procedures. 4Baxter, W., N. Skadsberg, W. B. Johnson, G. Crossley, and B. Foreman. New unanticipatied insights on peak lead bending during pectoralis ﬂexure. Heart Rhythm 7:309, 2010. 5Cooke, D. J., A. Himes, and C. D. Swerdlow. Improved engineering standards for transvenous cardiac leads: a progress report from the Association for the Advancement of Medical Instrumentation Cardiac Rhythm Management Device Committee Leads Working Group. Heart Rhythm 16(6):958–959, 2019. 6 6Ha, A. C., B. Z. Vezi, A. Keren, H. Alanazi, M. H. Gollob, M. S. Green, R. Lemery, P. B. Nery, E. Posan, and D. H. i i di f f i k f ll lib 6Ha, A. C., B. Z. Vezi, A. Keren, H. Alanazi, M. H. Gollob, M. S. Green, R. Lemery, P. B. Nery, E. Posan, and D. H. Birnie. Predictors of fracture risk of a small caliber implantable cardioverter deﬁbrillator lead. Pacing Clin. Electrophysiol. 33:437–443, 2010. ACKNOWLEDGMENTS Birnie. Predictors of fracture risk of a small caliber implantable cardioverter deﬁbrillator lead. Pacing Clin. Electrophysiol. 33:437–443, 2010. This study was supported by Biotronik SE & Co. KG, Berlin, Germany. p y , 7Hameiri, E., and I. Shimshoni. Estimating the principal curvatures and the Darboux frame from real 3D range data. In: Proceedings First International Symposium on 3D Data Processing Visualization and Transmission 2002 p y , 7Hameiri, E., and I. Shimshoni. Estimating the principal curvatures and the Darboux frame from real 3D range data. In: Proceedings First International Symposium on 3D Data Processing Visualization and Transmission, 2002. 8Harrigan, T., R. Kirkeeide, S. Jalal, T. Beveridge, and B. D. Montgomery. Assessment of pacing lead curvature and strain with three dimensional reconstruction of biplane cineangiographic images in vivo. J. Am. Coll. Cardiol. 27:345A, 1996. Clinical Implications and Limitations Using our methodology may open future perspec- tives. Previous trials failed to accurately reconstruct three-dimensional lead behavior in vivo. The data derived from our phantom study and the clinical evaluation are very promising and may pave the road for future possible clinical implementation. Although BIOMEDICAL ENGINEERING SOCIETY BIOMEDICAL ENGINEERING SOCIETY Three-Dimensional Analysis of the In Vivo Motion 137 we have no follow up data, it is very appealing that the patient with the most excessive loop and lead stress had lead dysfunction. It raises the possibility of implementing this as a guiding tool for implantation. Obviously, the major limitation is the limited avail- ability of biplane ﬂuoroscopy in the EP lab. However, the data can be used to initiate adjustments in the regulatory process and more adequate testing methods can be developed. Moreover, during lead design the susceptible regions can be identiﬁed and strengthened and may result in the development of more durable leads. 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://crea tivecommons.org/licenses/by/4.0/. OPEN ACCESS 10Hoffmann, K. R., B. B. Williams, J. Esthappan, S. Y. Chen, J. D. Carroll, H. Harauchi, V. Doerr, G. N. Kay, A. Eberhardt, and M. Overland. Determination of 3D posi- tions of pacemaker leads from biplane angiographic se- quences. Med. Phys. 24:1854–1862, 1997. 11 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 appro- priate 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 q y 11Kovalevsky, V. Curvature in digital 2D images. Int. J. Pattern Recognit. Artif. Intell. 15:1183–1200, 2001. 12 12Krahn, A., J. Morisette, R. Lahm, T. Haddad, W. W. Baxter, R. McVenes, E. Crystal, F. Ayala-Paredes, D. Cameron, A. Verma, C. S. Simpson, D. V. Exner, and D. H. Birnie. Radiographic predictors of lead conductor fracture. Circ. Arrhythm Electrophysiol. 7:1070–1077, 2014. 13 13Lagarias, J. C., J. A. Reeds, M. H. Wright, and P. E. Wright. Convergence properties of the nelder-mead simplex method in low dimensions. SIAM J. Optim. 9(1):112–147, 1998. B E S 138 SZILI-TOROK et al. 17Worring, M., and A. W. M. Smeulders. Digital curvature estimation. CVGIP 58:366–382, 1993. 14Liu, L., J. Wang, W. Yang, and S. J. Chen. In vivo stress analysis of a pacing lead from an angiographic sequence. J. Biomech. Eng. 133:041004, 2011. 15 18Zhang, Y. Analysis of intracardiac lead bending stress and motion characteristics of RVA versus RVOS pacing leads using an innovative 3-D reconstruction technique. In: Summer Bioengineering Conference, 2003. g 15Stoyan, G., and C. de Boor. A Practical Guide to Splines. Applied Mathematical Sciences 27, 1978, XXIV, 392 S., vol. 60. Berlin: Springer, pp. 62–62, 1980. 16 16Utcke, S. Error-bounds on curvature estimation. In: Scale Space Methods in Computer Vision: 4th International Conference, Scale Space 2003 Isle of Skye, UK, June 10–12, 2003 Proceedings, edited by L. D. Grifﬁn and M. Lillholm. Berlin: Springer, pp. 657–666, 2003. Publisher’s Note Springer Nature remains neutral with re- gard to jurisdictional claims in published maps and institu- tional afﬁliations. BIOMEDICAL ENGINEERING SOCIETY BIOMEDICAL ENGINEERING SOCIETY
https://openalex.org/W2986797282	https://elib.dlr.de/129535/2/sensors-19-04802-1.pdf	English	null	Localization and Tracking of Discrete Mobile Scatterers in Vehicular Environments Using Delay Estimates	Sensors	2,019	cc-by	10,535	sensors sensors Article Martin Schmidhammer * , Christian Gentner , Benjamin Siebler and Stephan Sand j p German Aerospace Center (DLR), Institute of Communications and Navigation, 82234 Wessling, Germany; christian.gentner@dlr.de (C.G.); benjamin.siebler@dlr.de (B.S.); stephan.sand@dlr.de (S.S.) * Correspondence: martin.schmidhammer@dlr.de; Tel.: +49-8153-28-1539; Fax: +49-8153-28-1871 † This paper is an extended version of our paper: Schmidhammer, M., Gentner, C., Siebler, B. Localization of Discrete Mobile Scatterers in Vehicular Environments Using Delay Estimates, Proceedings of the 9th International Conference on Localization and GNSS (ICL-GNSS), Nurenberg, Germany, 4–6 June 2019. ed: 11 October 2019; Accepted: 31 October 2019; Published: 5 November 2019 Abstract: This paper describes an approach to detect, localize, and track moving, non-cooperative objects by exploiting multipath propagation. In a network of spatially distributed transmitting and receiving nodes, moving objects appear as discrete mobile scatterers. Therefore, the localization of mobile scatterers is formulated as a nonlinear optimization problem. An iterative nonlinear least squares algorithm following Levenberg and Marquardt is used for solving the optimization problem initially, and an extended Kalman ﬁlter is used for estimating the scatterer location recursively over time. The corresponding performance bounds are derived for both the snapshot based position estimation and the nonlinear sequential Bayesian estimation with the classic and the posterior Cramér–Rao lower bound. Thereby, a comparison of simulation results to the posterior Cramér–Rao lower bound conﬁrms the applicability of the extended Kalman ﬁlter. The proposed approach is applied to estimate the position of a walking pedestrian sequentially based on wideband measurement data in an outdoor scenario. The evaluation shows that the pedestrian can be localized throughout the scenario with an accuracy of 0.8 m at 90% conﬁdence. Keywords: mulitlateration; localization; nonlinear least-squares; Levenberg–Marquardt; tracking; extended Kalman ﬁlter; Bayesian performance bounds; posterior Cramér–Rao lower bound Sensors 2019, 19, 4802; doi:10.3390/s19214802 1. Introduction With the trends of increasing urbanization and increasing automation in road transportation, the demand for improvements in vehicular safety technologies is steadily growing. In particular the mixed-trafﬁc environment shared by many different users including cars, motorcycles, cyclists, and pedestrians is challenging for any means of automated transport. To safely route through vehicular environments, therefore, timely and reliable information about other road users is required. In this regard, the exchange of user speciﬁc information, like position and velocity, enhances safety on roads by supporting mutual awareness [1–3]. This cooperative approach requires road users to be equipped with actively probing devices to determine their user speciﬁc information and to exchange the data. However, many road users do not carry any devices, i.e., they are non-cooperative. Thus, for a reliable and comprehensive situational awareness, further methods and sensor technologies are needed, accounting for non-cooperative road users. With regard to current automated and autonomous vehicles, the perception of their surrounding environment mainly relies on locally mounted sensors, including radar and lidar sensors, as well as camera based systems [4]. Due to physical properties, Sensors 2019, 19, 4802; doi:10.3390/s19214802 www.mdpi.com/journal/sensors www.mdpi.com/journal/sensors 2 of 19 Sensors 2019, 19, 4802 however, locally mounted perception sensors exhibit a series of critical limitations, like the limited performance in adverse lighting conditions [3]. Thus, to improve the reliability and to extend the awareness range of the local perception sensors, infrastructure based systems have been suggested, for instance based on dedicated radar sensors [5] and cameras [6]. Apart from remaining technical challenges, the main drawback of these systems is their deployment. For sufﬁciently sensing only a limited area like an urban intersection, several sensors would have to be mounted, which results in high deployment and maintenance costs. g p y Therefore, a localization system is introduced in [7], which reuses signals from vehicular communications infrastructure for passive radar application. It is assumed that road users and other objects affect the radio spectrum by inducing delayed and Doppler shifted multipath components (MPCs). This means that the characteristics of the MPCs correspond to location and dynamics of the road users. Thus, sensing the wireless propagation characteristics between the links of existing communication networks allows for detecting and localizing road users. Similarly, the authors in [8] propose the usage of existing vehicular radio links for detecting and localizing road users. 1. Introduction In addition to static links from communication infrastructure, they propose to include also mobile devices as possible network nodes. The localization accuracy of both systems strongly depends on precise location information of the individual network nodes. Thus, an incorporation of mobile nodes requires to account for location uncertainties, which can degrade the localization performance. The proposed sensing systems of [7] and [8] follow the idea of passive coherent location (PCL), i.e., the usage of arbitrary signals as illuminators of opportunity [9]. With its origin in aeronautics, PCL was mainly used for observing targets far from the sensing network [9–12]. The localization requirements, for instance, from aeronautic and maritime applications allow high integration times and low bandwidths. Recently, the interest in passive localization approaches for short-range and passive indoor localization is increasing [13–15]. The main challenge of short-range PCL are the dynamics of targets moving in the proximity of the sensing network. Thereby, signals from Wi-Fi access points with comparatively large bandwidths are beneﬁcial for localization, since lower integration times are required [7,13]. In this regard, the authors of [16] propose to use even ultra-wideband signals for PCL. Furthermore, they introduce different target tracking algorithms and analyze the algorithms in simulations. Another localization approach was introduced in [17], applying an iterative, nonlinear least-squares approach for position estimation. The applicability of the approach is demonstrated using wideband measurement data. On the basis of [17], this work extends the localization approach by sequential target tracking, namely an extended Kalman ﬁlter (EKF). For the evaluation of the tracking algorithm, the posterior Cramér-Rao lower bound (PCRLB) is derived. Finally, the proposed localization approach is applied to wideband measurement data. The main contributions of this paper are: • The signal processing for a localization approach to localize moving, non-cooperative objects recursively, using delay estimates from a network of spatially distributed transmitting and receiving nodes. g • The derivation of performance bounds including the Cramér-Rao lower bound (CRLB) on position estimation and the PCRLB on nonlinear sequential Bayesian estimation. q y The validation of the applicability of an EKF for the introduced localization problem via Monte Carlo simulations and a comparison to the PCRLB. p • The application of the proposed localization approach to wideband measurement data of an outdoor experiment for localizing a walking pedestrian. The remainder of this paper is structured as follows. 2. Network and Measurement Model This section introduces the network structure and the measurement model used for the position estimation. Refer to a widely distributed network of K transmitting and L receiving nodes, where both transmitting and receiving nodes are assumed to be static at known locations rTx k , k ∈{1, . . . , K}, and rRx l , l ∈{1, . . . , L}, respectively. Receiving nodes can be collocated with transmitting nodes or individually placed. The network link conﬁguration determines the index set P, where each link (k, l) ∈P is composed of the k-th transmitting node and the l-th receiving node. Accordingly, fully meshed networks result in a maximum number of \|P\| = KL links. Each transmitting node emits known signals sk(t) with period Tp, which allows for measuring the channel impulse response (CIR) at the receivers [18]. Thereby, the received signal of each link in the network is modeled as a superposition of a ﬁnite number of scaled and delayed replica of the transmit signal. These comprise the line-of-sight (LoS) and a discrete number of MPCs due to reﬂection and scattering. Following [19], the MPCs are further differentiated according to the dynamics of the scattering objects, i.e., static and mobile. Eventually, the CIR for a pair of transmitting and receiving nodes Txk and Rxl can be expressed as hkl(t, τ) = hLoS kl (t, τ) + hS kl(t, τ) + hM kl (t, τ), (1) (1) where hLoS kl (t, τ), hS kl(t, τ), and hM kl (t, τ) denote the contribution of the LoS, the sum of Rkl discrete static MPCs and of Qkl discrete mobile MPCs to the CIR, respectively. The static network allows for attributing mobile MPCs to moving scattering objects in the observed area. Here, the number of mobile MPCs is assumed to be identical for each link in the network, i.e., Qkl = Q, and the mobile MPCs can be uniquely assigned to the moving objects. Consequently, for time and phase synchronized transmitting and receiving nodes, discrete mobile scatterers are contributing to the CIR by hM kl (t, τ) = Q ∑ q=1 αM klq(t)e−j2π fcτM kl (rq(t))δ(τ −τM kl (rq(t))), (2) (2) with αM klq(t) and τM kl (rq(t)) representing the complex amplitude and the propagation delay associated with the qth mobile scatterer at location rq(t) = [xq(t), yq(t)]T, and δ(·) representing the Dirac function. 1. Introduction Section 2 introduces the network structure together with the measurement model used for position estimation. In Section 3, the signal processing is described in order to localize mobile scatterers, which comprises a procedure to extract the measurement vector, a snapshot based position estimator, as well as a nonlinear sequential Bayesian estimator. Corresponding performance bounds for positioning and for nonlinear sequential Bayesian estimation are derived in Section 4. Based on an exemplary measurement setup, the localization 3 of 19 Sensors 2019, 19, 4802 approach is evaluated in Section 5, ﬁrst theoretically by using the derived performance bounds and second by applying the approach to channel measurements. Finally, Section 6 concludes the paper by summarizing the ﬁndings. 3.1. Calibration Stage In order to deduce location information of moving objects from CIRs measurements, the propagation effects of the static environment need to be known. Therefore, the channel of every link in the network is initially observed over a calibration period Tcal. During this period, the environment is assumed to be devoid of any moving object, i.e., Q = 0. With Tg as the time interval between two adjacent CIR measurements, a total of ⌊Tcal/Tg⌋consecutive CIRs are collected for calibration. Given the set of recorded CIRs, ﬁrst, the channel parameters are estimated using KEST. Second, the parameter estimates are clustered with regard to amplitude and delay [20]. For static environments, the resulting clusters correspond to the LoS and to static MPCs. Thereby, the vectors ¯τklr and ¯σklr, r ∈{0, . . . , Rkl}, contain cluster mean and standard deviation for the LoS (r = 0) and Rkl static MPCs. Eventually, the vectors ¯τkl = [ ¯τkl0, . . . , ¯τklRkl]T and ¯σkl = [¯σkl0, . . . , ¯σklRkl]T uniquely characterize the static propagation environment between k-th transmitter Txk and l-th receiver Rxl. Note that, even though the static environment is only typically changing very slowly, any modiﬁcation in the propagation condition requires to recalibrate the system. Particularly, objects with strongly reﬂecting characteristics, will change the conditions signiﬁcantly, as for example, the placement of a car in the network environment. 3. Localization and Tracking As shown above, propagation delays of time-variant MPCs inherently contain location information of mobile scatterers being reﬂected in the CIRs. Besides time-variant MPCs, the CIRs of each link in the network also comprise a static LoS and static MPCs as stated in Equation (1). Thus, for localizing and tracking mobile scatterers, time-variant channel components need to be initially extracted. Therefore, the proposed localization approach is composed of three stages—ﬁrst, a calibration stage to identify and characterize the static LoS and MPCs, second, an estimation stage to extract mobile MPCs, and third, a tracking stage to estimate the scatterer position recursively. Both calibration and estimation stage rely on estimates of channel parameters, including complex amplitude and propagation delays of the LoS and MPCs. To estimate and track the channel parameters of the CIRs, Kalman enhanced super resolution tracking (KEST) is used [18]. 2. Network and Measurement Model The corresponding delay-induced phase shift for center frequency fc is expressed by the term e−j2π fcτM klq(rq(t)). For convenience, the notation for time dependence will be omitted in the remainder of this paper and only applied where explicitly needed. For any pair of transmitting and receiving node, the propagation delay induced by scatterer q is determined by the physical propagation path from transmitter to scatterer and from scatterer to receiver. Thus, given the distances dTx kq and dRx lq between scatterer and Txk and Rxl, respectively, the propagation delay can be expressed as τM kl (rq) = 1 c dTx kq + dRx lq = 1 c ∥rq −rTx k ∥+∥rq −rRx l ∥ , (3) (3) where c denotes the speed of light and the operator ∥·∥denotes the Euclidean norm. For \|P\| network links, linear independent propagation delays induced by the qth scatterer compose a vector τq, with Sensors 2019, 19, 4802 4 of 19 each vector element τM klq = τM kl (rq). Thus, for the mapping h(rq) = τq with each element deﬁned according to Equation (3), the measurement model is given by each vector element τM klq = τM kl (rq). Thus, for the mapping h(rq) = τq with each element deﬁned according to Equation (3), the measurement model is given by ˆτq = τq + wq = h(rq) + wq, (4) (4) with ˆτq as corresponding vector of measured delays and wq as zero-mean white Gaussian noise with covariance matrix Rq deﬁned as diagonal matrix with elements {σ2 klq}(k,l)∈P. 3.2. Estimation Stage Other than during the calibration period, now, additional objects can move within the observed environment. Thus, the CIRs and therefore also the estimated channel parameter comprise the propagation effects of moving objects. Similar to before, KEST is applied for estimating channel parameters from incoming CIRs. The resulting vector of propagation delay estimates is composed as ˆτfull kl = [ ˆτS kl0, . . . , ˆτS klRkl \| {z } LoS and static MPC , ˆτM kl1, . . . , ˆτM klQ]T. (5) (5) Consequently, extracting mobile MPCs from ˆτfull kl means to sort out static components. The sorting is based on ¯τkl determined in the preceding calibration stage. Particularly, all elements of ˆτfull kl lying in an interval of ¯τklp ± 3¯σklp are excluded. Here, the 3-σ interval ensures that delay estimates assigned as static are not considered as mobile MPC with a probability higher than 99%, assuming that the Sensors 2019, 19, 4802 5 of 19 previously determined amount of static MPCs remains constant. Eventually, the measurement vector of mobile MPCs for the link Txk and Rxl is previously determined amount of static MPCs remains constant. Eventually, the measurement vector of mobile MPCs for the link Txk and Rxl is ˆτkl = [ ˆτM kl1, . . . , ˆτM klQ]T. (6) (6) Associated elements of ˆτkl to scatterer q are rearranged over all links in the network, P, which ﬁnally determine the measurement vector ˆτq. Associated elements of ˆτkl to scatterer q are rearranged over all links in the network, P, which ﬁnally determine the measurement vector ˆτq. Based on the measurement model in Equation (4), scatterer q can be localized applying maximum likelihood estimation. In particular, a weighted nonlinear least-squares approach [21,22] is used to minimize the cost function (7) L(rq) = ( ˆτq −τq)TR−1 q ( ˆτq −τq), (7) L(rq) = ( ˆτq −τq)TR−1 q ( ˆτq −τq), (7) with respect to the unknown position rq, which is expressed as with respect to the unknown position rq, which is expressed as with respect to the unknown position rq, which is expressed as ˆrq = arg min rq L(rq). (8) (8) The two-dimensional, nonlinear optimization problem of Equation (8) needs to be solved by an iterative approach, since no closed-form solution is existing. 3.2. Estimation Stage Thus, in order to minimize the cost function in Equation (7), the proposed localization approach applies the Levenberg–Marquardt algorithm [22] due to high robustness and fast convergence characteristics. Particularly, the iterative procedure can be written as r(i+1) q = r(i) q + JT(r(i) q )R−1 q J(r(i) q ) + λ(i)I −1 JT(r(i) q )R−1 q ˆτq −τ(i) q , (9) (9) with J(rq), I and λ(i) denoting the Jacobian matrix, the identity matrix, and the dampening parameter for iteration step i, respectively. Individual elements of vector τ(i) q are calculated according to Equation (3) as τM kl (r(i) q ). The elements of Jacobian matrix J(rq) ∈R\|P\|×2, i.e., the partial derivatives of τq with respect to rq, are given as J(rq) =   xq−xTx 1 dTx 1q + xq−xRx 1 dRx 1q yq−yTx 1 dTx 1q + yq−yRx 1 dRx 1q ... ... xq−xTx K dTx Kq + xq−xRx 1 dRx 1q yq−yTx K dTx Kq + yq−yRx 1 dRx 1q xq−xTx 1 dTx 1q + xq−xRx 2 dRx 2q yq−yTx 1 dTx 1q + yq−yRx 2 dRx 2q ... ... xq−xTx K dTx Kq + xq−xRx 2 dRx 2q yq−yTx K dTx Kq + yq−yRx 2 dRx 2q ... ... xq−xTx 1 dTx 1q + xq−xRx L dRx Lq yq−yTx 1 dTx 1q + yq−yRx L dRx Lq ... ... xq−xTx K dTx Kq + xq−xRx L dRx Lq yq−yTx K dTx Kq + yq−yRx L dRx Lq   . (10) (10) 3.3. Tracking Stage 3.3. Tracking Stage The previous stage provides an approach to localize a moving scatterer based on delay measurements at one speciﬁc time instance t. Since the goal of this work is to localize moving ensors 2019, 19, 4802 6 6 of 19 Sensors 2019, 19, 4802 scatterers, it is reasonable to additionally take into account the mobility of the object by ﬁltering the state evolution over time. For any mobile scatterer q, the state is deﬁned by xq(tn) = h rT q (tn), vT q (tn) iT , (11) (11) where rq(tn) and vq(tn) = vq,x(tn), vq,y(tn) T denote the position and the velocity of the scatterer at time instant tn. To describe the state evolution from time instant tn−1 to time instant tn, a transition model is applied. Accounting for the mobility of scatterers induced by different road users, such as cars, bikes, and pedestrians, a white noise acceleration model is used [23]. The state equation results in xq(tn) = Aqxq(tn−1) + nq(tn), (12) (12) with transition matrix Aq and zero-mean white Gaussian process noise nq(tn) with covariance matrix Qq. Being Tg = tn −tn−1, the transition matrix is given by with transition matrix Aq and zero-mean white Gaussian process noise nq(tn) with covariance matrix Qq. 3.3. Tracking Stage Being Tg = tn −tn−1, the transition matrix is given by Aq =   1 0 Tg 0 0 1 0 Tg 0 0 1 0 0 0 0 1   (13) and the covariance matrix by Aq =   1 0 Tg 0 0 1 0 Tg 0 0 1 0 0 0 0 1   (13) and the covariance matrix by Qq =   σ2 q T3g 3 0 σ2 q T2g 2 0 0 σ2 q T3g 3 0 σ2 q T2g 2 σ2 q T2g 2 0 σ2 q Tg 0 0 σ2 q T2g 2 0 σ2 q Tg   , (14) (13) 0 0 1 0 0 0 0 1  and the covariance matrix by Qq =   σ2 q T3g 3 0 σ2 q T2g 2 0 0 σ2 q T3g 3 0 σ2 q T2g 2 σ2 q T2g 2 0 σ2 q Tg 0 T2   , (14) Qq =   σ2 q T3g 3 0 σ2 q T2g 2 0 0 σ2 q T3g 3 0 σ2 q T2g 2 σ2 q T2g 2 0 σ2 q Tg 0 0 σ2 q T2g 2 0 σ2 q Tg   , (14) (14) with σ2 q as process noise intensity of physical dimension m2/s3 , which needs to be set according to application requirements [23]. Given state vector xq(tn), the nonlinear mapping of the measurement model in Equation (4) can be expressed as h(rq(tn)) = h(xq(tn)), equivalently. Accordingly, the measurement model is given by Given state vector xq(tn), the nonlinear mapping of the measurement model in Equation (4) can be expressed as h(rq(tn)) = h(xq(tn)), equivalently. Accordingly, the measurement model is given by (15) zq(tn) = h(xq(tn)) + wq(tn). (15) Due to the nonlinearity of the measurement model, a recursive Bayesian ﬁlter is required, which is able to handle general nonlinear problems. A common implementation of such recursive Bayesian ﬁlters is the EKF. 4.1. Cramér–Rao Lower Bound on Position Estimation 4.1. Cramér–Rao Lower Bound on Position Estimation The CRLB is deﬁned as the inverse of the Fisher information matrix (FIM) [21]. This means, given the vector parameter rq, the elements of the unbiased estimator ˆrq = ˆxq, ˆyq T satisfy Var( ˆxq) ≥[F(rq)−1]1,1 = σ2 xq (18) (18) and and Var( ˆyq) ≥[F(rq)−1]2,2 = σ2 yq, (19) (19) where Var(·) denotes the variance of an estimator and the terms [F(rq)−1]n,n, n = {1, 2}, denote the diagonal elements of the inverse FIM F(rq). Inherently, received signals are a function of propagation delays τq. This holds for every pair of transmitting and receiving nodes in the network. Thus, by applying the chain rule, the FIM F(rq) can be alternatively expressed as [21] F(rq) = J(rq)TF(τq)J(rq), (20) (20) with the Jacobian matrix J(rq) as deﬁned in Equation (10) and the FIM F(τq) ∈R\|P\|×\|P\| with respect to delay vector τq (Please note also that path loss and phase information can be taken into account for calculating the FIM but is out of the scope of this work). For the linear independent time delays τq, the Fisher information is well known [21] and the diagonal elements of F(τq) can be written as with the Jacobian matrix J(rq) as deﬁned in Equation (10) and the FIM F(τq) ∈R\|P\|×\|P\| with respect to delay vector τq (Please note also that path loss and phase information can be taken into account for calculating the FIM but is out of the scope of this work). For the linear independent time delays τq, the Fisher information is well known [21] and the diagonal elements of F(τq) can be written as [F(τq)]p,p = 8π2β2SNRpq c2 , (21) (21) with β2 as effective bandwidth of the transmit signal. The index p = 1, . . . , \|P\| provides an enumeration of index set P. Therewith, SNRpq expresses the signal-to-noise-ratio (SNR) for the MPC caused by scatterer q in network link p. With Txk and Rxl deﬁning link p, the SNR can be written as SNRpq = PTxk GTxkGRxlσqc2 (4π)3 f 2c (dTx kq )2(dRx lq )2 ! P−1 n , (22) (22) where PTxk and Pn denote transmit power and receiver noise power, GTxk and GRxl express antenna gains at transmitter and receiver, and σq refers to the radar cross-section (RCS) of scatterer q [7]. 3.3. Tracking Stage Given the nonlinear measurement model in Equation (4) and the linear state model in Equation (12), the EKF results in the following set of equations: ˆxq(tn\|tn−1) = Aq ˆxq(tn−1\|tn−1), Pq(tn\|tn−1) = AqPq(tn−1\|tn−1)AT q + Qq, K(tn) = Pq(tn\|tn−1)HT q (tn) Hq(tn)Pq(tn\|tn−1)HT q (tn) + Rq(tn) −1 , ˆxq(tn\|tn) = ˆxq(tn\|tn−1) + K(tn) ˆτq(tn) −h(ˆxq(tn\|tn−1)) , Pq(tn\|tn) = Pq(tn\|tn−1) −K(tn)Hq(tn)Pq(tn\|tn−1), (16) (16) where K(tn) is the Kalman gain, ˆxq(tn\|tn−1) and Pq(tn\|tn−1) are the predicted state and covariance matrix at time tn, and ˆxq(tn\|tn) and Pq(tn\|tn) are the corrected state estimate and covariance matrix after the measurement update. The local linearization of the measurement model around ˆxq(tn\|tn−1) where K(tn) is the Kalman gain, ˆxq(tn\|tn−1) and Pq(tn\|tn−1) are the predicted state and covariance matrix at time tn, and ˆxq(tn\|tn) and Pq(tn\|tn) are the corrected state estimate and covariance matrix after the measurement update. The local linearization of the measurement model around ˆxq(tn\|tn−1) Sensors 2019, 19, 4802 7 of 19 is denoted in the observation matrix Hq(tn). Since the elements of measurement vector τq(tn) only depend on position and not on velocity, the linearized observation matrix consists of the Jacobian in Equation (10) at the predicted position ˆrq(tn\|tn−1) and of a \|P\| × 2 zero matrix 0, written as is denoted in the observation matrix Hq(tn). Since the elements of measurement vector τq(tn) only depend on position and not on velocity, the linearized observation matrix consists of the Jacobian in Equation (10) at the predicted position ˆrq(tn\|tn−1) and of a \|P\| × 2 zero matrix 0, written as Hq(tn) = J(ˆrq(tn\|tn−1)), 0 . (17) (17) 4. Performance Bounds For evaluating estimators, typically, theoretical lower bounds on the estimation performance can be used. With regard to the proposed localization approach of this paper, this section ﬁrst provides the classic CRLB on the error of the position estimate and second the PCRLB as performance bound for unbiased sequential Bayesian estimators. Thereby, the latter bound allows for evaluating recursive Bayesian ﬁlters like the proposed EKF. 4.1. Cramér–Rao Lower Bound on Position Estimation Hence, the elements of the FIM in Equation (21) strongly depend on the position of the scatterer and are proportional to [F(τq)]p,p ∝((dTx kq )2(dRx lq )2)−1. (23) (23) Sensors 2019, 19, 4802 8 of 19 Finally, the CRLB of Equations (18) and (19) can be used to evaluate an estimator by comparing it to the root mean square error (RMSE) according to inequality Finally, the CRLB of Equations (18) and (19) can be used to evaluate an estimator by comparing it to the root mean square error (RMSE) according to inequality RMSEq = q E ∥ˆrq −rq∥2 ≥ q σ2xq + σ2yq. (24) (24) 4.2. Posterior Cramér–Rao Lower Bound for Nonlinear Sequential Bayesian Estimation 4.2. Posterior Cramér–Rao Lower Bound for Nonlinear Sequential Bayesian Estimation The CRLB, as introduced above, allows for evaluating the positioning performance for a speciﬁc time instant. Thereby, the system is assumed to be time-invariant. An evaluation of Bayesian ﬁltering and tracking approaches, however, requires a performance bound, which accounts for time-variant systems, for underlying stochastic state space models, and for the incorporation of prior knowledge [24]. Such a performance bound is provided by the PCRLB. The PCRLB is the theoretical performance bound for sequential Bayesian estimators [24,25]. Equivalently to the CRLB, the PCRLB is calculated by the inverse of the posterior FIM FB,q(tn) deﬁning the inequality Exq(tn) hˆxq(tn) −xq(tn) ˆxq(tn) −xq(tn) Ti = Mq(tn) ≥FB,q(tn)−1, (25) (25) with Ea [·] as expectation with respect to the probability density of random variable a. The inequality in Equation (25) implies that the difference between the mean square error (MSE) matrix Mq(tn) and the inverse of the posterior FIM is a positive semi-deﬁnite matrix. Following [24–27], the posterior FIM FB,q(tn) can be calculated recursively as FB,q(tn) = D22,q(tn) −D21,q(tn) FB,q(tn−1) + D11,q(tn) −1 D12,q(tn), (26) (26) with D11,q(tn) =Exq(tn−1),xq(tn) h −∆ xq(tn−1) xq(tn−1) ln p xq(tn)\|xq(tn−1) i , (27) D12,q(tn) =Exq(tn−1),xq(tn) h −∆ xq(tn) xq(tn−1) ln p xq(tn)\|xq(tn−1) i = D21,q(tn)T, (28) D22,q(tn) =Exq(tn−1),xq(tn) h −∆ xq(tn) xq(tn) ln p xq(tn)\|xq(tn−1) i + Exq(tn) h Ezq(tn)\|xq(tn) h −∆ xq(tn) xq(tn) ln p zq(tn)\|xq(tn) ii \| {z } classic FIM . (29) (27) (28) (29) {z classic FIM With ∇a denoting the partial derivatives with respect to a, the operator ∆a b = ∇b∇T a gives the corresponding second-order partial derivatives. Furthermore, p (a(tn)\|b(tn)) is the conditional probability distribution of random variable a(tn) given b(tn) at time instant tn. Note that the term inside the second expectation of D22,q(tn), as highlighted by a brace, refers to the deﬁnition of the classic FIM [25]. For state xq(tn) given in Equation (11), the 4 × 4 FIM is ˜F(xq(tn)) = " F rq(tn) 0 0 0 # , (30) (30) where F(rq(tn) denotes the FIM of Equation (20). 5. Case Study In this section, the proposed localization approach is analyzed for the example of a walking pedestrian. Together with the network structure, the measurement setup is presented ﬁrst. Based on the introduced static network, the localization performance is evaluated theoretically using the performance bounds of Section 4. Finally, the localization approach is applied to channel measurements. 4.2. Posterior Cramér–Rao Lower Bound for Nonlinear Sequential Bayesian Estimation Thus, referring to the linear transition model introduced in Section 3.3 with transition matrix Aq given in Equation (13) and white Gaussian process noise with covariance matrix Qq given in Equation (14), Equations (27)–(29) result in Sensors 2019, 19, 4802 9 of 19 D11,q(tn) =AT q Q−1 q Aq, (31) D11,q(tn) =AT q Q−1 q Aq, (31) D12,q(tn) = −AT q Q−1 q = D21,q(tn)T, (32) D22,q(tn) =Q−1 q + Exq(tn) ˜F(xq(tn)) . (33) (31) (32) (33) (32) D22,q(tn) =Q−1 q + Exq(tn) ˜F(xq(tn)) . (33) (33) Inserting Equations (31)–(33) into Equation (26) and applying the matrix inversion lemma results in the recursive expression Inserting Equations (31)–(33) into Equation (26) and applying the matrix inversion lemma results in the recursive expression FB,q(tn) = Qq + AqFB,q(tn−1)−1AT q −1 + Exq(tn) ˜F(xq(tn)) . (34) (34) Given prior information according to the probability density function p(xq(t0)), the initial FIM FB,q(t0) can be calculated as FB,q(t0) = Exq(t0) h −∆ xq(t0) xq(t0) ln p xq(t0) i , (35) (35) which is used to initialize the recursion in Equation (34). which is used to initialize the recursion in Equation (34). 5.1. Network and Measurement Setup The analyzed static network consists of K = 1 transmitting and L = 4 receiving nodes. As shown in Figure 1a, the network nodes are individually placed with an inter-node distance between 10 m and 15 m. The center of the network forms a transmitting node with a small directional antenna. Thereby, the main beam of the transmit antenna is oriented towards a designated observation area between the receiving nodes, as shown in Figure 1b, illustrating the gain of the transmit antenna in the network area [28]. With the transmit antenna Tx chosen as the origin of the coordinate system, the overall considered observation area spans from −7.5 m to 30 m in the x-direction and from −10 m to 26 m in the x-direction. The selected scenario considers a walking pedestrian as single mobile scatterer (Q = 1), walking a wide circle within the illuminated area of the static network setup. As the experiment was conducted on an apron close to surrounding buildings, an elevated tachymeter was used as ground truth system. Using the tachymeter together with a high precision reﬂector prism, the individual static antenna locations of the network were determined prior to the experiment. For recording the ground truth during the experiment, the pedestrian wore a helmet, on which the reﬂector prism was mounted. Sensors 2019, 19, 4802 10 of 19 10 of 19 (a) −5 0 5 10 15 20 25 −10 −5 0 5 10 15 20 25 Tx Rx1 Rx2 Rx3 Rx4 x [m] y [m] −30 −20 −10 0 Tx Antenna Gain [dBi] (b) Figure 1. Overview of the evaluated measurement network with transmitting and receiving nodes as circle and triangles. The dashed lines illustrate the trajectory traveled. The gray line highlights the hangar wall as the strongest static reﬂector in the environment. The vertical radiation pattern of Tx antenna is shown in light gray. (a) Scenario overview with starting position of moving pedestrian and movement direction; (b) Scenario overview illustrating gain of small directional transmit antenna [28]. −5 0 5 10 15 20 25 −10 −5 0 5 10 15 20 25 Tx Rx1 Rx2 Rx3 Rx4 x [m] −30 −20 −10 0 Tx Antenna Gain [dBi] (b) Tx Antenna Gain [dBi] (a) y [m] [m] (b) (a) Figure 1. Overview of the evaluated measurement network with transmitting and receiving nodes as circle and triangles. The dashed lines illustrate the trajectory traveled. 5.1. Network and Measurement Setup The gray line highlights the hangar wall as the strongest static reﬂector in the environment. The vertical radiation pattern of Tx antenna is shown in light gray. (a) Scenario overview with starting position of moving pedestrian and movement direction; (b) Scenario overview illustrating gain of small directional transmit antenna [28]. As a measurement system, the Medav RUSK-DLR wideband channel sounder was used [29]. Therefore, the measured data are CIRs between transmit and receive antennas, i.e., network nodes, respectively. A full summary of the corresponding measurement parameter settings provides Table 1. Table 1. Measurement parameters. Table 1. Measurement parameters. Parameter Value Center frequency fc 5.2 GHz Bandwidth B 120 MHz Signal period Tp 3.2 µs Measurement rate Tg 2.048 ms Transmit power PTx 37 dBm Antenna gain GTx 9 dBi (small directional [28]) Antenna gain GRx1−4 8 dBi (toroidal, omni-directional) Table 1. Measurement parameters. Table 1. Measurement parameters. 5.2. Theoretical Performance Evaluation For illustrating the localization performance of the measurement setup, the RMSE is calculated for the considered observation area according to the CRLB as deﬁned in Equation (24). To determine the RMSE, the parameters provided in Table 1 are used. Assuming a transmit signal with rectangular power spectral density, the effective bandwidth is deﬁned as β2 = B2/12 [21]. The value of the RCS accounts for an object’s reﬂectivity characteristics, which is inﬂuenced by its size, shape, and material. Thus, for a pedestrian being considered as moving object, a typical RCS is 1 m2 [7]. Eventually, the 11 of 19 Sensors 2019, 19, 4802 localization RMSE for the measurement setup is shown in Figure 2a. Overall, the shape of the RMSE reveals good localization performance for scatterers located in areas of high transmit antenna gain; see also Figure 1b. Hence, scatterers located close to the transmitting node can be localized very precisely. Qualitatively, this holds also for scatterers located close to receiving nodes. For these areas, the high localization performance can be explained by the received signal strength of the backscattered signal, since the RMSE strongly depends on the SNR. Additionally, the decreasing localization performance for far positions from the network nodes conﬁrms this observation. Besides SNR, the localization performance depends on the system geometry. Effects due to system geometry can be explained by interpreting the scatterer location as the intersection point of multiple ellipses, which correspond to the propagation delays of respective transmitting and receiving nodes. To achieve high localization performance in both x- and y-directions, ideally, ellipses should intersect perpendicular to each other. For scatterers located far from the localization network, the shapes of the individual ellipses differ only marginally from each other. Thus, the intersection angle is very low, which results in an increased localization uncertainty in the direction parallel to the tangent at the intersection point. With a widely distributed network covering a large observation area, the effects of low intersection angles can be avoided. A further geometrical effect can be observed for scatterers located in the proximity of the baseline between transmitting and receiving nodes. Regarding a single link, the location information in the direction parallel to the baseline is very low, and, thus, the localization performance also decreases. Accordingly, in the localization network, the localization performance decreases in the proximity of the individual baselines; see Figure 2a. 5.2. Theoretical Performance Evaluation This performance degradation is independent of the network topology. −5 0 5 10 15 20 25 −10 −5 0 5 10 15 20 25 Tx Rx1 Rx2 Rx3 Rx4 x [m] y [m] 0 0.1 0.2 0.3 0.4 RMSE [m] (a) −5 0 5 10 15 20 25 −10 −5 0 5 10 15 20 25 Tx Rx1 Rx2 Rx3 Rx4 x [m] y [m] Scenario I Scenario II Scenario III (b) Figure 2. Overview of the measurement network, showing (a) the resulting CRLB on position estimation with dashed line as experiment trajectory, and (b) the three scenario trajectories for evaluating the PCRLB and the EKF; the trajectories represent the mean values of the position over a time period of 10 s and indicate starting positions and moving directions. −5 0 5 10 15 20 25 −10 −5 0 5 10 15 20 25 Tx Rx1 Rx2 Rx3 Rx4 x [m] 0 0.1 0.2 0.3 0.4 RMSE [m] (a) y [m] −5 0 5 10 15 20 25 −10 −5 0 5 10 15 20 25 Tx Rx1 Rx2 Rx3 Rx4 x [m] Scenario I Scenario II Scenario III (b) y [m] (b) (a) Figure 2. Overview of the measurement network, showing (a) the resulting CRLB on position estimation with dashed line as experiment trajectory, and (b) the three scenario trajectories for evaluating the PCRLB and the EKF; the trajectories represent the mean values of the position over a time period of 10 s and indicate starting positions and moving directions. Please note that the derived CRLB of Section 4 only depends on waveform and SNR. This means that the inﬂuence of any superposition of LoS and MPCs on the parameter estimation is not considered [27], and is out of the scope of this paper. Since the superposition of LoS and MPCs Sensors 2019, 19, 4802 12 of 19 strongly impacts the estimation capabilities of its inﬂuence on parameter estimation will be included in future research. In order to evaluate the performance of the EKF proposed in Section 3.3, the PCRLB is calculated for different scenarios within the measurement setup and compared to simulation results of the tracking ﬁlter. Three different scenarios are considered, each for a single scatterer moving with constant velocity. The initial absolute velocity is 1.41 m/s for every scenario. However, each scenario possesses an individual starting position and movement direction. 5.2. Theoretical Performance Evaluation Figure 2b provides an overview over the considered scenarios. Referring to Figure 2a, Scenario I is characterized by high localization capabilities throughout the whole trajectory. Scenario II, in contrast, crosses an area of poor localization capabilities between Tx and Rx3. In Scenario III, the trajectory starts in an area of poor localization capabilities and moves towards an area of very high localization capabilities in the main beam the transmitting node. For the calculation of the PCRLB as well as the EKF simulations, similar system and signal parameters as for calculating the static positioning CRLB are used. As process noise intensity for the covariance in Equation (14), a value of σ2 q = 0.01 m2/s3 is deﬁned [23]. The ﬁlter state is initialized randomly according to the initial state covariance P(t0) around the initial state x(t0) given by the respective scenario. Thereby, P(t0) is deﬁned as 4 × 4 diagonal matrix, with an initial variance of 0.1 m2 on position and 0.01 m2/s2 on velocity, in both x- and y-directions. Determining the PCRLB and the EKF performance results requires performing multiple Monte Carlo runs. In each run, the system equations of Section 3.3 are simulated with different samples of the process noise for 10 s, which equals an average walking distance of approximately 14 m. Thus, the multitude of simulation runs allows for approximating the expectation in Equation (33). In order to estimate the MSE matrices in Equation (25), the measurement noise also needs to be sampled for each run. In this study, for every scenario, 5000 Monte Carlo runs are performed with 50 realizations of measurement noise. Hence, the EKF is evaluated 2.5 × 105 times in each scenario. Thereby, the number of Monte Carlo runs was chosen to achieve statistically stable results for the PCRLB and to achieve results for the EKF, which ﬂuctuate only marginally compared to the absolute RMSE values. y g y p Figure 3 shows the simulation results for the three scenarios. The results for both PCRLB and EKF are given in terms of RMSE, i.e., as the square root of the MSE in Equation (25). The RMSE of PCRLB and EKF are denoted by ϵPCRLB and ϵEKF. Given P(t0), the initial RMSE of the positioning error is approximately 0.45 m for all scenarios. 5.3. Measurement Based Evaluation As stated in Section 3, the channel parameters for all links in the network are estimated using KEST. The estimation results of KEST for the measured CIR over time are shown in Figure 4a–d for each link individually. The ﬁgures show the consecutive vectors of delay estimates in Equation (5) over the full measurement time. Static delay estimates, including LoS and static MPCs, are shown in gray. These static delays together with corresponding standard deviations are determined during a preceding calibration phase according to the procedures described in Section 3.1. This characterization of the static propagation environment allows for extracting the mobile MPC, highlighted in color according to the estimated amplitude level. For each link, the delay estimates ﬁt the ground truth data as indicated by black dashed lines. Hence, the results qualitatively conﬁrm the point scattering assumption for pedestrians [29]. Due to the geometrical arrangement of the receiving antennas and limits in the dynamic range of the measurement system, the detection, estimation, and tracking capabilities of KEST differ. With the small directional antenna gain and the orientation of the transmitting antenna, as shown in Figure 1b, the LoS signal power is reduced for all links. The reduced LoS signal power avoids an elevation of the noise ﬂoor due to limits of the dynamic range. Noticeably, link Tx-Rx3 exhibits particularly good parameter estimation throughout the measurement. On the one hand, this can be explained by the advantageous placement of the receive antenna with respect to the transmit antenna gain, i.e., very low Tx gain towards Rx3. On the other hand, Rx3 and also Rx4 are not impacted by strong multipath fading like Rx1 and Rx2. Since Rx1 and Rx2 are located closer to the hangar wall, signal reﬂections off the wall are received with higher power. Examples for such multipath fading provide the estimation results of Rx1 and Rx2, where double reﬂections from the pedestrian and the hangar wall are clearly visible (see Figure 4a,b). However, apart from the perturbing effect regarding parameter estimation, assignable double reﬂections from mobile scatterers would contain additional location information. Even though in this work double reﬂections are not considered as source of information, an exploitation of reﬂected mobile MPCs will be included in future research. 5.2. Theoretical Performance Evaluation Due to sequential ﬁltering characteristics, the localization RMSE strongly decreases shortly after the initialization and follows the static performance capabilities along the respective scenario trajectory. Overall, it can be observed that the ﬁlter RMSE values are very close to those of the theoretical bound. This holds for the simulation results of each scenario. Particularly, for Scenario I, the results of the EKF converge very fast and reach the bound in less than 1 s; see Figure 3a. In addition, for Scenario II, the EKF converges to the bound after about 3 s as shown in Figure 3b. With the trajectory crossing an area of poor localization capabilities between Tx and Rx3, both the PCRLB and the ﬁlter RMSE show a temporary increase. For scatterer positions close to the baseline between transmitting and receiving nodes, the elements of the Jacobian in Equation (10) and therewith of the observation matrix in Equation (17) are close to zero. Due to singularities caused by the inversion of the observation matrix, the region close to the baselines between Tx and Rx3, as well as Tx and Rx4, is particularly challenging for the ﬁlter. For calculating the bound in Equation (34), the expectation of the FIM averages these singularities. The results in Figure 3b conﬁrm this effect, with the ﬁlter RMSE diverging from the bound between 1 s and 2 s. Finally, the ﬁlter results for Scenario III are provided in Figure 3c. The ﬁlter RMSE shows a fast convergence to the bound after approximately 2 s. As described above, however, the effects of singularities on the EKF in regions close to Tx-Rx baselines also inﬂuences the ﬁlter results of the third scenario. Here, the mean trajectory crosses Tx-Rx4 and travels along Tx-Rx3. With increasing state covariance, many simulation trajectories lie on or are close to the baselines. Thus, the ﬁlter RMSE does not fully converge to the bound in the period after 2 s of simulation time. 5.2. Theoretical Performance Evaluation Sensors 2019, 19, 4802 13 of 19 13 of 19 0 2 4 6 8 10 0 5 10 15 20 25 30 35 40 45 50 t [s] RMSE [mm] ǫP CRLB ǫEKF (a) Scenario I 0 2 4 6 8 10 0 5 10 15 20 25 30 35 40 45 50 t [s] RMSE [mm] ǫP CRLB ǫEKF (b) Scenario II 0 2 4 6 8 10 0 5 10 15 20 25 30 35 40 45 50 t [s] RMSE [mm] ǫP CRLB ǫEKF (c) Scenario III Figure 3. Simulation results for Scenarios I–III as provided in Figure 2b. Results for both PCRLB and EKF are given in terms of RMSE, referred to as ϵPCRLB and ϵEKF. (b) Scenario II (a) Scenario I Figure 3. Simulation results for Scenarios I–III as provided in Figure 2b. Results for both PCRLB and EKF are given in terms of RMSE, referred to as ϵPCRLB and ϵEKF. 5.3. Measurement Based Evaluation Besides dynamic range and multipath fading, the estimation results of Figure 4 clearly show how the presence of LoS and static MPCs impact the composition of measurement vector ˆτq and therewith the localization of the mobile scatterer. An unambiguous solution of Equation (8) requires at least Sensors 2019, 19, 4802 14 of 19 three independent measurements. The required static environment mitigation, i.e., the displacement of any delay estimates close to static components (see Section 3.2), however, reduces the amount of delay estimates assigned to a speciﬁc mobile scatterer. Thus, a rich static MPC environment reduces the overall localization capabilities with the current approach. This holds particularly for very sparse networks, such as the four-link network considered in this paper, since an outage in two links impedes localization. Besides static MPCs, the LoS also impacts the localization capability due to the so-called blind zone problem [30]. This means that MPCs caused by scatterers located close to the baseline between a transmitting and receiving node is hardly detectable. Exemplary, the parameter estimation results for link Tx-Rx4 conﬁrm these blind zones, since it is not possible to extract mobile MPCs when the pedestrian is located in the proximity to the link baseline. Thus, for Rx4, no parameter estimates are available from 33 s to 44 s. Finally, the localization approach proposed in Section 3 is used to estimate and track the location of the walking pedestrian using the CIR measurements described above. For initialization, the iterative localization procedure of Equation (9) is applied to determine an initial location estimate based on available parameter estimates of the four links at t0 = 0 s. The initial velocity components in x- and y-directions are randomly chosen and taken from a uniform distribution U(0 m/s, 1 m/s), respectively. Given the initial state, the EKF described in Section 3.3 is used for tracking the location of the pedestrian. According to [23], the covariance matrix of the process noise in Equation (14) is determined by the process noise intensity of σ2 q = 0.01 m2/s3. The resulting positioning error for the experiment over time is given in Figure 5 and the corresponding cumulative distribution function (CDF) in Figure 6. Overall, except for a few outliers, the positioning error remains below 1 m throughout the experiment. In particular, according to the CDF, more than 97% of positioning errors are below 1 m. 5.3. Measurement Based Evaluation The ﬁrst outlier appears between 20 s and 25 s. This time period coincides with a sharp left curve, as shown in Figure 1a. Thus, a delayed adaption of the velocity vector by the EKF can be a possible explanation for the exceeding positioning error. A very strong localization performance can be observed between 32 s and 37 s, when the pedestrian passes the main beam of the transmit antenna. Due to strong reﬂection and therewith high SNR, the parameters of each link and thus the position can be estimated very well. For the period from 37 s to 44 s, the error steadily increases. In this period, parameter estimates are almost solely available for Rx3. Hence, the few simultaneous and even erroneous delay estimates can explain the increasing positioning error and the outlier at about 44 s. After leaving the blind zone of Tx-Rx4 at about 44 s, the additional link parameter estimates support the positioning performance thereafter. The increasing error after 53 s is again due to few simultaneous parameter estimates from different network links. Similarly to before, the position estimation is mainly driven by parameter estimates for Rx3. Sensors 2019, 19, 4802 15 of 19 15 of 19 ✁✂✄✁☎ ✆ ✁✝✝ ☎✞✟✞✠✡☛☞ ✂ 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 ✡ ✌✍✎ ✏ τ ✑ ✒ ✓ −24 −22 −20 −18 −16 −14 −12 −10 −8 −6 −4 −2 ✔✕✖ ✝ ☛✡ ✗✘ ✞ ✌ ✘✙✕ ✎ (a) Tx-Rx1 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 t [s] c τ [m] (b) Tx-Rx2 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 t [s] c τ [m] (c) Tx-Rx3 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 t [s] c τ [m] (d) Tx-Rx4 Figure 4. Estimation results of KEST for CIR over time of moving pedestrian. Extracted mobile MPC i colored according to the estimated amplitude level. LoS and static MPCs are shown in gray. Dashed black lines indicate delays from ground truth data. Other mobile MPCs deviating from ground truth can be referred to double reﬂections of moving pedestrian and hangar wall. 5.3. Measurement Based Evaluation ✁✂✄✁☎ ✆ ✁✝✝ ☎✞✟✞✠✡☛☞ ✂ 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 ✡ ✌✍✎ ✏ τ ✑ ✒ ✓ −24 −22 −20 −18 −16 −14 −12 −10 −8 −6 −4 −2 ✔✕✖ ✝ ☛✡ ✗✘ ✞ ✌ ✘✙✕ ✎ (a) Tx-Rx1 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 t [s] c τ [m] ✁✂✄✁☎ ✆ ✁✝✝ ☎✞✟✞✠✡☛☞ ✂ 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 ✏ τ ✑ ✒ ✓ −24 −22 −20 −18 −16 −14 −12 −10 −8 −6 −4 −2 ✔✕✖ ✝ ☛✡ ✗✘ ✞ ✌ ✘✙✕ ✎ ✁✂✄✁☎ ✆ ✁✝✝ ☎✞✟✞✠✡☛☞ ✂ 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 t [s] c τ [m] 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 t [s] c τ [m] (c) Tx-Rx3 0 5 10 15 20 25 30 35 40 45 50 55 0 20 40 60 80 t [s] c τ [m] [m] Figure 4. Estimation results of KEST for CIR over time of moving pedestrian. Extracted mobile MPC is colored according to the estimated amplitude level. LoS and static MPCs are shown in gray. Dashed black lines indicate delays from ground truth data. Other mobile MPCs deviating from ground truth can be referred to double reﬂections of moving pedestrian and hangar wall. Sensors 2019, 19, 4802 16 of 19 16 of 19 0 5 10 15 20 25 30 35 40 45 50 55 0 0.2 0.4 0.6 0.8 1 1.2 t [s] Positioning error [m] Figure 5. Positioning results of localization approach based on channel measurements—absolute positioning error of moving pedestrian over time. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 0 20 40 60 80 100 Positioning error [m] CDF [%] Figure 6. CDF of absolute positioning error for a moving pedestrian scenario. l 0 5 10 15 20 25 30 35 40 45 50 55 0 0.2 0.4 0.6 0.8 1 1.2 t [s] Positioning error [m] Figure 5. Positioning results of localization approach based on channel measurements—absolute positioning error of moving pedestrian over time. 5.3. Measurement Based Evaluation 0 5 10 15 20 25 30 35 40 45 50 55 0 0.2 0.4 0.6 0.8 1 1.2 t [s] Positioning error [m] Positioning error [m] Figure 5. Positioning results of localization approach based on channel measurements—absolute positioning error of moving pedestrian over time. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 0 20 40 60 80 100 Positioning error [m] CDF [%] Figure 6. CDF of absolute positioning error for a moving pedestrian scenario. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 0 20 40 60 80 100 Positioning error [m] CDF [%] Figure 6. CDF of absolute positioning error for a moving pedestrian scenario. 6. Conclusions This paper presented a localization approach to detect, locate, and track moving, non-cooperative objects by means of a network of spatially distributed transmitting and receiving nodes. With moving objects affecting the radio spectrum as time-variant MPCs, i.e., discrete mobile scatterers, multipath propagation could be exploited for passive localization. Therefore, the localization of mobile scatterers was formulated as a nonlinear optimization problem. For initialization, the proposed approach uses an iterative nonlinear least squares algorithm following Levenberg and Marquardt to solve the optimization problem. Subsequently, an EKF is applied to estimate the scatterer location recursively over time. For both the snapshot based localization and the nonlinear sequential Bayesian estimation problem performance bounds, the classic CRLB on position estimation and the PCRLB were derived. This paper presented a localization approach to detect, locate, and track moving, non-cooperative objects by means of a network of spatially distributed transmitting and receiving nodes. With moving objects affecting the radio spectrum as time-variant MPCs, i.e., discrete mobile scatterers, multipath propagation could be exploited for passive localization. Therefore, the localization of mobile scatterers was formulated as a nonlinear optimization problem. For initialization, the proposed approach uses an iterative nonlinear least squares algorithm following Levenberg and Marquardt to solve the optimization problem. Subsequently, an EKF is applied to estimate the scatterer location recursively over time. For both the snapshot based localization and the nonlinear sequential Bayesian estimation problem performance bounds, the classic CRLB on position estimation and the PCRLB were derived. The proposed approach is evaluated based on a case study, which considers a pedestrian as a single mobile scatterer walking within a network of one transmitting and four receiving nodes. Thereby, a simulation study has shown that the EKF achieves a localization performance very close to the PCRLB. Even though the EKF performance results diverged from the theoretical bound in The proposed approach is evaluated based on a case study, which considers a pedestrian as a single mobile scatterer walking within a network of one transmitting and four receiving nodes. Thereby, a simulation study has shown that the EKF achieves a localization performance very close to the PCRLB. Even though the EKF performance results diverged from the theoretical bound in Sensors 2019, 19, 4802 17 of 19 areas close to Tx-Rx baselines due to singularities, the study overall conﬁrmed the applicability of the EKF for the localization problem. Abbreviations CDF cumulative distribution function CIR channel impulse response CRLB Cramér–Rao lower bound EKF extended Kalman ﬁlter FIM Fisher information matrix KEST Kalman enhanced super resolution tracking LoS line-of-sight MPC multipath component MSE mean square error PCL passive coherent location PCRLB posterior Cramér–Rao lower bound RCS radar cross-section RMSE root mean square error SNR signal-to-noise ratio CDF cumulative distribution function CIR channel impulse response CRLB Cramér–Rao lower bound EKF extended Kalman ﬁlter FIM Fisher information matrix KEST Kalman enhanced super resolution tracking LoS line-of-sight MPC multipath component MSE mean square error PCL passive coherent location PCRLB posterior Cramér–Rao lower bound RCS radar cross-section RMSE root mean square error SNR signal-to-noise ratio 6. Conclusions Particularly, for scatterers remote from any Tx-Rx baseline, the localization performance of the ﬁlter was shown to converge to the PCRLB. Moreover, the approach was applied to wideband measurement data corresponding to the case study of the theoretical analysis. With a resulting localization accuracy of 0.8 m at 90% conﬁdence, the case study proved that mobile scatterers can be localized using CIRs, i.e., exploiting multipath propagation. The performance of the proposed approach strongly depends on the underlying parameter estimation algorithm and the mitigation of the LoS and static MPCs. Since both the parameter estimation and the mitigation of the static components are prone to errors, future work will focus on a direct multilateration approach. The direct usage of CIRs for localization and tracking could help to overcome the dependence on individual parameter estimation and would link an object’s mobility information implicitly to the recursive position estimation. Author Contributions: M.S., C.G., and B.S. worked on the conceptualization and the methodology. S.S. provided scientiﬁc support and helped with the formal analysis. The original draft of the manuscript was prepared and written by M.S. and revised by C.G., B.S., and S.S. Funding: This research received no external funding Acknowledgments: The authors want to thank Siwei Zhang for the fruitful discussions on Bayesian performance bounds. Acknowledgments: The authors want to thank Siwei Zhang for the fruitful discussions on Bayesian performance bounds. 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. References 1. Biswas, S.; Tatchikou, R.; Dion, F. Vehicle-to-vehicle wireless communication protocols for enhancing highway trafﬁc safety. IEEE Commun. Mag. 2006, 44, 74–82. [CrossRef] 2. Hartenstein, H.; Laberteaux, L.P. A Tutorial Survey on Vehicular Ad Hoc Networks. IEEE Commun. Mag. 2008, 46, 164–171. [CrossRef] 3. De Ponte Müller, F. Survey on Ranging Sensors and Cooperative Techniques for Relative Positioning of Vehicles. Sensors 2017, 17, 271. [CrossRef] [PubMed] 4. Jo, K.; Kim, J.; Kim, D.; Jang, C.; Sunwoo, M. Development of Autonomous Car—Part II: A Case Study on the Implementation of an Autonomous Driving System Based on Distributed Architecture. IEEE Trans. Ind. Electron. 2015, 62, 5119–5132. [CrossRef] 18 of 19 Sensors 2019, 19, 4802 18 of 19 5. Munoz-Ferreras, J.M.; Perez-Martinez, F.; Calvo-Gallego, J.; Asensio-Lopez, A.; Dorta-Naranjo, B.P.; Blanco-del-Campo, A. Trafﬁc Surveillance System Based on a High-Resolution Radar. IEEE Trans. Geosci. Remote Sens. 2008, 46, 1624–1633. [CrossRef] 6. Sivaraman, S.; Trivedi, M.M. Looking at Vehicles on the Road: A Survey of Vision-Based Vehicle De Tracking, and Behavior Analysis. IEEE Trans. Intell. Transp. Syst. 2013, 14, 1773–1795. [CrossRef] 7. Schmidhammer, M.; Sand, S.; Soliman, M.; de Ponte Müller, F. 5G Signal Design for Road Surveillance. In Proceedings of the 14th Workshop on Positioning, Navigation and Communications (WPNC), Bremen, Germany, 25–26 October 2017. y 8. Thomä, R.S.; Andrich, C.; Del Galdo, G.; Döbereiner, M.; Hein, M.A.; Käske, M.; Schäfer, G.; Schieler, S.; Schneider, C.; Schwind, A.; et al. Cooperative Passive Coherent Location: A Promising 5G Service to Support Road Safety. IEEE Commun. Mag. 2019, 57, 86–92. [CrossRef] 9. Grifﬁths, H.D.; Baker, C.J. Passive coherent location radar systems. Part 1: Performance prediction. IEE Proc. Radar Sonar Navig. 2005, 152, 153–159. [CrossRef] 10. Haimovich, A.M.; Blum, R.S.; Cimini, L.J. MIMO Radar with Widely Separated Antennas. IEEE Signal Process. Mag. 2008, 25, 116–129. [CrossRef] 11. Grifﬁths, H.D.; Al-Ashwal, W.A.; Ward, K.D.; Tough, R.J.A.; Baker, C.J.; Woodbridge, K. Measurement and modelling of bistatic radar sea clutter. IET Radar Sonar Navig. 2010, 4, 280–292. [CrossRef] 12. Filip, A.; Shutin, D. Cramér-Rao bounds for L-band digital aeronautical communication system type 1 based passive multiple-input multiple-output radar. IET Radar Sonar Navig. 2016, 10, 348–358. [CrossRef] 13. Falcone, P.; Colone, F.; Lombardo, P. Potentialities and challenges of WiFi-based passive radar. IEEE Aerosp. Electron. Syst. Mag. 2012, 27, 15–26. [CrossRef] 14. Pastina, D.; Colone, F.; Martelli, T.; Falcone, P. Parasitic Exploitation of Wi-Fi Signals for Indoor Radar Surveillance. IEEE Trans. Veh. Technol. References 2015, 64, 1401–1415. [CrossRef] 15. Colone, F.; Martelli, T.; Bongioanni, C.; Pastina, D.; Lombardo, P. WiFi-based PCL for monitoring private airﬁelds. IEEE Aerosp. Electron. Syst. Mag. 2017, 32, 22–29. [CrossRef] 16. Sobhani, B.; Paolini, E.; Giorgetti, A.; Mazzotti, M.; Chiani, M. Target Tracking for UWB Multistatic Radar Sensor Networks. IEEE J. Sel. Top. Signal Process. 2014, 8, 125–136. [CrossRef] 17. Schmidhammer, M.; Gentner, C.; Siebler, B. Localization of Discrete Mobile Scatterers in Vehicular Environments Using Delay Estimates. In Proceedings of the 2019 International Conference on Localization and GNSS (ICL-GNSS), Nuremberg, Germany, 4–6 June 2019; pp. 1–6. 18. Jost, T.; Wang, W.; Fiebig, U.; Perez-Fontan, F. Detection and Tracking of Mobile Propagation Channel Paths. IEEE Trans. Antennas Propag. 2012, 60, 4875–4883. [CrossRef] 19. Karedal, J.; Tufvesson, F.; Czink, N.; Paier, A.; Dumard, C.; Zemen, T.; Mecklenbrauker, C.F.; Molisch, A.F. A geometry-based stochastic MIMO model for vehicle-to-vehicle communications. IEEE Trans. Wirel. Commun. 2009, 8, 3646–3657. [CrossRef] 20. Pelleg, D.; Moore, A. X-means: Extending k-means with efﬁcient estimation of the number of clusters. In Proceedings of the 17th International Conference on Machine Learning (ICML), Stanford, CA, USA, 29 June–2 July 2000; pp. 727–734. 21. Kay, S.M. Fundamentals of Statistical Signal Processing: Estimation Theory; Prentice-Hall: Upper Saddle River, NJ, USA, 1993. 22. Mensing, C.; Plass, S. Positioning algorithms for cellular networks using TDOA. In Proceedings of the 2006 IEEE International Conference on Acoustics Speech and Signal Processing, Toulouse, France, 14–19 May 2006; Volume 4, pp. 513–516. 23. Bar-Shalom, Y.; Li, X.; Kirubarajan, T. Estimation with Applications to Tracking and Navigation: Theory, Algorithms and Software; John Wiley & Sons: Hoboken, NJ, USA, 2004. 24. Tichavsky, P.; Muravchik, C.H.; Nehorai, A. Posterior Cramér-Rao Bounds for Discrete-Time Nonlinear Filtering. IEEE Trans. Signal Process. 1998, 46, 1386–1396. [CrossRef] 25. Van Trees, H.L.; Bell, K.L. Bayesian Bounds for Parameter Estimation and Nonlinear Filtering/Tracking; Wiley-IEEE Press: Hoboken, NJ, USA, 2007. 26. Siebler, B.; Sand, S. Posterior Cramér-Rao Bound and Suboptimal Filtering for IMU/GNSS based Cooperative Train Localization. In Proceedings of the 2016 IEEE/ION Position, Location and Navigation Symposium (PLANS), Savannah, GA, USA, 11–14 April 2016; pp. 353–358. 19 of 19 Sensors 2019, 19, 4802 27. Gentner, C.; Jost, T.; Wang, W.; Zhang, S.; Dammann, A.; Fiebig, U. Multipath Assisted Positioning with Simultaneous Localization and Mapping. IEEE Trans. Wirel. Commun. 2016, 15, 6104–6117. [CrossRef] 28. Huber & Suhner. c⃝2019 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/). References SENCITY Spot-S WiFi Dual-Band Antenna 1399.17.0210; Huber & Suhner: Herisau, S it l d 2019 27. Gentner, C.; Jost, T.; Wang, W.; Zhang, S.; Dammann, A.; Fiebig, U. Multipath Assisted Positioning with Simultaneous Localization and Mapping. IEEE Trans. Wirel. Commun. 2016, 15, 6104–6117. [CrossRef] 28. Huber & Suhner. SENCITY Spot-S WiFi Dual-Band Antenna 1399.17.0210; Huber & Suhner: Herisau, Switzerland, 2019. 29. Schmidhammer, M.; de Ponte Müller, F.; Sand, S.; Rashdan, I. Detection and Localization of Non-Cooperative Road Users based on Propagation Measurements at C-Band. In Proceedings of the 12th European Conference on Antennas and Propagation (EuCAP 2018), London, UK, 9–13 April 2018; pp. 1–5. 29. Schmidhammer, M.; de Ponte Müller, F.; Sand, S.; Rashdan, I. Detection and Localization of Non-Cooperative Road Users based on Propagation Measurements at C-Band. In Proceedings of the 12th European Conference on Antennas and Propagation (EuCAP 2018), London, UK, 9–13 April 2018; pp. 1–5. 30. Chiani, M.; Giorgetti, A.; Paolini, E. Sensor Radar for Object Tracking. Proc. IEEE 2018, 106, 1022–1041. [CrossRef] c⃝2019 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/W2970028299	https://ieeexplore.ieee.org/ielx7/6287639/8600701/08812699.pdf	English	null	Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor Tucker Decomposition	IEEE access	2,019	cc-by	5,075	The associate editor coordinating the review of this article and approving it for publication was Md Jahangir Hossain. Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor Tucker Decomposition This work was supported in part by the China Southern Power Grid Science and Technology Project Funding unde Grant 0002200000038772. ABSTRACT In order to deal with the problem of massive historical multi-station SCADA data storage in distribution management system, this paper proposes a data compression method for power distribution system based on tensor Tucker decomposition. Firstly, to maintain the high-dimensional spatial structure of multi-station SCADA data of distribution management system in the presentation stage, we establish a third- order tensor representation model of multi-station SCADA data. Then, a historical SCADA data compression method for distribution management system based on tensor tucker decomposition is proposed. This method can compress the data while maintaining the spatial intrinsic structure of multi-station SCADA data. Finally, the effectiveness of the method is veriﬁed using real data. And, the results of comparison with singular value decomposition method and principal component analysis method show that the proposed method is superior to the traditional method. INDEX TERMS Multi-station SCADA data, tensor tucker decomposition, data compression, power distri- bution system, spatial intrinsic structure. Received July 29, 2019, accepted August 21, 2019, date of publication August 26, 2019, date of current version September 13, 2019. Received July 29, 2019, accepted August 21, 2019, date of publication August 26, 2019, date of current version September 13, 2019. Digital Object Identifier 10.1109/ACCESS.2019.2937383 I. INTRODUCTION compressed and decompressed the real-time data of the fault transient process of the power system by using the linear integer transformation wavelet bi-orthogonal ﬁlter combina- tion Huffman coding method based on the lifting format, and the data volume after compression was 11% of the original data volume [5]. On this basis, the literature [6] improved the compression accuracy of this method by performing cubic spline re-sampling of the original data. Some researchers, in order to solve the problem of large data volume in the wide- area monitoring system(WAMS), an algorithm combining outliers and revolving doors is adopted to compress the real-time data of the wide-area monitoring system, and the data volume after compression is 10% to 20% [7]. In addi- tion, the revolving door compression algorithm based on effective estimation is also used in historical SCADA data compression to reduce the historical data volume of SCADA system and ensure the safe and stable operation of SCADA system [8], [9]. And, principal component analysis (PCA) method and singular value decomposition (SVD) method show their advantages when compressing traditional SCADA steady-state data [9], [10]. Souza uses singular value decom- position method to compress steady state data, and its results show that SVD method better than WT method [9]. The distribution management system (DMS) of power distri- bution network continuously receives the status monitoring data from a large number of distribution SCADA systems. These data are accumulated over time, putting a lot of pres- sure on the computer storage space [1]. Therefore, the scheme of using data compression technology to effectively reduce data volume and improve storage efﬁciency has attracted attention [2]. In terms of power system data compression, Ringwelski proposed the method of Huffman coding (HC) to compress the smart meter data, which could compress the data to 25% of the original data volume [3]. In literature [4], a com- pression algorithm suitable for the load data transmission requirements of smart electricity meters is introduced, and its performance is better than the traditional encoding transmis- sion scheme. For the compression of power system transient data, most researches focus on how to use the multi-resolution feature of wavelet transform (WT) and its improved algorithm to compress signals [5], [6]. Yan Changyou et al. The associate editor coordinating the review of this article and approving it for publication was Md Jahangir Hossain. 124390 under a Creative Commons Attribution 4.0 License. I. INTRODUCTION For more information, see http://creativecommons.org/licenses/by/4.0/ VOLUME 7, 2019 This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see http://creativecommons.org/licenses/by/4.0 H. Zhao et al.: Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor TD In reference [10], the error of PCA method in compressed steady-state SCADA data is also very small. According to (2), the unfolding process will sample all the orders of the tensor in staggered order. This acquisition process realizes the transmission and fusion of various order features of the tensor, thereby preserving the spatial eigen- structure of the data. For example, a third-order tensor is unfolded from three orders, as shown in Fig. 1. y y However, traditional compression algorithms (HC, WT, SVD, PCA) need to ‘‘hard dimensionality reduction’’ in the process of high-dimension data compression. This will destroy the intrinsic structure of high-dimensional data space and cover up the original redundant information and high- order dependence of data [11]–[16]. In practice, the main station data of distribution system is composed of large quantum sub-station data. Therefore, when the traditional data compression algorithm is used to compress the data of multiple sub-stations, the correlation between the data of each sub-station will be destroyed and the compression performance will decline. In order to overcome the damage of traditional 2-D compression to the high-dimensional intrinsic structure of SCADA data, this paper proposes a data com- pression method based on tensor Tucker decomposition (TD) for the multi-stations SCADA data compression of power distribution system. Firstly, we establish a third-order tensor representation model of multi-station SCADA data. Then, the tensor Tucker decomposition method is used to compress the data while preserving the spatial intrinsic structure of the data. In order to ensure the accuracy of compressed data to meet the application requirements, we establish the compression ratio and three information loss level indicators. The balance between data compression and information loss can be realized by setting compression ratio according to the accuracy requirement of application. Finally, real distribution system data are used to verify that the proposed method can effectively reduce the amount of distribution system data to deal with the problem of data storage. Compared with other two existing methods (PCA, SVD), the results show that the proposed method is superior. FIGURE 1. Unfolding of a third-order tensor. FIGURE 1. Unfolding of a third-order tensor. FIGURE 1. Unfolding of a third-order tensor. I. INTRODUCTION Deﬁnition 2: The multiplication of the N order tensor A ∈ RI1×I2×···×Ii−1×Ii×Ii+1···×IN by the matrix U∈RJ×Ii is called the i-mode multiplication of the tensor. The formula is (A ×i U)x1···xi−1jxi+1···xN = XIn xi=1 ax1x2···xN ujxi (3) (3) According to (3), if the number of rows J of the matrix UJ×Ii is less than the number of columns Ii, the dimension of the original tensor of the i-th order will be compressed from Ii to J, which is the key operation for realizing data compression. In Fig. 2, the dimensionality of the second order of the original tensor is reduced from 8 to 2 via the 2-mode product. II. TENSOR DECOMPOSITION A. TENSOR OPERATION FIGURE 2. Tensor dimensionality reduction with the i-mode product. Tensors are high-dimensional arrays. The spatial dimension- ality of a tensor is often called the order of the tensor. Zero-order tensors are scalars, which are represented by scalar a; ﬁrst-order tensors are vectors, which are represented by boldface a; second-order matrix tensors are represented by capital A; tensors of order three and above are called higher order tensors and are represented by the cursive capital letter A. FIGURE 2. Tensor dimensionality reduction with the i-mode product. Deﬁnition 1: The operation of unfolding a tensor to a matrix is called the i-mode unfolding operation. The i-mode unfolding operation of a N-order tensor A ∈ RI1×I2×···×IP···×IN is [17] B. TENSOR TUCKER DECOMPOSITION PRINCIPLE This form is called the tensor Tucker decomposi- tion (TD). FIGURE 5. The tensor representation model of multi-station SCADA data. For example, a third-order tensor A admits a tensor decom- position and its Tucker decomposition result consists of a core tensor G and three left singular value matrices U(1), U(2), and U(3), as shown in Fig. 3. FIGURE 5. The tensor representation model of multi-station SCADA data. FIGURE 5. The tensor representation model of multi-station SCADA data. FIGURE 3. Third-order tensor tucker decomposition. B. TENSOR TUCKER DECOMPOSITION PRINCIPLE For the tensor A ∈RI1×I2×Ii×···×IN , according to (1), the ten- sor A i-mode unfolded matrix from each order A(i) is A(i) ∈RIi×(I1×···×Ii−1×Ii+1×···×IN ) (1) A(i) ∈RIi×(I1×···×Ii−1×Ii+1×···×IN ) (1) f (A, i) = A(i) ∈RIi×(I1×···×Ii−1×Ii+1×···×IN ) (4) where 1 ≤i ≤N, i ∈Z. f (A, i) = A(i) ∈RIi×(I1×···×Ii−1×Ii+1×···×IN ) (4) The equation for unfolding the tensor element A(x1, x2, · · · , xN) to the matrix element A(i)(xi, j) is The equation for unfolding the tensor element A(x1, x2, · · · , xN) to the matrix element A(i)(xi, j) is (4) where 1 ≤i ≤N, i ∈Z. j = 1 + XN k=1 k̸=i (xk −1)Jk (2) The singular value decomposition of i-mode unfolded matrix A(i) is where Jk = Qk−1 m=1 m̸=i Im svd(A(i)) = UiSiV T i (5) 124391 (5) 124391 VOLUME 7, 2019 VOLUME 7, 2019 H. Zhao et al.: Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor TD H. Zhao et al.: Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor TD where Si is the singular value matrix and Ui and Vi are left and right singular value matrices. and measurement. Therefore, the distribution system multi- station data can be represented by a third-order tensor A: According to (3), the left singular value matrix is multiplied with the original tensor A via tensor mode mul- tiplication and the core tensor G is A ∈RIt×In×Im (8) (8) where It, In and Im respectively represent the time, measure- ment sub-station and measurement of multi-station SCADA system data. G = A ×1 UT 1 ×2UT 2 ×3 . . . ×nUT n (6) (6) Compared with the traditional two-dimensional matrix representation method, the third-order tensor representation model can not only represent all the element values of SCADA data, but also retain the interaction among the time of SCADA data, measuring sub-stations and measurement, as shown in Fig. 5. According to the operations of multi-linear algebra, the tensor A can be obtained via the tensor mode multiplica- tion of the core tensor G and n left singular value matrices U(1), · · · , U(n). A = G ×1 U1 ×2 U2 ×3 . . . ×n Un (7) (7) In (7), an n-order tensor is expressed in the form of the mode product of a core tensor and n left singular value matrices. B. TUCKER TENSOR DECOMPOSITION COMPTRSSION OF POWER DISTRIBUTION SYSTEM MULTI-STATION SCADA DATA The tensor unfolding operation of the power distribution SCADA data tensor A ∈RIt×In×Im is performed by equa- tion (1), and the corresponding three unfolded matrices are as follows: FIGURE 3. Third-order tensor tucker decomposition. An ∈RIn×(It×Im) (9a) At ∈RIt×(In×Im) (9b) Am ∈RIm×(In×It) (9c) III. MULTI-STATION SCADA DATA COMPRESSION PRINCIPLE BASED ON TENSOR TUCKER DECOMPOSITION 0 0 0 0 0   (11a) 6It×InIm =   σIt,1 · · · · · · · · · 0 0 σIt,2 · · · · · · 0 ... ... ... ... ... 0 · · · · · · σIt,In 0 ... ... ... ... ... 0 0 0 0 0   (11b) 6Im×InIt =   σIm,1 · · · · · · · · · 0 0 σIm,2 · · · · · · 0 ... ... ... ... ... 0 · · · · · · σIm,In 0 ... ... ... ... ... 0 0 0 0 0   (11c) 6Im×InIt are the singular value matrix, which are expressed as as FIGURE 6. SCADA data tensor compression and reconstruction. original tensor and truncated orthogonal basis. Therefore, the order dimensions of the core tensor are reduced from IN, It and Im to P, Q and R. According to equation (4), the core tensor and truncated orthogonal basis can be used to reconstruct the SCADA data tensor ˆA: ˆA = G ×1 UIn×P ×2 UIt×Q×3UIm×R (14) (14) Generally speaking, the reconstructed data tensor ˆA is an approximate tensor of the original tensor. The schematic dia- gram of SCADA data tensor compression and reconstruction is shown in Fig. 7. where σIn,In, σIt,In, σIm,In denotes the positive singular values of the unfolded matrix, and σIn,1 ≥σIn,2 ≥· · · ≥σIn,In > 0, σIt,1 ≥σIt,2 ≥· · · ≥σIt,In > 0, σIm,1 ≥σIm,2 ≥· · · ≥ σIm,In > 0. FIGURE 7. The result of F-norm error ratio. The small singular values mainly represent the noise and data interdependence. In practice, small singular values make up a large proportion. Ignoring the small singular values can not only rapidly reduce the dimension of the singular value matrix but also yield a satisfactory data approximation. Therefore, when the singular values satisfy σ k ≫σ k+1, the singular values after σ s will be truncated. We assume that the unfolded of 6In×ItIm, 6It×InIm and 6Im×InIt are truncated at P, Q and R respectively, and the approximate singular values are as follows: AR(n) ≈UIN ×P6P×PV T ItIm×P (12a) AR(t) ≈UIt×Q6Q×QV T IN Im×Q (12b) AR(m) ≈UIm×R6R×RV T IN It×R (12c) FIGURE 7. The result of F-norm error ratio. When storing, we only need to store the core tensor and truncated orthogonal bases after Tucker decomposition. III. MULTI-STATION SCADA DATA COMPRESSION PRINCIPLE BASED ON TENSOR TUCKER DECOMPOSITION When using the original data, use the stored data to recon- struct the original tensor. where AR(n), AR(t), and AR(m) are singular value decompo- sition approximation matrices of unfolded matrices. UIN ×P, UIt×Q, and UIm×R are approximate left singular value matrixes. V T ItIm×PV T IN Im×Q, and V T IN It×R are approximate right singular value matrixes. IV. DATA COMPRESSION RATIO AND INFORMATION LOSS According to equation (5), tensor Tucker decomposition and compression of the SCADA data is as III. MULTI-STATION SCADA DATA COMPRESSION PRINCIPLE BASED ON TENSOR TUCKER DECOMPOSITION ( ) (9b) (9c) (9c) A. TENSORT REPRESENTATION MODEL OF POWER DISTRIBUTION SYSTEM MULTI-STATION SCADA DATA where An, At and Am are respectively the tensor unfolded matrix of the power distribution system SCADA data after staggered sampling by time, measuring sub-station and measuring. As shown in Fig. 4, the distribution management system of power distribution system, as the main station of SCADA sys- tems, continuously collects data information from n SCADA sub-stations. It is assumed that each SCADA system collects m measurements, and the collection time of each measure- ment is t time intervals. The collected SCADA data is trans- mitted to the main station of SCADA. Compared with the traditional two-dimensional matrix rep- resentation method, the advantage of this method is that ten- sor unfolding can realize the characteristic transfer and fusion of time, measurement sub-station and measurement orders. Therefore, the high-dimensional spatial feature structure of SCADA data is preserved. FIGURE 4. Hierarchical structure of power distribution SCADA system. The singular value decomposition transformation of the three unfolded matrices of tensor A is as An = UIn×In6In×ItImV T ItIm×ItIm (10a) At = UIt×It6It×InImV T InIm×IN Im (10b) Am = UIm×Im6Im×InItV T InIt×InIt (10c) (10a) (10b) (10c) (10b) (10c) ( ) (10c) (10c) FIGURE 4. Hierarchical structure of power distribution SCADA system. where An, At and Am are the unfolded matrices corresponding to tensor A. UIn, UIt and UIm are the corresponding left sin- gular value matrices; VItIm, VInIm and VInIt are corresponding right singular value matrices; And 6In×ItIm, 6It×InIm and Thus, the data collected in the main station of SCADA sys- tem has three characteristics: time, measurement sub-station 124392 Thus, the data collected in the main station of SCADA sys- tem has three characteristics: time, measurement sub-station 124392 VOLUME 7, 2019 DA Data Compression of Distribution Management System Based on Tensor TD H. Zhao et al.: Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor TD FIGURE 6. SCADA data tensor compression and reconstruction. 6Im×InIt are the singular value matrix, which are expressed as 6In×ItIm =   σIn,1 · · · · · · · · · 0 0 σIn,2 · · · · · · 0 ... ... ... ... ... 0 · · · · · · σIn,In 0 ... ... ... ... ... 2) MEAN ABSOLUTE PERCENT ERROR Fig. 7-Fig. 9 respectively shows the F-norm Error Ratio (FER), mean absolute percentage error (MAPE) and mean absolute error (MAE) when three methods compressed the test set data. Mean absolute percent error (MAPE) measures the data com- pression effect of multi-station SCADA data each measure- ment from the relative value. And the reconstruction mean absolute percent error of the k-th measurement of data tensor A is deﬁned as As can be seen from Fig. 7, when the compression ratio is less than about 2.5:1, the errors of the three methods are not much different. However, with the compression ratio increases, the errors of the PCA method and the SVD method increase exponentially and proportionally, respectively. The TD method compression error tends to be stable after slowly increasing. For example, when the maximum compression ratio of the PCA method is 25.66:1, the TD method error and the PCA method error are 0.2642 and 0.043, respectively. So, the TD method error is only 1/18 of the PCA method error; similarly, when the maximum compression ratio of SVD is 46.77:1, the TD method error is only 1/5 of the SVD method; in addition, when the compression ratio of the TD method is close to twice the maximum compression ratio of the SVD method(80:1), the error of the TD method is equal to error of the SVD method when the compression ratio is 15:1.Therefore, the TD method not only has a small error when the compression ratio is low, but is also more suitable for the high compression ratio than the other two methods. MAPE (k) = 1 It ∗In XIt j XIn i \|A(i, j, k) −ˆA(i, j, k) A(i, j, k) \| (17) (17) where A is the original tensor; ˆA is the approximation tensor. It is the time order. In is the sub-station order. B. COMPARATIVE ANALYSIS FER = \|A −ˆA\|F \|A\|F (16) (16) In literature [13] and [14], their authors show that SVD and PCA methods have good performance in SCADA data com- pression of the power distribution system. To demonstrate the superiority of the TD compression method, the proposed method is compared with the two data compression methods of SVD and PCA in this section. where \| · \|F is a function of the F-norm; A is the original tensor; ˆA is the approximation tensor. B. INFORMATION LOSS Three indicators of information loss level are shown as follows: A. COMPRESSION RATIO Data compression ratio(CR) is usually one of the impor- tant indexes to represent compression effect. In this paper, the compression ratio of data tensor A is deﬁned as G = A ×1 UT In×P ×2 UT It×Q×3UT Im×R (13) (13) where G is the compressed core tensor. CR = NOrg NRec = num(A) num(G) + Pn i=1 num(Ui) (15) 124393 CR = NOrg NRec = num(A) num(G) + Pn i=1 num(Ui) (15) It can be known from the equation (14) that the core tensor G ∈RP×Q×R is obtained by mode multiplication of the (15) VOLUME 7, 2019 H. Zhao et al.: Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor TD 3,136 measurements include three-phase current, three-phase voltage, active value, reactive value and power factor. where num(·) is a function of the number of elements; A is the original tensor; G is the core tensor; and Ui is the mode i truncated orthogonal basis. The measurement of the incoming line is three-phase current; the measurement of the transformer includes three- phase voltage and current, active value, reactive power and power factor. The 10kV bus measurement includes three- phase voltage, line voltage, zero sequence voltage; The 10kV reactive compensation capacitor measurement includes reac- tive value and three-phase current value; The measurement of load line includes active value, reactive value, three-phase current and load power factor. V. TEST RESULT The proposed data compression methodology has been implemented using MATLAB and evaluated on real measure- ment data from 10 kV distribution systems. The tests have been performed on a PC Intel core i5 processor at 3.20 GHz with 8 GB of RAM. The data description and the obtained results are presented next. 1) F-NORM ERROR RATIO F-norm error ratio(FER) represents the data compression effect from the global loss level, and the reconstruction F- norm Error Ratio of data tensor A is deﬁned as 3) MEAN ABSOLUTE ERROR Mean absolute error (MAE) measures the data compression effect of multi-station SCADA data each measurement from the absolute value. And the reconstruction mean absolute error of the k-th measurement of data tensor A is deﬁned as MAE (k) = 1 It ∗IN XIt j XIN i \|A(i, j, k) −ˆA(i, j, k)\| (18) (18) In the practical application, the constraint conditions of information loss index are ﬁrstly set to ensure the accuracy of compressed data. Then, on the premise of ensuring accuracy, the maximum compression ratio is selected to complete the data compression. It can be seen from Fig. 8 that the error curve of the SVD method is higher than the error curve of the TD method as a whole, indicating that the TD method is superior to the other two methods at the same compression ratio. In addition, SVD and PCA methods have different compression performance for different quantity measurements. For example, the error of SVD method for voltage measurement is smaller than that of the PCA method. However, the error of SVD for current measurement is larger than that of the PCA method. For the TD method, the performance for all measurements increases slowly and remains stable. REFERENCES [1] V. C. Gungor, D. Sahin, T. Kocak, S. Ergut, C. Buccella, C. Cecati, and G. P. Hancke, ‘‘A survey on smart grid potential applications and communication requirements,’’ IEEE Trans. Ind. Informat., vol. 9, no. 1, pp. 28–42, Feb. 2013. [2] D. Geng, C. Zhang, C. Xia, X. Xia, Q. Liu, and X. Fu, ‘‘Big data-based improved data acquisition and storage system for designing industrial data platform,’’ IEEE Access, vol. 7, pp. 44574–44582, 2019. [3] M. Ringwelski, C. Renner, A. Reinhardt, A. Weigel, and V. Turau, ‘‘The Hitchhiker’s guide to choosing the compression algorithm for your smart meter data,’’ in Proc. IEEE Int. Energy Conf. Exhibit. (ENERGYCON), Florence, Italy, Sep. 2012, pp. 935–940. FIGURE 9. The result of mean absolute error. [4] A. Unterweger and D. Engel, ‘‘Resumable load data compression in smart grids,’’ IEEE Trans. Smart Grid, vol. 6, no. 2, pp. 919–929, Mar. 2015. [5] N. C. F. Tse, J. Y. C. Chan, W.-H. Lau, J. T. Y. Poon, and L. L. Lai, ‘‘Real-time power-quality monitoring with hybrid sinusoidal and lifting wavelet compression algorithm,’’ IEEE Trans. Power Del., vol. 27, no. 4, pp. 1718–1726, Oct. 2012. power measurement, the error of the three methods is very small. However, the TD method is still less than the other two methods. [6] J. Ning, J. Wang, W. Gao, and C. Liu, ‘‘A wavelet-based data compres- sion technique for smart grid,’’ IEEE Trans. Smart Grid, vol. 2, no. 1, pp. 212–218, Mar. 2011. A. DATA DESCRIPTION It can be seen from Fig. 9 that the curve trend of the mean absolute error is substantially the same as the curve trend of the mean absolute percent error. In addition, for the reactive The adopted test data set comes from 32 10kV power distribu- tion stations in a region from 0 to 23.55 on March 1, 2018. The sampling frequency of SCADA data is 0.5 Hz. And a total of 124394 VOLUME 7, 2019 VOLUME 7, 2019 H. Zhao et al.: Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor TD FIGURE 8. The result of mean absolute percentage error. FIGURE 9. The result of mean absolute error. This method can effectively deal with the storage problems caused by the long-term accumulation of SCADA data in the power distribution system. Firstly, we establish a third-order tensor representation model for multi-station SCADA data in the data representation stage. This representation model can not only represent all the element values of SCADA data, but also retain the interaction among the time of SCADA data, measuring sub-stations and measurement. Then, the data compression method for power distribution system multi- station SCADA data based on Tucker decomposition is pro- posed. This method can reduce the amount of distribution system data while preserving the spatial intrinsic structure of high-dimensional data space to realize the compression of massive SCADA data. In addition, the compression ratio and three information loss level indexes are established, and the tradeoff between compression ratio and information loss is realized according to the actual compression requirements. Finally, the effectiveness of the method is veriﬁed using real data. And, the results of comparison with SVD and PCA methods show that the proposed method is superior to the traditional algorithm. FIGURE 8. The result of mean absolute percentage error. FIGURE 9. The result of mean absolute error. VI. TIME CONSUMPTION [7] F. Zhang, L. Cheng, X. Li, Y. Sun, W. Gao, and W. Zhao, ‘‘Application of a real-time data compression and adapted Protocol technique for WAMS,’’ IEEE Trans. Power Syst., vol. 30, no. 2, pp. 653–662, Mar. 2015. The tests have been performed on a PC Intel core i5 processor at 3.20 GHz with 8 GB of RAM. For the compression of 32 × 289 × 98 SCADA data, TD method, SVD method and PCA method took about 0.0107s, 0.0496s and 0.3725s respectively. The reconstruction time was 0.0078s, 0.05992s and 0.0586s, respectively. And, with the fast tensor Tucker decomposition calculation methods proposed, the compu- tational time continues to be reduced. For example, litera- ture [18]–[20] can greatly reduce the solving time through different solving methods. Therefore, the time consumption of the proposed method is very small. [8] M. Fayong, L. Qipeng, and M. Zhibin, ‘‘The research of historical data compression and storage strategy in power dispatch SCADA system,’’ Power Syst. Technol., vol. 38, no. 4, pp. 1109–1114, 2014. [9] J. C. S. de Souza, T. M. L. Assis, and B. C. Pal, ‘‘Data compression in smart distribution systems via singular value decomposition,’’ IEEE Trans. Smart Grid, vol. 8, no. 1, pp. 275–284, Jan. 2017. [10] S. Das and P. S. N. Rao, ‘‘Principal component analysis based compres- sion scheme for power system steady state operational data,’’ in Proc. ISGT2011-India, Kollam, India, Dec. 2011, pp. 95–100. [11] G.-Y. Cho, G.-Y. Lee, and T.-R. Lee, ‘‘Efﬁcient real-time lossless EMG Data transmission to monitor pre-term delivery in a medical information system,’’ Appl. Sci., vol. 7, no. 4, p. 366, Apr. 2017. [12] D. Wang, J. Zhou, K. He, C. Liu, and J. Xia, ‘‘Using tucker decomposition to compress color images,’’ in Proc. 2nd Int. Congr. Image Signal Process., Tianjin, China, Oct. 2009, pp. 1–5. VII. CONCLUSION This paper presents a data compression method for histor- ical multi-station SCADA data compression of distribution management system based on tensor Tucker decomposition. [13] S. Gao, N. Guo, M. Zhang, J. Chi, and C. Zhang, ‘‘Image denoising based on HOSVD with iterative-based adaptive hard threshold coefﬁcient shrinkage,’’ IEEE Access, vol. 7, pp. 13781–13790, 2019. VOLUME 7, 2019 124395 H. Zhao et al.: Historical Multi-Station SCADA Data Compression of Distribution Management System Based on Tensor TD [14] S. Holtz, T. Rohwedder, and R. Schneider, ‘‘The alternating linear scheme for tensor optimization in the tensor train format,’’ SIAM J. Sci. Comput., vol. 34, no. 2, pp. A683–A713, Jan. 2012. LIBO MA was born in Shijiazhuang, Hebei, China, in 1994. He is currently pursuing the master’s degree with the School of Electrical and Elec- tronic Engineering, North China Electric Power University (NCEPU). LIBO MA was born in Shijiazhuang, Hebei, China, in 1994. He is currently pursuing the master’s degree with the School of Electrical and Elec- tronic Engineering, North China Electric Power University (NCEPU). [15] F. van Belzen and S. Weiland, ‘‘A tensor decomposition approach to data compression and approximation of ND systems,’’ Multidimensional Syst. Signal Process., vol. 23, nos. 1–2, pp. 209–236, Jun. 2012. His research interests include analysis and con- trol of power systems and smart distribution system big data compression. His supervisor is Hongshan Zhao. [16] L. Kuang, F. Hao, L. T. Yang, M. Lin, C. Luo, and G. Min, ‘‘A tensor-based approach for big data representation and dimensionality reduction,’’ IEEE Trans. Emerg. Topics Comput., vol. 2, no. 3, pp. 280–291, Sep. 2014. [17] T. G. Kolda and B. W. Bader, ‘‘Tensor decompositions and applications,’’ SIAM Rev., vol. 51, no. 3, pp. 455–500, Aug. 2009. [18] C. A. Andersson and R. Bro, ‘‘Improving the speed of multi-way algo- rithms: Part I. Tucker3,’’ Chemometrics Intell. Lab. Syst., vol. 42, nos. 1–2, pp. 93–103, Aug. 1998. [19] L. De Lathauwer, B. De Moor, and J. Vandewalle, ‘‘On the best rank-1 and Rank-(R1, R2,. . .,RN ) approximation of higher-order tensors,’’ SIAM J. Matrix Anal. Appl., vol. 21, no. 4, pp. 1324–1342, Jan. 2000. XIHUI YAN received the bachelor’s degree from North China Electric Power University, Baoding, China, in 2017, where he is currently pursuing the master’s degree with the School of Electrical and Electronic Engineering. His research interests include demand response and non-intrusive load monitoring. [20] T. G. Kolda and J. VII. CONCLUSION Sun, ‘‘Scalable tensor decompositions for multi-aspect data mining,’’ in Proc. 8th IEEE Int. Conf. Data Mining, Pisa, Italy, Dec. 2008, pp. 363–372. HONGSHAN ZHAO was born in Cangzhou, Hebei, China, in 1965. He received the M.S. and Ph.D. degrees in power system and its automa- tion from North China Electric Power Univer- sity (NCEPU), Beijing, China, in 1996 and 2004, respectively. YANG ZHAO was born in Cangzhou, Hebei, China, in 1994. She received the bachelor’s degree from North China Electric Power Uni- versity, Baoding, China, in 2017, where she is currently pursuing the master’s degree with the School of Electrical and Electronic Engineering. Her research interests include condition monitor- ing and fault diagnosis of electrical equipment. Since 1989, he has been with the Department of Electrical Engineering, NCEPU, where he is cur- rently a Professor. His research interests include power system automation, fault diagnosis and opti- mal maintenance of wind turbine, dynamic analysis of hybrid power system, dynamic analysis of power system discrete events, and the early warning theory of power systems. 124396 VOLUME 7, 2019
https://openalex.org/W2045619663	https://link.springer.com/content/pdf/10.1007/s11277-013-1570-5.pdf	English	null	Iterative Overlap QRM-ML Block Detection for Single-Carrier MIMO Transmission Without CP Insertion	Wireless personal communications	2,014	cc-by	7,260	Wireless Pers Commun (2014) 74:1163–1177 DOI 10.1007/s11277-013-1570-5 Wireless Pers Commun (2014) 74:1163–1177 DOI 10.1007/s11277-013-1570-5 Iterative Overlap QRM-ML Block Detection for Single-Carrier MIMO Transmission Without CP Insertion Hideyuki Moroga · Tetsuya Yamamoto · Fumiyuki Adachi Published online: 7 January 2014 Published online: 7 January 2014 © The Author(s) 2014. This article is published with open access at Springerlink.com Abstract QR decomposition and M-algorithm based near maximum likelihood block detec- tion (QRM-MLBD) signiﬁcantly improves the single-carrier (SC) multiple-input multiple- output (SC-MIMO) transmission performance in a frequency-selective fading channel. In the conventional QRM-MLBD, the cyclic preﬁx (CP) is inserted in order to avoid the inter-block interference (IBI). However, CP insertion reduces the transmission efﬁciency. In this paper, an iterative overlap QRM-MLBD is proposed for SC-MIMO transmission with no CP inser- tion. It is conﬁrmed by computer simulation that the iterative overlap QRM-MLBD with no CP insertion provides improved throughput performance while reducing the computational complexity over the conventional QRM-MLBD with CP insertion. Keywords Single-carrier · MIMO · No cyclic preﬁx · QR decomposition · M-algorithm H. Moroga · T. Yamamoto (B) · F. Adachi Department of Communications Engineering, Graduate School of Engineering, Tohoku University, 6-6-05 Aza-Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan e-mail: yamamoto@mobile.ecei.tohoku.ac.jp H. Moroga e-mail: moroga@mobile.ecei.tohoku.ac.jp F. Adachi e-mail: adachi@ecei.tohoku.ac.jp 1 Introduction In the next generation mobile communication systems, broadband data services are demanded. Multiple-input multiple-output (MIMO) spatial multiplexing [1] provides broad- band data transmissions without increasing the signal bandwidth. For uplink (mobile terminal to base-station) application, single-carrier (SC) transmission is suitable because of its lower peak-to-average power ratio (PAPR) property [2,3] compared with multi-carrier transmis- sion, e.g., orthogonal frequency division multiplexing (OFDM) [4]. Thus, SC-MIMO has H. Moroga · T. Yamamoto (B) · F. Adachi Department of Communications Engineering, Graduate School of Engineering, Tohoku University, 6-6-05 Aza-Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan e-mail: yamamoto@mobile.ecei.tohoku.ac.jp H. Moroga e-mail: moroga@mobile.ecei.tohoku.ac.jp F. Adachi e-mail: adachi@ecei.tohoku.ac.jp H. Moroga · T. Yamamoto (B) · F. Adachi Department of Communications Engineering, Graduate School of Engineering, Tohoku University, 6-6-05 Aza-Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan e-mail: yamamoto@mobile.ecei.tohoku.ac.jp 123 1164 H. Moroga et al. been adopted for the uplink transmission in 3rd generation partnership project long term evolution-advanced (3GPP LTE-A) systems [5]. been adopted for the uplink transmission in 3rd generation partnership project long term evolution-advanced (3GPP LTE-A) systems [5]. y For broadband signal transmission, wireless channel is severely frequency-selective [6]. The broadband SC-MIMO spatial multiplexing suffers from inter-symbol interference (ISI) arising from the severe frequency-selectivity of the channel. The use of the cyclic preﬁx (CP) and frequency-domain block detection such as a computational efﬁcient minimum mean square error (MMSE) based linear detection [7] can improve the transmission performance ofSC-MIMOspatialmultiplexing.However,abigperformancegapfromthemaximumlikeli- hood (ML) performance still exists due to the presence of residual ISI and inter-antenna inter- ference(IAI).Recently,QRdecompositionandM-algorithmbasednear-maximumlikelihood block detection (QRM-MLBD) was proposed [8,9] for broadband SC-MIMO spatial mul- tiplexing. QRM-MLBD signiﬁcantly improves the transmission performance of SC-MIMO spatial multiplexing in a frequency-selective fading channel while signiﬁcantly reducing the computational complexity compared with ML detection. The conventional block detection schemes, such as MMSE based linear detection and QRM-MLBD,requiretheinsertionofCPtoavoidtheinter-blockinterference(IBI).However, CP insertion reduces the transmission efﬁciency. As a linear detection, overlap frequency- domain linear detection was proposed [10–12] in which the received symbol stream is divided into a sequence of blocks of X symbols each and then, frequency-domain block detection is applied to an extended block of Nc symbols centering the X-symbol block of interest (X ≤Nc). Note that Nc-symbol block is ﬁrst transformed by the discrete Fourier transform (DFT) into the frequency-domain signal. 123 1 Introduction Knowing that the residual IBI is signiﬁcant near both ends of Nc-symbol block after MMSE based linear detection, only X symbols are picked up in order to avoid the residual IBI. However, a big performance gap from the ML performance is observed due to insufﬁcient suppression of interferences, i.e., IAI, ISI, and also IBI. In [13,14], we presented the SC transmission using time-domain iterative overlap QRM- MLBD with no CP insertion for single-input single-output (SISO) systems. In overlap QRM- MLBD, the concept of overlap processing is applied to QRM-MLBD, in which the received symbol stream is divided into a sequence of blocks of X symbols each and then, QRM- MLBD is applied to an extended block of Nc + L −1 symbols to detect an Nc-symbol block including the X-symbol block of interest at the beginning, where L denotes the channel length in symbols (in this paper, the channel is assumed to be composed of symbol-spaced L propagation paths). The residual IBI is signiﬁcant near the end of Nc-symbol block after QRM-MLBD. Based on the above observation, only X symbols are picked up in order to avoid the residual IBI. To improve the IBI suppression, iterative processing and IBI cancellation are also introduced. Note that the proposed iterative overlap QRM-MLBD is implemented in the time-domain (no DFT is used in block detection). This is because time-domain overlap QRM- MLBD is equivalent to the frequency-domain overlap QRM-MLBD [13] and time-domain processing has an advantage in terms of the computational complexity. In this paper, we extend the previously proposed iterative overlap QRM-MLBD to SC-MIMO spatial multiplexing with no CP insertion. To extend our previously proposed algorithm to the MIMO systems, we introduce an appropriate modiﬁcation of the received signal vector for SC-MIMO spatial multiplexing. It is conﬁrmed by computer simulation that the iterative overlap QRM-MLBD with no CP insertion can achieve signiﬁcant performance improvement while reducing the computational complexity compared to the conventional QRM-MLBD with CP insertion. The rest of the paper is organized as follows. Section 2 describes the iterative overlap QRM-MLBD for SC-MIMO spatial multiplexing with no CP insertion. Simulation results are presented in Sect. 3. The achievable throughput performance with the iterative overlap 123 1165 Iterative Overlap QRM-ML Block Detection QRM-MLBD is compared with the conventional QRM-MLBD with CP insertion and the computational complexity of the iterative overlap QRM-MLBD is discussed as well. 1 Introduction Finally, we conclude the paper in Sect. 4. QRM-MLBD is compared with the conventional QRM-MLBD with CP insertion and the computational complexity of the iterative overlap QRM-MLBD is discussed as well. Finally, we conclude the paper in Sect. 4. 2.1 Transmission System Model Figure 1 illustrates the transmitter/receiver structure of SC-MIMO spatial multiplexing using the iterative overlap QRM-MLBD, where the numbers of transmit antennas and receive antennas are denoted by Nt and Nr, respectively. At the transmitter, the information bit sequence is transformed into a data-modulated symbol sequence and then, serial-to-parallel (S/P) converted to Nt parallel symbol sequences and each symbol sequence is transmitted from a different antenna. The transmitted symbol sequence of each transmit antenna propagates through different channel and received by Nr receive antennas at the receiver. The received symbol sequence on each receive antenna is divided into a sequence of blocks of X symbols each. To detect an Nc-symbol block including the X-symbol block of interest at the beginning, block signal processing is applied to an extended block of Nc + L −1 symbols (referred to as observation window). Iterative processing is applied to reduce the residual IBI. In the ith iteration stage, the replica of IBI from the previous block is generated by using the ith stage decision of the pre- vious block. The replica of IBI from the next block is generated by the (i −1)th stage decision of the next block. The IBIs are removed by subtracting their replicas from the received sym- bol sequence of interest over the observation window before applying QRM-MLBD. The received symbol sequence over the observation window at each receive antenna after IBI (a) (b) Fig. 1 Transmission system model. a Transmitter. b Receiver (a) (b) Fig. 1 Transmission system model. a Transmitter. b Receiver 123 H. Moroga et al. 1166 Received symbol sequence X symbols X symbols X symbols X symbols Nc L 1 symbols observation window IBI canceller + Generation of received signal vector + QRM-MLBD Pick up X symbols Detected symbol sequence time Detection of Nc-symbol X symbols X symbols X symbols X symbols Number I of iterations X symbols X symbols X symbols X symbols Nr antennas Detection of Nc-symbol Detection of Nc-symbol Detection of Nc-symbol Detection of Nc-symbol Detection of Nc-symbol X symbols X symbols X symbols X symbols Nr antennas Nt antennas + – Fig. 2 Iterative overlap QRM-MLBD X symbols Nc L 1 symbols observation window + – IBI canceller + Generation of received signal vector + QRM-MLBD Detection of Nc-symbol Pick up X symbols Fig. 2 Iterative overlap QRM-MLBD suppression is represented by a column vector of Nc + L −1 elements. 2.1 Transmission System Model Nr received symbol vectors are stacked to form a stacked received symbol vector. Then, by appropriately mod- ifying the stacked received symbol vector, QRM-MLBD is applied to detect Nt transmitted blocks of Nc symbols each. After QRM-MLBD, the ﬁrst X-symbol block is picked up for each transmit antenna from each detected block of Nc symbols. To detect the next X-symbol block, the observation window is shifted by X symbols as shown in Fig. 2. By repeating this process, the continuously transmitted symbol sequence from each transmit antenna is detected. The above overlap QRM-MLBD processing is repeated I times (I = 0 represents the initial iteration stage) to suppress the residual IBI sufﬁciently. 2.2 Received Signal Representation A frequency-selective fading channel is assumed to be composed of symbol-spaced L distinct propagation paths with different time delays. The channel impulse response between the ntth transmit antenna and the nrth receive antenna hnr,nt (τ) is expressed as hnr,nt (τ) = L−1 l=0 hnr,nt,lδ(τ −τnr,nt,l), (1) (1) where hnr,nt,l and τnr,nt,l are respectively the complex-valued path gain with E[\|hnr,nt,l\|2]=1 and the time delay of the lth path between the ntth transmit antenna and the nrth receive antenna. The received Nc+L−1-symbol sequence, ynr = [ynr (0), . . . , ynr (t), . . . , ynr (Nc+ Iterative Overlap QRM-ML Block Detection 1167 L −2)]T with (.)T denoting the transpose operation, over the observation window on the nrth receive antenna can be expressed using the matrix form as ynr = 2Es Ts Nt Nt−1 nt=0 hnr,nt dnt + 2Es Ts Nt Nt−1 nt=0 hnr,nt,−1dnt,−1 + 2Es Ts Nt Nt−1 nt=0 hnr,nt,+1dnt,+1 + nnr , (2) (2) where Es and Ts are respectively the symbol energy and duration. dnt = [dnt (0), . . . , dnt (t), . . . , dnt (Nc −1)]T represents the transmit symbol vector of interest from the ntth trans- mit antenna. dnt,−1(+1) = [dnt,−1(+1)(0), . . . , dnt,−1(+1)(t), . . . , dnt,−1(+1)(Nc −1)]T rep- resents the previous (next) transmit symbol vector. The ﬁrst term of (2) denotes the desired signal and the second and third terms denote the IBI from the previous and the next blocks, respectively. nnr = [nnr (0), . . . , nnr (t), . . . , nnr (Nc + L −2)]T is the zero-mean addi- tive white Gaussian noise (AWGN) vector with the one-sided power spectrum density N0. hnr,nt , hnr,nt,−1, and hnr,nt,+1 are respectively (Nc + L −1) × Nc channel impulse response matrixes between the ntth transmit antenna and the nrth receive antenna, given as ⎧ ⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨ ⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩ hnr ,nt = ⎡ ⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎣ hnr ,nt,0 0 ... ... hnr ,nt,L−1 ... ... hnr ,nt,0 ... ... 0 hnr ,nt,L−1 ⎤ ⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎦ hnr ,nt,−1 = ⎡ ⎢⎢⎢⎣ hnr ,nt,L−1 · · · hnr ,nt,1 ... ... hnr ,nt,L−1 0 ⎤ ⎥⎥⎥⎦, hnr ,nt,+1 = ⎡ ⎢⎢⎢⎢⎢⎣ 0 hnr ,nt,0 ... ... hnr ,nt,L−2 · · · hnr ,nt,0 ⎤ ⎥⎥⎥⎥⎥⎦ . (3) (3) 2.3 Iterative Overlap QRM-MLBD 2.3 Iterative Overlap QRM-MLBD 2.3 Iterative Overlap QRM-MLBD 2.3.1 Stacked Received Symbol Vector 2.3.1 Stacked Received Symbol Vector In the ith iteration stage, the IBI replica from the previous block is generated by using the decision of ˆd(i) nt,−1 = [ ˆd(i) nt,−1(0), . . . , ˆd(i) nt,−1(t), . . . , ˆd(i) nt,−1(Nc −1)]T , nt = 0 ∼Nt −1, of the previous block. When i ≥1, the IBI replica from the next block is also generated by using the decision of ˆd(i−1) nt,+1 = [ ˆd(i−1) nt,+1(0), . . . , ˆd(i−1) nt,+1(t), . . . , ˆd(i−1) nt,+1(Nc −1)]T , nt = 0 ∼Nt −1, of the next block. The IBI cancellation is performed by subtracting the IBI replicas from the received signal as ˜y(i) nr = ynr − 2Es Ts Nt Nt−1 n=0 hnr,nt,−1 ˆd(i) nt,−1 + 2Es Ts Nt Nt−1 n=0 hnr,nt,+1 ˆd (i−1) nt,+1 . (4) (4) 123 123 123 1168 H. Moroga et al. The received symbol vector after the IBI cancellation at each receive antenna is stacked to form an Nr(Nc + L −1)×1 stacked received symbol vector ˜Y(i) = [ {˜y(i) 0 }T · · · {˜y(i) Nr−1}T ]T as The received symbol vector after the IBI cancellation at each receive antenna is stacked to form an Nr(Nc + L −1)×1 stacked received symbol vector ˜Y(i) = [ {˜y(i) 0 }T · · · {˜y(i) Nr−1}T ]T as ˜Y(i) = [ {˜y(i) 0 }T · · · {˜y(i) Nr−1}T ]T = 2Es Ts Nt ⎡ ⎢⎣ h0,0 · · · h0,Nt−1 ... ... ... hNr−1,0 · · · hNr−1,Nt−1 ⎤ ⎥⎦ ⎡ ⎢⎣ d0 ... dNt−1 ⎤ ⎥⎦+ ⎡ ⎢⎣ n0 ... nNr−1 ⎤ ⎥⎦ + 2Es Ts Nt ⎡ ⎢⎣ h0,0,−1 · · · h0,Nt−1,−1 ... ... ... hNr−1,0,−1 · · · hNr−1,Nt−1,−1 ⎤ ⎥⎦ ⎡ ⎢⎢⎣ d0,−1 −ˆd(i) 0,−1 ... dNt−1,−1 −ˆd(i) Nt−1,−1 ⎤ ⎥⎥⎦ + 2Es Ts Nt ⎡ ⎢⎣ h0,0,+1 · · · h0,Nt−1,+1 ... ... ... hNr−1,0,+1 · · · hNr−1,Nt−1,+1 ⎤ ⎥⎦ ⎡ ⎢⎢⎣ d0,+1 −ˆd(i−1) 0,+1 ... dNt−1,+1 −ˆd(i−1) Nt−1,+1 ⎤ ⎥⎥⎦ = 2Es Ts Nt HD + 2Es Ts Nt H−1 D−1 −ˆD(i) −1 + 2Es Ts Nt H+1 D+1 −ˆD(i−1) +1 + N, (5) where N = [ {n0}T · · · {nNr−1}T ]T is the Nr(Nc + L −1) × 1 stacked noise vector. 2.3.1 Stacked Received Symbol Vector D = [ {d0}T · · · {dNt−1}T ]T , D−1 = [ {d0,−1}T · · · {dNt−1,−1}T ]T , and D+1 = [ {d0,+1}T · · · {dNt−1,+1}T ]T are the Nt Nc × 1 stacked transmit symbol vectors. H,H−1, and H+1 are equivalent channel matrixes of size Nr(Nc + L −1) × Nt Nc, which represent space and time-domain channel, given by ⎧ ⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨ ⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩ H= ⎡ ⎢⎣ h0,0 · · · h0,Nt−1 ... ... ... hNr −1,0 · · · hNr −1,Nt−1 ⎤ ⎥⎦ H−1 = ⎡ ⎢⎣ h0,0,−1 · · · h0,Nt−1,−1 ... ... ... hNr −1,0,−1 · · · hNr −1,Nt−1,−1 ⎤ ⎥⎦, H+1 = ⎡ ⎢⎣ h0,0,+1 · · · h0,Nt−1,+1 ... ... ... hNr −1,0,+1 · · · hNr −1,Nt−1,+1 ⎤ ⎥⎦ . (6) (6) The second and the third terms of (5) are the residual IBIs from the previous and the next blocks, respectively. The second and the third terms of (5) are the residual IBIs from the previous and the next blocks, respectively. 2.3.2 Modiﬁcation of the Stacked Received Symbol Vector Overlap QRM-MLBD for SISO systems [13,14] utilizes the property that the IBI from the next block, which cannot be removed in the initial iteration stage, exists only on the elements near the bottom of the received symbol vector. QRM-MLBD is applied to the received symbol vector and then, the residual IBI is signiﬁcant near the end of Nc-symbol block after QRM- MLBD. Therefore, symbol error rate near the beginning of the block is lower while symbol error rate near the end of the block is higher. Based on the above observation, overlap QRM- MLBD can effectively suppress the IBI by picking up only the reliable ﬁrst X-symbol block from the Nc-symbol block. 123 Iterative Overlap QRM-ML Block Detection 1169 However, in (5), the IBI from the next block exists on the elements near the bottom of received symbol vector at each receive antenna. Therefore, if QRM-MLBD is applied to (5) directly, the effect of IBI spreads over the all symbols in the entire block. To extend the previously proposed overlap QRM-MLBD to the MIMO systems, we modify the stacked received symbol vector as ˜Y′(i) = [{ ˜Y(i)(0)}T , . . . , { ˜Y(i)(t)}T , . . . , { ˜Y(i)(Nc −L −2)}T ]T , (7) ˜Y′(i) = [{ ˜Y(i)(0)}T , . . . , { ˜Y(i)(t)}T , . . . , { ˜Y(i)(Nc −L −2)}T ]T , (7) where ˜Y(i)(t) = [ ˜Y(i) 0 (t), . . . , ˜Y(i) Nr−1(t)]T denotes the tth Nr× 1 size received symbol vector after IBI cancellation. After the above modiﬁcation, the equivalent channel matrixes corresponding to the IBIs from the previous and the next blocks are permutated as ˜Y′(i) = [{ ˜Y(i)(0)}T , . . . , { ˜Y(i)(t)}T , . . . , { ˜Y(i)(Nc −L −2)}T ]T , (7) where ˜Y(i)(t) = [ ˜Y(i) 0 (t), . . . , ˜Y(i) Nr−1(t)]T denotes the tth Nr× 1 size received symbol vector after IBI cancellation. After the above modiﬁcation, the equivalent channel matrixes corresponding to the IBIs from the previous and the next blocks are permutated as (7) ⎧ ⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨ ⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩ H ′ −1 = ⎡ ⎢⎢⎢⎢⎢⎢⎢⎣ H′ 0,L−1 · · · H′ 0,1 ... ... H′ 0,L−1 0 H′ Nt−1,L−1 · · · H′ Nt−1,1 ... ... H′ Nt−1,L−1 0 ⎤ ⎥⎥⎥⎥⎥⎥⎥⎦ H ′ +1 = ⎡ ⎢⎢⎢⎢⎢⎢⎢⎣ 0 H′ 0,1 ... ... 2.3.2 Modiﬁcation of the Stacked Received Symbol Vector Therefore, the previously proposed overlap QRM-MLBD can be applied in a similar way to the MIMO system. 2.3.2 Modiﬁcation of the Stacked Received Symbol Vector H′ 0,L−2 · · · H′ 0,0 0 H′ Nt−1,0 ... ... H′ Nt−1,L−2 · · · H′ Nt−1,0 ⎤ ⎥⎥⎥⎥⎥⎥⎥⎦ , (8) (8) where H′nt,l = [h0,nt,l . . . , hNr−1,nt,l]T . It can be seen from (7) and (8) that the IBI from the next block exists only in the elements near the bottom of the modiﬁed stacked received symbol vector. where H′nt,l = [h0,nt,l . . . , hNr−1,nt,l]T . It can be seen from (7) and (8) that the IBI from the next block exists only in the elements near the bottom of the modiﬁed stacked received symbol vector. In QRM-MLBD, M-algorithm [15] is performed starting from the last symbol in the stacked transmit symbol vector. Since overlap QRM-MLBD outputs only X-symbol block for each transmit antenna which suffers less of the IBI from the next symbol block, the Nt Nc symbols in the stacked transmit symbol vector D is changed as D′ = [DT (Nc −1), . . . , DT (t), . . . , DT (0)]T , (9) (9) where D(t) = [dNt−1(t), . . . , d0(t)]T denotes the tth Nt×1 size transmit symbol vector. By the above modiﬁcation of the stacked received symbol vector and ordering of the stacked transmit symbol vector, the equivalent channel matrix for the desired signal component H is given as where D(t) = [dNt−1(t), . . . , d0(t)]T denotes the tth Nt×1 size transmit symbol vector. By the above modiﬁcation of the stacked received symbol vector and ordering of the stacked transmit symbol vector, the equivalent channel matrix for the desired signal component H is given as H ′′ = ⎡ ⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎣ 0 H′′ L−1 ... ... ... H′′ 0 H′′L−1 ... ... ... H′′0 0 ⎤ ⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎦ , (10) 1 3 (10) 123 1170 H. Moroga et al. where H′′ = ⎡ ⎢⎣ h0,Nt−1,l · · · h0,0,l ... ... ... hNr−1,Nt−1,l · · · hNr−1,0,l ⎤ ⎥⎦. (11) (11) It can be seen from (7), (10), and (11) that the modiﬁed stacked received symbol vector is similar to the SISO case [14] as shown in Fig. 3. Therefore, the previously proposed overlap QRM-MLBD can be applied in a similar way to the MIMO system. It can be seen from (7), (10), and (11) that the modiﬁed stacked received symbol vector is similar to the SISO case [14] as shown in Fig. 3. 2.3.3 QRM-MLBD QR decomposition is applied to the equivalent channel matrix H ′′ to obtain H ′′ = QR, where Q is an Nr(Nc + L −1) × Nt Nc unitary matrix and R is an Nt Nc × Nt Nc upper triangular matrix. Since the equivalent channel matrix for the desired signal component H ′′ is shown in (10), the unitary matrix Q and the upper triangular matrix R obtained by QR decomposition of H ′′ are represented as ⎧ ⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎨ ⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎪⎩ Q = ⎡ ⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎣ 0 Q0,Nt (Nc−1) · · · Q0,Nt Nc−1 ... ... ... QNt ,Nt (Nc−2) · · · QNt ,Nt (Nc−1) · · · QNt ,Nt Nc−1 ... ... QNt (Nc−1),0 ... ... ... QNt (Nc+L−2),0 · · · · · · · · · · · · · · · QNt (Nc+L−2),Nt Nc−1 ⎤ ⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎦ R = ⎡ ⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎢⎣ R0,0 · · · · · · · · · R0,Nt L−1 ... ... RNt −1,Nt −1 · · · RNt −1,Nt L−1 RNt ,Nt · · · · · · RNt −1,Nt (L+1)−1 ... ... ... RNt (Nc−L),Nt Nc−1 ... ... ... ... RNt Nc−1,Nt Nc−1 ⎤ ⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎥⎦ . (12) (12) The transformed received signal vector ˆY(i) is obtained as The transformed received signal vector ˆY(i) is obtained as The transformed received signal vector ˆY(i) is obtained as ˆY(i) = QH ˜Y′(i) = 2Es Ts Nt RD′ + 2Es Ts Nt ˆH−1 D−1 −ˆD(i) −1 + 2Es Ts Nt ˆH+1 D+1 −ˆD(i−1) +1 + ˆN, (13) ˆY(i) = QH ˜Y′(i) = 2Es Ts Nt RD′ + 2Es Ts Nt ˆH−1 D−1 −ˆD(i) −1 (13) 123 123 123 1171 Iterative Overlap QRM-ML Block Detection Fig. 3 Modiﬁed received signal vector (Nt = Nr = 2) where ˆH−1 = QHH ′ −1, ˆH+1 = QH ¯H′ +1, ˆN = QHN, and (.)H denotes the Hermitian transpose operation. Fig. 3 Modiﬁed received signal vector (Nt = Nr = 2) Fig. 3 Modiﬁed received signal vector (Nt = Nr = 2) where ˆH−1 = QHH ′ −1, ˆH+1 = QH ¯H′ +1, ˆN = QHN, and (.)H denotes the Hermitian transpose operation. where ˆH−1 = QHH ′ −1, ˆH+1 = QH ¯H′ +1, ˆN = QHN, and (.)H denotes the Hermitian transpose operation. 2.3.3 QRM-MLBD From (13), the ML solution is to select the path with the minimum Euclidean distance in a Nt Nc-stages tree diagram. This can be realized by M-algorithm. At each stage, the best M surviving paths selected from all the paths are passed to the next stage. The squared Euclidean distance is used for branch metric calculation. Data demodulation is carried out by tracing back the path having the smallest path metric at the last stage. In this paper, the stopping criterion [14] can be applied similarly to the SISO case to stop the tree search at an earlier stage to reduce the detection complexity. It can be seen from (12) and (13) that the branch metric beyond the (Nt L+n)th stage does not affect the detection of the ⌊n/Nt⌋th data symbol transmitted from the (n mod Nt)th transmit antenna, where ⌊x⌋represents the largest integer smaller than or equal to x. Therefore, the tree search using M-algorithm can be stopped at the (Nt(X + L)-1)th stage in order to output an X-symbol block for each transmit antenna. 2.3.4 IBI power distribution 16QAM Nt=2, Nr=2 Nc=64 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] Average received Es/N0 (dB) 5 10 15 20 25 30 35 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Throughput (bps/Hz) X=4 X=8 X=16 X=24 X=32 X=48 16QAM Nt=2, Nr=2 Nc=64 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] Average received Es/N0 (dB) 5 35 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Throughput (bps/Hz) X=24 X=32 X=48 (a) I=0 (b) I=1 10 15 20 25 30 Fig. 4 Throughput performance. a I = 0. b I = 1 16QAM Nt=2, Nr=2 Nc=64 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] Average received Es/N0 (dB) 5 35 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Throughput (bps/Hz) X=24 X=32 X=48 10 15 20 25 30 16QAM Nt=2, Nr=2 Nc=64 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] Average received Es/N0 (dB) 5 10 15 20 25 30 35 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Throughput (bps/Hz) X=4 X=8 X=16 X=24 X=32 X=48 Throughput (bps/Hz) (b) I=1 (b) Fig. 4 Throughput performance. a I = 0. b I = 1 to the beginning of the transformed signal vector while the IBI power from the next block is less signiﬁcant at an element near the end of the transformed signal vector. Hence, in the M-algorithm, the probability of erroneously removing the correct path is higher at last stages due to the stronger IBI while it is lower at early stages, resulting in higher error rate for the symbols near the end of the Nc-symbol block. Therefore, only the reliable X-symbol block at early stages is picked up from the detected block. to the beginning of the transformed signal vector while the IBI power from the next block is less signiﬁcant at an element near the end of the transformed signal vector. Hence, in the M-algorithm, the probability of erroneously removing the correct path is higher at last stages due to the stronger IBI while it is lower at early stages, resulting in higher error rate for the symbols near the end of the Nc-symbol block. Therefore, only the reliable X-symbol block at early stages is picked up from the detected block. 2.3.4 IBI power distribution Without CP insertion, the previous and the next blocks produce IBI. In the initial iteration stage (i = 0), IBI from the previous block can be suppressed by using the decision of the previous block as (5). However, IBI from the next block cannot be removed. Here, we consider the distribution of IBI from next block after multiplying QH [third term of (13)]. It can be seen from (12) that since QH is an unitary matrix, the absolute value of each element of QH is larger in upper row vector. Therefore, the absolute value of each element of ˆH+1 = QHH ′ +1 of (13) is larger in an upper row vector and is smaller in a lower row vector. As a result, the IBI power from the next block is more signiﬁcant at an element closer 123 1172 H. Moroga et al. Table 1 Computer simulation condition Transmitter Modulation 16QAM Number of transmit antennas Nt = 2 Number of symbols per packet Nt Np = 384 symbols GI length Ng = 16 Channel Fading type Frequency-selective block Rayleigh Power delay proﬁle L = 16-path uniform Time delay τl = l (l = 0 ∼L −1) Receiver Number of receive antennas Nr = 2 Channel estimation Ideal 16QAM Nt=2, Nr=2 Nc=64 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] Average received Es/N0 (dB) 5 10 15 20 25 30 35 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Throughput (bps/Hz) X=4 X=8 X=16 X=24 X=32 X=48 16QAM Nt=2, Nr=2 Nc=64 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] Average received Es/N0 (dB) 5 35 0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 Throughput (bps/Hz) X=24 X=32 X=48 (a) I=0 (b) I=1 10 15 20 25 30 Fig. 4 Throughput performance. a I = 0. b I = 1 to the beginning of the transformed signal vector while the IBI power from the next block is less signiﬁcant at an element near the end of the transformed signal vector. Hence, in the M-algorithm, the probability of erroneously removing the correct path is higher at last stages due to the stronger IBI while it is lower at early stages, resulting in higher error rate for the symbols near the end of the Nc-symbol block. Therefore, only the reliable X-symbol block at early stages is picked up from the detected block. 3 Computer Simulation Results The performance of SC-MIMO spatial multiplexing using iterative overlap QRM-MLBD is evaluated by computer simulation. The simulation condition is summarized in Table 1. 16QAM is used for data modulation. We assume Nt = 2, Nr = 2, and a frequency-selective quasi-static Rayleigh fading channel with an L = 16-path uniform power delay proﬁle. Ideal channel estimation is assumed. In this paper, SC-MIMO packet transmission is considered, where one packet is composed of 384 symbols. 123 1173 Iterative Overlap QRM-ML Block Detection 3.1 Throughput Performance Figure 4 plots the throughput performance as a function of average received Es/N0 for Nc = 64, X = 4 ∼48, I = 0 and 1, and M = 16; X is the number of symbols to be picked up and M is the number of surviving paths in the M-algorithm. In this paper, the throughput is deﬁned as Ntlog2Z × (1 −PER)/(1 + Ng/Nc), where Z is the modulation level and PER denotes the packet error rate. The throughput performance of the conventional QRM- MLBD with CP insertion is also plotted for comparison. The training sequence (TS) aided QRM-MLBD with TS length of 16 symbols [9] is used, which is similar to the conventional QRM-MLBD with CP insertion. It can be seen from Fig. 4 that iterative overlap QRM-MLBD improves the throughput performance when smaller X is used. This is because at early stages of M-algorithm, the IBI from the next block is less signiﬁcant. However, the use of smaller X increases the detection complexity. It can also be seen that iterative processing improves the throughput even if large X is used. With no iteration (I = 0), X = 4 should be used to improvethethroughputperformancesufﬁciently.However,when I = 1,amuchlarger X (e.g, X = 48) can be used. Since iterative overlap QRM-MLBD does not require the CP insertion, the peak throughput is higher than that of the conventional QRM-MLBD with CP insertion. 3.2 Computational Complexity 3.2 Computational Complexity The computational complexity of iterative overlap QRM-MLD is discussed. The compu- tational complexity here is deﬁned as the number of complex multiplications per symbol. The computational complexity for the iterative overlap QRM-MLBD and the conventional QRM-MLBD with CP insertion (TS-aided QRM-MLBD) are shown in Table 2. Time-domain signal processing is also implemented in TS-aided QRM-MLBD. The overall computational complexity per symbol is shown in Fig. 5 as a function of the number of iterations I for Nc = 64 and M = 16. As we mentioned in the previous section, X = 4 should be used to sufﬁciently improve the throughput performance with no iteration (I = 0) and X = 48 can be used when I = 1. Therefore, X is set to 4 and 48 for I = 0 and 1, respectively. When I = 2, X = 48 is also required. The overall computational complexity of the conventional QRM-MLBD with CP insertion is also plotted for reference. Since iter- ative processing improves the throughput even if large X is used, the overall computational complexity can be reduced even if iterative processing is introduced when Nc = 64. In the next, we discuss the relationship between the observation window size (Nc + L-1 symbols) and the overall computational complexity. Figure 6 plots the overall computational complexity as a function of Nc when the best combination of I and Xis used to achieve a throughput of 8bps/Hz at Es/N0 = 22dB. When a smaller Nc is used, larger I and smaller X are needed. Therefore, the computational complexity of the IBI cancellation, computation Table 2 Number of complex multiplications per symbol Iterative overlap QRM-MLBD TS-aided QRM-MLBD [9] IBI cancellation (I + 1)L(L −1)Nr/X QR decomposition Nr(Nc + L −1)(Nt Nc)2/Np Nr N2t (Nc + Ng)N2c /Np Computation of ˆY(i) (I + 1)Nr(Nc + L −1)Nc/X Nr(Nc + Ng) Path metric calc. (I + 1){(1 + M)Nt(X + L)Z + (1 −M)Z + M Nt(X + L) (Nt(X + L) −1)/2}/Nt X {(1 + M)Nt Nc Z + (1 −M)Z +M Nt Nc(Nt Nc−1)/2}/Nt Nc Table 2 Number of complex multiplications per symbol 1174 H. Moroga et al. g Fig. 5 Impact of I on the overall computational complexity 16QAM Nt=2, Nr=2 Nc=64 M=16 L=16 path uniform Conv. 3.2 Computational Complexity QRM-MLBD with CP insertion [9] Number I of iterations 0 1 2 105 Number of complex multiplications per symbol Iterative overlap QRM-MLBD 106 X=4 X=48 X=48 104 Fig. 6 Impact of Nc on the overall computational complexity 16QAM Nt=2, Nr=2 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] (Nc=64) Nc 16 103 Number of complex multiplications per symbol 106 24 32 40 48 56 64 104 105 Path metric calc. QR decomposition Total of ˆY(i) as well as path metric computation increases (because of larger I and smaller X). On the other hand, since the size of the channel matrix is small (the size of equivalent channel matrix is Nr(Nc + L −1) × Nt Nc), the computational complexity of QR decomposition reduces. It is understood from Fig. 6 that the overall computational complexity to achieve a peak throughput of 8bps/Hz is lowest when Nc = 28 and is about 50% of the conventional QRM-MLBD (Nc = 64) with CP insertion. 4 Conclusion 16QAM Nt=2, Nr=2 Nc=64 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] Number I of iterations 0 1 2 105 Number of complex multiplications per symbol Iterative overlap QRM-MLBD 106 X=4 X=48 X=48 104 Conv. QRM-MLBD with CP insertion [9] Number I of iterations Fig. 6 Impact of Nc on the overall computational complexity 16QAM Nt=2, Nr=2 M=16 L=16 path uniform Conv. QRM-MLBD with CP insertion [9] (Nc=64) Nc 16 103 Number of complex multiplications per symbol 106 24 32 40 48 56 64 104 105 Path metric calc. QR decomposition Total Conv. QRM-MLBD with CP insertion [9] ( 64) of ˆY(i) as well as path metric computation increases (because of larger I and smaller X). On the other hand, since the size of the channel matrix is small (the size of equivalent channel matrix is Nr(Nc + L −1) × Nt Nc), the computational complexity of QR decomposition reduces. It is understood from Fig. 6 that the overall computational complexity to achieve a peak throughput of 8bps/Hz is lowest when Nc = 28 and is about 50% of the conventional QRM-MLBD (Nc = 64) with CP insertion. References 1. Foschini, G. J., & Gans, M. J. (1998). On limits of wireless communications in a fading environment when using multiple antennas. Wireless Personal Communications, 6(3), 311–335. 2. Ekstrom, H., Furuskar, A., Karlsson, J., Meyer, M., Parkvall, S., Torsner, J., et al. (2006). Technical solutions for the 3G long-term evolution. IEEE Communications Magazine, 44(3), 38–45. 3. Benjamin, N., Chan-Tong, L., & Falconer, D. (2007). Turbo frequency domain equalization for sing carrier broadband wireless systems. EEE Transactions on Wireless Communications, 6(2), 759–767. 4. Higuchi, K., Kawai, H., Maeda, N., Taoka, H., & Sawahashi, M. (2006). Experiments on real-time 1-Gb/s packet transmission using MLD-based signal detection in MIMO-OFDM broadband radio access. IEEE Journal on Selected Areas Communications, 24(6), 1141–1153. 5. 3GPP, TS 36.211 (V10.4.0) (2011). 3rd Generation partnership project (3GPP); evolved universal restrial radio access (E-UTRA); physical channels and modulation (release 10). 6. Proakis, J. G., & Salehi, M. (2008). Digital communications (5th ed.). New York: McGraw-Hill. 7. Nakajima, A., Garg, D., & Adachi, F. (2005). Throughput of turbo coded hybrid ARQ using single-carrier MIMO multiplexing. In Proceedings of IEEE 61st vehicular technology conference (VTC2005-Spring) (Vol. 1, pp. 610–614). Stockholm, Sweden, 30 May–1 June. 8. Nagatomi, K., Higuchi, K., & Kawai, H. (2009). Complexity reduced MLD based on QR decomposition in OFDM MIMO multiplexing with frequency domain spreading and code multiplexing. In Proceedings of IEEE wireless communications and networking conference (WCNC 2009). Budapest, Hungary, 5–8 Apr. 9. Yamamoto, T., Takeda, K., & Adachi, F. (2011). Training sequence-aided QRM-MLD block signal detec- tion for single-carrier MIMO spatial multiplexing. In Proceedings of 2011 IEEE international conference on communications (ICC 2011). Kyoto, Japan, 5–9 June. 10. Martoyo, I., Weiss, T., Capar, F., & Jondral, F. K. (2003). Low complexity CDMA downlink receiver based on frequency domain equalization. In Proceedings of IEEE 58th vehicular technology conference (VTC2003-Fall). Orlando, Florida, USA, 6–9 Sept. 11. Obara, T., Takeda, K., Lee, K., & Adachi, F. (2011). Performance comparison of overlap FDE and sliding-window chip equalization for multi-code DS-CDMA in a frequency-selective fading channel. IEICE Transactions on Communications, E94–B(3), 750–757. 12. Ishihara, K., Takatori, Y., Kubota, S., & Adachi, F. (2009). Multiuser detection for asynchronous bro band single-carrier transmission systems. IEEE Transactions on Vehicular Technology, 58(6), 3055–30 13. Moroga, H., Yamamoto, T., & Adachi, F. (2012). Overlap QRM-ML block signal detection for single- carrier transmission without CP insertion. In Proceedings of IEEE 75th vehicular tecnnology conference (VTC2012-Spring). 4 Conclusion In this paper, we proposed a time-domain iterative overlap QRM-MLBD which requires no CP insertion for SC-MIMO spatial multiplexing. To extend our previously proposed iterative Iterative Overlap QRM-ML Block Detection 1175 overlap QRM-MLBD to the MIMO systems, we introduce an appropriate modiﬁcation of the stacked received symbol vector of SC-MIMO spatial multiplexing. Remembering that the IBI is signiﬁcant near the bottom of the elements in the modiﬁed stacked received symbol vector and the residual IBI is signiﬁcant near the end of the detected symbol block after QRM-MLBD, only the reliable X symbols are picked up after performing QRM-MLBD. To detect a continuously transmitted symbol stream, the present observation window for performing QRM-MLBD is overlapped with the previous and the next observation windows. To further improve the detection performance, iterative processing was introduced. It has been showed that the proposed iterative overlap QRM-MLBD improves the throughput by 25% than the conventional QRM-MLBD with CP insertion (Nc = 64, Ng = 16). The overall computational complexity can be reduced to 50% when compared with the conventional QRM-MLBD with CP insertion (Nc = 64). overlap QRM-MLBD to the MIMO systems, we introduce an appropriate modiﬁcation of the stacked received symbol vector of SC-MIMO spatial multiplexing. Remembering that the IBI is signiﬁcant near the bottom of the elements in the modiﬁed stacked received symbol vector and the residual IBI is signiﬁcant near the end of the detected symbol block after QRM-MLBD, only the reliable X symbols are picked up after performing QRM-MLBD. To detect a continuously transmitted symbol stream, the present observation window for performing QRM-MLBD is overlapped with the previous and the next observation windows. To further improve the detection performance, iterative processing was introduced. It has been showed that the proposed iterative overlap QRM-MLBD improves the throughput by 25% than the conventional QRM-MLBD with CP insertion (Nc = 64, Ng = 16). The overall computational complexity can be reduced to 50% when compared with the conventional QRM-MLBD with CP insertion (Nc = 64). Open Access This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited. 14. Moroga, H., Yamamoto, T., & Adachi, F. (2012). Iterative overlap TD-QRM-ML block signal detection for single-carrier transmission without CP insertion. In Proceedings of IEEE 76th vehicular technology conference (VTC2012-Fall). Québec City, Canada, 3–6 Sept. Tetsuya Yamamoto received his B.S. degree in Electrical, Informa- tion and Physics Engineering in 2008 and M.S. and Dr. Eng. degrees in communications engineering from Tohoku University, Sendai Japan, in 2010 and 2012, respectively. From April 2010 to March 2013, he was a Japan Society for the Promotion of Science (JSPS) research fel- low. Since April 2013, he has been with Panasonic Corporation. He was a recipient of the 2008 IEICE RCS (Radio Communication Systems) Active Research Award and Ericsson Best Student Award 2012. References Yokohama, Japan, 6–9 May. 123 1176 H. Moroga et al. 14. Moroga, H., Yamamoto, T., & Adachi, F. (2012). Iterative overlap TD-QRM-ML block signal detection for single-carrier transmission without CP insertion. In Proceedings of IEEE 76th vehicular technology conference (VTC2012-Fall). Québec City, Canada, 3–6 Sept. 14. Moroga, H., Yamamoto, T., & Adachi, F. (2012). Iterative overlap TD-QRM-ML block signal detection for single-carrier transmission without CP insertion. In Proceedings of IEEE 76th vehicular technology conference (VTC2012-Fall). Québec City, Canada, 3–6 Sept. 15. Anderson, J. B., & Mohan, S. (1984). Sequential coding algorithms: A survey and cost analysis. IEEE Transactions on Communications, 32(2), 169–176. 15. Anderson, J. B., & Mohan, S. (1984). Sequential coding algorithms: A survey and cost analysis. IEEE Transactions on Communications, 32(2), 169–176. Hideyuki Moroga received his B.S. degree in Information System Engineering from Kanazawa University, Kanazawa, Japan, in 2010 and M.S. degree in communications engineering, from Tohoku Univer- sity, Sendai, Japan, in 2012. Currently he is with NTT DoCoMo. His research interests include time-domain equalization and signal detec- tion techniques for mobile communication systems. Hideyuki Moroga received his B.S. degree in Information System Engineering from Kanazawa University, Kanazawa, Japan, in 2010 and M.S. degree in communications engineering, from Tohoku Univer- sity, Sendai, Japan, in 2012. Currently he is with NTT DoCoMo. His research interests include time-domain equalization and signal detec- tion techniques for mobile communication systems. Tetsuya Yamamoto received his B.S. degree in Electrical, Informa- tion and Physics Engineering in 2008 and M.S. and Dr. Eng. degrees in communications engineering from Tohoku University, Sendai Japan, in 2010 and 2012, respectively. From April 2010 to March 2013, he was a Japan Society for the Promotion of Science (JSPS) research fel- low. Since April 2013, he has been with Panasonic Corporation. He was a recipient of the 2008 IEICE RCS (Radio Communication Systems) Active Research Award and Ericsson Best Student Award 2012. 123 1177 Iterative Overlap QRM-ML Block Detection Fumiyuki Adachi received the B.S. and Dr. Eng. degrees in electri- cal engineering from Tohoku University, Sendai, Japan, in 1973 and 1984, respectively. In April 1973, he joined the Electrical Communi- cations Laboratories of Nippon Telegraph & Telephone Corporation (now NTT) and conducted various types of research related to digital cellular mobile communications. From July 1992 to December 1999, he was with NTT Mobile Communications Network, Inc. References (now NTT DoCoMo, Inc.), where he led a research group on wideband/broadband CDMA wireless access for IMT-2000 and beyond. Since January 2000, he has been with Tohoku University, Sendai, Japan, where he is a Professor of Communications Engineering at the Graduate School of Engineering. In 2011, he was appointed a Distinguished Professor. His research interest is in the areas of wireless signal processing and networking including broadband wireless access, equalization, trans- mit/receive antenna diversity, MIMO, adaptive transmission, channel coding, etc. From October 1984 to September 1985, he was a United Kingdom SERC Visiting Research Fellow in the Department of Electrical Engineering and Electronics at Liverpool University. Dr. Adachi is an IEEE Fellow and a VTS Distinguished Lecturer for 2011 to 2013. He was a co-recipient of the IEEE Vehicular Technology Transactions Best Paper of the Year Award 1980 and again 1990 and also a recipient of Avant Garde award 2000. He is a Fellow of Institute of Electronics, Infor- mation and Communication Engineers of Japan (IEICE) and is a recipient of IEICE Achievement Award 2002 and a co-recipient of the IEICE Transactions Best Paper of the Year Award 1996, 1998 and again 2009. He is a recipient of Thomson Scientiﬁc Research Front Award 2004, Ericsson Telecommunications Award 2008, Telecom System Technology Award 2009, and Prime Minister Invention Prize 2010, British Royal Academy of Engineering Distinguished Visiting Fellowship 2011, and KDDI Foundation Research Award 2012. 123
W3186896710.txt	null	de		Einsätze für eine Genealogie des erwachsenenpädagogischen Blicks	Debatte. Beiträge zur Erwachsenenbildung	2,019	cc-by-sa	7,395	Klingovsky, Ulla Einsätze für eine Genealogie des erwachsenenpädagogischen Blicks Debatte : Beiträge zur Erwachsenenbildung 2 (2019) 1, S. 5-22 Quellenangabe/ Reference: Klingovsky, Ulla: Einsätze für eine Genealogie des erwachsenenpädagogischen Blicks - In: Debatte : Beiträge zur Erwachsenenbildung 2 (2019) 1, S. 5-22 - URN: urn:nbn:de:0111-pedocs-227906 - DOI: 10.25656/01:22790 https://nbn-resolving.org/urn:nbn:de:0111-pedocs-227906 https://doi.org/10.25656/01:22790 in Kooperation mit / in cooperation with: https://www.budrich.de Nutzungsbedingungen Terms of use Dieses Dokument steht unter folgender Creative Commons-Lizenz: http://creativecommons.org/licenses/by-sa/4.0/deed.de Sie dürfen das Werk bzw. den Inhalt vervielfältigen, verbreiten und öffentlich zugänglich machen sowie Abwandlungen und Bearbeitungen des Werkes bzw. Inhaltes anfertigen, solange sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen und die daraufhin neu entstandenen Werke bzw. Inhalte nur unter Verwendung von Lizenzbedingungen weitergeben, die mit denen dieses Lizenzvertrags identisch, vergleichbar oder kompatibel sind. Mit der Verwendung dieses Dokuments erkennen Sie die Nutzungsbedingungen an. This document is published under following Creative Commons-License: http://creativecommons.org/licenses/by-sa/4.0/deed.en - You may copy, distribute and transmit, adapt or exhibit the work or its contents in public and alter, transform, or change this work as long as you attribute the work in the manner specified by the author or licensor. New resulting works or contents must be distributed pursuant to this license or an identical or comparable license. By using this particular document, you accept the above-stated conditions of use. Kontakt / Contact: peDOCS DIPF \| Leibniz-Institut für Bildungsforschung und Bildungsinformation Informationszentrum (IZ) Bildung E-Mail: pedocs@dipf.de Internet: www.pedocs.de Debatte 2019 · Jg. 2 · H. 1 Debatte Beiträge zur Erwachsenenbildung Erwachsenenbildung: Kategorial stillgelegt? Repliken zu vorherigen Beiträgen Disziplinäre Unbestimmtheit Andreas Seiverth Lernen als historischer Begriff Einsätze für eine Genealogie des erwachsenenpädagogischen Blicks Ulla Klingovsky Sebastian Manhart Repliken zum Themenbeitrag Geschichtsschreibung der Erwachsenen- und Weiterbildungsforschung Neoliberale Subjektivität. Sei du! Wiltrud Gieseke & Bernd Käpplinger Offenheit als Stärke von Erwachsenenbildung Vom stummen Sprechen und lauten Schweigen Sebastian Lerch Katharina Herrmann „Genealogie des erwachsenenpädagogischen Blicks“ durch eine relationslogische Optik geschärft Stefan Vater Ortfried Schäffter Debatte. Beiträge zur Erwachsenenbildung · 2019 · Jg. 2 · H. 1 Herausgeber*innen- & Redaktionskollektiv Erscheinen & Bezugsbedingungen Christoph Damm (Otto-von-Guericke-Universität Magdeburg), Malte Ebner von Eschenbach (Martin-Luther-Universität Halle-Wittenberg), Maria Kondratjuk (Martin-Luther-Universität Halle-Wittenberg), Hannah Rosenberg (Universität Koblenz-Landau), Maria Stimm (Humboldt-Universität zu Berlin), Jana Trumann (Universität Duisburg-Essen), Farina Wagner (Martin-Luther-Universität Halle-Wittenberg) Die Zeitschrift erscheint 2 x jährlich mit einem Jahresumfang von ca. 200 Seiten (Print und Online). Privat: Kombi-Abonnement Print+Online – 69,90 €, Kombi-Abonnement Print+Online Studierende – 49,90 €, Online-Only-Abonnement – 69,90 €, OnlineOnly-Abonnement Studierende – 49,90 €, Print-Abonnement – 59,90 €, Print-Abonnement Studierende – 39,90 €; Institutionen: Print-Abonnement – 79,00 €, OnlineOnly-Abonnement – 119,90 €, Kombi-Abonnement Print+Online – 119,90 €, Einzelheft Print – 29,90 €, Einzelheft PDF – 29,90 €, Einzelbeitrag (PDF) im Download – 6,00 € Alle Preise inkl. MwSt. und zzgl. Versandkosten (pro Jahr): Inland: 4,00 €, Ausland: 8,00 €. Abonnements verlängern sich automatisch um ein Jahr. Abonnement-Kündigungen bitte schriftlich an den Verlag. Kündigungsfrist bis drei Monate zum Jahresende. Mit einem Kombi-Abonnement Print+Online haben Sie zusätzlich zu den PrintHeften freien Online-Zugriff auf das gesamte Volltext-Archiv der Zeitschrift, solange Ihr Abonnement besteht. Für ein ermäßigtes Abonnement ist eine Studienbescheinigung als PDF erforderlich. Das digitale Angebot finden Sie auf: debatte.budrich-journals.de. Bestellungen bitte an den Buchhandel oder an: Verlag Barbara Budrich, Stauffenbergstr. 7, D-51379 Leverkusen-Opladen Tel.: (+49) (0)2171 79491 50; Fax: (+49) (0)2171 79491 69; info@budrich.de www.budrich.de • www.budrich-journals.de Aktuelle Mediadaten/Anzeigenpreisliste: DEBATTE919 vom 01.02.2019 Lektorat Stephanie Freide, Berlin Design Svenja Klau, Berlin © 2019 Verlag Barbara Budrich Opladen • Berlin • Toronto Die Zeitschrift sowie alle in ihr enthaltenen Beiträge sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung des Verlags. Dies gilt insbesondere für Vervielfältigungen, Übersetzungen, Mikroverfilmungen und die Einspeicherung und Verarbeitung in elektronischen Systemen. Namentlich gekennzeichnete Beiträge geben nicht in jedem Fall die Meinung der Redaktion wieder. ISSN 2567-5966 Online-ISSN 2567-5974 Debatte. Beiträge zur Erwachsenenbildung Debatte 2019 · Jg. 2 · H. 1 Heftthema: Erwachsenenbildung: Kategorial stillgelegt? Einsätze für eine Genealogie des erwachsenenpädagogischen Blicks Ulla Klingovsky 3 Editorial 5 Repliken zum Themenbeitrag 23 33 Redaktionskollektiv Neoliberale Subjektivität. Sei du! Stefan Vater Offenheit als Stärke von Erwachsenenbildung Sebastian Lerch 40 „Genealogie des erwachsenenpädagogischen Blicks“ durch eine relationslogische Optik geschärft. Eine Replik auf Ulla Klingovsky Ortfried Schäffter Diszipliniert und disziplinierend? Anschlüsse an die Debatte um das Selbstverständnis der Erwachsenenbildungswissenschaft Hannah Rosenberg 64 65 Repliken zu vorherigen Beiträgen Disziplinäre Unbestimmtheit. Anregungen für eine veränderte Perspektive auf das Konstitutionsdilemma der Erwachsenenbildung als wissenschaftliche Disziplin Andreas Seiverth Historische Erwachsenenbildungsforschung: Geprägt von Kontroversität und Ambivalenz Christine Zeuner 82 83 Lernen als historischer Begriff. Zur historischen Kompetenz nicht nur in der Erwachsenenbildungsforschung Sebastian Manhart 97 Geschichtsschreibung der Erwachsenen- und Weiterbildungsforschung – Ein Überblick mit Reflexionen für den wissenschaftlichen Nachwuchs Wiltrud Gieseke & Bernd Käpplinger sprechen. schweigen. ignorieren. Echokammer-Effekte, Machtmanifestationen und Schweigespiralen in Debattenunkulturen Daniela Holzer 114 115 Vom stummen Sprechen und lauten Schweigen – ein Plädoyer für eine pädagogische Politisierung der Debattenunkultur Katharina Herrmann 128 Call for Replies Debatte 2019 · Jg. 2 · H. 1 · 5-22 · https://doi.org/10.3224/debatte.v2i1.02 Einsätze für eine Genealogie des erwachsenenpädagogischen Blicks Ulla Klingovsky Zusammenfassung In diesem Beitrag werden erste Sondierungen für eine Genealogie des erwachsenenpädagogischen Blicks vorgenommen und die These entfaltet, wonach in der Erwachsenenbildung eine systematische Verständigung über die Frage der ,Bildung‘ in der Erwachsenenbildung aussteht. Unter Rückgriff auf den jüngeren bildungstheoretischen Diskurs wird Bildung als Problematisierungsformel vorgestellt, die sich weder bei den funktionalen Entleerungen noch bei den normativen Vereindeutungen von Bildung beruhigt, sondern die Frage nach der Bildung in Erwachsenenbildung offen zu halten sucht. Erwachsenenbildung · Bildungstheorie · Genealogie · Kritik · Versprechen der Bildung abstract This contribution explores possibilities of a genealogy from the perspective of adult education. It develops the thesis that adult education lacks a systematic understanding of, and agreement on the question of ‘education’ in adult education. Based on the recent discourse in educational theories, the author suggests introducing education as formula of problematization, which resolves in neither a functional emptying, nor a normative disambiguation of education. The aim of such an approach is to keep the meaning of education in adult education in suspense and open to further interrogation. adult education · theories of education (Bildung) · genealogy · critique/criticism · promise of education Debatte Beiträge zur Erwachsenenbildung 5 Einsätze für eine Genealogie des erwachsenenpädagogischen Blicks Ulla Klingovsky Die Genealogie als kritisches Programm der Sozial- und Kulturwissenschaften untersucht gesellschaftliche Lebensformen oder diskursive Formationen, kurz: soziale Praktiken der Gegenwart auf historische Weise. Insofern die genealogische Analyse1 bestehende Überzeugungen und Perspektiven im Rückgriff auf historische Entwicklungslinien distanziert, eröffnet sie auch der Erwachsenenbildung einen problematisierenden Blick. Dieser ‚erwachsenenpädagogische Blick‘ vermag Ordnungsstrukturen zu distanzieren, welche die Disziplin in ihr Wahrnehmungsfeld bringt. Ein wesentliches Kennzeichen dieser Ordnung ist der Begriff der ‚Bildung‘, den die Erwachsenenbildung konstitutiv in ihrem Namen trägt. Obwohl mit ordnungsbildender Funktion versehen, wird der Begriff der ‚Bildung‘ in der Erwachsenenbildung in der Regel allerdings nicht weiter thematisiert. Die Bewegung, die durch die Begriffe ‚Genealogie‘ und ‚Blick‘ im Titel dieses Beitrags markiert werden soll, kreist um die Frage, was es mit dieser ‚Bildung‘ in Erwachsenenbildung eigentlich auf sich hat. Die Bewegung bezieht sich dabei nun weder auf die „Suche nach dem Ursprung“ (Foucault 1978a, S. 84) dessen, was der ‚Bildung‘ in Erwachsenen- bildung genuin ist oder sein sollte, noch zielt sie auf die Ausgrabung normativer Kriterien, von denen aus sich ‚das Erwachsenenpädagogische‘ bestimmen ließe. Mit den in diesem Beitrag entwickelten Einsätzen für eine genealogische Analyse ist vielmehr die kritische Absicht verbunden, der Stellung der Bildung in Erwachsenenbildung analytisch auf die Spur zu kommen, um bestimmte Problemlagen sichtbar resp. kenntlich zu machen und auf diese Weise den erwachsenenpädagogischen Blick auf gegenwärtige Entwicklungen zu schärfen. 1. Erste Sondierungen Ausgangspunkt der Sondierungen für eine Genealogie des erwachsenenpädagogischen Blicks ist die spannungsreiche Beobachtung, wonach sich die Erwachsenenbildung gegenwärtig in einer eigentümlichen Lage befindet: Auf der einen Seite vergeht kaum ein Tag, an dem in öffentlichen Debatten nicht auf die immense Bedeutung von ‚Bildung‘ hingewiesen wird. Gegenwärtig scheint kaum etwas bedeutsamer als ‚Bildung‘ – und das ein Leben lang. Die Erwachsenenbildung erfährt in einer sich zunehmend rasch wandelnden Gesellschaft, in der die Menschen auf eine dauerhafte Weiterqualifizierung angewiesen zu sein scheinen, einen enormen Bedeutungsaufschwung. 1 Im Anschluss an Nietzsches Ausführungen Zur Genealogie der Moral (1954) und Foucaults Aufsatz Nietzsche, die Genealogie, die Historie ist es im deutschsprachigen Raum v. a. Martin Saar, der die Entwicklung der Genealogie als kritische Methode unter dem Titel Genealogie als Kritik (2007) rekonstruiert hat. Ebenfalls umfassendere Arbeiten dazu bilden jene von Michael Mahon Foucault’s Nietzschean Genealogy (1992) und Brian Lightbody Philosophical Genealogy (2010). 6 Debatte Beiträge zur Erwachsenenbildung Auf der anderen Seite spielt die Erwachsenenbildungswissenschaft in diesen Kontexten bei weitem nicht die Rolle, die angesichts der offiziellen Wertschätzung des ‚Lebenslangen Lernens in der Wissensgesellschaft‘ erwartbar wäre. Es scheint ihr als Bezugswissenschaft kaum zu gelingen, sich als Reflexionsinstanz dieser gesellschaftlichen Aktivitäten zu profilieren oder Entwicklungen in ihrem Feld anzuregen. Zunehmend scheint es schwieriger, Stellenprofilierungen für Professuren der Erwachsenenbildung und Weiterbildung zu behaupten, und Studierende beginnen und beenden ihr Studium nicht selten mit der Frage, was den disziplinär eigenständigen erwachsenenpädagogischen Blick denn eigentlich kennzeichne. Es scheint außerordentlich schwierig, einen disziplinären ‚Kern‘ zu bestimmen, von dem aus sich die ‚Eigenlogik‘ des Erwachsenenpädagogischen rekonstruieren ließe. Mit welchem Fokus blickt ein_e Erwachsenenpädagog_in/e auf ihr_sein Handlungsfeld, wo sind die Grenzen dieses Blicks und was unterscheidet das Erwachsenenpädagogische von anderen disziplinären Wahrnehmungen und Zugriffen auf das Feld? Bei all diesen Fragen scheint der gelegentlich in der Debatte vorgetragene komplexitätsreduzierende Hinweis darauf, dass es doch erwachsenenpädagogische Institutionen und professionelle erwachsenenpädagogische Praktiken gäbe, die auf ‚das Erwachsenenpädagogische‘ verweisen, offensichtlich kaum weiterzuhelfen. Die eigentümliche Diskrepanz zwischen der gesellschaftlichen Bedeutung der Erwachsenenbildung und der disziplinären Lage allein mit innerdisziplinären Verfehlungen zu erklären, würde allerdings wohl ebenso zu kurz greifen wie der Versuch, die Diskrepanz auf die vielfältigen außerpädagogischen Einflüsse resp. deren Unverständnis zurückzuführen. Vielmehr ist sie – so die leitende These der folgenden Ausführungen – Ausdruck einer kategorialen Stilllegung, denn gerade die Omnipräsenz des Bildungsgeredes verweist auf die Notwendigkeit, sich über das Verständnis von ‚Bildung‘ in Erwachsenenbildung systematisch zu verständigen. Einige Facetten dieser Stilllegung sind Gegenstand der nun folgenden Ausführungen. Dabei wird in einem ersten Schritt über die wissenschaftliche Auseinandersetzung mit der ‚Entleerung‘ der Bildungsvokabel eine vertraute Frontstellung nachgezeichnet, die die Beschäftigung mit ‚Bildung‘ in Erwachsenenbildung gerade nicht zu befördern scheint (Kap. 1.1). Demgegenüber wird in einem zweiten Schritt eine Annäherung an ‚Bildung‘ in Erwachsenenbildung über den Problemgehalt des Bildungsgedankens vorgeschlagen (Kap. 1.2). Der darin eröffnete Differenzsinn lässt erkennen, dass sich das Potenzial von ‚Bildung‘ gerade dann zu entfalten scheint, wenn es sich jeglicher begrifflichen Definition und empirischen Bestimmung entzieht. Von hier aus sollen schließlich drei Perspektiven eines erwachsenenpädagogischen Blicks konturiert werden (Kap. 2), die im erwachsenpädagogischen Feld bislang noch wenig Beachtung gefunden haben. 1.1 Die ‚Entleerung‘ der Bildungsvokabel Wer sich im wissenschaftlichen Diskurs und der Alltagspraxis für mehr ‚Bildung‘ ausspricht, kann sich auf der richtigen Seite Debatte Beiträge zur Erwachsenenbildung 7 wähnen. Die Erwachsenen- und Weiterbildung gilt gegenwärtig als der am stärksten expandierende Bildungsbereich, und niemand scheint ernsthaft Einwände gegen ihre vermeintlich große Bedeutung vorzubringen. ‚Bildung‘ wird in aktuellen öffentlichen Debatten als zentrale Ressource für den Erhalt der eigenen Lebenschancen, des individuellen Lebensstandards und damit verbunden auch als eine Versicherung gegen die Gefahren sozialer Exklusion aufgerufen. Mit der richtigen ‚Bildung‘ sichert man sich den angemessenen Platz im sozialen Gefüge. Dabei scheint ‚Bildung‘ nicht nur für die_den Einzelne_n von enormer Bedeutung zu sein. Sie wird darüber hinaus auch als wichtigste Ressource in rohstoffarmen Ländern betrachtet, mit der die Wettbewerbsfähigkeit einer expandierenden Wirtschaft auf sich globalisierenden Weltmärkten gesichert werden soll. Die Steigerung der ‚Bildung‘ und der ‚Bildungschancen‘ wird als gesamtgesellschaftliche und staatliche Zukunftsaufgabe deklariert und dient zugleich als Lösungschiffre bzw. Heilsversprechen für hochgradig heterogene, soziale und gesellschaftliche Konfliktzusammenhänge. Aus dieser Optik scheint ‚Bildung‘ mit Blick auf den Fachkräftemangel und die Beschäftigungsfähigkeit älterer Menschen ebenso unerlässlich wie bezogen auf die Konkurrenzfähigkeit der Wirtschaft, die Integration fremdländischer Arbeitskräfte, die Beseitigung von Fluchtursachen und die Begrenzung des Klimawandels durch ‚Verbraucherbildung‘. Nun kann dieser inflationäre Gebrauch der Bildungsvokabel aus einer erwachsenenpädagogischen wie bildungstheoretischen 8 Debatte Beiträge zur Erwachsenenbildung Perspektive durchaus Anlass zur Beunruhigung geben. Die in zahlreichen öffentlichen Verlautbarungen, bildungspolitischen Programmen und gesellschaftlichen Diskursen ubiquitäre Beanspruchung und Indienstnahme der Bildungsvokabel scheint doch – zumindest kategorial – die Abgrenzung von ‚Bildung‘ gegenüber anderen Konzepten, wie Lernen, Fähigkeitsentwicklung, Normalisierung, Ausbildung, Qualifikation, institutionelle und außerinstitutionelle Aneignungsform etc., zu unterlaufen, die für den spezifischen Gehalt des Bildungsbegriffs gerade wesentlich sind und waren. Dieses eigentümliche Auseinandertreten von öffentlicher Inanspruchnahme und wissenschaftlicher Beschäftigung mit ‚Bildung‘ ist mittlerweile auch zum Gegenstand disziplinärer Auseinandersetzung geworden. Dabei wird die ‚Entleerung‘ der Bildungsvokabel in aktuellen erwachsenpädagogischen Perspektivierungen auf unterschiedliche Weise theoretisch gerahmt. Es lassen sich mindestens zwei Varianten der ‚Entleerung‘ der Bildungsvokabel in der Erwachsenenbildung unterscheiden. Auf der einen Seite ist im Feld eine Position auszumachen, die die gesellschaftliche Bedeutungssteigerung grundsätzlich begrüßt, aber die Befürchtung artikuliert, den ‚eigentlichen Kern‘ der erwachsenenpädagogischen Herangehensweise im Zuge ihrer Funktionalisierung zu verlieren. Aus dieser Binnenperspektive scheint ‚das Erwachsenenpädagogische‘ programmatisch derart identifiziert, dass vor einer ‚Entleerung‘ des Begriffs der ‚Bildung‘ in Erwachsenenbildung von seiner eigentlichen Bestimmung gewarnt werden muss. Sigrid Nolda beobachtet nicht nur bei Teilnehmer_innen, sondern auch bei beruflich in der Erwachsenenbildung Tätigen eine „Hintanstellung des spezifisch Erwachsenenbildnerischen“ (Nolda 2008, S. 124). Die Bildungsaspiration und genuin erwachsenenpädagogische Handlungslogik müsste nicht zuletzt deshalb z. B. über professionelles Programmplanungshandeln in eine zunehmend betriebswirtschaftliche Organisationslogik wieder ‚eingeschrieben‘ werden (vgl. Gieseke 2000, S. 38; Gieseke & Robak 2004, S. 38). Mit der Übernahme betriebswirtschaftlicher Konzepte „verschwindet und dreht sich der Begriff der Bildung bis zur Unkenntlichkeit“ (Faulstich & Zeuner 2015, S. 32), weshalb „gegensteuernd und widerständig“ (Faulstich & Zeuner 2015, S. 32) an dem Begriff ‚Bildung‘ festgehalten werden müsse, um ihn im Kampf gegen seine Ökonomisierung und Funktionalisierung in Anschlag zu bringen. Auf der anderen Seite findet sich eine gegenteilige Position: Nicht wenige fragen, ob diese ‚Entleerung‘ in Anbetracht der Tatsache, dass die Begründungen der Erwachsenenbildung traditionell normativ aufgeladen waren, denn überhaupt so gravierend sei (vgl. Kade 1999; Kade, Seitter & Dinkelaker 2011). Die Leitvorstellungen und Bezugspunkte der Disziplin, wie Selbstbildung, Autonomie etc., seien demnach immer schon zu radikal gedacht gewesen und die Bildungstheorie habe solche Konzepte als normativ-metaphysische Fixpunkte des realen pädagogischen Handelns stets überhöht. In einem solchen Verständnis ist das, was als ‚erwachsenenpädagogisch‘ etikettiert werden kann, nur in konkreten sozialen Kontexten bestimmbar. Die Erwachsenenbildungswissenschaft habe sich demnach von normativ-ideologischen Fragen zu lösen, und das Handlungsfeld der Erwachsenenbildung sei als Funktionssystem mit eigener Problematik und Rationalität zu beobachten. Dabei seien schlicht die Formen zu analysieren, mit denen das System operiert (Dinkelaker 2017). Wieder andere sehen das Ende einer Bildungstheorie gekommen, die der empirischen Forschung keine operationalisierbaren Kategorien anzubieten habe. Hier wird Bildungsforschung als „Maßnahmen- und Steuerungsforschung“ (Tippelt 2006) konfiguriert, die den zweckgerichteten Transfer von Wissen und die praxisnahe Verwendbarkeit entsprechender Forschungsergebnisse sicherzustellen habe (Tippelt 2006, S. 143). Aus einer solchen Perspektive erscheint die gegenwärtige ‚Entleerung‘ des Bildungsverständnisses den Traditionalisten zwar ruinös, erweise sich aber als produktiv für die Entwicklung von Wissenschaft und Gesellschaft. Mit einer derartigen Gegenüberstellung befindet man sich unversehens in einem auch wissenschaftspolitisch verminten Gelände. Es organisiert sich nach dem lange etablierten und verhärteten Muster von ‚Traditionalisten‘ resp. ‚Idealisten‘ auf der einen und produktiven ‚Erneuerern‘ und ‚Modernisierern‘ auf der anderen Seite. Während die einen rückwärtsgewandt an den von der gesellschaftlichen Gegenwart längst überholten Idealen festhalten, scheinen die anderen tatkräftig und mit objektiv-empirischen wissenschaftlichen Mitteln an der Bewältigung gegebener Bildungsprobleme zu arbeiten. In der Debatte über die potenzielle ‚Entleerung‘ der Bildungsvokabel beziehen sich beide beschriebenen Positionen allerdings auf ein Substrat, eine identifizierbare Grundlage oder Basis von Erwachsenenbildung, Debatte Beiträge zur Erwachsenenbildung 9 so als wüsste man bereits, was sie ist oder könne sie in aufwendigen Studien mindestens identifizieren. Gerade indem Erwachsenenbildung als eine identifizierbare Größe vorausgesetzt wird, bleibt der Blick auf ihre normativen Sollens- und Zielbestimmungen gerichtet, an denen entweder festgehalten wird oder die verworfen resp. modifiziert werden sollten. Aus der hier eingenommenen Perspektive lässt sich in dieser – etwas zugespitzten – Gegenüberstellung nun gerade kein hinreichender Beitrag zu der Frage nach der Verfasstheit von Bildung in der Erwachsenenbildung gewinnen. Um den erwachsenenpädagogischen Blick auf die zu konstatierende kategoriale Stilllegung weiter zu schärfen, soll im Folgenden deshalb eine dritte Position entwickelt werden. Im Verweis auf jüngere Theorieentwicklungen in der Erziehungswissenschaft nähert sich diese dem „Bildungsproblem nach der humanistischen Illusion“ (Schäfer 1996) und damit der Verfasstheit von Bildung in Erwachsenenbildung auf spezifische Weise. Entwickelt man die Figur der ‚Entleerung‘ der Bildungsvokabel aus der hier eingenommenen Perspektive, dann bedeutet dies weder eine bereits entschiedene Vorstellung von ‚Bildung‘ zu behaupten, noch sie grundsätzlich zu verwerfen. ‚Bildung‘ soll hier zunächst als Problem betrachtet werden. Um die Bedeutungshöfe der ‚Bildung als Problem‘ zu veranschaulichen, soll in einem nächsten Schritt die „Erfindung des Pädagogischen“ (Schäfer 2009a) und damit der theoriestrategische Einsatzort des Bildungsdenkens (Wimmer 2016) skizzenhaft rekonstruiert werden. 10 Debatte Beiträge zur Erwachsenenbildung 1.2 Bildung als Problem Einige Vertreter_innen bildungstheoretischer Bemühungen um eine Neujustierung pädagogischer Sachverhalte, zentraler Begriffe und theoretischer Konzepte veranschaulichen, wie sich die Entstehungsbedingungen des pädagogischen Denkens auf die Begründungsproblematik der Neuzeit beziehen lassen. Mit dem Beginn der Moderne komme es zur ‚eigentlichen‘ Erfindung des Sozialen. Auch wenn sich Menschen zu allen Zeiten Vorstellungen über ihr soziales Zusammenleben gemacht haben, werde hier der Ursprung und die Begründung der sozialen Ordnung durch ‚Gott‘ resp. ein feudales Herrschaftssystem fraglich. Die Begründungsproblematik kulminiere in der Frage, auf welche Weise eine neue soziale Ordnung begründet werden kann und soll. Es galt, die Leerstelle, die ‚Gott‘ hinterlassen hat, neu zu füllen. Zu dieser Zeit etabliere sich der konstitutive Anspruch einer bürgerlichen Gesellschaft, wonach eine Sozialordnung zu schaffen sei, die sich alleine aus der Bestimmung des Menschen (also durch praktische Selbstbestimmung) zu begründen habe. Diese Bestimmung des Menschen könne fortan durch „keinerlei Rückgriff auf eine vorgegebene Ordnung angemessen begriffen und befriedigend gelöst werden“ (Benner 2009[1987], S. 5), und jede legitime Bestimmung des Menschen müsse sich daran orientieren, „dass die Menschen selbst die Frage nach ihrer Bestimmung stellen“ (Benner 2009[1987], S. 5). Das Problem der Unbestimmtheit des Menschen sollte in der Folge zum Kern pädagogischer Gegenstandsbestimmung werden (Wimmer 2016, S. 12). Ein Bezugspunkt, der die Vergesellschaftung des Einzelnen stets als kritikwürdig ausweist: „Das, was zunächst nach einer Bedrohung der sozialen und symbolischen Ordnung aussieht, ist nun auf einmal genau das, von dem man sich eine bessere soziale Ordnung verspricht“ (Schäfer 2012, S. 71). Im modernen Diskurs über Bildung artikuliert sich demzufolge die Bestimmungssuche des Menschen nach einer sozialen Ordnung, die ihm entspricht. „Wo die Zukunft nicht mehr durch Herkunft bestimmt wird, verliert auch das Selbstverständnis des Menschen seinen vormals festen Bestimmungsgrund“ (Wimmer 2016, S. 11). Gegen die Verpflichtung auf eine soziale Ordnung und deren akzeptierte symbolische Repräsentationsmuster, die über Mythen, Riten und religiöse Praktiken z. B. ein Geschöpf des Schöpfers behaupten, werden in der Moderne Konzepte wie Individualität, Selbstbestimmung, Autonomie und Gleichheit in Stellung gebracht. Derartige Konzepte sollten den Blick öffnen für Selbstverhältnisse jenseits sozialer Vereinnahmung, sozialer Erwartungen und selbstverständlich scheinender Lebensweisen und Gewohnheiten. Im Rahmen der Begründungsproblematik der Neuzeit enthalten pädagogische Denkmuster damit nicht nur pädagogische, sondern immer auch gesellschaftliche Erneuerungsentwürfe. Sie sind entschieden auf eine nicht absehbare Zukunft verwiesen – wie Michael Wimmer bemerkt. Für ihn erhält der Begriff der ‚Bildung‘ erst durch diesen offenen Zukunftsbezug seine Begründung (Wimmer 2016, S. 11). Denn einerseits zählt und zählte es zu den Aufgaben der Institutionen des Bildungswesens, den Menschen eine Auseinandersetzung mit der Welt, die sie umgibt, mit ihren Wissensvorräten und kulturellen Traditionen zu ermöglichen, dies aber dergestalt, dass ihnen stets eine eigene Gestaltungsmöglichkeit von Zukunft eröffnet wird. Damit können – so Wimmer weiter – konstitutive Bezugspunkte des pädagogischen Diskurses der Moderne ausgewiesen werden. Es handelt sich bei diesen allerdings um ein doppeltes Nicht-Wissen: Wir wissen nichts über die Bestimmung des Menschen und wir wissen nichts über die zukünftige Bestimmung der gesellschaftlichen Ordnung, die dieser entspricht. In der hier eingenommenen genealogischen Perspektive gerät erst in dieser doppelten Kontingenz ein pädagogischer Raum in den Blick, der mit seinen Konzepten von Individualität, Selbstbestimmung, Autonomie und Gleichheit Bildungsprozesse verspricht, die sich jeder Einbindung in gesellschaftliche Imperative, funktionale Erfordernisse oder Normalitätsvorstellungen entziehen. Mit der Aufklärung etabliert sich ein pädagogisches Nachdenken darüber, wie ein solcher Raum vorstellbar ist: er soll keine staatlich organisierbare Veranstaltung sein und keine Verpflichtung auf die soziale Ordnung oder deren akzeptierte symbolische Repräsentationsmuster implizieren. Bildung wurde als Gegenkonzept gegen die vorfindlichen Lebensbedingungen und ihre individuelle Aneignung postuliert. Dem Einzelnen sollte die Möglichkeit eröffnet werden, sich zu den gegebenen Bedingungen in ein ‚freies‘ Verhältnis zu setzen und diese – wo mit Blick auf eine andere Zukunft immer nötig – auch zu problematisieren. Als Selbstentfaltung sollte Bildung keinem anderen Zweck untergeordnet sein, sondern ihren Zweck in sich selbst Debatte Beiträge zur Erwachsenenbildung 11 tragen, d. h. als offener Prozess resp. als Suchbewegung verstanden werden, die sich nicht auf spezifische, sozial erwünschte Resultate engführen lässt. Ein solcher Bildungsraum muss – so Alfred Schäfer – zunächst einmal als ein „Versprechen“ (Schäfer 2011) verstanden werden. Dieses eigentümliche ‚Versprechen der Bildung‘ weiß zugleich selbst, dass es die eigene Möglichkeit gegen die realen Bedingungen verspricht. Denn es ist offensichtlich, dass sich das Pädagogische als soziale Praxis in das unauflösbare Paradoxieproblem verstrickt, wie sich diese „Freiheit bei all dem Zwange“ (Kant 1995, S. 711) kultivieren lasse. Es ist ja davon auszugehen, dass alle Bildungsprozesse in sozialen Kontexten stattfinden, dass sie bestimmter Sozialverhältnisse oder bestimmter Einrichtungen bedürfen – und als solche sind sie stets in den Reproduktionsprozess der sozialen Ordnung eingebunden. ‚Bildung als Versprechen‘ zu verstehen bedeutet, dass hier gerade nicht die Einlösung einer erwarteten oder erhofften Verbesserung behauptet wird, sondern deren „Zielund Orientierungspunkte in ihrer realen Möglichkeit unbestimmt bleiben […]. Dieses Versprechen verspricht sich selbst. Es verspricht etwas, das man eigentlich nicht versprechen kann, aber es verspricht sich gegen diese Unmöglichkeit als dennoch sinnvolles, weil mögliches Versprechen“ (Schäfer 2011, S. 22). Zentraler theoriestrategischer Einsatzpunkt des modernen pädagogischen Denkens ist folglich das Problem, wie die Möglichkeit eines solchen Bildungsraums ohne identifizierende Vorgriffe konstelliert werden kann, obwohl er eigentlich (noch) undenkbar oder nur als imaginärer vorstellbar ist.2 Vor diesem Hintergrund sind alle klassischen Konzeptionen des Pädagogischen als „Entwürfe von Vorstellungsräumen [zu verstehen, U. K.], in denen das Unmögliche als möglich darstellbar wird“ (Bünger 2013, S. 115).3 Die genealogische Skizze dieser Ideengeschichte pädagogischen Denkens veranschaulicht, dass in entproblematisierender Weise von ‚Bildung‘ nicht die Rede sein kann. ‚Bildung‘ als Problem offenzuhalten, bedeutet einen stets uneindeutigen Raum als erwachsenenpädagogischen zu imaginieren, der die Möglichkeit von Bildung verspricht und zugleich die Differenz zu Verortungen, Bestimmungen und Steuerungsabsichten der sozialen Wirklichkeit betont. Der besondere Clou des ‚Versprechens der Bildung‘ liegt in eben dieser Uneindeutigkeit, die allerdings nicht als abstrakte Unbestimmbarkeit verstanden werden sollte, sondern als eine Ambivalenz, die das Verhältnis von Bildung und sozialer Ordnung durchzieht (Bünger 2013, S. 17). Auf der einen Seite erhält das ‚Versprechen der Bildung‘ genau hieraus sein Problematisierungspotenzial, das die je spezifische Verfasstheit von 2 Schäfer zeigt, dass bereits die von Jean-Jacques Rousseau gewählte Form des Romans darauf verweist, dass Versuche der gezielten Überschreitung nur im ästhetischen Möglichkeitsraum formulierbar sind und Rousseau selbst auch darum wusste (Schäfer 2009a, S. 239). 3 Zur Auseinandersetzung mit den erwachsenenpädagogischen Möglichkeitsbedingungen derartiger Bildungsräume vergleiche auch Klingovsky & Pfruender (2017). 12 Debatte Beiträge zur Erwachsenenbildung Bildungsprogrammen und Bildungsaspirationen befragbar macht. Auf der anderen Seite scheint das ‚Versprechen der Bildung‘ aber gerade indem es mit der Überschreitung der Ordnung des Gegebenen kalkuliert, auch offen zu sein für historisch sich wandelnde Konzepte und Programme, die eine spezifische Realisierung versprechen. So kommt auch Jürgen Oelkers zu dem Schluss: Die moderne pädagogische Theoriebildung werde hier „progressiv unwahrscheinlich und verliert den Kontakt zur Realität. Genau aus diesem Grunde ist sie erfolgreich“ (Oelkers 1983, S. 813). 2. Konturen eines erwachsenenpädagogischen Blicks Betrachtet man die gegenwärtige ‚Entleerung‘ der Bildungsvokabel von hier aus, werden die aktuellen Programmierungen von ‚Bildung‘ zum Ausgangspunkt der Analyse. Ein zentrales Kennzeichen dieser Programmierung ist die Aspiration, das eigentlich nicht einlösbare Versprechen zu realisieren. Dies betrifft die Beschreibung von Bildungsprozessen als Individualakte, d. h. die Verortung von Bildungsprozessen auf der Ebene des Individuums ebenso wie die Vorstellung einer erwachsenenpädagogischen Praxis als mechanisches Wirkungshandeln. Diese nur stellvertretend genannten Auffassungen werfen die Frage auf, ob sie als identifizierende Zugriffe die Verfasstheit der ‚Bildung‘ in Erwachsenenbildung schon angemessen beschreiben und verstehen. Demgegenüber soll hier der Versuch unternommen werden, das Problematisierungspotenzial der ‚Bildung‘ in der Erwachsenenbildung neu zu fassen. Diese Neu-Justierung vermag sich allerdings nicht länger auf vertraute Maßstäbe der Kritik und nicht minder vertraute Oppositionen von Freiheit und Macht zu stützen. Stattdessen widersteht sie den nicht erst gegenwärtig wahrzunehmenden Tendenzen, in denen sich eine zunehmende Vergessenheit der sozialen Dimensionen von ‚Bildung‘ in Erwachsenenbildung abzeichnet. Im Versuch, die mit einer quasi transhistorischen Selbstverständlichkeit ausgestattete ‚Bildung‘ zu dekonstruieren, betrachtet ein derart informierter erwachsenenpädagogischer Blick die Gegenwart der ‚Bildung‘ weder als unterwerfende Disziplinar- noch als hervorbringende Ermöglichungs- und Humanisierungsgeschichte. Demgegenüber werden ‚Bildung‘ und die in ihrem Namen aufgerufenen Programme, Praktiken, Verordnungen und Institutionalisierungsprozesse als Regulative analysiert, die sich auf gesellschaftliche Transformationsprozesse beziehen lassen. In dieser Bewegung können die spezifischen Gebrauchsweisen der Bildungsvokabel sichtbar und damit das je gegebene Versprechen im Namen der ‚Bildung‘ einer Analyse zugänglich gemacht werden. Diese analytischen Problematisierungen werden im Folgenden an den gegenwärtig unter dem Label ‚Bildung‘ artikulierten Steigerungs- und Steuerungsphantasien (2.1) veranschaulicht, um schließlich die Bezugspunkte des ‚Versprechens der Bildung‘ selbst zu problematisieren (2.2) und nicht zuletzt die politischen Dimensionen von ‚Bildung‘ neu zu beschreiben (2.3). Debatte Beiträge zur Erwachsenenbildung 13 2.1 Allgegenwärtig: Das Versprechen von Steigerung und Steuerung Im Zuge des Umbaus des gesamten Bildungswesens unter den Vorzeichen einer neuen Steuerungslogik seit etwa Mitte der 1990er Jahre gerät auch das angebotsorientierte System der Erwachsenenbildung unter erhöhten Modernisierungsdruck. Angesichts der mit den gesellschaftlichen Transformationsprozessen zunehmenden Anforderungen an subjektive, wirtschaftliche und gesellschaftliche Anschlussfähigkeit werden exzellente Organisationsformen gesucht, die sich auf dem Bildungsmarkt bewähren und ihre Leistungsfähigkeit und Qualität einer auf Produkte bzw. Wirkungen bezogenen vergleichenden Überprüfung unterziehen. Der Erwachsenenbildung wird darin die Aufgabe zuteil, „immer wieder neu entstehende Problemlagen moderner Gesellschaften bildungspraktisch zu bearbeiten und so gesellschaftliche Problemlagen in individuelle Entwicklungsprojekte zu transformieren“ (Forneck & Wrana 2005, S. 196). Um die gesellschaftliche Integrationsfähigkeit des zu entwickelnden Humanpotenzials sicherzustellen, benötigen diese individuellen Entwicklungsprojekte im Zuge der angeblich beschleunigten Veränderungsdynamik gesellschaftlicher Anforderungen und der wachsenden Anpassungsnotwendigkeit immer auch umfassende professionelle erwachsenenpädagogische Begleitung. In Anbetracht der Steigerungs- und Steuerungsrhetorik, in der sich ein derartiges Programm der kontrollierten und erschöpfenden Ausnutzung aller institutionellen und menschlichen Ressour- 14 Debatte Beiträge zur Erwachsenenbildung cen artikuliert, stellt sich die Frage, warum überhaupt auf ‚Bildung‘ Bezug genommen und z. B. nicht schlicht von gegenwärtig erhöhtem Lern-, Qualifikations- oder Normalisierungsbedarf gesprochen wird? Ein problematisierender erwachsenenpädagogischer Blick beunruhigt sich an dieser Frage und eröffnet eine erweiternde Perspektive auf die gegenwärtige Attraktivität des ‚Versprechens der Bildung‘. Appelle der Veränderungs- und Anpassungsbereitschaft an das ‚Versprechen der Bildung‘ zu knüpfen, scheint immens attraktiv, da sie hierin mit einer harmonistischen Vorstellung der möglichen Versöhnung mit den gesellschaftlichen Verhältnissen einhergehen. Das mit dem ‚Versprechen der Bildung‘ verbundene Assoziationsfeld von Selbstbestimmung und frei wählbaren Optimierungsoptionen evoziert den Eindruck, im Rahmen der (neoliberalen) Neustrukturierung des Bildungssystems ginge es zugleich um Persönlichkeitsund Selbstbildung (Schäfer 2009b, S. 45). Auch in seiner trivialsten Form scheint das Bildungsversprechen eine ‚Aura‘ zu besitzen: Es verspricht nicht nur einen Zusammenschluss der permanenten Selbstentwicklung des Einzelnen mit der Bewältigung zentraler Zukunftsherausforderungen, sondern auch seine soziale Wirkmächtigkeit. In anderen Worten: ‚Bildung‘ wird zum Motor erklärt, mit dem die Steigerungspotenziale des Einzelnen für die Neustrukturierung der sozialen Ordnung fruchtbar gemacht werden sollen. In dieser Aspiration, das Bildungsversprechen sozial organisiert einzulösen, zeigt sich allerdings auch, auf welche Weise das erwachsenenpädagogische Handlungsfeld adressiert wird. Eingebunden in die komplexe (Neu-) Ordnung des Sozialen, werden soziale Erwartungen und bildungspolitische Vorgaben an die Bildungsarbeit mit Erwachsenen gerichtet. Sie soll ihren Beitrag zur Neustrukturierung der sozialen Ordnung bei gleichzeitiger Steigerung der individuellen Handlungsfähigkeit leisten. Die Adressierung geht mit einem Appell einher: Der bislang bildungstheoretisch nur imaginäre Raum der ‚Bildung‘ muss nun konkret umgesetzt, also realisiert werden. Der Wissenschaft wird entsprechend einer politökonomischen Logik die Aufgabe zuteil, opportunes Steuerungswissen zu generieren. Kaum jemals zuvor wurden derart große Anstrengungen unternommen, den Raum der ‚Bildung‘ zu vermessen: Wir verfügen über Erkenntnisse global angelegter Studien (z. B. PIAAC), über Bildungsberichterstattungen und flächendeckende Evaluationen von Qualität in Institutionen. Die Governanceforschung liefert neues Wissen zur Optimierung und Steuerung von Organisationen und die Professionsforschung fokussiert die Erweiterung individualisierter erwachsenenpädagogischer Handlungskompetenzen, von denen – so wird behauptet – die Effizienz der Erwachsenen- und Weiterbildung entscheidend abhänge. All diese Projekte – unabhängig davon, ob sie Kompetenzen, institutionelles Handeln, exkludierte Individuen oder Biografien zu operationalisieren versuchen – organisieren eine Empirie im Horizont jener pädagogischen Bezugspunkte (Individualität, Selbstbestimmung, Gleichheit und Autonomie), die schon das traditionelle Bildungsversprechen aufgerufen hatte. Der bedeutsame Unterschied zur gegenwärtigen Bildungsmetaphorik ist nun der, dass in klassischen päda- gogischen Entwürfen ‚Bildung‘ jenseits ihrer Einbindung in soziale Reproduktionsprozesse noch als Problem virulent war und gerade gegen gesellschaftliche Brauchbarkeits- und Normalitätsvorstellungen in Stellung gebracht werden sollte. Diese kategorialen Differenzierungen scheinen in gegenwärtigen Untersuchungen allerdings kaum mehr erkennbar. Obwohl das Problem, über die Wirkungen und Ziele der eigenen Anstrengungen nicht verfügen zu können, zu den Ausgangskonstellationen erwachsenenpädagogischer Selbstverständigung zählt, erhält die ‚Bildung‘ Erwachsener im Zuge der mit den gesellschaftlichen Transformationsprozessen einhergehenden Herausforderungen den Status einer prioritären gesellschaftlichen und politischen Steuerungsaufgabe. Je ungewisser in der erwachsenenpädagogischen Selbstverständigung die Möglichkeit der Steuerung von Personwerdungsprozessen scheint, desto selbstverständlicher wird diese Möglichkeit im öffentlichen Diskurs (gelegentlich auch in der wissenschaftlichen Diskussion und nicht zuletzt auch der erwachsenenpädagogischen Praxis) in Anspruch genommen. Damit unterläuft der Versuch einer sozial organisierten Einlösung des Bildungsversprechens allerdings den „Differenzsinn dieser [mit ‚Bildung‘ verbundenen, U. K.] Konzepte“ (Schäfer 2011, S. 75). Im Sinne eines erwachsenenpädagogischen Grundlagendiskurses wäre es vielversprechend, all jene kulturellen, politischen und wissenschaftlichen Vorgegebenheiten offenzulegen und zu problematisieren, die zu hegemonialen ‚Schließungen‘ führen, in denen ‚Bildung‘ in Erwachsenenbildung als etwas ‚Bestimmtes‘ identifiziert ist. Eine wichtige Debatte Beiträge zur Erwachsenenbildung 15 Funktion der Erwachsenenbildungswissenschaft könnte vor diesem Hintergrund gerade darin bestehen, durch kategoriale Reflexionen das Verhältnis von „Begriff und Sache“ (Adorno 2007, S. 15) nicht normativ stillzustellen, sondern in seiner Kontingenz offenzuhalten und ihre wissenschaftliche Erkenntnisproduktion stärker im Kontext gesellschaftlicher Entwicklungen zu reflektieren, aus deren Kriterienkatalog sie sich speist (Effizienz, Leistungsfähigkeit, Erfolgs- und Steigerungsfähigkeit). 2.2 Desillusionierungen des Versprechens aus macht- und identitätskritischer Perspektive Schon in der hier nur angedeuteten Analyse zeigt sich, dass die kategorialen Bezugspunkte Selbstbestimmung, Individualität, Autonomie und Gleichheit des ‚Versprechens der Bildung‘ zu einer scheinbar unproblematischen Selbstverständlichkeit geworden und – obwohl als Einsatz gegen die gesellschaftliche Ordnung versprochen – längst in gesellschaftliche Normalitätsvorstellungen eingewoben sind. „Die ‚freie Entfaltung des Individuums‘ zu ermöglichen, scheint nicht nur zum Anliegen der sozialen Ordnung selbst geworden zu sein; zugleich drückt sich darin eine Erwartung an die Einzelnen aus, sich stets auf ‚neue Erfahrungen‘ einzulassen und Veränderungen als lebenslang auftretende Lerngelegenheiten wahrzunehmen“ (Bünger 2013, S. 125). Versteht man das ‚Versprechen der Bildung‘ in Erwachsenenbildung von hier aus, werden deren traditionelle Referenzpunkte selbst problematisch. So fragt 16 Debatte Beiträge zur Erwachsenenbildung Käte Meyer-Drawe bereits 1996, ob sich vor dem Hintergrund einer poststrukturalistischen Subjekt- und Identitätskritik und den damit verbundenen Irritationen überhaupt noch sinnvoll über Autonomie oder Individualität jenseits sozialer Verstrickungen in einer akzeptierten symbolischen Ordnung sprechen lasse (Meyer-Drawe 1996, S. 662). Poststrukturalistische und kulturwissenschaftliche Einsätze nach dem linguistic resp. cultural turn opponieren gegen die Evidenz von Identitäten, über die sich die Innerlichkeit des Menschen entschlüsseln ließe. Das ‚selbstbestimmte Subjekt‘ der Gegenwart ist aus machtanalytischer Perspektive im Anschluss an Michel Foucault keine unabhängige und moralische Instanz von Autonomie, sondern ein ‚Effekt‘ sich historisch transformierender Machtverhältnisse. Es scheint mehr als fragwürdig, ob von einem vernünftigen, sich selbst gegenwärtigen Subjekt, das jenseits der sozialen Ordnung seine ‚eigentliche‘ Bestimmung sucht, überhaupt noch die Rede sein kann. Foucault analysiert gouvernementale Regierungspraxen, die diese ‚Innerlichkeit‘ des Menschen als komplexes Zusammenspiel von Freiheit und Unterwerfung, von Macht und Wissen und von Individualisierung und Normalisierung hervorbringen (Foucault 2004). Im Anschluss an diese Subjekt- und Identitätskritik scheint gegenwärtig kein Vernunftsubjekt mehr gedacht werden zu können, das sich ‚bilden‘ und damit nach und nach an der Überwindung gesellschaftlicher Problemlagen arbeiten könnte, sondern nur mehr ein subjektiviertes Individuum, das den Verhältnissen nichts entgegen zu setzen hat, weil es selbst als deren Effekt auftritt. Vor dem Hintergrund einer solch machtanalytischen Reflexion verliert das ‚Versprechen der Bildung‘ in der Erwachsenenbildung nicht nur an Überzeugungskraft, sondern erwachsenenpädagogische Praktiken geraten gerade mit ihrer Hypostasierung des subjektiven Faktors selbst als Technologien des Selbst in den Blick (Klingovsky 2009). Gerade indem die Erwachsenenbildung das Subjekt ins Zentrum der Bearbeitung stellt, entwickelt sie Instrumente und Verfahren, mit denen das ‚Innen‘ der Subjekte zur Operationsbasis für Veränderungsprozesse wird. Ein zentrales Strukturmoment der qua Erwachsenenbildung eröffneten Handlungsfelder ist es, dass die Lernenden in ihnen angehalten werden, sich selbst zu objektivieren, und indem sie dies tun, machen sie sich zum Gegenstand einer subjektivierenden Arbeit an sich selbst (Klingovsky 2013, S. 7-9). Dabei aber versprechen Bildungsprozesse alles andere als die Möglichkeit einer freien Entwicklung und individuellen Selbstbestimmung jenseits sozialer Zumutungen (Klingovsky 2011, S. 170). Unter den Bedingungen der gesellschaftlichen Instrumentalisierung des Bildungsversprechens sollte die ‚Bildung‘ Erwachsener nicht länger als ein Versprechen auf Freiheit jenseits sozialer Zumutungen konzipiert werden. Im Gegenteil: Die ‚Bildung‘ Erwachsener findet als Formierung von Subjektivität entlang gesellschaftlicher Zuschreibungen und Erwartungen an die Lebensführung jedes Einzelnen bereits statt. All das, was sich der Normalisierung des Menschen entziehen sollte, wird nun zur Anforderung an die Selbstbestimmung, Emanzipation und Partizipation im sozialen Raum. Indem die ‚Bildung‘ der Individuen entlang dieser Konzepte auf eine spezifische Weise reguliert wird, wird es – vereinfacht gesagt – fragwürdig, ob die klassischen pädagogischen Entwürfe von Individualität, Selbstbestimmung, Autonomie und Gleichheit überhaupt noch als Referenzpunkte gelesen werden können. Als Reservoir einer dynamischen Selbstoptimierung führen eben jene pädagogischen Grundkategorien unter dem Signum der Responsibilisierung mittlerweile längst zu neuen sozialen Hierarchisierungen resp. Ein- und Ausschlüssen. Von hier aus gilt es jenseits der traditionellen Bildungsmetaphorik einen Blick auf ‚Bildung‘ in Erwachsenenbildung zurückzugewinnen. Dieser orientiert sich nicht an normativen Bezugspunkten jenseits seiner sozialen und gesellschaftlichen Beanspruchung, sondern entfaltet sich als Problematisierungsformel, mit der gegenwärtige Machtverhältnisse und Subjektivierungsmuster ebenso in den Blick genommen werden können wie Verengungen von scheinbar alternativlosen Perspektiven auf sozialen Wandel oder die Anstrengungen und „Strapazen des Selbstseins“ (Ehrenberg 2015). Hierfür gilt es in einem letzten Zug zu überprüfen, ob und wie das Politische der Bildung verstanden werden kann. 2.3 Die Politizität von Bildung Es spricht einiges dafür, den Begriff der ‚Bildung‘ angesichts seiner gegenwärtig entgrenzten Beanspruchung als Schauplatz einer hegemonialen Auseinandersetzung zu betrachten. Sowohl die Vermessung von Bildungsräumen wie die Figur der Subjektivierung verweisen in hohem Maße auf ‚politische‘ Einsätze um Debatte Beiträge zur Erwachsenenbildung 17 ‚Bildung‘. Indem der Begriff abwechselnd ausgelegt, vereinnahmt oder negiert wird, wirkt er im politischen Raum geradezu ‚leer‘. Erst derart entleert lässt er sich als zentrale Größe individueller wie gesellschaftlicher Bestrebungen um Leistungs- und Wettbewerbsfähigkeit beanspruchen (Jergus & Thompson 2011). Nun wäre es aus machtanalytischer Perspektive allerdings zu kurz gegriffen, auf eine ‚Idee der Bildung‘ zu bestehen und diese in Anbetracht der gegenwärtigen sozialen Relevanz von Erwachsenenbildung gegen Strategien der ‚Ökonomisierung‘ und damit verbunden der ‚Entpolitisierung‘ zu verteidigen. Im Anschluss an Thomas Höhne (2004) lässt sich umgekehrt vielmehr eine „Pädagogisierung sozialer Machtverhältnisse“ und damit verbunden eine zunehmende Politisierung der ‚Bildung‘ konstatieren. Damit ist zum einen die Entgrenzung von Bildungsverhältnissen markiert, die längst nicht mehr nur an klassische Institutionen gebunden sind. Bildungsverhältnisse haben sich als verallgemeinerte Sozialform etabliert, in denen „Individuen zu Bildungssubjekten“ (Höhne 2004, S. 35) geworden sind. ‚Bildungsprozesse‘ lassen sich als flexible Modi der Disziplinierung und Steigerung individueller und sozialer Ressourcen darstellen. Wenn Individuen an ihren Bildungsbemühungen bemessen und auf kontrollierende Weise individuelle, institutionelle oder soziale Entwicklungsmöglichkeiten organisiert werden, die ‚öffentliche Bildung‘ folglich als zentraler Modus sozialer Veränderung angegeben wird, erfährt die Macht in der Moderne eine eminent pädagogische Ausrichtung. ‚Bildung‘ lässt sich dann als „strategischer Imperativ“ (Foucault 1978b, S. 120) bezeichnen, 18 Debatte Beiträge zur Erwachsenenbildung dem man sich kaum entziehen kann. „Diesen Modi des Lernens, der Entwicklung und der Bildung kann sich auf Dauer auch kein ,vernünftiger Mensch‘ verschließen, wenn er sich selbst nicht dadurch ausschließen will“ (Höhne 2004, S. 5). Für diesen Befund spricht, dass die gesellschaftliche Aufmerksamkeit gegenüber Bildung und Lernen im Erwachsenenalter (als einheitliche Lösungsformel für sehr unterschiedliche individuelle wie kollektive Problemlagen) in den vergangenen Dekaden beispielsweise über den Diskurs des Lebenslangen Lernens gar noch gesteigert wurde (Klingovsky 2013). Neben der Beobachtung einer sozial verallgemeinerten Beanspruchung der Erwachsenen- und Weiterbildung erlaubt der Terminus ‚Pädagogisierung sozialer Machtverhältnisse‘ zum zweiten eine Beschreibung des veränderten Zugriffs auf Bildungsprozesse Erwachsener und damit Rückschlüsse auf die soziale Bedingtheit der qua ‚Bildung‘ induzierten Selbstformierungen und -verhältnisse. Hierin werden Selbsttätigkeit, Eigenverantwortung, Teilhabe und Autonomie nicht zwingend gegen soziale Machtverhältnisse in Stellung gebracht. Im Gegenteil: Indem die Subjekte aufgerufen sind, ein Leben lang mit all ihrem Eigensinn und ihrer Persönlichkeit engagiert und dauerhaft an ihrem subjektiven Lernund Kompetenzprofil zu arbeiten, werden im Sinne Foucaults spezifische Selbstverhältnisse etabliert, entlang derer soziale Identitäten und Differenzen ausgebildet werden. Testierung, Prüfung, Evaluation und Dauerbeobachtung sind als soziale Kontrollpraktiken dechiffrierbar, die eine ‚Subjekt-Bildung‘ mit sozialisierender und zugleich integrierender Funktion organisieren. Indem die Individuen lernen, integriert und zugleich flexibel an der permanenten Steigerung ihrer eigenen Kapazitäten zu arbeiten, wird diese Subjekt-Bildung als immanenter Bestandteil machtvoller Subjektivierungsformationen und damit als politische Strategie erkennbar. Betrachtet man die ‚Bildung‘ in Erwachsenenbildung von hier aus, müsste sie auf gesellschaftliche Entwicklungen bzw. sozial immanente Praktiken und Prozesse bezogen werden. Damit ließe sich mit Carsten Bünger eine neue Perspektive auf die „Politizität von Bildung“ (Bünger 2013) eröffnen, die darin liegt, das Politische nicht (allein) als äußere Bedingung der Bildung zu verstehen, sondern als ein für Lern- und Bildungsprozesse konstitutives Moment. „Nicht, dass ‚Bildung‘ ein programmatischer Punkt der gegenwärtigen Parteienpolitik ist, sondern dass mit Bildung ein subjektkonstitutives Verhältnis zum Gegebenen angezeigt ist, kennzeichnet deren Politizität“ (Bünger 2013, S. 17). Eine bildungstheoretisch informierte Reflexion der ‚Politizität von Bildung‘ entfaltet ihr kritisches Potenzial nicht über wünschenswerte Ideale oder normative Bezugspunkte des Bildungsbegriffs, sondern als immanente Kritik an den bestehenden Zu- und Übergriffen. Mit ihren vernunftkritischen Reflexionspotenzialen ist sie gleichermaßen allergisch gegenüber einer rationalen Technisierung wie gegenüber einer Moralisierung von Bildung. Sie gewinnt ihren Gegenstand schließlich aus den Verhältnissen, in welche die ‚Bildung‘ in Erwachsenenbildung selbst eingelassen ist, und ist bestrebt, die allgegenwärtigen praktischen Ansprüchlichkeiten zu dezentrieren. Mit einem derart veränderten Begriff von Kritik (Butler 2002) gerieten eben jene machtvollen politischen Programmierungen der ‚Bildung‘ in Erwachsenenbildung in den Blick. 3. Schlussbetrachtungen Die hier vorgenommene Sondierung für eine Genealogie des erwachsenenpädagogischen Blicks soll zu einem erweiternden Verständnis von ‚Bildung‘ in Erwachsenenbildung beitragen. Entlang einer Diskussion der Probleme und Gehalte des Bildungsbegriffs und der mit ihm verbundenen Konzepte von Individualität, Selbstbestimmung, Gleichheit und Autonomie sollte anschaulich werden, wie sich der „Differenzsinn dieser Konzepte“ (Schäfer 2011, S. 75) gegenwärtig verflüchtigt, obwohl oder gerade indem das ‚Versprechen der Bildung‘ als Lösungsformel für alle möglichen hochgradig heterogenen, sozialen und gesellschaftlichen Konfliktzusammenhänge in Anspruch genommen wird. Von hier aus sollte ein erwachsenenpädagogischer Blick konturiert werden, der die quasi transhistorisch bestimmte Selbstverständlichkeit von ‚Bildung‘ in Erwachsenenbildung zersetzt und ‚Bildung‘ als das zurückzugewinnen versucht, was sie sein kann: Eine Problematisierungsformel, die das Verhältnis von Bildung und sozialer Ordnung ohne jeden identifizierenden Zugriff stets neu befragt. Indem das ‚Versprechen der Bildung‘ den Einzelnen eine Versöhnungsperspektive im Rahmen der sozialen Ordnung verspricht, geht in der gegenwärtigen Konstellation eine inhärente Spannung verloren, eine Potenz von Einwänden, so dass sich die Privilegierung Debatte Beiträge zur Erwachsenenbildung 19 des Selbst unangefochten durchsetzen und als Unterwerfungspraxis unkenntlich halten kann. Damit geht der Stachel der Kritik verloren, der notwendig wäre, um die affirmierende Kraft des ‚Versprechens der Bildung‘ für die erwachsenenpädagogische Theorie und Praxis zu irritieren (Thompson 2004). Ein derart kritischer erwachsenenpädagogischer Blick wäre ein Einsatzpunkt, der sich weder mit funktionalen Entleerungen noch mit normativen Vereindeutungen von Bildung beruhigt, sondern die Frage nach der Bildung in Erwachsenenbildung offenzuhalten und immer wieder neu zu problematisieren sucht. Es wäre dies der Versuch, der ausgeführten kategorialen Stilllegung zu begegnen und damit zu verhindern, dass die Disziplin nicht mehr nur nichts Neues zu ‚Bildung‘ in Erwachsenenbildung zu sagen hat, sondern darüber hinaus zu erkennen, wie die Disziplin an der Konstitution dessen, was sie verspricht, selbst mitwirkt, ohne sich ihrer impliziten Annahmen und kategorialen Voraussetzungen zu vergewissern. Literatur Adorno, T. W. (2007). Vorlesung über Negative Dialektik. Fragmente zur Vorlesung 1965/66. Frankfurt/Main: Suhrkamp. Benner, D. (2009[1987]). Allgemeine Pädagogik. Eine systematisch-problemgeschichtliche Einführung in die Grundstruktur pädagogischen Denkens und Handelns. 6., überarbeitete Auflage. Weinheim et al.: Juventa. Butler, J. (2002). Was ist Kritik? Ein Essay über Foucaults Tugend. Deutsche Zeitschrift für Philosophie, 50 (2), 249-265. 20 Debatte Beiträge zur Erwachsenenbildung Bünger, C. (2013). Die offene Frage der Mündigkeit. Studien zur Politizität der Bildung. Paderborn: Ferdinand Schöningh. Dinkelaker, J. (2017). Aufmerksamkeit als Kategorie einer Empirie pädagogischer Situationen. In W. Meseth, J. Dinkelaker, O. Dörner, K. Kunze, S. Neumann & K. Rabenstein (Hrsg.), Empirie des Pädagogischen und Empirie der Erziehungswissenschaft. Beobachtungen erziehungswissenschaftlicher Forschung (S. 110-124). Bad Heilbrunn: Klinkhardt. Ehrenberg, A. (2015). Das erschöpfte Selbst. Depression und Gesellschaft in der Gegenwart. Frankfurt/Main: Campus. Faulstich, P. & Zeuner, C. (2015). Ökonomisierung und Politisierung des Feldes der Erwachsenenbildung: Rolle der Wissenschaft. Erziehungswissenschaft, 50 (26), 25-35. Forneck, H. J. & Wrana, D. (2005). Ein parzelliertes Feld. Einführung in die Weiterbildung. Bielefeld: W. Bertelsmann Verlag. Foucault, M. (1978a). Nietzsche, die Genealogie, die Historie. In W. Seitter (Hrsg.), Michel Foucault: Von der Subversion des Wissens (S. 83-109). Frankfurt/Main et al.: Ullstein. Foucault, M. (1978b). Dispositive der Macht. Berlin: Merve. Foucault, M. (2004). Geschichte der Gouvernementalität I. Sicherheit, Territorium, Bevölkerung. Frankfurt/Main: Suhrkamp. Gieseke, W. (2000). Programmplanung und Bildungsmanagement. In W. Gieseke (Hrsg.), Programmplanung als Bildungsmanagement? Qualitative Studie in Perspektivverschränkung (S. 30-58). Recklinghausen: Bitter. Gieseke, W. & Robak, S. (2004). Programmplanung und Management aus der Bildungsforschungsperspektive. Empirische Befunde und konzeptionelle Wendungen. Literatur- und Forschungsreport Weiterbildung, 27 (2), 33-41. Höhne, T. (2004). Pädagogisierung sozialer Machtverhältnisse. Pädagogisierung – Die Kunst Menschen mittels Lernen immer dümmer zu machen. Schulheft 112, 30-44. Jergus, K. & Thompson, C. (2011). Die Politik der Bildung. Eine theoretische und empirische Analyse. In R. Reichenbach, N. Ricken & H.-C. Koller (Hrsg.), Erkenntnispolitik und die Konstruktion pädagogischer Wirklichkeit (S. 103-121). Paderborn: Ferdinand Schöningh. Kade, J. (1999). System, Protest und Reflexion. Gesellschaftliche Referenzen und theoretischer Status der Erziehungswissenschaft/Erwachsenenbildung. Zeitschrift für Erziehungswissenschaft, 2 (4), 527-544. Kade, J., Seitter, W. & Dinkelaker, J. (2011). Wissen(stheorie) und Erwachsenenbildung/Weiterbildung. In R. Tippelt & A. von Hippel (Hrsg.), Handbuch Erwachsenenbildung/Weiterbildung (S. 197-212). 5. Auflage. Wiesbaden: VS Verlag für Sozialwissenschaften. Kant, I. (1995). Über Pädagogik. In I. Kant: Schriften zur Anthropologie, Geschichtsphilosophie, Politik und Pädagogik 2 (S. 695761). Werkausgabe Band XII: 9. Auflage. Frankfurt/Main: Suhrkamp. Klingovsky, U. (2009). Schöne Neue Lernkultur. Transformationen der Macht in der Weiterbildung. Eine gouvernementalitätstheoretische Analyse. Bielefeld: transcript. Klingovsky, U. (2011). Praxiskonzepte in der Weiterbildung: Ein machtanalytischer Zugang zur „Frei“-Setzung der Subjekte. In C. Hof, J. Ludwig & B. Schäffer (Hrsg.), Steuerung – Regulation – Gestaltung. Dokumentation der Jahrestagung der Sektion Erwachsenenbildung der DGfE (S. 162175). Baltmannsweiler: Schneider Verlag Hohengehren. Klingovsky, U. (2013). Lebenslanges Lernen im Postfordismus. Magazin erwachsenenbildung.at, Ausgabe 18, 2-9. Verfügbar unter https://www.pedocs. de/volltexte/2013/7352/pdf/Erwachsenenbildung_18_2013_Klingovsky_Lebenslanges_Lernen_Postfordismus.pdf [12.09.2018]. Klingovsky, U. & Pfruender, G. (2017). Critical Diversity Literacy through Arts & Further Education – Konturen und Grundzüge neuer Formen der Begegnung mit dem Fremden und dem Eigenen. Hessische Blätter für Volksbildung, 67 (4), 367-375. Lightbody, B. (2010). Philosophical Genealogy, Volume I. An epistemological reconstruction of Nietzsche and Foucault's Genealogical Method. New York: Peter Lang. Mahon, M. (1992). Foucault’s Nietzschean Genealogy. Truth, Power, and the Subject. Albany: State University of New York Press. Meyer-Drawe, K. (1996). Versuch einer Archäologie des pädagogischen Blicks. Zeitschrift für Pädagogik, 42 (5), 655-666. Nietzsche, F. (1954[1887]). Zur Genealogie der Moral. In K. Schlechta (Hrsg.), Friedrich Nietzsche: Werke in drei Bänden. München: Carl Hanser Verlag. Nolda, S. (2008). Einführung in die Theorie der Erwachsenenbildung. Darmstadt: Wissenschaftliche Buchgesellschaft. Debatte Beiträge zur Erwachsenenbildung 21 Oelkers, J. (1983). Rousseau und die Entwicklung des Unwahrscheinlichen im pädagogischen Denken. Zeitschrift für Pädagogik, 30 (5), 801-816. Saar, M. (2007). Genealogie als Kritik. Geschichte und Theorie des Subjekts nach Nietzsche und Foucault. Frankfurt/Main et al.: Campus Verlag. Schäfer, A. (1996). Das Bildungsproblem nach der humanistischen Illusion. Weinheim: Deutscher Studien Verlag. Schäfer, A. (2009a). Die Erfindung des Pädagogischen. Paderborn: Ferdinand Schöningh. Schäfer, A. (2009b). Bildung. In G. Opp & G. Theunissen (Hrsg.), Handbuch schulische Sonderpädagogik (S. 44-53). Stuttgart: UTB. Schäfer, A. (2011). Das Versprechen der Bildung. Paderborn: Ferdinand Schöningh. Schäfer, A. (2012). Das Pädagogische und die Pädagogik. Annäherungen an eine Differenz. Paderborn: Ferdinand Schöningh. Tippelt, R. (2006). Bildung und Handeln – Möglichkeiten empirischer Bildungsforschung. In L. Pongratz, M. Wimmer & W. Nieke (Hrsg.), Bildungsphilosophie und Bildungsforschung (S. 138-155). Bielefeld: Janus. Thompson, C. (2004). Diesseits von Authentizität und Emanzipation. Verschiebungen kritischer Erziehungswissenschaft zu einer ‚kritischen Ontologie der Gegenwart‘. In N. Ricken & M. Rieger-Ladich (Hrsg.), Michel Foucault: Pädagogische Lektüren (S. 39-56). Wiesbaden: VS Verlag für Sozialwissenschaften. 22 Debatte Beiträge zur Erwachsenenbildung Wimmer, M. (2016). Dekonstruktion und Erziehung. Studien zum Paradoxieproblem in der Pädagogik. Paderborn: Ferdinand Schöningh. Ulla Klingovsky, Prof. Dr., Professorin für Erwachsenenbildung und Weiterbildung an der Pädagogischen Hochschule der Fachhochschule Nordwestschweiz. Forschungsschwerpunkte: Theorie und Empirie von Lern- und Bildungsprozessen, Professionalisierung erwachsenenpädagogischer Denkund Handlungspraxen, Diskursanalyse und Gouvernementalitätsstudien, Wissenschaftliche Weiterbildung und Hochschuldidaktik. ulla.klingovsky@fhnw.ch
https://openalex.org/W2593410154	https://molecularneurodegeneration.biomedcentral.com/track/pdf/10.1186/s13024-017-0166-z	English	null	Caspase vinyl sulfone small molecule inhibitors prevent axonal degeneration in human neurons and reverse cognitive impairment in Caspase-6-overexpressing mice	Molecular neurodegeneration	2,017	cc-by	16,449	* Correspondence: andrea.leblanc@mcgill.ca 1Bloomfield Center for Research in Aging, Lady Davis Institute for Medical Research, Jewish General Hospital, 3999 Ch. Cote Ste-Catherine, Montreal, QC H3T 1E2, Canada 2Department of Neurology and Neurosurgery, McGill University, 845 Sherbrooke O, Montreal, QC H3A 0G4, Canada Full list of author information is available at the end of the article Caspase vinyl sulfone small molecule inhibitors prevent axonal degeneration in human neurons and reverse cognitive impairment in Caspase-6-overexpressing mice Caspase vinyl sulfone small molecule inhibitors prevent axonal degeneration in human neurons and reverse cognitive impairment in Caspase-6-overexpressing mice Prateep Pakavathkumar1,2, Anastasia Noël1,2, Clotilde Lecrux3, Agne Tubeleviciute-Aydin1,2, Edith Hamel3, Jan-Eric Ahlfors4 and Andrea C. LeBlanc1,2,5* © The Author(s). 2017 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 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 DOI 10.1186/s13024-017-0166-z Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 DOI 10.1186/s13024-017-0166-z Open Access Background impairment in the Huntington mouse model [13–18]. Casp6-dependent tubulin fragmentation is associated with neuritic degeneration in mouse sympathetic, reti- nocollicular, dorsal root ganglion sensory, commis- sural, and motor neurons following nerve growth factor (NGF)-deprivation [19–21]. Casp6 participates in axonal degeneration of neurturin-deprived dorsal root ganglion cells [22], myelin-mediated sympathetic and septal cholinergic neurons [23], ischemic neurons [24, 25], and retinal ganglion cells in models of optic nerve injury in rodents [26]. In these conditions, Casp6 is activated in the presence of other caspases, including Casp9 and Casp3. Similarly, human neonatal, infant, and adult hypoxic-ischemic brain injury results in in- creased levels of active Casp6, active Casp3, and tubu- lin cleaved by Casp6/3 (TubΔCasp6/3) [4, 27]. Alzheimer disease (AD) is a neurodegenerative condition characterized by cognitive impairments leading to demen- tia with no disease-modifying treatments. Pathologically, AD is defined by an accumulation of extracellular plaques containing mostly amyloid-beta peptide (Aβ) and intracel- lular neurofibrillary tangles (NFT) composed of a hyper- phosphorylated form of the microtubule-associated protein Tau. Clinical trials targeting Aβ plaques have been unsuc- cessful in restoring cognitive function [1], while trials on disaggregating NFTs are currently ongoing [2]. The results from the current clinical trials suggest that therapeutic intervention against AD could be improved by targeting earlier pathogenic events. One emerging potential disease-modifying therapeutic target is Caspase-6 (Casp6), a cysteinyl protease that cleaves protein substrates specifically after an aspartic acid residue [3]. Casp6, but not Casp3 or Casp7, is acti- vated in neurites interspersing Aβ plaques, NFTs, and neuropil threads of familial and sporadic AD brains [4–6]. In brains from some aged non-cognitively impaired indi- viduals, Casp6 activity levels correlates negatively with epi- sodic and semantic memory performance [7], two types of memory first affected in AD. Tau-cleaved by Casp6 (TauΔCasp6) levels in post-mortem cerebrospinal fluid correlate inversely with episodic, semantic, and working memory performance [8]. Overexpression of human Casp6 in the CA1 region of mice hippocampi results in age-dependent episodic and spatial memory loss [9]. These findings suggest that early Casp6 activation in the hippocampus of aged pre-symptomatic individuals leads to cognitive impairment. Natural caspase inhibitors either do not inhibit Casp6 or are non-selective. Viral proteins p35 and CrmA inhibit sev- eral caspases [28, 29]. The mammalian inhibitors of apop- tosis proteins (IAP) do not inhibit Casp6 [30, 31]. Abstract Background: The activation of the aspartate-specific cysteinyl protease, Caspase-6, is proposed as an early pathogenic event of Alzheimer disease (AD) and Huntington’s disease. Caspase-6 inhibitors could be useful against these neurodegenerative diseases but most Caspase-6 inhibitors have been exclusively studied in vitro or show acute liver toxicity in humans. Here, we assessed vinyl sulfone small molecule peptide caspase inhibitors for potential use in vivo. Methods: The IC50 of NWL vinyl sulfone small molecule caspase inhibitors were determined on Caspase-1 to 10, and Caspase-6-transfected human colon carcinoma HCT116 cells. Inhibition of Caspase-6-mediated axonal degeneration was assessed in serum-deprived or amyloid precursor protein-transfected primary human CNS neurons. Cellular toxicity was measured by phase contrast microscopy, mitochondrial and lactate dehydrogenase colorimetric activity assays, or flow cytometry. Caspase inhibition was measured by fluorogenic activity assays, fluorescence microscopy, and western blot analyses. The effect of inhibitors on age-dependent cognitive deficits in Caspase-6 transgenic mice was assessed by the novel object recognition task. Liquid chromatography coupled to tandem mass spectrometry assessed the blood-brain barrier permeability of inhibitors in Caspase-6 mice. Results: Vinyl sulfone NWL-117 caspase inhibitor has a higher selectivity against Caspase-6, −4, −8, −9, and −10 whereas NWL-154 has higher selectivity against Caspase-6, −8, and −10. The half-maximal inhibitory concentrations (IC50) of NWL-117 and NWL-154 is 192 nM and 100 nM against Caspase-6 in vitro, and 4.82 μM and 3.63 μM in Caspase-6-transfected HCT116 cells, respectively. NWL inhibitors are not toxic to HCT116 cells or to human primary neurons. NWL-117 and NWL-154 inhibit serum deprivation-induced Caspase-6 activity and prevent amyloid precursor protein-mediated neurite degeneration in human primary CNS neurons. NWL-117 crosses the blood brain barrier and reverses age-dependent episodic memory deficits in Caspase-6 mice. (Continued on next page) © The Author(s). 2017 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. Page 2 of 19 Pakavathkumar et al. Abstract Molecular Neurodegeneration (2017) 12:22 Conclusions: NWL peptidic vinyl methyl sulfone inhibitors are potent, non-toxic, blood-brain barrier permeable, and irreversible caspase inhibitors with neuroprotective effects in HCT116 cells, in primary human CNS neurons, and in Caspase-6 mice. These results highlight the therapeutic potential of vinyl sulfone inhibitors as caspase inhibitors against neurodegenerative diseases and sanction additional work to improve their selectivity against different caspases. Keywords: Alzheimer disease, Caspases, Caspase-6, Axonal degeneration, Peptide inhibitors, Primary human neurons, Caspase-6 transgenic mice, Vinyl sulfone inhibitors Cell cultures and NWL inhibitor treatments Cell cultures and NWL inhibitor treatments HCT116: Human colon carcinoma (HCT116) cells (ATCC, Manassas, VA, USA) were cultured in McCoy’s 5A modified media (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA, USA) and transfected with 1 μg of pCep4βCasp6p20p10 mixed with 8 μg of polyethyleneimine (Polysciences Inc., Warrington, PA, USA) [47]. Protein expression was allowed for 24 h before treating with vehicle (PBS), 100 μM NWL-117, or 100 μM NWL-154 for 2 h. For NWL inhibitor kinetic ex- periments, 100 μM of NWL-117 or NWL-154 was added at 120, 90, 60, 30, 15, or 0 min before harvest. For the ex- tended time course from 2 to 48 h, NWL inhibitors were dissolved in PEG400 (Sigma-Aldrich, Oakville, ON, CA) (50% v/v), anhydrous ethanol (20% v/v), and 154 mM NaCl (BioShop Canada Inc, Burlington, Ontario, CA) (30% v/v)). For recovery time course experiments, after the initial treatment with 100 μM of NWL-117 or −154 for 2 h, the media was replaced with fresh media for 120, 90, 60, 30, 15, or 0 min before harvest. Primary Human CNS neurons: Cortical tissues were obtained from the Birth Defects Research Laboratory (BDRL, University of Washington, Seattle, USA) in accordance with NIH ethical guidelines approved by McGill University’s institutional review board and primary neurons cultured as previously described [48]. Primary human neurons were seeded on poly-L-lysine-coated (5 μg/mL) 6-well plates or glass Q , , Q , ) Recombinant Casp6 Expression and purification: Casp6 was expressed from the pET23b(+)-Casp6-His plasmid in E. coli BL21(DE3)pLysS strain (Promega, Fitchburg, WI, USA) at 37 °C in 2xYT medium (16 g/l tryptone, 10 g/l yeast extract, 5 g/l NaCl) supplemented with 0.1 mg/ml ampicillin and 0.034 mg/ml chloram- phenicol under vigorous shaking according to [46]. Casp6 expression was induced with 50 μM isopropyl β- D-1-thiogalactopyranoside (IPTG) when cell cultures reached OD595 nm of 0.6 and cells cultured at 22 °C for 16 h under vigorous shaking. Cells were harvested by centrifugation, resuspended in buffer A (50 mM Tris pH 8.5, 300 mM NaCl, 5% glycerol, 2 mM imidazole), and lysed by sonicating on ice with a Vibra-Cell ultra- sonic processor (Sonics and Materials, Newtown, CT, USA) for 2 min at 50% duty with output control set to four. Background There are two natural Casp6 protein inhibitors: the alternatively spliced Casp6β isoform, which only prevents Casp6 activa- tion and the caspase inhibitory factor (CIF), which is react- ive against other caspases [32, 33]. Many competitive small molecule Casp6 inhibitors have been developed but most have not been tested for cellular toxicity, blood brain bar- rier permeability and in vivo inhibition. Aza-peptides spe- cifically inhibit caspases and not other cysteine proteases [34]. Casp6 specificity is improved with sulfonamide isatin Michael acceptors [35]. Aldehyde or fluoromethyl ketones (fmk)-conjugated peptides obtained from positional scan- ning libraries or natural AP-2α and Lamin A Casp6 sub- strates have been used as Casp6 inhibitors [18, 36–41]. The commercially available Casp6 inhibitor benzyloxycarbonyl- Val-Glu-Ile-Asp-fmk (Z-VEID-fmk) is toxic to mammals because the fmk moiety can be metabolized into fluoroci- trate, an inhibitor of aconitase that depletes tricarboxylic acid cycle intermediates [42]. Nevertheless, a Huntington- based peptide inhibitor conjugated to TAT to enhance membrane permeability and delivered to the brain with an osmotic pump protects against behavioral and motor defi- cits in a mutant Huntingtin mouse model [18]. When activated, Casp6 can impair the microtubule network within neuronal axons and lead to degener- ation. Casp6 cleaves the C-terminus of several neuronal cytoskeletal or associated proteins including Tau and α- tubulin [4, 10]. In human CNS neuron cultures, overex- pression of wild type amyloid precursor protein (APPWT), a condition associated with familial AD [11], results in Casp6-dependent, but Aβ-independent, neuritic degener- ation [12]. Therefore, inhibiting Casp6 activity could pre- vent axonal degeneration. New World Laboratories Inc. (NWL) has developed novel peptidomimetic irreversible small molecule inhibitors Caspases play an important role in other neurode- generative conditions. Casp6 is associated with motor Page 3 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 with buffer B (50 mM Tris pH 8.5, 500 mM NaCl, 5% glycerol, 20 mM imidazole), and bound proteins eluted with a 50-300 mM linear imidazole gradient in buffer A. Fractions were assessed for recombinant Casp6 purity by SDS-PAGE and Coomassie blue staining. Fractions containing pure Casp6 were pooled together, dialyzed against storage buffer (20 mM Tris pH 8.5, 200 mM NaCl, 10 mM DTT, 5% glycerol), concentrated by dialysis against polyethylene glycol (PEG) 20,000 (Sigma-Aldrich, Oakville, ON, CA), and stored at -80 °C in small aliquots. Protein concentration was measured using Quick Start Bradford 1x Dye Reagent (Bio-Rad Laboratories, Hercules, CA, USA). Background Active site titration assay: The concentration of Casp6 active sites was determined by active site titration assay using Z-VAD-fmk (N-benzyloxycarbonyl- Val-Ala-Asp-(O-methyl)-fluoromethylketone, MP Biomedi- cals, Santa Ana, CA, USA) inhibitor [46]. Casp6 (398 nM) was incubated in SB with 0 to 1.25 μM Z-VAD-fmk for 2 h at room temperature in a final volume of 10 μl, diluted 20- fold with SB, and 25 μl of aliquots transferred to a black clear bottom 96-well microplate (Costar, Corning, NY, USA). Casp6 VEIDase activity (see below) was plotted as a function of Z-VAD-fmk concentration; the intersection at the X-axis in the linear region of the curve indicates the concentration of active sites of Casp6. that retain Casp6’s Z-VEID preferred substrate, but have a methyl vinyl sulfone chemical warhead which 1) is selective for cysteinyl proteases, 2) is unreactive with circulating thiols or non-active site cysteines, 3) forms a hydrogen bond with the active site histidine [43], and (4) is safe in rats, dogs, and primates [44]. Here, we describe a non-toxic and blood-brain permeable NWL caspase inhibitor that prevents axonal degeneration of primary human neurons, and reverses Casp6-dependent episodic memory impair- ment in mice. These findings highlight vinyl sulfones as viable caspase inhibitors for pre-clinical studies. DNA constructs The mammalian constructs encoding human Casp6p20p10 in the pCep4β vector (Thermo Fisher Scientific, Waltham, MA, USA) [45], and enhanced green flurorescent protein (EGFP) or EGFP and amyloid precursor protein (APPWT) in the double promoter-containing pBudCE4.1 vector (Thermo Fisher Scientific, Waltham, MA, USA) [12] were previously cloned in our laboratory. A synthetic Escherichia coli codon-optimized gene (GenScript, Piscataway, NJ, USA) coding for human Casp6 large subunit (amino acids 24-179, flanked by start (ATG) and stop (TAA) codons) and small subunit (amino acids 194-293, preceded by a start codon), separated by GAATTCAATAATTTTGTT TAACTTTAAGAAGGAGATATACAT containing an in- ternal ribosome binding site (underlined), was ligated into the XbaI/XhoI sites of the pET23b(+)-Casp6-His plasmid (a kind gift from Dr. Guy Salvesen, Sanford Burnham Pre- bys Medical Discovery Institute, CA, USA), under the con- trol of a single T7 promoter. All plasmids were sequenced by the Sanger method (McGill University and Genome Quebec Innovation Center, Montreal, Quebec, CA). Microscopy analyses py y Immunofluorescence on human neuron cultures: Primary human neurons were pre-treated 2 h and serum-deprived in the presence of 100 μM NWL-117, 100 μM NWL-154, or 5 μM Z-VEID-fmk (Biomol, Plymouth meeting, PA, USA) for 24 h. Following treatment, human neurons were washed once with warm PBS, fixed for 20 min at room temperature with 4% paraformaldehyde (Sigma, Oakville, ON, CA)/4% sucrose (BioRad, Mississauga, ON, CA) for TubΔCasp6 or 2% formaldehyde (Thermo Fisher Scien- tific, Waltham, MA, USA)/0.2% glutaraldehyde (Sigma, Oakville, ON, CA) for pBudEGFP or pBudEGFP/APPWT- transfected neurons [50], incubated in permeabilization buffer (0.1% Triton X-100, 0.1 sodium citrate) for 1 min on ice, washed with PBS, blocked for 20 min at room temperature with 10% goat serum (Sigma, Oakville, ON, CA), and incubated with primary antibodies diluted in 10% goat serum in PBS overnight at 4 °C in a humid chamber. The glass coverslips were washed with PBS, and incubated with goat anti-rabbit secondary antibody coupled to Alexa 488 (Molecular Probes, Eugene, OR, USA) or Cy3 (GE Healthcare Life Sciences, Baie D’Urfe, QC, CA) and Hoechst 33342 (ImmunoChemistry, Bloomington, MN, USA) at 1 μg/mL for 2 h at room temperature. The coverslips were washed with PBS and rinsed in Milli-Q water before mounting in fluoromount Cell cultures and NWL inhibitor treatments Cells were washed once with 1 mL ice-cold PBS, incubated on ice for 5 min with 200 μL cell lysis buffer (CLB) (50 mM HEPES, 0.1% CHAPS, 0.1 mM EDTA), and gently scraped off. Protein con- centrations were determined by Bradford assay (BioRad, Mississauga, ON, CA) by measuring the absorbance at 595 nm using the BioTek Synergy H4 plate reader. Caspase-6 activity was measured in 40–60 μg total protein as described above. The IC50 for recombinant and cellular caspases were determined using GraphPad Prism 5.0 (La Jolla, CA, USA) using a log (inhibitor) – response curve with a Hill slope of −1. coverslips coated with poly-L-lysine and laminin (5 μg/ mL) (Sigma-Aldrich, Oakville, ON, CA) at a density of 3x106 cells/mL. Primary human neurons were pre-treated with 0.1 μM epoxomicin (Enzo LifeSciences, Farmingdale, NY, USA) and 100 μM NWL-117, 100 μM NWL-154, or vehicle (PBS) for 2 h, and serum-deprived for 2 h in the presence of epoxomicin and vehicle (PBS), 100 μM NWL- 117, or 100 μM NWL-154 before harvesting or measuring caspase activity with FLICA, as described below. coverslips coated with poly-L-lysine and laminin (5 μg/ mL) (Sigma-Aldrich, Oakville, ON, CA) at a density of 3x106 cells/mL. Primary human neurons were pre-treated with 0.1 μM epoxomicin (Enzo LifeSciences, Farmingdale, NY, USA) and 100 μM NWL-117, 100 μM NWL-154, or vehicle (PBS) for 2 h, and serum-deprived for 2 h in the presence of epoxomicin and vehicle (PBS), 100 μM NWL- 117, or 100 μM NWL-154 before harvesting or measuring caspase activity with FLICA, as described below. Casp6 activity assays Recombinant or extracted cellular Casp6 activity: Casp6 activity was assessed by in vitro fluorogenic assays using Ac-Val-Glu-Ile-Asp-(7-Amino-4-trifluoromethylcouramin) (Ac-VEID-AFC: Enzo LifeSciences, NY, USA) as the Casp6 substrate. The activity was measured in Stennicke’s buffer (SB) (20 mM piperazine-N, N-bis (2-ethanesulfonic acid) (PIPES: BioShop Canada Inc, Burlington, Ontario, CA) pH 7.2, 30 mM NaCl, 1 mM ethylenediaminetetraacetic acid (EDTA), 0.1% 3-[(3-cholamidopropyl)-dimethylammo- nio]-2-hydroxy-1-propanesulfonic acid (CHAPS), 10% su- crose) [49]. The reaction mix consisted of either 20 nM RCasp6 or 20–30 μg cellular protein extracts, SB, 10 mM DTT, 10 μM VEID-AFC substrate and deionized water. The activity was measured in a black clear bottom 96-well plate (Costar, Corning, NY, USA) at 50 μL/well in triplicate at 37 °C in the Synergy H4 plate reader (BioTek) at excita- tion 380 nm and emission 505 nm every two minutes for 100 min. Fluorescence units were converted to the moles of AFC released based on a standard curve of 0–625 pico- moles of free AFC. Cleavage rates were calculated from the linear phase of the assay. The activity is considered on a percentage scale where no inhibitor present is equated to 100% activity of the enzyme. Cellular Casp6 activity assay by FLICA: Active Casp6 was labeled within primary human neurons using the fluorescent inhibitor of Casp6 (FLICA) (FAM-VEID-fmk, ImmunoChemistry, Bloomington, MN, USA) following the manufacturer’s protocol. Briefly, FLICA reagent and Hoechst 33342 were added to a black clear bottom 96-well plate containing 100,000 of the treated- primary human neurons for 2 h at 37 °C in 5% CO2. The cells were rinsed twice with wash buffer and fresh media was added to the cells. The fluorescence of FAM-VEID- fmk was measured at 490 nm excitation and 520 nm emis- sion with bandwidth reduced to 8 nm in the Synergy H4 plate reader (BioTek). The Hoechst signal was measured by excitation at 360 nm and emission at 485 nm. Cell cultures and NWL inhibitor treatments The lysate was clarified by centrifugation (30,000 x g for 30 min at 4 °C) and loaded on Ni Sepharose Fast Flow 6 medium (GE Healthcare Life Sciences, Baie D’Urfe, QC, CA) pre-equilibrated with buffer A, washed Page 4 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Page 4 of 19 154 with 20 nM of active site-titrated Casp6 in SB at room temperature for 5 minutes. Then, 10 μM Ac-VEID-AFC was added and fluorescence measured for 20 min at 37 °C as described above. New World Laboratories performed the IC50 determination for NWL inhibitors dissolved in di- methyl sulfoxide (DMSO) against recombinant Casp1-10 using the Caspase Inhibitor Drug Screening Kits (BioVision, San Francisco, CA) following the manufacturer’s instruc- tions and the preferred substrates for the caspases (Cas- pase-1: WAD-AFC, Caspase-2: VDVAD-AFC, Caspase-3: DEVD-AFC, Caspase-4: LEVD-AFC, Caspase-5: WEHD- AFC, Casp6:VEID-AFC, Caspase-7: DEVD-AFC, Caspase- 8: IETD-AFC, Caspase-9: LEHD-AFC, Caspase-10: AEVD-AFC). In transfected HCT116 cells, 0 to 100 μM NWL inhibitors were added in the culture media and left on the cells for 2 h. Cells were washed once with 1 mL ice-cold PBS, incubated on ice for 5 min with 200 μL cell lysis buffer (CLB) (50 mM HEPES, 0.1% CHAPS, 0.1 mM EDTA), and gently scraped off. Protein con- centrations were determined by Bradford assay (BioRad, Mississauga, ON, CA) by measuring the absorbance at 595 nm using the BioTek Synergy H4 plate reader. Caspase-6 activity was measured in 40–60 μg total protein as described above. The IC50 for recombinant and cellular caspases were determined using GraphPad Prism 5.0 (La Jolla, CA, USA) using a log (inhibitor) – response curve with a Hill slope of −1. 154 with 20 nM of active site-titrated Casp6 in SB at room temperature for 5 minutes. Then, 10 μM Ac-VEID-AFC was added and fluorescence measured for 20 min at 37 °C as described above. New World Laboratories performed the IC50 determination for NWL inhibitors dissolved in di- methyl sulfoxide (DMSO) against recombinant Casp1-10 using the Caspase Inhibitor Drug Screening Kits (BioVision, San Francisco, CA) following the manufacturer’s instruc- tions and the preferred substrates for the caspases (Cas- pase-1: WAD-AFC, Caspase-2: VDVAD-AFC, Caspase-3: DEVD-AFC, Caspase-4: LEVD-AFC, Caspase-5: WEHD- AFC, Casp6:VEID-AFC, Caspase-7: DEVD-AFC, Caspase- 8: IETD-AFC, Caspase-9: LEHD-AFC, Caspase-10: AEVD-AFC). In transfected HCT116 cells, 0 to 100 μM NWL inhibitors were added in the culture media and left on the cells for 2 h. IC50 determination of NWL inhibitors on recombinant proteins or in cells The half-maximal inhibitory concentrations (IC50) for NWL inhibitors (Patent publication # WO/2009/140765, WO/2010/133000, and WO/2012/140500) was deter- mined by incubating 0 to 20 μM NWL-117 and NWL- Page 5 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 (Dako, Burlington, ON, CA). Images were acquired by fluorescence microscopy and quantified using the ImageJ software (NIH, Bethesda, MD, USA) for TubΔCasp6 and manually counted for transfected EGFP(+)-neurons. Time lapse-imaging by live fluorescence microscopy: Primary hu- man neurons were transfected with gold beads coated with pBudEGFP or pBudEGFP/APPWT using a Helios Gene gun (BioRad, Mississauga, ON, CA) [12]. Briefly, cells were pre- treated with 100 μM NWL inhibitors or 5 μM Z-VEID-fmk for 2 h. The media was removed and the neurons were shot at 100 psi. The media was quickly replaced with the inhibi- tors present. The plasmid was expressed for 16 h before setting up the fluorescence microscope (Nikon Eclipse Ti) to acquire 20 images per condition every hour for 72 h at 37 °C with 5% CO2. The images were analyzed by counting the total number of neurons (50–100 neurons per condi- tion per independent experiment), and the number of beaded, swollen soma, and healthy neurons. In addition, the time at which cells beaded was noted. Phase contrast microscopy: HCT116 cells, plated at a density of 1×105 cells/well and human neurons, plated at a density of 6×106 cells/well on poly-L-lysine were treated with PBS, 100 μM NWL-117, 100 μM NWL-154, or 2 μM staurosporine (Bio- mol, Plymouth meeting, PA, USA) for 24 h. Images were acquired with the Nikon Eclipse Ti microscope and the NIS-Elements (Version 3.10) software. cells were also used to correct the absorbance. SubG1 population analysis: HCT116 cells were seeded in a 6-well plate at a density of 1x105 cells/well and treated the fol- lowing day with 100 μM NWL-117, 100 μM NWL-154, equal volumes of vehicle (PBS), or 2 μM staurosporine for 24 h. The media was recovered and the cells were trypsi- nized in 0.25% Trypsin-EDTA (Thermo Fisher Scientific, Waltham, MA, USA). Both the media and the cells were combined and centrifuged for 5 min at 4 °C and washed with cold PBS-EDTA (5 mM). Cells were resuspended in 1 mL cold PBS-EDTA (5 mM), fixed by the dropwise addition of 3 mL of ice cold 100% ethanol and stored at −20 °C overnight. IC50 determination of NWL inhibitors on recombinant proteins or in cells After centrifugation, the ethanol was re- moved and the cells were washed with cold PBS-EDTA (5 mM). Then, 1 mL of staining solution was added (5 mM PBS-EDTA, 50 μg/mL propidium iodide, 20 μg/ mL RNAse A (Sigma-Aldrich, Oakville, ON, CA)). The samples were analyzed by flow cytometry using the FACS Calibur II instrument (BD Biosciences, Mississauga, ON, Canada). The data were interpreted using the cell cycle analysis tool in FlowJo Version 10.0, which determined the DNA content in cells based on propidium iodide intensity (Ashland, OR, USA). Western blot analyses y Protein extracts from HCT116 cells and human primary neurons were subjected to western blotting analyses. The 10630 (1:10 000) and GN60622 (1:10 000) neoepi- tope antibodies against the p20 subunit of active Casp6 (Casp6p20) and α-tubulin cleaved by Casp6 (TubΔCasp6) were generated previously in our laboratory [4, 10, 51]. The β-actin clone AC-15 (1:5 000, Sigma-Aldrich, Oakville, ON, CA), Casp6 (1:1 000), Synapsin (1:5 000), full-length α-tubulin (1:1 000, Cell Signalling Technol- ogy Inc., Danvers, MA, USA), GFAP (1:3 000, Dako, Burlington, ON, CA), synaptophysin (1:5 000, Sigma, Oakville, ON, CA), and PSD95 clone K28143 (1:5 000, UC Davis/NIH NeuroMab Facility) were purchased. All anti- bodies were diluted in 5% non-fat dry milk. Secondary anti-mouse (1:5 000, GE Healthcare Life Sciences, Baie D’Urfe, QC, CA) and anti-rabbit antibodies (1:5 000, Dako, Burlington, ON, CA) conjugated to horseradish peroxidase were used to detect immunoreactive proteins using ECL prime western blotting detection reagent (GE Healthcare Life Sciences, Baie D’Urfe, QC, CA) and Kodak BioMax MR film (Kodak, Rochester, NY, USA). Secondary anti-mouse conjugated to alkaline phosphatase (Jackson Immunoresearch Laboratories Inc., West Grove, PA) was developed with nitro-blue tetrazolium (Thermo Fisher Scientific, Waltham, MA, USA) and 5-bromo-4- chloro-3-indolylphosphate (Thermo Fisher Scientific, Waltham, MA, USA) for chromogenic detection of pro- teins. The western blots were scanned with an HP scanner and the images were not manipulated except to adjust the Cellular toxicity assays MTT assay: HCT116 cells or primary human neurons were seeded in 96-well plates at a density of 1x104 and 1x105 cells per well, respectively. The next day, cells and neurons were treated with vehicle (PBS), or 20, 50, or 100 μM of NWL-117, NWL-154, or 2 μM staurosporine for 24 or 48 h. The media was replaced with 0.5 μg/ml MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazo- lium bromide) (Sigma-Aldrich, Oakville, ON, CA) and the cells were incubated for 4 h at 37 °C in 5% CO2. The media was removed before dissolving the formazan crys- tals in 100 μL DMSO while shaking for 30 min. Once dissolved, the absorbance of each sample was measured at 560 nm and 670 nm using the Synergy H4 plate reader from BioTek (Winooski, VT, USA). LDH Assay: HCT116 cells were seeded in a 6-well plate at a density of 1×105 cells/well and treated the following day with 100 μM NWL-117 or −154 or equal volumes of vehicle (PBS) for 24 h. As a positive control, some cells were lysed for 2 h in 0.9% Triton X-100 (BioShop Canada Inc, Burlington, Ontario, CA). Media was collected and stored at −20 °C or assayed right away using the Cytotox 96 kit (Promega, Madison, WI, USA) following the man- ufacturer’s protocol. Hydrochloric acid (1 N) was added for 10 min to stop the reaction, which was read at 490 nm (signal) and 520 nm (background) using the Synergy H4 plate reader from BioTek. Media without Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Page 6 of 19 Page 6 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 brightness/contrast and this was done simultaneously to the entire blot. Quantification was performed with the ImageJ software (NIH, Bethesda, MD, USA). The integrity of the blood-brain barrier was confirmed by injecting 3% Evan’s blue solution in 20 month old Casp6 mice. Mouse cognitive analysis by novel object recognition: Mice were handled during 5 min for one week prior to be- havioral tests. Novel object recognition (NOR) task was administered in three phases: habituation, familiarization (pre-exposure), and test phase. For habituation, mice were placed in the NOR box (80 cm x 80 cm, Stoelting Co, Wood Dale, IL, USA) for 5 min. After 24 h, the pre- exposure phase was initiated by allowing the animals to explore two identical objects inside the NOR box. NWL caspase inhibitors and Casp6 transgenic mice NWL caspase inhibitors and Casp6 transgenic mice Casp6 transgenic mouse model: All animal procedures followed the Canadian Council on Animal Care guidelines and were approved by the McGill Animal care committees. Sixteen to 20 month old C57BL/6 J mice were bred and aged in the pathogen-free Goodman Cancer Research Centre Mouse Transgenic Facility at McGill University. Mice were housed in a temperature-controlled room at 22 °C and were kept on a 12 h light/dark cycle. Food and water were available at libitum. Casp6 overexpressing mice (KI/Cre) express Casp6p20p10 under the CAG pro- moter (CMV immediate early enchancer/chicken β- actin promoter fusion) in the CA1 pyramidal cell layer of the hippocampus under the control of calmodulin kinase IIa (CAMKIIa)-regulated Cre expression [9]. No obvious toxicity was observed after a one month treat- ment of NWL-117 on mice (Additional file 1). Cellular toxicity assays Then, following a 2 h gap, mice were re-introduced to the NOR box which now contained a familiar and a novel object. The position of the novel object was counterbalanced be- tween animals to avoid any bias related to a preference in the location of the new object and the use of potential confounding spatial cues. The objects were located in the middle of the NW and SE quadrants of the box, equidis- tantly from the box corners and from each other. The mice were placed in the middle of the SW quadrant. Washing the box and objects with 70% ethanol eliminated odour cues. The number of times touching each object was manually recorded, while the total distance, percent time moving, and number of entries into virtual cells were recorded using the HVS 2100 automated video tracking system (HVS Image, Buckingham, UK). Animals whose exploration was considered insufficient to allow recogni- tion (<10 s per object) during the familiarization phase were excluded from analysis. Different object sets were used in the pre- and post-tests. Immunohistochemistry on mouse brain slices: Following behavioural analysis, animals were anaesthetized under isoflurane and perfused intra- cardially with ice cold saline for 7 min and 4% paraformal- dehyde for 20 min. Mice brains were removed and stored in 10% neutral-buffered formalin (Thermo Fischer Scien- tific, Waltham, MA, USA) for 24 h then dehydrated in 70% ethanol for 24 h or less. Brains were embedded in paraffin and cut using a vibratome at the histology plat- form of the Institute for Research on Immunology and Cancer (U Montreal). Slides containing 4 μm thick sec- tions of the anterior hippocampus were deparaffinized in xylene (Thermo Fisher Scientific, Waltham, MA, USA), and rehydrated before demasking in antigen retrieval buffer (10 mM Tris, 1 mM EDTA, pH 9; or 10 mM tri-sodium citrate, pH 6 for synaptophysin) for 20 min at 97 °C in the Pascal Dako Cytomation (Dako, Burlington, ON, CA). The immunostaining procedure was automated using the Dako Autostainer Plus slide processor and the EnVision Flex system (Dako, Burlington, ON, CA). Slides were treated with peroxidase for 5 min, then rinsed, blocked with Serum-Free Protein Block (Dako, Burlington, ON, CA) for 30 min, and incubated with either Iba1 (1: 2 000, Wako, Richmond, VA, USA), TubΔCasp6 (1: 5 000), or synaptophysin (1: 8 000, Sigma, Oakville, ON, CA) NWL treatments Only males were used and littermates from each geno- type (wild type (WT)/WT, WT/Cre, & knock-in (KI/ )Cre) were tested together. The experimenter was blind to genotype and treatment groups. Mice were adminis- tered 20 mg/kg NWL-117 or physiological saline (0.9% NaCl) by intraperitoneal injections two times 48 h apart. Injection volumes did not exceed 150 μL of 10 mg/mL NWL-117 prepared in physiological saline. Blood brain barrier permeability of NWL-117 caspase inhibitor: Briefly, 18–22 month old Casp6 mice were anesthetised with isoflurane, warmed with a heating blanket, and their physiological vitals (heart rate, body temperature, and res- piration) monitored. The skin over the mouse’s neck was shaved, xylocaine applied, and an incision was made along the midline. Under a surgical microscope, the right carotid artery was gently separated from surrounding tissue. Two suture threads were placed at the proximal and distal ends of the carotid, and a small incision was made along the ca- rotid wall. A micro-catheter (attached to a 1 ml syringe controlled by a micropump) was inserted and pushed to the entrance of the internal carotid artery. The catheter was secured in place using suture thread. NWL-117 (20 mg/kg) was infused via the micro-catheter using a pump set at 50 μL/min (total volume between 64 μL and 130 μL). After 5 min, blood was collected by intra-cardiac puncture, the mouse was perfused through the heart with ice cold saline for 2–3 min. The brain was removed and hippocampi were dissected and frozen on dry ice and stored. Samples were sent to the Biopharmacy platform at the University of Montreal (Quebec, Canada) for liquid chromatography and tandem mass spectrometry analysis. Page 7 of 19 Page 7 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 antibodies diluted in EnVision Flex Antibody Diluent (Dako, Burlington, ON, CA) for 30 min. After rinsing, the mouse brain slices were incubated with secondary rabbit-horseradish peroxidase antibody (Dako, Burlington, ON, CA) for 30 min and diaminobenzidine (Dako, Burlington, ON, CA) for 10 min before counterstain- ing with hematoxylin (Dako, Burlington, ON, CA). Slides were scanned using the MIRAX SCAN (Zeiss, Oberkochen, Germany) and analyzed using the ImageJ software (NIH, Bethesda, MD, USA) by measuring the area of positive immunoreactivity over the total area in square microns. a dose-dependent Casp6 inhibition with half-maximal inhibitory concentrations (IC50) of 192 nM and 100 nM, respectively (Fig. 1b). NWL treatments The peptide backbone of NWL in- hibitors binds to the active site of Casp6, which allows the vinyl sulfone warhead to hydrogen bond with the protonated imidazole ring of histidine, while the cata- lytic cysteine attacks the β-carbon of the vinyl group (Fig. 1c). The reaction is irreversibly stabilized under physiological conditions by histidine de-protonation by the α-carbon of the vinyl group [43]. Thus, peptide- based vinyl sulfones are potent, irreversible, and com- petitive Casp6 inhibitors. NWL inhibitors inhibit recombinant initiator caspases and Casp6 at sub-micromolar concentrations Statistical analysis of data was performed using Graph- pad Prism 5.0 (La Jolla, CA, USA). The analyses were done with ANOVA followed by post-hoc analyses as in- dicated for each test in the figure legends. Alternatively, a student t-test was done to compare between two sam- ples as indicated in figure legends. Significance was set at p < 0.05 for all experiments. NWL inhibitor specificity was tested on recombinant Casp1 to Casp10 (Table 1). NWL-117 inhibited effector Casp6, but not Casp3 or −7. NWL-117 inhibited initi- ator Casp8, Casp9, and Casp10, and inflammatory Casp4 but not Casp1 and Casp5. NWL-154 inhibited most strongly Casp6, Casp8 and Casp10, but not Casp3, Casp7, Casp9, Casp1, Casp4, or Casp5. Neither compounds showed inhibitory activity on Casp2. Casp6 IC50 in Fig. 1b differs slightly from these results be- cause the recombinant Casp1-10 were not active site titrated as done for Casp6 purified in-house. These re- sults indicate that NWL-117 and NWL-154 are strong NWL-117 and −154 are potent peptide-based vinyl methyl sulfone inhibitors of recombinant active Casp6 NWL-117 and −154 are potent peptide-based vinyl methyl sulfone inhibitors of recombinant active Casp6 NWL-117 and NWL-154 are peptide-based inhibitors flanked by a lipophilic moiety and a vinyl methyl sulfone chemical warhead (Fig. 1a). NWL-117 and −154 showed Fig. 1 NWL inhibitors are irreversible peptide vinyl sulfone inhibitors of Casp6. a Chemical structures of NWL-117 and NWL-154 highlighting the lipophilic moiety, the peptide backbone, and the chemical warhead. b Dose-response curve for NWL-117 (closed circle, IC50 = 192 nM) or NWL-154 (open circle, IC50 = 100 nM) against 20 nM recombinant active site-titrated Casp6. Data represent the mean ± S.D. for three independent experiments. c Schematic representation of the mechanism for covalent linkage of vinyl sulfone warheads to the catalytic cysteine of Casp6. 1 Peptide (VEID) binds to the substrate-binding pocket. 2 This interaction allows the sulfone moiety to form a hydrogen bond with the protonated imidazole ring of histidine. 3 The sulfur from the catalytic cysteine performs a nucleophilic attack on the β-carbon of the vinyl group, which triggers a movement of electrons leading to the protonation of the α-carbon. The result is a covalent link between the vinyl sulfone inhibitor and Casp6 Fig. 1 NWL inhibitors are irreversible peptide vinyl sulfone inhibitors of Casp6. a Chemical structures of NWL-117 and NWL-154 highlighting the lipophilic moiety, the peptide backbone, and the chemical warhead. b Dose-response curve for NWL-117 (closed circle, IC50 = 192 nM) or NWL-154 (open circle, IC50 = 100 nM) against 20 nM recombinant active site-titrated Casp6. Data represent the mean ± S.D. for three independent experiments. c Schematic representation of the mechanism for covalent linkage of vinyl sulfone warheads to the catalytic cysteine of Casp6. 1 Peptide (VEID) binds to the substrate-binding pocket. 2 This interaction allows the sulfone moiety to form a hydrogen bond with the protonated imidazole ring of histidine. 3 The sulfur from the catalytic cysteine performs a nucleophilic attack on the β-carbon of the vinyl group, which triggers a movement of electrons leading to the protonation of the α-carbon. The result is a covalent link between the vinyl sulfone inhibitor and Casp6 Fig. 1 NWL inhibitors are irreversible peptide vinyl sulfone inhibitors of Casp6. a Chemical structures of NWL-117 and NWL-154 highlighting the lipophilic moiety, the peptide backbone, and the chemical warhead. NWL inhibitors are non-toxic and prevent Casp6- dependent neuritic degeneration in APPWT-transfected human CNS neurons Human primary neurons are primary targets for caspase inhibitors in neurodegenerating brains, and therefore these were cells of choice to examine the potential tox- icity and caspase inhibition by these vinyl sulfone cas- pase inhibitors. Treatment with 100 μM NWL-117 or NWL-154 for 24 h did not change neuronal morphology (Fig. 4a), and showed mitochondrial reductive potential comparable to vehicle-treated neurons at 24 (Fig. 4b) or 48 h (Additional file 2: Figure S2a). These results indi- cate that neither NWL-117 nor −154 is toxic at concen- trations up to 100 μM in human neurons. Casp6 overexpression in HCT116 increases Casp6 ac- tivity fivefold (Fig. 2f), and 100 μM NWL-117 or −154 treatments for 2 h decreased Casp6 activity by 80% (IC50 4.82 μM; r2 = 0.94) and 96% (IC50 3.63 μM; r2 = 0.91), respectively (Fig. 2g-h). Functional inhibition of Casp6 activity was confirmed by a substantial decrease in TubΔCasp6 and a modest reduction of active Casp6 p20 subunit in cells treated with 100 μM NWL inhibitors (Fig. 2i). These findings suggest a functional inhibition of Casp6 by NWL inhibitors within HCT116 cells. p μ Casp6 fluorogenic (Fig. 4c) and fluorescent Casp6 in- hibitor (FLICA)-measured activity (Fig. 4d) and TubΔ- Casp6 (Fig. 4e) were decreased by both inhibitors in serum-deprived human neurons, thus supporting the ability of NWL-117 and −154 to inhibit intraneuronal Casp6 activity. Co-expression of APPWT and enhanced green fluorescent protein (EGFP) induced neuritic beading and somatic swelling after 48 h (Fig. 4f, g), as previously observed [12]. NWL-117, NWL-154, or 5 μM Z-VEID- fmk prevented neuritic beading (Fig. 4h). To study this ef- fect over time, pBudEGFP- (Additional file 2: Figure S3a) or pBudEGFP/APPWT − transfected neurons treated with ve- hicle (Fig. 4i) or NWL-117 (Additional file 2: Figure S3b) were assessed for beading by live fluorescent microscopy for a 72 h period. The percentage of beaded neurons increased significantly by 1.52 fold ± 0.07 with APPWT compared to EGFP-alone (1.00 ± 0.05) and decreased to 0.84 fold ± 0.13, 1.07 fold ± 0.08, 0.97 fold ± 0.14 with NWL-117, −154, or 5 μM Z-VEID-fmk treatment, respectively (Fig. 4j). Analyses with time indicate that the NWL-117-treated neurons have less beading com- pared to both pBudEGFP-transfected and pBudEGFP/ NWL-117 and −154 are non-toxic and potent inhibitors of Casp6 activity in Casp6-transfected HCT116 cells To address NWL inhibitor efficacy in a cellular context, human colon carcinoma HCT116 cells were treated with 100 μM NWL-117, NWL-154 or staurosporine as a cell death control for 24 h (Fig. 2a). NWL inhibitor-treated cells looked morphologically normal. The mitochondrial reductive potential of cells treated with 20, 50, or 100 μM NWL-117 or −154 for 24 (Fig. 2b) or 48 h (Fig. 2c) was comparable to vehicle-treated cells. NWL- treated cells did not release lactase dehydrogenase (LDH) (Fig. 2d) nor did the cells show increased sub-G1 levels of DNA (Fig. 2e), excluding necrosis and apop- tosis, respectively. Therefore, unlike staurosporine, NWL inhibitors are not toxic to HCT116 cells at the tested concentrations. NWL inhibitors are non-toxic and prevent Casp6- dependent neuritic degeneration in APPWT-transfected human CNS neurons NWL-117 and −154 are potent peptide-based vinyl methyl sulfone inhibitors of recombinant active Casp6 b Dose-response curve for NWL-117 (closed circle, IC50 = 192 nM) or NWL-154 (open circle, IC50 = 100 nM) against 20 nM recombinant active site-titrated Casp6. Data represent the mean ± S.D. for three independent experiments. c Schematic representation of the mechanism for covalent linkage of vinyl sulfone warheads to the catalytic cysteine of Casp6. 1 Peptide (VEID) binds to the substrate-binding pocket. 2 This interaction allows the sulfone moiety to form a hydrogen bond with the protonated imidazole ring of histidine. 3 The sulfur from the catalytic cysteine performs a nucleophilic attack on the β-carbon of the vinyl group, which triggers a movement of electrons leading to the protonation of the α-carbon. The result is a covalent link between the vinyl sulfone inhibitor and Casp6 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Page 8 of 19 Table 1 Half-maximal inhibitory concentrations (IC50)a of NWL-117 and NWL-154 against recombinant Caspase-1 to −10 Cpd No. MWb Caspase- 1 2 3 4 5 6 7 8 9 10 NWL-117 668 2.7 >100 6.96 0.38 19.54 0.60 >100 0.66 0.79 0.13 NWL-154 703 3.85 >100 1.34 1.80 4.19 0.23 >100 0.53 6.40 0.19 aIC50 values are measured in μM bMolecular weight in g/mol Table 1 Half-maximal inhibitory concentrations (IC50)a of NWL-117 and NWL-154 against recombinant Caspase-1 to −10 Cpd No MWb Caspase Table 1 Half-maximal inhibitory concentrations (IC50)a of NWL-117 and NWL-154 against recombinant Caspase-1 to −10 Cpd No. MWb Caspase- 1 2 3 4 5 6 7 8 9 NWL-117 668 2.7 >100 6.96 0.38 19.54 0.60 >100 0.66 0.79 NWL-154 703 3.85 >100 1.34 1.80 4.19 0.23 >100 0.53 6.40 aIC l d i M Table 1 Half-maximal inhibitory concentrations (IC50)a of NWL-117 and NWL-154 against recombinant Caspase-1 to −10 Casp6 inhibitors although their selectivity requires improvement. NWL-117 removal and recovered 85% ± 2.5 of the initial activity after 2 h (Fig. 3d). In contrast, NWL-154 with- drawal increased by 21% ± 2.2 within 15 min, reaching 30% ± 15 after 2 h. Western blot analysis showed that the levels of TubΔCasp6 and Casp6p20 (Fig. 3e–f) were consistent with the VEIDase activity restoration (Fig. 3d). Together, these results indicate that the NWL inhibitors rapidly inhibit Casp6 activity and that this inhibition can be rapidly washed out in transfected HCT116 cells. NWL inhibitors block Casp6 activity within minutes in Casp6-transfected HCT116 cells No statistical differences were found by two-way ANOVA with Bonferroni post-tests. d Lactate dehydrogenase activity in untreated cells, or treated with PBS, 100 μM NWL inhibitors, or lysed with 0.9% Triton X-100 for 24 h. e Quantification of the sub-G1 population following cell cycle analysis in PBS, 100 μM NWL inhibitors, or 2 μM staurosporine treatment for 24 h. f VEIDase activity in pCep4β-transfected (mock) or pCep4β-Casp6p20p10 transfected HCT116 cells. Data represent the mean ± SEM of five independent experiments. Statistical analysis was performed using an unpaired two-tailed t-test (* p < 0.01). g Casp6 VEIDase activity in cellular extracts from pCep4β-Casp6p20p10-transfected HCT116 cells treated with PBS, NWL-117 or NWL-154 at 100 μM for 2 h. h Dose-response curve for NWL-117 (closed circle, IC50 = 4.82 μM) and NWL-154 (open circle, IC50 = 3.63 μM) in pCep4β-Casp6p20p10-transfected HCT116 cells treated for 2 h. i Western blot analysis of samples from panel f and g for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. Casp6 expression was allowed for 24 h before treatment with inhibitors in all transfection experiments. For panels b-g, data represent the mean of three independent experiments ± SEM and were analyzed by one-way ANOVA (p < 0.0001) with post hoc Dunnett’s multiple comparison test comparing to vehicle-treated (denotes p < 0.001) unless specified otherwise than NWL-154 against neuronal beading and rounding in the APPWT -transfected human neurons. APPWT − transfected neurons (Fig. 4k). This protection is at- tenuated after 24 h because media was not replenished with the NWL-117 to avoid loosing the settings for the time-lapse microscopy. Similarly, neuronal soma rounding was inhibited by NWL-117 (Fig. 4l). Finally, a measure of neurons with homogeneously distributed EGFP as a meas- ure of health shows that NWL-117 significantly increases neuronal survival under these conditions (Fig. 4m) There- fore, NWL-117 showed stronger neuroprotective effects NWL inhibitors block Casp6 activity within minutes in Casp6-transfected HCT116 cells To assess the kinetics of NWL-117 and −154-mediated Casp6 inhibition, Casp6-expressing HCT116 cells were treated with 100 μM NWL-117 or NWL-154 for 15 to 120 min after 24 h of transfection. Within 15 min of treatment, Casp6 VEIDase activity decreased signifi- cantly by 88% with NWL-117 and 95% with NWL-154 and the inhibition continued for 2 (Fig. 3a) and 48 h (Additional file 2: Figure S1). TubΔCasp6 and Casp6p20 levels were decreased within 15 min of treatment (Fig. 3b, c). To assess Casp6 activity recovery, transfected cells were treated with 100 μM NWL-117 or NWL-154 for 2 h and subsequently replaced with fresh media. Casp6 activity rose by 43% ± 5.0 within the first 15 min of Page 9 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Fig. 2 Non-toxic concentrations of NWL-117 and −154 inhibit Casp6 activity in HCT116 cells. a Phase contrast microscope images of HCT116 cells treated with phosphate-buffered saline, 100 μM NWL inhibitors, or 2 μM staurosporine for 24 h. Scale bar represents 10 μm. b & c MTT absorbance following treatment with PBS, NWL-117, or NWL-154 at 20, 50, or 100 μM for 24 h (b) or 48 h (c). No statistical differences were found by two-way ANOVA with Bonferroni post-tests. d Lactate dehydrogenase activity in untreated cells, or treated with PBS, 100 μM NWL inhibitors, or lysed with 0.9% Triton X-100 for 24 h. e Quantification of the sub-G1 population following cell cycle analysis in PBS, 100 μM NWL inhibitors, or 2 μM staurosporine treatment for 24 h. f VEIDase activity in pCep4β-transfected (mock) or pCep4β-Casp6p20p10 transfected HCT116 cells. Data represent the mean ± SEM of five independent experiments. Statistical analysis was performed using an unpaired two-tailed t-test ( p < 0.01). g Casp6 VEIDase activity in cellular extracts from pCep4β-Casp6p20p10-transfected HCT116 cells treated with PBS, NWL-117 or NWL-154 at 100 μM for 2 h. h Dose-response curve for NWL-117 (closed circle, IC50 = 4.82 μM) and NWL-154 (open circle, IC50 = 3.63 μM) in pCep4β-Casp6p20p10-transfected HCT116 cells treated for 2 h. i Western blot analysis of samples from panel f and g for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. Casp6 expression was allowed for 24 h before treatment with inhibitors in all transfection experiments. NWL inhibitors block Casp6 activity within minutes in Casp6-transfected HCT116 cells For panels b-g, data represent the mean of three independent experiments ± SEM and were analyzed by one-way ANOVA (p < 0.0001) with post hoc Dunnett’s multiple comparison test comparing to vehicle treated (denotes p < 0001) unless specified otherwise Fig. 2 Non-toxic concentrations of NWL-117 and −154 inhibit Casp6 activity in HCT116 cells. a Phase contrast microscope images of HCT116 cells treated with phosphate-buffered saline, 100 μM NWL inhibitors, or 2 μM staurosporine for 24 h. Scale bar represents 10 μm. b & c MTT absorbance following treatment with PBS, NWL-117, or NWL-154 at 20, 50, or 100 μM for 24 h (b) or 48 h (c). No statistical differences were found by two-way ANOVA with Bonferroni post-tests. d Lactate dehydrogenase activity in untreated cells, or treated with PBS, 100 μM NWL inhibitors, or lysed with 0.9% Triton X-100 for 24 h. e Quantification of the sub-G1 population following cell cycle analysis in PBS, 100 μM NWL inhibitors, or 2 μM staurosporine treatment for 24 h. f VEIDase activity in pCep4β-transfected (mock) or pCep4β-Casp6p20p10 transfected HCT116 cells. Data represent the mean ± SEM of five independent experiments. Statistical analysis was performed using an unpaired two-tailed t-test ( p < 0.01). g Casp6 VEIDase activity in cellular extracts from pCep4β-Casp6p20p10-transfected HCT116 cells treated with PBS, NWL-117 or NWL-154 at 100 μM for 2 h. h Dose-response curve for NWL-117 (closed circle, IC50 = 4.82 μM) and NWL-154 (open circle, IC50 = 3.63 μM) in pCep4β-Casp6p20p10-transfected HCT116 cells treated for 2 h. i Western blot analysis of samples from panel f and g for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. Casp6 expression was allowed for 24 h before treatment with inhibitors in all transfection experiments. For panels b-g, data represent the mean of three independent experiments ± SEM and were analyzed by one-way ANOVA (p < 0.0001) with post hoc Dunnett’s multiple comparison test comparing to vehicle-treated (denotes p < 0.001) unless specified otherwise Fig. 2 Non-toxic concentrations of NWL-117 and −154 inhibit Casp6 activity in HCT116 cells. a Phase contrast microscope images of HCT116 cells treated with phosphate-buffered saline, 100 μM NWL inhibitors, or 2 μM staurosporine for 24 h. Scale bar represents 10 μm. b & c MTT absorbance following treatment with PBS, NWL-117, or NWL-154 at 20, 50, or 100 μM for 24 h (b) or 48 h (c). NWL-117 penetrates the blood-brain barrier and reaches high nanomolar concentrations in mouse brains To assess the blood-brain barrier permeability of NWL- 117, 18 month old Casp6-expressing transgenic mice were injected via the carotid artery. Liquid chromatog- raphy/tandem mass spectrometry (LC/MS-MS) analyses Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Page 10 of 19 Fig. 3 NWL-117 and −154 rapidly inhibit Casp6 activity in HCT116 cells. a Percent VEIDase activity from cellular protein extracts of Casp6-transfected HCT116 cells treated with either NWL-117 (closed triangle) or −154 (closed circle) at 100 μM for 0, 15, 30, 60, 90, or 120 min. No statistical significant differences were obtained between NWL-117 and NWL-154. b & c Western blot analysis of samples from panel (a) of NWL-117 (b) and NWL-154 (c) for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. d Percent VEIDase activity from cellular extracts after 2 h of treatment with 100 μM NWL-117 (closed triangle) or −154 (closed circle) in Casp6-transfected HCT116 cells followed by the removal of the inhibitors for 0, 15, 30, 60, 90, or 120 min. e & f Western blot analysis of samples from panel (d) of NWL-117 (e) and NWL-154 (f) for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. For panels (a) & (d), data represent the mean ± SEM of three independent experiments. Statistical analysis was performed by two-way ANOVA (((compound (p = 0.0010), T116 cells a Percent VEIDase activity from cellular protein extracts of Casp6 transfected Fig. 3 NWL-117 and −154 rapidly inhibit Casp6 activity in HCT116 cells. a Percent VEIDase activity from cellular protein extracts of Casp6-transfected HCT116 cells treated with either NWL-117 (closed triangle) or −154 (closed circle) at 100 μM for 0, 15, 30, 60, 90, or 120 min. No statistical significant differences were obtained between NWL-117 and NWL-154. b & c Western blot analysis of samples from panel (a) of NWL-117 (b) and NWL-154 (c) for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. d Percent VEIDase activity from cellular extracts after 2 h of treatment with 100 μM NWL-117 (closed triangle) or −154 (closed circle) in Casp6-transfected HCT116 cells followed by the removal of the inhibitors for 0, 15, 30, 60, 90, or 120 min. e & f Western blot analysis of samples from panel (d) of NWL-117 (e) and NWL-154 (f) for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. NWL-117 penetrates the blood-brain barrier and reaches high nanomolar concentrations in mouse brains For panels (a) & (d), data represent the mean ± SEM of three independent experiments. Statistical analysis was performed by two-way ANOVA (((compound (p = 0.0010), time (p < 0.0001), interaction (p = 0.4546)); (compound (p < 0.0001), time (p < 0.0001), interaction (p = 0.0054))), respectively, with Bonferroni post-tests (p < 0.05, p < 0.01, p < 0.001) Fig. 3 NWL-117 and −154 rapidly inhibit Casp6 activity in HCT116 cells. a Percent VEIDase activity from cellular protein extracts of Casp6-transfected HCT116 cells treated with either NWL-117 (closed triangle) or −154 (closed circle) at 100 μM for 0, 15, 30, 60, 90, or 120 min. No statistical significant differences were obtained between NWL-117 and NWL-154. b & c Western blot analysis of samples from panel (a) of NWL-117 (b) and NWL-154 (c) for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. d Percent VEIDase activity from cellular extracts after 2 h of treatment with 100 μM NWL-117 (closed triangle) or −154 (closed circle) in Casp6-transfected HCT116 cells followed by the removal of the inhibitors for 0, 15, 30, 60, 90, or 120 min. e & f Western blot analysis of samples from panel (d) of NWL-117 (e) and NWL-154 (f) for α-tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), active Casp6 p20 subunit (Casp6p20), and β-actin. For panels (a) & (d), data represent the mean ± SEM of three independent experiments. Statistical analysis was performed by two-way ANOVA (((compound (p = 0.0010), time (p < 0.0001), interaction (p = 0.4546)); (compound (p < 0.0001), time (p < 0.0001), interaction (p = 0.0054))), respectively, with Bonferroni post-tests (p < 0.05, p < 0.01, p < 0.001) confirmed with Evan’s blue. These results demonstrate the ability of NWL-117 to cross the blood-brain barrier in mice. showed hippocampal concentrations ranging from 67.9 nM to 879 nM, while plasma concentrations ranged from 3.4 μM to 48 μM, after 5 minutes of injection (Table 2). The ratio between hippocampal and plasma concentrations suggested low brain penetrance, although levels greater than the in vitro IC50 against Casp6 (192 nM) and some initiator caspases (Table 1) were reached. Variability was expected due to the unpredictable effects of surgery on old mice. Blood-brain barrier integrity was Treatment of human Casp6 knock-in mice with NWL-117 improves their performance in the novel object recognition (NOR) task Treatment of human Casp6 knock-in mice with NWL-117 improves their performance in the novel object recognition (NOR) task Human Casp6 overexpression in the CA1 region of the hippocampus results in age-dependent episodic memory Page 11 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Fig. 4 Non-toxic concentrations of NWL-117 and −154 inhibit Casp6 activity in primary human neurons. a Micrographs of primary human neurons treated with PBS vehicle, 100 μM NWL inhibitors, or 2 μM staurosporine for 24 h. Scale bar represents 10 μm. b MTT absorbance following treatment with PBS, NWL-117, or NWL-154 at 20, 50, or 100 μM or 2 μM staurosporine for 24 h (n = 4, one-way ANOVA (p = 0.0169), Tukey’s multiple comparison test (p < 0.05)). c VEIDase activity in neuronal extracts following treatment with PBS, 100 μM NWL-117 (n = 5), or 154 (n = 2) for 2 h (one-way ANOVA (p < 0.0001)). d Casp6 FLICA assay (one-way ANOVA (p = 0.0041)). e Quantification of the number of TubΔCasp6 beads/nuclei in Additional file 1: Figure S2b (one-way ANOVA (p = 0.0095)). f–h Fluorescence micrographs following transfections with pBudEGFP (f) or pBudEGFP/APPWT (g) stained for α-tubulin (Cy3), Hoechst, and quantified (h) (one-way ANOVA (p = 0.0385)). Scale bar represents 100 μm for merge and Hoechst panels, and 50 μm for EGFP panel. i Live-imaging fluorescence micrographs from 0 to 60 h of a human neuron transfected with pBudEGFP/APPWT and pre-treated with vehicle. The inset highlights the neurite extending upward. Arrowheads indicate agglomerates of EGFP protein within the axonal membrane while arrows mark a rounded cell body. Scale bar represents 10 μm. j–m Quantification of panel (i) for overall fold increased beaded neurites (j), fold increased beaded neurons at specific times (k), overall fold increase swollen neuronal soma (l), or normal EGFP positive neurons (m). For panels (c–e, h), and (j–k) data represent the mean ± SEM (n ≥3), one-way ANOVA (p = 0.0005 for j, p =0.0264 for k, and post hoc tests were performed with Dunnett’s multiple comparison test (compares to serum (+) or EGFP-vehicle: p < 0.05, p < 0.01, p < 0.001; # compares to serum (−) with vehicle or to EGFP/APPWT-vehicle: # p < 0.05, ## p < 0.01, ### p < 0.001) unless stated otherwise. For panel (m), log-rank Mantel-Cox test was performed to compare between curves Fig. Discussion Our study demonstrates that NWL inhibitors are 1) non-toxic, but non-selective, strong Casp6 inhibitors in vitro, in colon cancer cells, and in primary CNS human neurons, 2) protective against Casp6-mediated neuritic degeneration in serum-deprived or APPWT expressing human neurons, 3) blood-brain barrier permeable, and 4) reversing episodic memory impairments in transgenic Casp6 mice. impairments measured by NOR [9]. Casp6 KI/Cre mice were tested following the experimental paradigm shown in Fig. 5a. Control mice spent more time with the novel object (70% ± 1.6), while KI/Cre mice did not (47% ± 3.6) (Fig. 5b) in the pre-test. Total path length (Fig. 5c), the percentage of time moving (Fig. 5d), and total number of entries in each part of the arena (Fig. 5e) were equiva- lent in control and KI/Cre mice indicating comparable locomotor and exploratory activities. Following two in- traperitoneal injections, NWL-117-treated control mice performed equally to saline-treated mice (Fig. 5f). In contrast, saline-treated KI/Cre mice remained impaired (Fig. 5g), whereas NWL-117-treated mice regained nor- mal NOR performance (Fig. 5h). Injections had no effect on the locomotor or exploratory activities (Fig. 5i–k). Hence, age-dependent Casp6-mediated deficits in NOR can be overcome by an acute treatment with NWL-117. There are several advantages to these vinyl sulfone in- hibitors. NWL-117 and NWL-154 are potent, non-toxic, but non-selective Casp6 inhibitors. Both NWL-117 and NWL-154 inhibited 1) recombinant Casp6 activity (IC50 = 192 nM and 100 nM, respectively), 2) Casp6 activity in Casp6-transfected HCT116 cells (IC50 = 4.82 μM and 3.63 μM, respectively), and 3) Casp6 activity in serum- deprived human neurons. In contrast to Z-VEID-fmk [42], vinyl sulfone inhibitors remain intact after target engagement [43], and are not toxic to mammals [44]. Similarly, NWL vinyl sulfone inhibitors do not cause cel- lular toxicity measured by mitochondrial activity, cell morphology, LDH release, or sub-G1 populations, or any gross physiological or anatomical changes in vivo (supplementary document: pathology report). Further- more, NWL inhibitors non-covalently interact with the substrate-binding pocket of Casp6 and effectively block other substrates from entering the active site. In addition, by bringing the weak vinyl sulfone electrophilic warhead, near the catalytic histidine and cysteine resi- dues of Casp6, Casp6 enzymatic activity is irreversibly blocked. Compared to reversible inhibitors, irreversible inhibitors can achieve higher potency by completely inhibiting their target and require less frequent and lower doses resulting in higher safety profiles [52]. Treatment of human Casp6 knock-in mice with NWL-117 improves their performance in the novel object recognition (NOR) task 4 Non-toxic concentrations of NWL-117 and −154 inhibit Casp6 activity in primary human neurons. a Micrographs of primary human neurons treated with PBS vehicle, 100 μM NWL inhibitors, or 2 μM staurosporine for 24 h. Scale bar represents 10 μm. b MTT absorbance following treatment with PBS, NWL-117, or NWL-154 at 20, 50, or 100 μM or 2 μM staurosporine for 24 h (n = 4, one-way ANOVA (p = 0.0169), Tukey’s multiple comparison test (p < 0.05)). c VEIDase activity in neuronal extracts following treatment with PBS, 100 μM NWL-117 (n = 5), or 154 (n = 2) for 2 h (one-way ANOVA (p < 0.0001)). d Casp6 FLICA assay (one-way ANOVA (p = 0.0041)). e Quantification of the number of TubΔCasp6 beads/nuclei in Additional file 1: Figure S2b (one-way ANOVA (p = 0.0095)). f–h Fluorescence micrographs following transfections with pBudEGFP (f) or pBudEGFP/APPWT (g) stained for α-tubulin (Cy3), Hoechst, and quantified (h) (one-way ANOVA (p = 0.0385)). Scale bar represents 100 μm for merge and Hoechst panels, and 50 μm for EGFP panel. i Live-imaging fluorescence micrographs from 0 to 60 h of a human neuron transfected with pBudEGFP/APPWT and pre-treated with vehicle. The inset highlights the neurite extending upward. Arrowheads indicate agglomerates of EGFP protein within the axonal membrane while arrows mark a rounded cell body. Scale bar represents 10 μm. j–m Quantification of panel (i) for overall fold increased beaded neurites (j), fold increased beaded neurons at specific times (k), overall fold increase swollen neuronal soma (l), or normal EGFP positive neurons (m). For panels (c–e, h), and (j–k) data represent the mean ± SEM (n ≥3), one-way ANOVA (p = 0.0005 for j, p =0.0264 for k, and post hoc tests were performed with Dunnett’s multiple comparison test (compares to serum (+) or EGFP-vehicle: p < 0.05, p < 0.01, p < 0.001; # compares to serum (−) with vehicle or to EGFP/APPWT-vehicle: # p < 0.05, ## p < 0.01, ### p < 0.001) unless stated otherwise. For panel (m), log-rank Mantel-Cox test was performed to compare between curves Pakavathkumar et al. Discussion Therefore, the potential that NWL vinyl sulfone inhibi- tors could be used in humans is high. On the other hand, as with other active-site directed Casp6 inhibitors [18, 34–37, 41, 53, 54], NWL inhibitors remain non- selective for Casp6 since they inhibit many other cas- pases. Reducing the concentration of NWL inhibitors can increase selectivity, but significant enhancements in potency and specificity are still required before these in- hibitors reach clinical trials. To overcome this limitation, the unique inactive conformation of Casp6 was targeted by others [55]. The peptide inhibitor, pep419, targets and stabilizes the tetrameric inactive form of Casp6 in a pH-dependent non-competitive manner in vitro and in Treatment of human Casp6 knock-in mice with NWL-117 improves their performance in the novel object recognition (NOR) task Molecular Neurodegeneration (2017) 12:22 Page 12 of 19 Page 12 of 19 Table 2 NWL-117 levels in mice hippocampi and plasma following carotid artery injectionsa Animal # Hippocampus Plasma Ratiob 1 193 48521 0.004 2 816 3416 0.239 3 67.9 8998 0.008 4 356 7408 0.048 5 879 25425 0.035 aConcentrations are reported in nM bRatio of [hippocampal]/[plasma] Table 2 NWL-117 levels in mice hippocampi and plasma following carotid artery injectionsa treated or not with NWL-117 (Fig. 6m-o). Astroglial glial fibrillary acidic protein (GFAP) levels also did not change with NWL-117 treatment (Fig. 6p-q). Thus, analyses could not detect changes in synaptic proteins, Casp6 substrates, or inflammatory markers that account for the behavioral improvement seen in NWL-117-treated KI/Cre mice. Casp6 substrates, synaptic proteins, and glial inflammation markers are unchanged in mice hippocampi following NWL-117 treatment In our study, the possibil- ity that NWL inhibitors are acting on Casp3 to protect neurons was excluded because only Casp1 and Casp6 are co-activated in our cellular model [12, 68, 69], and both NWL-117 and NWL-154 are more effective against Casp6 than Casp1 and Casp3. Moreover, in AD brains, active Casp6 is detected in the absence of Casp3 [6, 70]. Thus, although studies in mouse peripheral neuron cul- tures implicate Casp3 to be an important regulator of axonal degeneration, the pathways involved in human CNS neurons seem to converge on Casp6. Nevertheless, given the strong inhibition of initiator caspases by the NWL vinyl sulfone caspase inhibitors, it is not possible in these experiments to conclude that the effect ob- served was uniquely due to Casp6. The ability of short- term treatment with NWL-117 to reverse episodic The selectivity of VEID is controversial but it remains the best candidate for targeting the active site of Casp6. The study by McStay et al. [60] does suggest that VEID can be cleaved by recombinant Caspase-3 or by Caspase-3 in extracts from Jurkat cells undergoing in- trinsic apoptosis after the addition of cytochrome c and ATP, which activates the apoptosome pathway. However, other groups have shown that Casp6 is better at cleaving VEID than Caspase-3 [61, 62]. In fact, the Michaelis- Menten constant (Km) for VEID is 8-fold lower for re- combinant Casp6 (30 μm) than it is for Caspase-3 (250 μm) [37]. This suggests that VEID binds the active site of Casp6 with greater affinity than that of Caspase-3. Similarly, the IC50 and the inhibitory constant (Ki) of z- VEID-CHO (aldehyde) are 2-fold smaller for Casp6 than they are for Caspase-3 [58]. These data suggest that VEID is a preferred substrate of Casp6, although not specific. Apart from the peptide sequence, the other components of the small molecule also influence its se- lectivity. This is evident in [63] where screening of pep- tide acyloxymethyl ketones (AOMK) inhibitors resulted in the identification of TETD as the preferred peptide sequence for Casp6 over Caspase-3 [41]. Yet, in the same study, they found that Cy5 labeled VEID-AOMK was a better substrate for Casp6 than the TETD version. Similarly, chemical warheads can affect the selectivity of inhibitors bearing the same peptide sequence. In fact, VEID-CHO was more selective for Casp6 than −3 com- pared to the fluoromethyl ketone (FMK) counterpart [64]. Casp6 substrates, synaptic proteins, and glial inflammation markers are unchanged in mice hippocampi following NWL-117 treatment Statistical analysis was performed by repeated measures ANOVA (p = 0.0860) with Bonferroni’s multiple comparison test ( p < 0.05). i Distance traveled, (j) percent time moving, and (k) number of cell entries of control mice injected with saline (n = 8) or NWL-117 (n = 8) and KI/Cre injected with saline (n = 4) or NWL-117 (n = 5). Statistical analysis was performed by two-way ANOVA for panels (i) to (k) and no significant differences were found. For panels (a–k), data represent the mean ± SEM and post hoc analyzes were performed using Bonferroni’s multiple comparison test (p < 0.05, p < 0.01, and **p < 0.001) unless stated otherwise possible that exosites are responsible for limiting the cleavage of lamin A at the VEID sequence to Casp6 [62]. In our study, we find Z-VEID vinyl methyl sulfone inhib- itors are 10-fold more selective against Casp6 than Caspase-3 (Table 1: IC50). It is possible that interactions of the lipophilic moiety or the chemical warhead with natural substrate exosites increase selectivity of the NWL inhibitors further towards Casp6. cells [56]. Similarly, non-competitive ligands that stabilize the L2 loop of Casp6, which normally rearranges during activation [57], and a potent uncompetitive inhibitor tar- geting the caspase-substrate interface [58], in vitro, are effective Casp6 inhibitors. Through these innovative mechanisms, highly specific inhibitors have emerged, yet remain to be tested for toxicity and efficiency in cells and in mice. Nevertheless, active site inhibitors can be chem- ically modified to reach exquisite selectivity against spe- cific caspases [59]. Our results demonstrate that the NWL vinyl sulfone caspase inhibitors are non toxic to human neurons and neuroprotective against serum deprivation or APP over- expression. NWL inhibitors prevent serum-deprivation- or APPWT-expression induced TubΔCasp6 and neuritic degeneration in human neurons, as shown previously with Z-VEID-fmk and Casp6 dominant negative inhibi- tors [12]. Maintaining full-length α-tubulin is essential to stabilize microtubules [66], and intact microtubules are critical for neuronal function. Since active Casp6 or TubΔCasp6 are increased in human AD and hypoxia- induced ischemia [4, 10, 27], inhibition of Casp6 and other caspases in these conditions may help maintain neuronal function. Even if Casp6 has been implicated in axonal degeneration of NGF-dependent neurons, recent evidence suggests that Casp3 also participates in axonal degeneration [15, 19–26, 67]. Casp6 substrates, synaptic proteins, and glial inflammation markers are unchanged in mice hippocampi following NWL-117 treatment Western blot analyses (Fig. 6a) showed that old KI/Cre mice overexpressed Casp6 but TubΔCasp6 levels remained below detection, as expected since neurons are likely degenerated. TubΔCasp6 positive immunohistochemical staining in the CA1 region (Fig. 6b, c) was slightly higher in saline KI/Cre mice compared to controls (Fig. 6d). NWL-117-treatment non-significantly decreased TubΔ- Casp6 levels threefold in control, but increased non- significantly in KI/Cre mice. Valosin-containing protein p97 cleaved by Casp6 (Δp97) (Fig. 6e) was not different in saline and NWL-117 treatments (Fig. 6f). Synapsin, synaptophysin, and post-synaptic protein (PSD95) hippocampal protein levels were unchanged with NWL-117 treatment (Fig. 6g–i, Additional file 2: Figure S4). Synaptophysin immunoreactivity in the hippo- campal CA1 region was slightly higher in saline-treated control than KI/Cre mice, and unchanged in NWL-117- treated mice (Fig. 6j–l). Protein analyses of mouse hippocampi showed a non- significant reduction in microglial ionized calcium binding adapter molecule 1 (Iba1) levels in KI/Cre mice, whether Page 13 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Fig. 5 (See legend on next page.) Fig. 5 (See legend on next page.) Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Page 14 of 19 (See figure on previous page.) Fig. 5 Acute NWL-117 administration reverses novel object recognition deficits in Casp6-overexpressing KI/Cre mice. a Experimental design for the in vivo study highlighting the novel object recognition (NOR) tests and injections. b Percent touches of objects during the NOR task in WT/ WT and WT/Cre controls (n = 16), and Casp6-expressing KI/Cre (n = 9) mice prior to injections. Statistical analysis was performed by one-way ANOVA (p < 0.0001). c Distance traveled, (d) percent time moving, and (e) number of cell entries of control (n = 16) and KI/Cre (n = 9) mice. Statistical analysis was performed by unpaired two-tailed t test. No significant differences were found in C-E. f Percent touches of objects during the NOR task following saline (n = 8) or 20 mg/Kg NWL-117 (n = 8) injections in control mice. Statistical analysis was performed by one-way ANOVA (p < 0.0001). g Percent touches of objects during the NOR task in pre- and post-injections (n = 4) saline injections in KI/Cre mice. Statistical analysis was performed by repeated measures ANOVA (p = 0.7661). h Percent touches of objects during the NOR task in pre- and post- (n = 5) 20 mg/Kg NWL-117 injections in KI/Cre mice. Casp6 substrates, synaptic proteins, and glial inflammation markers are unchanged in mice hippocampi following NWL-117 treatment SO: Stratum Oriens, PCL: Pyramidal Cell layer, SR: Stratum Radiatum, SLM: Stratum Lacunosum memory impairments in our mice suggests that Casp6- mediated damage is reversible in aged mice. We did not determine whether other caspases are activated down- stream of Casp6 in our mouse model and it remains a possibility that the inhibition of other caspases such as Casp4, 8, 9, and 10 additionally contributed to the im- provement of cognitive deficits mediated by the over- expression of Casp6. Nevertheless, the development of specific Casp6 inhibitors and assessment of target en- gagement will help determine whether it is Casp6 inhib- ition alone, a combination of Casp6 with other caspases activated as a consequence of Casp6 activation in the mice brains, or a non-caspase effect that is responsible for the behavioural improvement. Most importantly, our findings suggest that vinyl sulfone NWL caspase inhibitors are permeable to the blood-brain barrier with concentra- tions in the hippocampus reaching in vitro IC50 values. Intra-carotid injections were used as a proof of principle as it limits exposure to peripheral tissues and is the most rapid path to the brain. Rapid exposure was necessary as the LC/MS-MS method can only detect free NWL-117. Pharmacokinetic and pharmacodynamic detailed analyses should be conducted in order to determine dosing regi- mens for chronic administration studies. Only one other Casp6 inhibitor, ED11, was shown to be brain permeable and have a significant effect against behavioral and cogni- tive deficits in an Huntington’s mouse model [18]. ED11 was delivered by subcutaneous pump delivery of 4 mg/kg/ day for 28 days or more. In contrast, NWL-117 reversed Casp6-induced memory deficits after only 2 intraperito- neal injections of 20 mg/kg within 72 h in the Casp6 transgenic mouse. These results support delivery of the NWL inhibitors to the brain and suggest that Casp6- mediated functional impairment can be rapidly reversed. to plasma membranes and the blood-brain barrier, while the huntingtin sequence is used to target Casp6. NWL in- hibitors have 1) lower molecular weight, 2) lower IC50 against VEID substrates, and 3) no activity enhancing ef- fect on caspases compared to ED11. Lower molecular weight is associated with several different parameters that determine the oral bioavailability of compounds as well as their production cost [71]. In addition, large peptides have lower half-lives in the body due to extensive degradation and clearance by the liver and kidneys. Casp6 substrates, synaptic proteins, and glial inflammation markers are unchanged in mice hippocampi following NWL-117 treatment Exosites also modulate substrate binding [65]. It is Page 15 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Fig. 6 (See legend on next page.) Fig. 6 (See legend on next page.) Page 16 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 (See figure on previous page.) Fig. 6 Hippocampal levels of synaptic and glial markers are unchanged with NWL-117 treatment. a Levels of Casp6 p20p10 (Casp6p20p10), α- tubulin-cleaved by Casp6 (TubΔCasp6), α-tubulin (Tubulin), and β-actin in hippocampal protein extracts WT/WT, WT/Cre, and KI/Cre mice treated with saline or 20 mg/Kg NWL-117. b & c Immunohistochemistry of TubΔCasp6-stained hippocampi from KI/Cre mice treated with saline (b) or NWL-117 (c). Arrowheads mark some positive immunoreactivity. d Quantification of positive immunostaining shown in panel (b) and (c). e Levels of p97 and p97-cleaved by Casp6 (p97ΔCasp6) in hippocampal protein extracts of KI/Cre mice treated with saline (n = 3) or NWL-117 (n = 3). f Quantification of the levels of p97ΔCasp6/p97 shown in panel (e). g & h Levels of synapsin in hippocampal extracts from KI/Cre mice treated with saline (n = 3) or NWL-117 (n = 3) (g) and quantified in (h). i Levels of Synaptophysin in hippocampal protein extracts from WT/WT, WT/Cre, and KI/ Cre (Casp6 overexpressing) mice treated with saline or NWL-117. j–k Brightfield scans of KI/Cre-Saline (j) and NWL-117 (k) mice brains stained for synaptophysin by immunohistochemistry and quantification (l). m & n Iba1-stained hippocampi from KI/Cre mice treated with saline (m) or NWL- 117 (n). Arrowheads mark some positive immunoreactivity. o Quantification of the area of positive immunostaining for Iba1 over the total area of the tissue. p Levels of GFAP in hippocampal extracts from KI/Cre mice treated with saline (n = 3) or NWL-117 (n = 3). q Quantification of the levels of GFAP/β-actin. For panels (d), (l), and (o), data represent the mean ± SEM for each group: Control-saline (n = 5), Control-NWL-117 (n = 5), KI/Cre-sa- line (n = 5), and KI/Cre-NWL-117 (n = 4). Statistical analysis was performed by two-way ANOVA and no significant differences were found. For panels (f), (h), and (q), data represent the mean ± SEM and statistical analysis was performed by unpaired two-tailed t test, no significant differences were found. Casp6 substrates, synaptic proteins, and glial inflammation markers are unchanged in mice hippocampi following NWL-117 treatment This often leads to the use of parenteral routes of administration, like injections for insulin, which subsequently leads to reduced patient compliance. Thus, the development of small mole- cules is often preferred. In addition, although the IC50 against mutant huntingtin cleavage determined by fluores- cence resonance energy transfer (FRET) was 12.12 nM for ED11, ED11 did not inhibit the cleavage of VEID- aminoluciferin as potently (>10 μM). Like NWL inhibi- tors, ED11 showed inhibition of other caspases, but the IC50 of ED11 against all caspases on their preferred sub- strates has not been determined. Therefore, it is not pos- sible to compare the selectivity of ED11 to that of NWL. Furthermore, ED11 showed a significant enhancement of Caspase-5 activity on the FRET assay. Deregulated Caspase-5 activation could perturb inflammasome signal- ling [72]. No activation of caspases was observed with NWL inhibitors. Finally, ED11 is a competitive reversible inhibitor that gets cleaved by Casp6. Although it has not been measured, the affinity of cleaved ED11 could be lower than the parent compound. Thus, the efficacy of ED11 will be reduced over time. In contrast, irreversible NWL inhibitors can completely inhibit the target enzyme, require less frequent dosing, and have better safety profiles than reversible inhibitors [52]. In theory, both these cas- pase inhibitors are in the early phases of development and will need much improvement before being considered for clinical use. Th d l i l l h i ( ) i l d i Compared to ED11 [18], NWL have several advan- tages. ED11 is a 24 amino acid peptide (GRKKRRQRRR PPQSSEIVLDGTDN) containing part of the human im- munodeficiency virus (HIV) TAT-peptide and huntingtin protein sequences. The TAT-peptide confers permeability The underlying molecular mechanism(s) involved in the restoration of cognitive function remain unclear. Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Page 17 of 19 Page 17 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 NWL-117 had no effect on hippocampal levels of TubΔ- Casp6 or p97ΔCasp6, synaptic protein expression, or glial inflammation markers. Our inability to detect changes may be a consequence of the short treatment period. Furthermore, Casp6 expression and activation is limited to the pyramidal neurons of the CA1 region of the hippocampus and this does not provide sufficient material to assess Casp6-cleaved protein substrates by western blot, especially since neurons will not all degen- erate at the same time. Funding Prateep Pakavathkumar is the recipient of a Fonds de recherche Québec-Santé scholarship (2011–2013) and the FRQ-S Alzheimer Society of Canada doctoral award (2014–2017). Anastasia Noël received a postdoctoral scholarship from the Alzheimer Society of Canada (2013–2015). This work was supported by funds from the Canadian Institutes of Health Research (CIHR) MOP-142417 to EH and from the Canadian Foundation for Innovation, CIHR MOP-243413-BCA-CGAG- 45097, and the JGH Foundation to ALB. The funding agencies did not play a role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript. Abbreviations Ac-VEID-AFC: Ac-Val-Glu-Ile-Asp-(7-Amino-4-trifluoromethylcouramin); AD: Alzheimer disease; ANOVA: Analysis of variance; APPWT: Wild type amyloid precursor protein; Aβ: Amyloid-beta peptide; BSA: Bovine serum albumin; CAG: CMV immediate early enchancer/chicken β-actin promoter fu- sion; CAMKIIa: Calmodulin kinase IIa; Casp6: Caspase-6; Casp6p20: p20 subunit of active Casp6; CHAPS: 3-[(3-cholamidopropyl)-dimethylammonio]- 2-hydroxy-1-propanesulfonic acid; CIF: Caspase inhibitory factor; CNS: Central nervous system; DMSO: Dimethyl sulfoxide; EDTA: Ethylenediaminetetraacetic acid; EGFP: Enhanced green fluorescent protein; FLICA: Fluorescent inhibitor of Casp6; Fmk: Fluoromethyl ketones; GFAP: Glial fibrillary acidic protein; HCT116: Human colon carcinoma cell line; IAP: Inhibitors of apoptosis proteins; Iba1: Ionized calcium binding adapter molecule 1; IC50: Half- maximal inhibitory concentrations; IPTG: Isopropyl β-D-1-thiogalactopyrano- side; KI: Knock in; LC/MS-MS: Liquid chromatography/tandem mass spectrometry; LDH: Lactate dehydrogenase; MTT: 3-(4,5-dimethylthiazol-2-yl)- 2,5-diphenyltetrazolium bromide; NFT: Neurofibrillary tangles; NGF: Nerve growth factor; NOR: Novel object recognition; NWL: New World Laboratories Inc.; PBS: Phosphate-buffered saline; PEG: Polyethylene glycol; PIPES: Piperazine-N, N-bis (2-ethanesulfonic acid); PSD95: Post-synaptic protein; SB: Stennicke’s buffer; TauΔCasp6: Tau-cleaved by Casp6; TubΔCasp6: Tubulin cleaved by Casp6; WT: Wild type; Z-VAD-fmk: N- benzyloxycarbonyl-Val-Ala-Asp-(O-methyl)-fluoromethylketone; Z-VEID- fmk: Benzyloxycarbonyl-Val-Glu-Ile-Asp-fmk; Δp97: p97 cleaved by Casp6 benzyloxycarbonyl Val Ala Asp (O methyl) fluoromethylketone; Z VEID fmk: Benzyloxycarbonyl-Val-Glu-Ile-Asp-fmk; Δp97: p97 cleaved by Casp6 Additional files Additional file 1: Pathology report. Contains pharmacokinetic and toxicological analysis of CD-1 mice treated with NWL-117 for 28 days every 3 days. (PDF 3742 kb) Additional file 1: Pathology report. Contains pharmacokinetic and toxicological analysis of CD-1 mice treated with NWL-117 for 28 days every 3 days. (PDF 3742 kb) Additional file 2: Figures S1-S4. Contains additional information to complement Figs. 3, 4, and 6 of the manuscript. (PDF 463 kb) Additional file 2: Figures S1-S4. Contains additional information to complement Figs. 3, 4, and 6 of the manuscript. (PDF 463 kb) Casp6 substrates, synaptic proteins, and glial inflammation markers are unchanged in mice hippocampi following NWL-117 treatment In addition, the rapid reversal of cognitive deficits suggest that the effect is possibly medi- ated through neuronal plasticity which re-establishes neuronal function, therefore, different tools encompass- ing synaptic plasticity need to be developed to assess how NWL reverses cognitive deficits in these mice. In other mice studies, cognitive amelioration was measured in the absence of changes in synaptic protein expression or brain volume [18, 73]. Future long-term prophylactic treatment studies may enlighten us to the effects of NWL-117 in the context of age and Casp6-dependent cognitive impairment and allow proper identification of target engagement. In addition, NWL inhibitors need to be administered to different AD mouse models which display an aggravated pathological phenotype, such as high Aβ load or NFTs, to determine the efficacy of this treatment in re-establishing memory function in other models of neurodegeneration. Conclusion This study reveals the potential for vinyl sulfone cas- pase inhibitors to effectively inhibit Casp6 activity and promote neuronal axonal integrity. Also, our results suggest that Casp6-mediated damage can be reversed in aged brains. Much work still needs to be done to confirm target engagement, measure selectivity, po- tency, and blood-brain-barrier permeability in animal models. However, with the increasing number of re- search groups focusing on Casp6 as a therapeutic target against neurodegenerative diseases, the possibility that Casp6 inhibitors will one day reach human trials is promising. Whether Casp6 inhibitors will be sufficient as a monotherapy, or whether they will become part of a combinatorial approach with Tau, amyloid, and other emerging therapies is a question that will be answered in the years to come. Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. Acknowledgements ld l k h k NWL inhibitors have advantages over current research tools as they are permeable to the blood brain barrier and are less toxic than the commercially available fluor- omethylketone based tools (cell permeable inhibitors and FLICA reagents) without any compromise in select- ivity. As NWL Inc. improves on their small molecules Casp6 inhibitors, experiments on non-human primates, clinical trials, and the development of radio-ligands for positron emission tomography will become reality. We would like to thank Dr. Vikas Kaushal, Dr. Joseph Flores, and Andrea Hébert- Losier for culturing human primary neurons, Dr. Benedicte Foveau for technical help with immunohistochemical staining of mice brains against synaptophysin, Dr. Xing-Kang Tong for performing the cardiac puncture and perfusion on mice for the blood-brain barrier permeability experiment, Martin Jutras for analyzing blood and hippocampal samples by liquid chromatography tandem mass spectrometry, the animal quarters staff for maintaining the mice, and New World Laboratories Inc. specifically, Richard Frenette and Dr. Benoit Bachand, for providing NWL-117 and NWL-154 inhibitors. References 1. Karran E, De Strooper B. The amyloid cascade hypothesis: are we poised for success or failure?. J Neurochem. 2016; Suppl. 2:237-52. 1. Karran E, De Strooper B. The amyloid cascade hypothesis: are we poised for success or failure?. J Neurochem. 2016; Suppl. 2:237-52. 23. Park KJ, Grosso CA, Aubert I, Kaplan DR, Miller FD. p75NTR-dependent, myelin-mediated axonal degeneration regulates neural connectivity in the adult brain. Nat Neurosci. 2010;13:559–66. 2. Wischik CM, Harrington CR, Storey JM. Tau-aggregation inhibitor therapy for Alzheimer’s disease. Biochem Pharmacol. 2014;88:529–39. 2. Wischik CM, Harrington CR, Storey JM. Tau-aggregation inhibitor therapy for Alzheimer’s disease. Biochem Pharmacol. 2014;88:529–39. 3. Fernandes-Alnemri T, Litwack G, Alnemri ES. Mch2, a new member of the apoptotic Ced-3/Ice cysteine protease gene family. Cancer Res. 1995;55: 2737–42. 24. Akpan N, Serrano-Saiz E, Zacharia BE, Otten ML, Ducruet AF, Snipas SJ, Liu W, Velloza J, Cohen G, Sosunov SA, et al. Intranasal delivery of caspase-9 inhibitor reduces caspase-6-dependent axon/neuron loss and improves neurological function after stroke. J Neurosci. 2011;31:8894–904. 4. Guo H, Albrecht S, Bourdeau M, Petzke T, Bergeron C, LeBlanc AC. Active caspase-6 and caspase-6-cleaved tau in neuropil threads, neuritic plaques, and neurofibrillary tangles of Alzheimer’s disease. Am J Pathol. 2004;165: 523–31. 25. Shabanzadeh AP, D’Onofrio PM, Monnier PP, Koeberle PD. Targeting caspase-6 and caspase-8 to promote neuronal survival following ischemic stroke. Cell Death Dis. 2015;6:e1967. 5. Albrecht S, Bogdanovic N, Ghetti B, Winblad B, LeBlanc AC. Caspase-6 activation in familial alzheimer disease brains carrying amyloid precursor protein or presenilin i or presenilin II mutations. J Neuropathol Exp Neurol. 2009;68:1282–93. 5. Albrecht S, Bogdanovic N, Ghetti B, Winblad B, LeBlanc AC. Caspase-6 activation in familial alzheimer disease brains carrying amyloid precursor protein or presenilin i or presenilin II mutations. J Neuropathol Exp Neurol. 2009;68:1282–93. 26. Monnier PP, D’Onofrio PM, Magharious M, Hollander AC, Tassew N, Szydlowska K, Tymianski M, Koeberle PD. Involvement of caspase-6 and caspase-8 in neuronal apoptosis and the regenerative failure of injured retinal ganglion cells. J Neurosci. 2011;31:10494–505. 6. LeBlanc AC. Caspase-6 as a novel early target in the treatment of Alzheimer’s disease. Eur J Neurosci. 2013;37:2005–18. 6. LeBlanc AC. Caspase-6 as a novel early target in the treatment of Alzheimer’s disease. Eur J Neurosci. 2013;37:2005–18. 27. Sokolowski JD, Gamage KK, Heffron DS, LeBlanc AC, Deppmann CD, Mandell JW. Caspase-mediated cleavage of actin and tubulin is a common feature and sensitive marker of axonal degeneration in neural development and injury. References Acta Neuropathol Commun. 2014;2:16. 7. Ramcharitar J, Afonso VM, Albrecht S, Bennett DA, LeBlanc AC. Caspase-6 activity predicts lower episodic memory ability in aged individuals. Neurobiol Aging. 2013;34:1815–24. 7. Ramcharitar J, Afonso VM, Albrecht S, Bennett DA, LeBlanc AC. Caspase-6 activity predicts lower episodic memory ability in aged individuals. Neurobiol Aging. 2013;34:1815–24. 28. Zhou Q, Snipas S, Orth K, Muzio M, Dixit VM, Salvesen GS. Target protease specificity of the viral serpin CrmA. Analysis of five caspases. J Biol Chem. 1997;272:7797–800. 8. Ramcharitar J, Albrecht S, Afonso VM, Kaushal V, Bennett DA, LeBlanc AC. Cerebrospinal fluid tau cleaved by caspase-6 reflects brain levels and cognition in aging and Alzheimer disease. J Neuropathol Exp Neurol. 2013;72:824–32. 29. Zhou Q, Krebs JF, Snipas SJ, Price A, Alnemri ES, Tomaselli KJ, Salvesen GS. Interaction of the baculovirus anti-apoptotic protein p35 with caspases. Specificity, kinetics, and characterization of the caspase/p35 complex. Biochemistry. 1998;37:10757–65. 9. LeBlanc AC, Ramcharitar J, Afonso V, Hamel E, Bennett DA, Pakavathkumar P, Albrecht S. Caspase-6 activity in the CA1 region of the hippocampus induces age-dependent memory impairment. Cell Death Differ. 2014;21: 696–706. 30. Roy N, Deveraux QL, Takahashi R, Salvesen GS, Reed JC. The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 1997;16: 6914–25. 10. Klaiman G, Petzke TL, Hammond J, LeBlanc AC. Targets of caspase-6 activity in human neurons and Alzheimer disease. Mol Cell Proteomics. 2008;7:1541–55. 11. Rovelet-Lecrux A, Hannequin D, Raux G, Le Meur N, Laquerriere A, Vital A, Dumanchin C, Feuillette S, Brice A, Vercelletto M, et al. APP locus duplication causes autosomal dominant early-onset Alzheimer disease with cerebral amyloid angiopathy. Nat Genet. 2006;38:24–6. 31. Deveraux QL, Takahashi R, Salvesen GS, Reed JC. X-linked IAP is a direct inhibitor of cell-death proteases. Nature. 1997;388:300–4. 32. Zhang Y, Tounekti O, Akerman B, Goodyer CG, LeBlanc A. 17-beta-estradio induces an inhibitor of active caspases. J Neurosci. 2001;21:RC176. 12. Sivananthan SN, Lee AW, Goodyer CG, LeBlanc AC. Familial amyloid precursor protein mutants cause caspase-6-dependent but amyloid beta-peptide- independent neuronal degeneration in primary human neuron cultures. Cell Death Dis. 2010;1:e100. induces an inhibitor of active caspases. J Neurosci. 2001;21:RC176 33. Lee AW, Champagne N, Wang X, Su XD, Goodyer C, LeBlanc AC. Alternatively spliced caspase-6B isoform inhibits the activation of caspase-6A. J Biol Chem. 2010;285:31974–84. 34. Ethical approval and consent to participate All animal procedures adhered to the guidelines of the Canadian Council on Animal Care and were approved by the McGill Animal care committees. Cortical tissues were obtained from the Birth Defects Research Laboratory (BDRL, University of Washington, Seattle, USA) operated under NIH ethical guidelines with consent to use the tissue for research, and with McGill University’s Institutional Review Board approval. All animal procedures adhered to the guidelines of the Canadian Council on Animal Care and were approved by the McGill Animal care committees. Cortical tissues were obtained from the Birth Defects Research Laboratory (BDRL, University of Washington, Seattle, USA) operated under NIH ethical id li ith t t th ti f h d ith M Gill 16. Waldron-Roby E, Ratovitski T, Wang X, Jiang M, Watkin E, Arbez N, Graham RK, Hayden MR, Hou Z, Mori S, et al. Transgenic mouse model expressing the caspase 6 fragment of mutant huntingtin. J Neurosci. 2012;32:183–93. ( , y g , , ) p guidelines with consent to use the tissue for research, and with McGill University’s Institutional Review Board approval. 17. Wong BK, Ehrnhoefer DE, Graham RK, Martin DD, Ladha S, Uribe V, Stanek LM, Franciosi S, Qiu X, Deng Y, et al. Partial rescue of some features of Huntington disease in the genetic absence of caspase-6 in YAC128 mice. Neurobiol Dis. 2015;76:24–36. Consent for publication Consent for publication All authors consent for the publication of this study. All authors consent for the publication of this study. 15. Uribe V, Wong BK, Graham RK, Cusack CL, Skotte NH, Pouladi MA, Xie Y, Feinberg K, Ou Y, Ouyang Y, et al. Rescue from excitotoxicity and axonal degeneration accompanied by age-dependent behavioral and neuroanatomical alterations in caspase-6-deficient mice. Hum Mol Genet. 2012;21:1954–67. Received: 24 November 2016 Accepted: 22 February 2017 Received: 24 November 2016 Accepted: 22 February 2017 22. Vohra BP, Sasaki Y, Miller BR, Chang J, DiAntonio A, Milbrandt J. Amyloid precursor protein cleavage-dependent and -independent axonal degeneration programs share a common nicotinamide mononucleotide adenylyltransferase 1-sensitive pathway. J Neurosci. 2010;30:13729–38. Authors’ contributions PP d i d f d PP designed, performed experiments, analysed data, and wrote the majority of the manuscript. CLL performed the intra-carotid injections and EH helped write the manuscript. ATA designed, performed experiments, analyzed data for IC50 determination against Caspase-6, and helped write the manuscript. AN designed, performed experiments, analyzed data for behavioural assessment of mice, and helped write the manuscript. JEA provided data for IC50 values against different caspases, analyzed data, and helped write the manuscript. ALB designed experiments, analyzed data, and helped write the manuscript. All authors read and approved the final manuscript. Page 18 of 19 Page 18 of 19 Page 18 of 19 Page 18 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 Competing interests huntingtin reduces toxicity and aggregate formation in neuronal and nonneuronal cells. J Biol Chem. 2000;275:19831–8. huntingtin reduces toxicity and aggregate formation in neuronal and nonneuronal cells. J Biol Chem. 2000;275:19831–8. huntingtin reduces toxicity and aggregate formation in neuronal and nonneuronal cells. J Biol Chem. 2000;275:19831–8. JEA is founder, Chief Executive Officer, and Chief Scientific Officer of New World Laboratories Inc. 14. Graham RK, Deng Y, Slow EJ, Haigh B, Bissada N, Lu G, Pearson J, Shehadeh J, Bertram L, Murphy Z, et al. Cleavage at the caspase-6 site is required for neuronal dysfunction and degeneration due to mutant huntingtin. Cell. 2006;125:1179–91. Author details 1 1Bloomfield Center for Research in Aging, Lady Davis Institute for Medical Research, Jewish General Hospital, 3999 Ch. Cote Ste-Catherine, Montreal, QC H3T 1E2, Canada. 2Department of Neurology and Neurosurgery, McGill University, 845 Sherbrooke O, Montreal, QC H3A 0G4, Canada. 3Laboratory of Cerebrovascular Research, Montreal Neurological Institute, 3801 University Street, Montreal, QC H3A 2B4, Canada. 4New World Laboratories, 500 Boulevard Cartier Ouest, Laval, QC H7V 5B7, Canada. 5Molecular and Regenerative Medicine Axis, Lady Davis Institute for Medical Research, Sir Mortimer B Davis Jewish General Hospital, 3755 ch. Côte Ste-Catherine, Montréal, QC H3T 1E2, Canada. 18. Aharony I, Ehrnhoefer DE, Shruster A, Qiu X, Franciosi S, Hayden MR, Offen D. A Huntingtin-based peptide inhibitor of caspase-6 provides protection from mutant Huntingtin-induced motor and behavioral deficits. Hum Mol Genet. 2015;24:2604–14. 19. Nikolaev A, McLaughlin T, O’Leary DD, Tessier-Lavigne M. APP binds DR6 to trigger axon pruning and neuron death via distinct caspases. Nature. 2009;457:981–9. 20. Cusack CL, Swahari V, Hampton Henley W, Michael Ramsey J, Deshmukh M. Distinct pathways mediate axon degeneration during apoptosis and axon- specific pruning. Nat Commun. 2013;4:1876. 21. Simon DJ, Pitts J, Hertz NT, Yang J, Yamagishi Y, Olsen O, Tesic Mark M, Molina H, Tessier-Lavigne M. Axon degeneration gated by retrograde activation of somatic Pro-apoptotic signaling. Cell. 2016;164:1031–45. References Ekici OD, Li ZZ, Campbell AJ, James KE, Asgian JL, Mikolajczyk J, Salvesen GS, Ganesan R, Jelakovic S, Grutter MG, Powers JC. Design, synthesis, and 13. Wellington CL, Singaraja R, Ellerby L, Savill J, Roy S, Leavitt B, Cattaneo E, Hackam A, Sharp A, Thornberry N, et al. Inhibiting caspase cleavage of Page 19 of 19 Page 19 of 19 Page 19 of 19 Pakavathkumar et al. Molecular Neurodegeneration (2017) 12:22 evaluation of aza-peptide Michael acceptors as selective and potent inhibitors of caspases-2, −3, −6, −7, −8, −9, and −10. J Med Chem. 2006;49:5728–49. 58. Heise CE, Murray J, Augustyn KE, Bravo B, Chugha P, Cohen F, Giannetti AM, Gibbons P, Hannoush RN, Hearn BR, et al. Mechanistic and structural understanding of uncompetitive inhibitors of caspase-6. PLoS One. 2012;7:e50864. evaluation of aza-peptide Michael acceptors as selective and potent inhibitors of caspases-2, −3, −6, −7, −8, −9, and −10. J Med Chem. 2006;49:5728–49. 35. Chu W, Rothfuss J, Chu Y, Zhou D, Mach RH. Synthesis and in vitro evaluation of sulfonamide isatin Michael acceptors as small molecule inhibitors of caspase-6. J Med Chem. 2009;52:2188–91. 59. Methot N, Huang J, Coulombe N, Vaillancourt JP, Rasper D, Tam J, Han Y, Colucci J, Zamboni R, Xanthoudakis S, et al. Differential efficacy of caspase inhibitors on apoptosis markers during sepsis in rats and implication for fractional inhibition requirements for therapeutics. J Exp Med. 2004;199:199–207. 36. Nyormoi O, Wang Z, Bar-Eli M. Sequence-based discovery of a synthetic peptide inhibitor of caspase 6. Apoptosis. 2003;8:371–6. 37. Talanian RV, Quinlan C, Trautz S, Hackett MC, Mankovich JA, Banach D, Ghayur T, Brady KD, Wong WW. Substrate specificities of caspase family proteases. J Biol Chem. 1997;272:9677–82. 60. McStay GP, Salvesen GS, Green DR. Overlapping cleavage motif selectivity of caspases: implications for analysis of apoptotic pathways. Cell Death Differ. 2008;15:322–31. 38. Thornberry NA, Rano TA, Peterson EP, Rasper DM, Timkey T, Garcia-Calvo M Houtzager VM, Nordstrom PA, Roy S, Vaillancourt JP, et al. A combinatorial approach defines specificities of members of the caspase family and granzyme B. Functional relationships established for key mediators of apoptosis. J Biol Chem. 1997;272:17907–11. 61. Ehrnhoefer DE, Skotte NH, Savill J, Nguyen YT, Ladha S, Cao LP, Dullaghan E, Hayden MR. A quantitative method for the specific assessment of caspase-6 activity in cell culture. PLoS One. 2011;6:e27680. 62. References Takahashi A, Alnemri ES, Lazebnik YA, Fernandes-Alnemri T, Litwack G, Moir RD, Goldman RD, Poirier GG, Kaufmann SH, Earnshaw WC. Cleavage of lamin A by Mch2 alpha but not CPP32: multiple interleukin 1 beta-converting enzyme- related proteases with distinct substrate recognition properties are active in apoptosis. Proc Natl Acad Sci U S A. 1996;93:8395–400. 39. Garcia-Calvo M, Peterson EP, Leiting B, Ruel R, Nicholson DW, Thornberry NA. Inhibition of human caspases by peptide-based and macromolecular inhibitors. J Biol Chem. 1998;273:32608–13. 40. Orth K, Chinnaiyan AM, Garg M, Froelich CJ, Dixit VM. The CED-3/ICE-like protease Mch2 is activated during apoptosis and cleaves the death substrate lamin A. J Biol Chem. 1996;271:16443–6. 63. Berger AB, Witte MD, Denault JB, Sadaghiani AM, Sexton KM, Salvesen GS, Bogyo M. Identification of early intermediates of caspase activation using selective inhibitors and activity-based probes. Mol Cell. 2006;23:509–21. 41. Edgington LE, van Raam BJ, Verdoes M, Wierschem C, Salvesen GS, Bogyo M. An optimized activity-based probe for the study of caspase-6 activation. Chem Biol. 2012;19:340–52. 64. Pereira NA, Song Z. Some commonly used caspase substrates and inhibitors lack the specificity required to monitor individual caspase activity. Biochem Biophys Res Commun. 2008;377:873–7. 42. Eichhold TH, Hookfin EB, Taiwo YO, De B, Wehmeyer KR. Isolation and quantification of fluoroacetate in rat tissues, following dosing of Z-Phe-Ala- CH2-F, a peptidyl fluoromethyl ketone protease inhibitor. J Pharm Biomed Anal. 1997;16:459–67. 65. Boucher D, Blais V, Denault JB. Caspase-7 uses an exosite to promote poly(ADP ribose) polymerase 1 proteolysis. Proc Natl Acad Sci U S A. 2012;109:5669–74. 66. Maccioni RB, Vera JC, Dominguez J, Avila J. A discrete repeated sequence defines a tubulin binding domain on microtubule-associated protein tau. Arch Biochem Biophys. 1989;275:568–79. 43. Palmer JT, Rasnick D, Klaus JL, Bromme D. Vinyl sulfones as mechanism- based cysteine protease inhibitors. J Med Chem. 1995;38:3193–6. 44. Abdulla MH, Lim KC, Sajid M, McKerrow JH, Caffrey CR. Schistosomiasis mansoni: novel chemotherapy using a cysteine protease inhibitor. PLoS Med. 2007;4:e14. 67. Schoenmann Z, Assa-Kunik E, Tiomny S, Minis A, Haklai-Topper L, Arama E, Yaron A. Axonal degeneration is regulated by the apoptotic machinery or a NAD + −sensitive pathway in insects and mammals. J Neurosci. 2010;30: 6375–86. 45. Klaiman G, Champagne N, LeBlanc AC. Self-activation of Caspase-6 in vitro and in vivo: Caspase-6 activation does not induce cell death in HEK293T cells. Biochim Biophys Acta. 1793;2009:592–601. 68. References Guo H, Petrin D, Zhang Y, Bergeron C, Goodyer CG, LeBlanc AC. Caspase-1 activation of caspase-6 in human apoptotic neurons. Cell Death Differ. 2006; 13:285–92. 46. Denault J, Salvesen G. Expression, Purification, and Characterization of Caspases. In Current Protocols in Protein Science. 2002: 21.13.1-21.13.5 69. Kaushal V, Dye R, Pakavathkumar P, Foveau B, Flores J, Hyman B, Ghetti B, Koller BH, LeBlanc AC. Neuronal NLRP1 inflammasome activation of caspase- 1 coordinately regulates inflammatory interleukin-1-beta production and axonal degeneration-associated caspase-6 activation. Cell Death Differ. 2015; 22:1676–86. 47. Boussif O, Lezoualc’h F, Zanta MA, Mergny MD, Scherman D, Demeneix B, Behr JP. A versatile vector for gene and oligonucleotide transfer into cells in culture and in vivo: polyethylenimine. Proc Natl Acad Sci U S A. 1995;92: 7297–301. 48. LeBlanc A. Increased production of 4 kDa amyloid beta peptide in serum deprived human primary neuron cultures: possible involvement of apoptosis. J Neurosci. 1995;15:7837–46. 70. Selznick LA, Holtzman DM, Han BH, Gokden M, Srinivasan AN, Johnson Jr EM, Roth KA. In situ immunodetection of neuronal caspase-3 activation in Alzheimer disease. J Neuropathol Exp Neurol. 1999;58:1020–6. 49. Stennicke HR, Salvesen GS. Biochemical characteristics of caspases-3, −6, −7, and −8. J Biol Chem. 1997;272:25719–23. 71. Veber DF, Johnson SR, Cheng HY, Smith BR, Ward KW, Kopple KD. Molecular properties that influence the oral bioavailability of drug candidates. J Med Chem. 2002;45:2615–23. 50. Nybo K. GFP imaging in fixed cells. Biotechniques. 2012;52:359–60. 51. Halawani D, Tessier S, Anzellotti D, Bennett DA, Latterich M, LeBlanc AC. Identification of Caspase-6-mediated processing of the valosin containing protein (p97) in Alzheimer’s disease: a novel link to dysfunction in ubiquitin proteasome system-mediated protein degradation. J Neurosci. 2010;30: 6132–42. 72. Baker PJ, Boucher D, Bierschenk D, Tebartz C, Whitney PG, D’Silva DB, Tanzer MC, Monteleone M, Robertson AA, Cooper MA, et al. NLRP3 inflammasome activation downstream of cytoplasmic LPS recognition by both caspase-4 and caspase-5. Eur J Immunol. 2015;45:2918–26. 73. Tong XK, Lecrux C, Rosa-Neto P, Hamel E. Age-dependent rescue by simvastatin of Alzheimer’s disease cerebrovascular and memory deficits. J Neurosci. 2012;32: 4705–15. 52. Singh J, Petter RC, Baillie TA, Whitty A. The resurgence of covalent drugs. Nat Rev Drug Discov. 2011;10:307–17. 53. Leyva MJ, Degiacomo F, Kaltenbach LS, Holcomb J, Zhang N, Gafni J, Park H, Lo DC, Salvesen GS, Ellerby LM, Ellman JA. Identification and evaluation of small molecule pan-caspase inhibitors in Huntington’s disease models. Chem Biol. 2010;17:1189–200. 54. References Henzing AJ, Dodson H, Reid JM, Kaufmann SH, Baxter RL, Earnshaw WC. Synthesis of novel caspase inhibitors for characterization of the active caspase proteome in vitro and in vivo. J Med Chem. 2006;49:7636–45. 55. Vaidya S, Hardy JA. Caspase-6 latent state stability relies on helical propensity. Biochemistry. 2011;50:3282–7. 56. Stanger K, Steffek M, Zhou L, Pozniak CD, Quan C, Franke Y, Tom J, Tam C, Krylova I, Elliott JM, et al. Allosteric peptides bind a caspase zymogen and mediate caspase tetramerization. Nat Chem Biol. 2012;8:655–60. 57. Murray J, Giannetti AM, Steffek M, Gibbons P, Hearn BR, Cohen F, Tam C, Pozniak C, Bravo B, Lewcock J, et al. Tailoring small molecules for an allosteric site on procaspase-6. ChemMedChem. 2014;9:73–7. 2.
https://openalex.org/W2101366234	https://tbiomed.biomedcentral.com/counter/pdf/10.1186/1742-4682-11-14	English	null	The role of the entorhinal cortex in epileptiform activities of the hippocampus	Theoretical biology and medical modelling	2,014	cc-by	11,763	© 2014 Ren 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 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. * Correspondence: pmzhang@sjtu.edu.cn 1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China Full list of author information is available at the end of the article The role of the entorhinal cortex in epileptiform activities of the hippocampus Hui Ren1, Ye-Jun Shi2, Qin-Chi Lu2, Pei-Ji Liang1 and Pu-Ming Zhang1* * Correspondence: pmzhang@sjtu.edu.cn 1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China Full list of author information is available at the end of the article RESEARCH Open Access Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Abstract Background: Temporal lobe epilepsy (TLE) is the commonest type of epilepsy in adults, and the hippocampus is indicated to have a close relationship with TLE. Recent researches also indicate that the entorhinal cortex (EC) is involved in epilepsy. To explore the essential role that the EC may play in epilepsy, a computational model of the hippocampal CA3 region was built, which consisted of pyramidal cells and two types of interneurons. By changing the input signals from the EC, the effects of EC on epileptiform activities of the hippocampus were investigated. Additionally, recent studies have found that the antiepileptic drug valproate (VPA) can block ictal discharges but cannot block interictal discharges in vitro, and the mechanism under this phenomenon is still confusing. In our model, the effects of VPA on epileptiform activities were simulated and some mechanisms were explored. Results: Interictal discharges were induced in the model without the input signals from the EC, whereas the model with the EC input produced ictal discharges when the EC input contained ictal discharges. The GABA-ergic connection strength was enhanced and the NMDA-ergic connection strength was reduced to simulate the effects of VPA, and the simulation results showed that the disappearance of ictal discharges in the model mainly due to the disappearance of ictal discharges in the input signals from the EC. Results: Interictal discharges were induced in the model without the input signals from the EC, whereas the model with the EC input produced ictal discharges when the EC input contained ictal discharges. The GABA-ergic connection strength was enhanced and p g g g the NMDA-ergic connection strength was reduced to simulate the effects of VPA, and the simulation results showed that the disappearance of ictal discharges in the model mainly due to the disappearance of ictal discharges in the input signals from the EC. Conclusions: Simulation results showed that ictal discharges in the EC were necessary for the hippocampus to generate ictal discharges, and VPA might block the ictal discharges in the EC, which led to the disappearance of ictal discharges in the hippocampus. Conclusions: Simulation results showed that ictal discharges in the EC were necessary for the hippocampus to generate ictal discharges, and VPA might block the ictal discharges in the EC, which led to the disappearance of ictal discharges in the hippocampus. Keywords: Computational model, Hippocampal CA3 region, Entorhinal cortex, Valproate, Temporal lobe epilepsy Introduction Temporal lobe epilepsy (TLE) is the commonest type of epilepsy in adults, and about 75% of patients with mesial TLE are considered to have drug-resistant epilepsy [1]. It has been commonly accepted that the hippocampus has a close relationship with the generation of TLE. In specimens from surgical resections and post-mortem studies of patients with TLE, neuronal loss, atrophy, and gliosis have been revealed in the hippocampus [2]. It has been reported that epileptiform activities depend crucially on intrinsic neuronal properties, and the organisation of the synaptic networks [3]. The hippocampus has massive recurrent excitatory connections, intrinsically burst-generating cells, as well as closely spaced cell bodies and dendrites, and these all make it easy to generate epileptiform activities [4,5]. Page 2 of 22 Page 2 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 In spite of the essential role that the hippocampus plays in the generation of epileptiform activities, observations in animal models also have indicated that the epileptogenic zone is broad, and some other limbic regions may be involved in the generation of TLE as well [2]. Now, many scientists have shown much interest in the role the entorhinal cortex (EC) plays in epilepsy [6]. Magnetic resonance imaging studies showed that the EC was damaged in patients with TLE [7], and observations in both rats and patients with TLE displayed preferential neuronal loss in the EC [8,9]. There exist extensive reciprocal connections between the EC, hippocampus, and other brain areas, which make the EC a potential candidate for generating and propagating TLE seizures [10]. Epileptiform activities appear as interictal discharges and ictal discharges. Interictal discharges are the simplest identifiable epileptiform activities, which last tens or hundreds of milliseconds [4], and ictal discharges (also termed seizures) represent the critical events and the primary clinical burden of an active epileptic condition, which usually last more than 10 seconds [4,11]. It has been suggested that the EC may contribute to the initiation of ictal discharges [6,12,13]. It was reported that cutting the perforant pathway blocked the ictal discharges in the hippocampus, but not in the EC, where ictal discharges continued to occur with similar features compared to those seen in the combined EC-hippocampal slices where the perforant pathway was preserved [6,14]. Introduction Our laboratory has done researches about epileptiform activities induced by Mg2+-free artificial cerebrospinal fluid (ACSF), which unblocks the N-methyl-D-aspartate (NMDA) receptors, in hippocampal slices and combined EC-hippocampal slices as well [15-19]. We found that ictal dis- charges were only induced in the combined EC-hippocampal slices, which indicated that the EC was very essential for the hippocampus to initiate ictal discharges [18,19]. Although the EC has been shown to have a close relationship with the generation of epileptiform activities, the exact role it plays in epilepsy is still unclear. Many experimental studies in hippocampal slices indicated that the CA3 field was the place that could generate interictal activities [6,20]. Pyramidal cells in the CA3 receive signals from the dentate gyrus (DG) via mossy fibers, and from the EC via the perforant pathway, and they project to the CA1 [21]. Divergent connections of the CA3 pyramidal cells with local cells and cells of the other fields allow for the expansion of synchronous population discharges [20]. Many models of the CA3 field had been built to investigate the mechanisms of different epileptiform activities, such as carbachol-driven rhythmic population oscillations [22], picrotoxin-induced synchronized after-discharges [23], or low-Mg2+ induced neuronal bursts and after-discharges [24]. But they only simulated interictal discharges and didn’t consider the impacts that the EC may make on epileptiform activities of the hippocampus. Valproate (VPA) is one of the major antiepileptic drugs (AEDs) used today, and its ability to control epileptiform activities mainly depends on enhancing the GABA-ergic inhibitory functions and reducing the NMDA-ergic excitatory functions of the nervous system [25]. Researchers have found that VPA could block ictal discharges but couldn’t completely block interictal discharges in vitro [26-28]. In our experiments, we also found that the application of 3 mM VPA suppressed the frequency of interictal discharges and completely blocked the ictal discharges in the combined EC-hippocampal slices [19]. However, the mechanism of this phenomenon is still confusing. In this work, we built a model of the hippocampal CA3 region to simulate interictal and ictal discharges based on NEURON [29]. The model contained 4 neurons including Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Page 3 of 22 Page 3 of 22 2 pyramidal cells, one basket cell and one oriens-lacunosum molecular (OLM) cell. The model contained 2 inputs which were from the DG and EC respectively. Simulation reproduces Mg2+-free-ACSF induced interictal discharges Simulation reproduces Mg2+-free-ACSF induced interictal discharges In our experiments, the application of Mg2+-free-ACSF consistently induced interictal discharges in adult rat hippocampal slices [15-17] and adult mouse hippocampal slices [18,19]. Figure 1A shows an example of Mg2+-free-ACSF induced interictal discharges in one mouse hippocampal slice. These signals were local field potentials (LFPs) recorded by one electrode of the micro-electrode array (MEA) located in the CA3 region. The onset of interictal discharges in different slices appeared with different time delays after Mg2+-free-ACSF perfusion began, ranging between 10 and 20 min (15.8 ± 2.6 min, n = 4 slices) [18]. Signals measured in the model were the membrane potentials of pyramidal cells, which seemed similar to the intracellular recordings and different from LFPs recorded in experiments as shown in Figure 1A. LFPs were the electric potentials recorded in the extracellular space and it had been shown that LFPs were synchronized with intracellular recordings in the same region during epileptiform activities [27,30]. As the hippocampal slices didn’t contain EC, we only considered the input from the DG, which followed a Poisson process (λ = 1) [31] in this model (Figure 1B). The somatic membrane potentials of one pyramidal cell under different Mg2+ concentrations were plotted in Figure 1C. When the Mg2+ concentration was 1.0 mM, which was the normal Mg2+ concentration in the extracellular solution [32], interictal discharges were absent. However, when the Mg2+ concentration was changed to 0 mM, interictal discharges occurred. Every single burst of interictal discharges consisted of a train of several spikes riding on a large depolarizing wave, which was similar to the paroxysmal depolarizing shift recorded in experiments [4,33]. What’s more, the interictal discharges in the simulation occurred when the input signals followed a Poisson process, which supported the viewpoint, indicated in many experimental studies, that the CA3 field could generate interictal activities [6,16,18-20]. Introduction Simulation results indicated that interictal discharges were induced in the model without the EC input, whereas the model with the EC input produced ictal discharges when the EC input contained ictal discharges. From the results, we supposed that ictal discharges from the EC were necessary for the hippocampus to generate ictal discharges. In our model, the GABA-ergic inhibitory connections were enhanced and the NMDA-ergic excitatory connections were reduced to simulate the effects of VPA. Simulation results showed that the disappearance of ictal discharges in the model mainly due to the disappearance of ictal discharges of the input signals from the EC. Thus, we supposed that VPA might block the ictal discharges in the EC of combined EC-hippocampal slices, which led to the disappearance of ictal discharges in the hippocampus. Ictal and interictal discharges induced by the EC input In our experiments, two types of epileptiform discharges were recorded in the mouse combined EC-hippocampal slices during the application of Mg2+-free-ACSF (Figure 2A). Interictal discharges occurred regularly in the CA3 region, and they generally appeared before an ictal discharge with a frequency of 0.22 ± 0.06 Hz, and the frequency of ictal Page 4 of 22 Page 4 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Figure 1 (See legend on next page.) Page 5 of 22 Page 5 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 (See figure on previous page.) Figure 1 Interictal discharges in experiments and in the model, and input signals from DG. (A) LFPs in CA3 region recorded by MEA in a hippocampal slice. Interictal discharges occurred with more than 10 minutes’ delays after Mg2+-free-ACSF perfusion. (B) Simulated input signals from the DG which followed a Poisson process (λ = 1). (C) Somatic membrane potentials of one pyramidal cell under different Mg2+ concentrations in the model without the input from the EC. Interictal discharges occurred when [Mg2+] was 0 mM. [Mg2+]: extracellular Mg2+ concentration. (See figure on previous page.) Figure 1 Interictal discharges in experiments and in the model, and input signals from DG. (A) LFPs in CA3 region recorded by MEA in a hippocampal slice. Interictal discharges occurred with more than 10 minutes’ delays after Mg2+-free-ACSF perfusion. (B) Simulated input signals from the DG which followed a Poisson process (λ = 1). (C) Somatic membrane potentials of one pyramidal cell under different Mg2+ concentrations in the model without the input from the EC. Interictal discharges occurred when [Mg2+] was 0 mM. [Mg2+]: extracellular Mg2+ concentration. discharges was about 0.004 ± 0.001 Hz (n = 4 slices) [18]. The epileptiform activities measured in the DG and EC were synchronized with those measured in the CA3 region [18,19,34]. Also, according to experimental studies, the EC was indicated to be the first site to produce ictal discharges in combined EC-hippocampal slices [6,18,19]. In this model, interictal and ictal discharges were added into the network to simulate inputs from the EC and DG (Figure 2B). As a result, the pyramidal cells in the model produced interictal and ictal discharges (Figure 2C). Ictal and interictal discharges induced by the EC input The interictal discharges appeared before an ictal discharge with a frequency of 0.2 Hz and the ictal discharges appeared with a frequency of 0.005 Hz. These patterns were synchronized with the input signals, which supported the viewpoint that synchronized firing occurred during Mg2+-free-ACSF induced epileptiform activities [24]. In the simulation, ictal discharges occurred when the input signals from the EC and DG contained ictal activities, and there were no reports or experimental results showing that the DG was the first place to generate ictal discharges. We supposed that ictal discharges in the hippocampus were induced by the EC. To confirm this supposition, we set the input from the DG as signals followed a Poisson process (λ = 1) as shown in Figure 1B, and didn’t change the input from the EC (Figure 2B). The simulation result showed that ictal discharges still occurred (Figure 2D). In this way, we deduced that the EC induced the hippocampus to generate ictal discharges in combined EC-hippocampal slices. Effects of VPA on the epileptiform activities In our experiments, the application of 3 mM VPA suppressed the frequency of interictal discharges induced by Mg2+-free-ACSF, but could not completely block them in the mouse hippocampal slices [19], and one example was shown in Figure 3A. In the model without the input from the EC, the GABA-ergic inhibitory connection strength was enhanced and the NMDA-ergic excitatory connection strength was reduced to simulate the effects of VPA on interictal activities. Figure 3B shows one example of the membrane potentials of one pyramidal cell when the GABA-ergic connection strength is tripled and the NMDA-ergic connection strength is reduced by 20% in the model. We found that the frequency of interictal discharges decreased about 22.2%. According to our experimental studies in the mouse combined EC-hippocampal slices, the application of 3 mM VPA completely blocked ictal discharges, and suppressed but didn’t completely block interictal discharges (Figure 4A). The frequency of interictal discharges changed from 0.22 ± 0.06 Hz to 0.15 ± 0.03 Hz (n = 4 slices) [19]. Suggested by the experimental data, the EC and DG produced interictal discharges with lower frequencies after the application of VPA in combined EC-hippocampal slices [27]. Thus, in the model with the input from the EC, interictal discharges with a frequency of 0.15 Hz were added Page 6 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Figure 2 (See legend on next page.) Page 7 of 22 Page 7 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 (See figure on previous page.) Figure 2 Interictal (asterisks) and ictal (continuous lines) discharges in experiments and the model, and input signals. (A) LFPs in the CA3 region recorded by MEA in a combined EC-hippocampal slice. Interictal discharges occurred regularly before an ictal discharge. (B) Simulated input signals from the DG or EC, which were alternate interictal and ictal discharges. (C) Somatic membrane potentials of one pyramidal cell in the model with the input from the EC. When input signals from the DG and EC were alternate interictal and ictal discharges as shown in (B), pyramidal cells in the model produced interictal and ictal discharges. (D) Somatic membrane potentials of one pyramidal cell in the model with the input from the EC. Effects of VPA on the epileptiform activities When the input signals from the DG followed a Poisson process as shown in Figure 1B and the input signals from the EC were alternate interictal and ictal discharges as shown in (B), pyramidal cells in the model produced interictal and ictal discharges. Figure 3 Effects of VPA on interictal activities in experiments and simulations. (A) Interictal discharges before and during 3 mM VPA application in one hippocampal slice, which represents the recording of one electrode in the CA3 region. (B) The somatic membrane potentials of one pyramidal cell in the model without the EC input. The input signals from the DG followed a Poisson process as shown in Figure 1B. When the GABA-ergic connection strength was enhanced and the NMDA-ergic connection strength was reduced, the frequency of interictal discharges decreased. Figure 3 Effects of VPA on interictal activities in experiments and simulations. (A) Interictal discharges before and during 3 mM VPA application in one hippocampal slice, which represents the recording of one electrode in the CA3 region. (B) The somatic membrane potentials of one pyramidal cell in the model without the EC input. The input signals from the DG followed a Poisson process as shown in Figure 1B. When the GABA-ergic connection strength was enhanced and the NMDA-ergic connection strength was reduced, the frequency of interictal discharges decreased. Page 8 of 22 Page 8 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Figure 4 (See legend on next page.) Figure 4 (See legend on next page.) Page 9 of 22 Page 9 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 (See figure on previous page.) Figure 4 Effects of VPA on interictal (asterisks) and ictal (continuous lines) activities in experiments and simulations. (A) Epileptiform discharges before and during 3 mM VPA application in a combined EC-hippocampal slice, which represents the recordings of one electrode in the CA3 region. (B) The somatic membrane potentials of a pyramidal cell in the model with the input from the EC. Effects of VPA on the epileptiform activities When the GABA-ergic connection strength was enhanced and the NMDA-ergic connection strength was reduced, and the input signals from the EC and DG were interictal discharges with a lower frequency compared to interictal discharges added previously as shown in Figure 2B, ictal discharges disappeared and interictal discharges still existed. (C) The somatic membrane potentials of a pyramidal cell in the model with the input from the EC. When the GABA-ergic connection strength was enhanced and the NMDA-ergic connection strength was reduced, and the input signals from the DG followed a Poisson process as shown in Figure 1B, and the input signals from the EC were alternate interictal and ictal discharges as shown in Figure 2B, ictal and interictal discharges still existed. into the network to simulate input signals from the EC and DG (Figure 4B). The pyramidal cells in the model produced interictal discharges synchronized with the input signals (data not shown). When the GABA-ergic connection strength was enhanced more than 10 times and the NMDA-ergic connection strength was reduced by more than 20%, the interictal discharges still existed and were synchronized with the input signals (Figure 4B). Thus, VPA blocked the ictal discharges but could not block the interictal discharges. The disappearance of ictal discharges in the model seemed due to the disappearance of ictal discharges in the input signals. In this way, we supposed that VPA might block the ictal discharges in the EC of the combined EC-hippocampal slices, which led to the disappearance of ictal discharges in the hippocampus. To confirm this opinion further, ictal and interictal discharges were added into the model to simulate the input signals from the EC, and the signals followed a Poisson process (λ = 1) were added into the model to simulate the input signals from the DG. The model generated interictal and ictal discharges as shown in Figure 2D. When the GABA-ergic connection strength was enhanced more than 50 times and the NMDA-ergic connection strength was reduced by more than 20%, the ictal and interictal discharges still existed (Figure 4C). The results suggested that VPA could not block ictal discharges in the hippocampus when the EC contained ictal discharges. Thus, we deduced that during the application of VPA in combined EC-hippocampal slices, the disappearance of ictal discharges in the hippocam- pus might result from the disappearance of ictal discharges in the EC. Discussion In this work, we established a model of the hippocampal CA3 region to characterize the effects of the EC on the hippocampal epileptiform activities. In the model without the input from the EC, interictal discharges were induced by changing the Mg2+ concentration to 0 mM, which supported the viewpoint that CA3 field could generate interictal activities [6]. In the model with the input from the EC, which contained interictal and ictal dis- charges, pyramidal cells of the model generated interictal and ictal discharges. Additionally, when the input signals from the DG followed a Poisson process and the input signals from the EC contained ictal discharges, pyramidal cells of the model generated ictal discharges. From the simulation results, we deduced that the EC was the first place to generate ictal discharges in combined EC-hippocampal slices, which was suggested in many previous experimental studies [6,12-14]. Finally, the GABA-ergic inhibitory connection strength was enhanced and the NMDA-ergic excitatory connection strength was reduced to simulate Page 10 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Page 10 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 the effects of VPA. As the ictal discharges of the model disappeared when the input signals from the EC didn’t contain ictal discharges, we supposed that the disappearance of ictal discharges in the hippocampus of combined EC-hippocampal slices might due to the disappearance of ictal discharges in the EC. In this way, we deduced that VPA might block the ictal discharges in the EC of combined EC-hippocampal slices, which led to the disappearance of ictal discharges in the hippocampus. Many people have built models of the CA3 region to investigate possible mechanisms of epileptiform activities [22-24,35]. Those models were very large and contained at least hundreds of neurons. Some other people built models of the CA3 region to do researches of the other problems such as schizophrenia [31,36,37]. Those models contained more than 10 neurons. Compared to the models built previously, the model built here was much smaller and only contained 4 neurons. Our model contained the pyramidal cell, the basket cell and the OLM cell, which were the commonest neurons in the CA3 model [31,36,37]. Connections between these neurons were based on anatomical study results [21,38]. The model also contained the commonest receptors, which were AMPA, NMDA and GABAA receptors [31,35]. Discussion Although our model was small, it could generate epilepti- form activities and simulate the effects of VPA. But, this small model has many limits. It has been reported that the epileptiform activities may firstly occur in a small number of pyramidal neurons, and then spread to other pyramidal neurons [23]. With our small model, we cannot simulate the spread of epileptiform activities. Additionally, it has been shown that during different periods of epileptiform activities, the number of pyramidal cells that fire synchronously is different [24]. Since there are only two pyramidal cells in our model, it cannot simulate that phenomenon. The pyramidal cells built in this model were burst generating cells, as it had been reported that most pyramidal cells in the CA3 field could generate bursts and the generation of epileptiform activities were related to the burst generating cells [4,5,39]. However, the CA3 field also contains nonbursting pyramidal cells, which show action potentials with a property of spike frequency adaption after the application of somatic current injections [21,39]. The discharges of bursting pyramidal cells always precede the population dis- charges, and bursting pyramidal cells may be the pacemakers of epileptiform activities, whereas nonbursting pyramidal cells only discharge simultaneously with the population discharges during epileptiform activities [40]. So, in our model, we didn’t consider the nonbursting pyramidal cells. But, if we want to research the details of the firing pattern of epileptiform discharges, we may need to consider the nonbursting pyramidal cells. In normal physiological conditions, the structure of the pyramidal cell is very com- plex [39], and many people have built different reconstructed CA3 pyramidal cell models. Some models contained more than 200 compartments [39,41], and some contained more than 10 compartments [22-24], or only several compartments [31,36]. The pyramidal cell built in our model consisted of 4 compartments: one somatic and 3 dendritic compartments (Figure 5). Although the structure of this cell model was different from many other models, its responses to the somatic current injections were similar to the firing patterns of pyramidal cells measured in experiments [39]. However, the distribution of ion channels in this cell model could not be the same as that in the cell models whose structures were similar with that of a real pyramidal cell, which made this model’s output have some differences with the real pyramidal cells’ firing activities. Discussion If we investigate the effects of the firing pattern or the distribution of some Page 11 of 22 Page 11 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Figure 5 Structure models for different cell types, and their firing properties in response to current injections. (A) Structure model of the pyramidal cell and its firing properties in response to 0.23 nA somatic current injections. Prox dend: proximal dendritic compartment; Med dend: medial dendritic compartment; Dist dend: distal dendritic compartment. (B) Structure model of the basket cell and its firing properties in response to 0.003 nA somatic current injections. (C) Structure model of the OLM cell and its firing properties in response to 0.003 nA somatic current injections. Figure 5 Structure models for different cell types, and their firing properties in response to current injections. (A) Structure model of the pyramidal cell and its firing properties in response to 0.23 nA somatic current injections. Prox dend: proximal dendritic compartment; Med dend: medial dendritic compartment; Dist dend: distal dendritic compartment. (B) Structure model of the basket cell and its firing properties in response to 0.003 nA somatic current injections. (C) Structure model of the OLM cell and its firing properties in response to 0.003 nA somatic current injections. Figure 5 Structure models for different cell types, and their firing properties in response to current injections. (A) Structure model of the pyramidal cell and its firing properties in response to 0.23 nA somatic current injections. Prox dend: proximal dendritic compartment; Med dend: medial dendritic compartment; Dist dend: distal dendritic compartment. (B) Structure model of the basket cell and its firing properties in response to 0.003 nA somatic current injections. (C) Structure model of the OLM cell and its firing properties in response to 0.003 nA somatic current injections. Figure 5 Structure models for different cell types, and their firing properties in response to current injections. (A) Structure model of the pyramidal cell and its firing properties in response to 0.23 nA somatic current injections. Prox dend: proximal dendritic compartment; Med dend: medial dendritic compartment; Dist dend: distal dendritic compartment. (B) Structure model of the basket cell and its firing properties in response to 0.003 nA somatic current injections. (C) Structure model of the OLM cell and its firing properties in response to 0.003 nA somatic current injections. Discussion ion channels of pyramidal cells on the epileptiform activities, we may need to build more detailed pyramidal cells in the future work. There are many types of interneurons in the CA3 field. In this model, we only con- sidered the basket cell and the OLM cell, which were 2 common types in the CA3 field [21,38]. However, there exist some other types of perisomatic-targeting interneurons (such as the axo-axonic cells) and dendritic-targeting interneurons (such as the bistratified cells) in the CA3 field [21]. The acting sites towards pyramidal cells and the firing patterns of these cells are similar to the basket cells or the OLM cells, but there still exist some differences. The axo-axonic cells project to the proximal dendrites of pyramidal cells and the bistratified cells project to the medial dendrites of pyramidal cells [21,38]. So considering these interneurons may make the CA3 model more reliable. However, our model is small and these cells only compose a small part of CA3 interneurons, so we didn’t consider these cells. If we expand the structure of our network, we’d better add these types of interneurons. VPA is one of the major AEDs used today, and has high efficiency in treating various seizure types such as myoclonic, generalized tonic-clonic seizures and partial seizures [25,42]. In this work, the GABA-ergic inhibitory connection strength was enhanced and the NMDA-ergic excitatory connection strength was reduced to simulate the effects of VPA. However, VPA may also control epileptiform activities by affecting some ion channels. It has been reported that VPA may affect calcium and potassium channels by interfering with calcium entry into the cell and activating the potassium conductance, which lead to the reduction of neuronal excitability [25,43]. VPA may also inhibit sodium channels [44], although this opinion has been questioned [45]. In this model, we didn’t consider the possible effects of VPA on these channels, which might make the simulated effects of VPA on epileptiform activities smaller than VPA’s real effects. In the future work, our model may be used to do other researches about epileptiform activities, such as the effects of 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX) or Page 12 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 bicuculline on the Mg2+-free-ACSF induced epileptiform discharges [24]. However, the model built here is very simple compared with what we have known of the CA3 field. Network structure of the CA3 model The network structure of the CA3 model is illustrated in Figure 6. The network consists of 2 pyramidal cells, one OLM cell and one basket cell. Pyramidal cells excite each other, and they also excite the basket cell and OLM cell. The basket cell inhibits itself, the OLM cell and both pyramidal cells. The OLM cell inhibits both pyramidal cells. As shown in Figure 6, the simulated region receives 2 inputs from the EC and DG respectively. The input from the DG excites the proximal dendrite of one pyramidal cell and the input from the EC ex- cites the distal dendrites of both pyramidal cells. Schematic representations of the simulated cells and their responses to current injections are described in Figure 5. The complete mathematical implementation of each cell is described in the Appendix. All simulations were performed using NEURON [29] running on a personal computer under Windows 7. Basket cells and OLM cells are 2 commonest interneurons in the CA3 models [31,35-37]. It has been reported that interneurons that control the firing activities of pyramidal cells have 2 types: perisomatic-targeting interneurons and dendritic-targeting interneurons. Basket cells belong to the former type and OLM cells belong to the latter [37]. Basket cells control the synchrony of action potentials of pyramidal cells [46]. OLM cells affect the membrane potentials of dendrites of pyramidal cells, where most The network structure of the CA3 model is illustrated in Figure 6. The network consists of 2 pyramidal cells, one OLM cell and one basket cell. Pyramidal cells excite each other, and they also excite the basket cell and OLM cell. The basket cell inhibits itself, the OLM cell and both pyramidal cells. The OLM cell inhibits both pyramidal cells. As shown in Figure 6, the simulated region receives 2 inputs from the EC and DG respectively. The input from the DG excites the proximal dendrite of one pyramidal cell and the input from the EC ex- cites the distal dendrites of both pyramidal cells. Schematic representations of the simulated cells and their responses to current injections are described in Figure 5. The complete mathematical implementation of each cell is described in the Appendix. All simulations were performed using NEURON [29] running on a personal computer under Windows 7. Basket cells and OLM cells are 2 commonest interneurons in the CA3 models [31,35-37]. Discussion To make the model more similar to the real CA3 network, the number of the pyramidal cells, basket cells and OLM cells in the model can be increased, and more cell types and their connectivity can be included in the future work. Pyramidal cell Each simulated pyramidal cell consists of 4 compartments: one somatic and 3 dendritic compartments [36]. Each pyramidal cell contains leak current, sodium current, delayed rectifier potassium (K+) current, calcium (Ca2+) activated K+ current, afterhyperpolarization (AHP) K+ current, M-type K+ current, L-type Ca2+ current, N-type Ca2+ current, and T-type Ca2+ current [24,31,39,41]. The leak, sodium and delayed rectifier K+ currents allow cells to generate action potentials [31]. The Ca2+-activated K+ current and AHP K+ current make the cells have the property of spike-frequency adaptation, which has been observed in most pyramidal cells [49]. The M-type K+ current plays an important role in the regulation of firing rate, and it has been reported that small changes of the M current seem to be sufficient to cause epileptic seizures [50]. Voltage-sensitive Ca2+ currents may contribute to epileptogenesis and the N-type and L-type Ca2+ current have been revealed to regulate a number of neuronal processes, including Ca2+-dependent K+ currents [51]. In addition, the T-type Ca2+ current can control the membrane potentials and intracellular Ca2+ concentrations [52]. The parameters of all ionic conductance used in the model are listed in Table 1. Sodium, delayed rectifier K+, N-type Ca2+ and T-type Ca2+ conductance is uniformly distributed throughout the entire neuron, Ca2+-activated K+ conductance decreases with distance from the soma, AHP K+ conductance is lower in distal than in proximal dendrites, and T-type Ca2+ conductance exists only in the soma and proximal dendrites. Network structure of the CA3 model It has been reported that interneurons that control the firing activities of pyramidal cells have 2 types: perisomatic-targeting interneurons and dendritic-targeting interneurons. Basket cells belong to the former type and OLM cells belong to the latter [37]. Basket cells control the synchrony of action potentials of pyramidal cells [46]. OLM cells affect the membrane potentials of dendrites of pyramidal cells, where most Figure 6 Simulated CA3 network. The basket cell (B) inhibits itself, the OLM cell (OLM) and the pyramidal cells (P). The OLM cell inhibits the distal dendrites of the pyramidal cells. Pyramidal cells excite each other and both inhibitory cells. The input from the DG excites the proximal dendrite of one pyramidal cell and the input from the EC excites the distal dendrites of both pyramidal cells. In this diagram, the filled circles represent inhibitory synapses, whereas the arrows represent excitatory synapses. Figure 6 Simulated CA3 network. The basket cell (B) inhibits itself, the OLM cell (OLM) and the pyramidal cells (P). The OLM cell inhibits the distal dendrites of the pyramidal cells. Pyramidal cells excite each other and both inhibitory cells. The input from the DG excites the proximal dendrite of one pyramidal cell and the input from the EC excites the distal dendrites of both pyramidal cells. In this diagram, the filled circles represent inhibitory synapses, whereas the arrows represent excitatory synapses. Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Page 13 of 22 glutamatergic receptors locate. As the conductance of glutamatergic receptors is related to the local membrane potentials [47,48], OLM cells control the efficacy of glutamatergic inputs [46]. Hence changes in firing patterns of the basket and OLM cells may result in pathological forms of population synchrony, which may lead to epilepsy [46]. cm: membrane capacity, Rm: membrane resistivity, Ra: intracellular resistivity, EL: equilibrium potential of the leak current, m p y m y a y L q p ENa: equilibrium potential of the sodium current, Ek: equilibrium potential of the K+ current. Pyramidal cell The distribution of the ionic conductance is similar with the Table 1 Passive parameters and ionic conductance of channels for all compartments of pyramidal cells Mechanism Soma Prox dend Med dend Dist dend Cm, μF/cm2 1 2 2 2 Rm, Ωcm2 60000 30000 30000 30000 Ra, Ωcm 200 200 200 200 Leak conductance [S/cm2] 0.000017 0.000033 0.000033 0.000033 Sodium conductance [S/cm2] 0.035 0.035 0.035 0.035 Delayed rectifier K+ conductance [S/cm2] 0.005 0.005 0.005 0.005 Ca2+-activated K+ conductance [S/cm2] 0.003 0.00225 0.00075 – AHP K+ conductance [S/cm2] 0.0009 – 0.000225 0.000225 M-type K+ conductance [S/cm2] 0 0.0023 0.00023 0.000023 L-type Ca2+ conductance [S/cm2] 0.0014 0.0014 – – N-type Ca2+ conductance [S/cm2] 0.0016 0.0016 0.0016 0.0016 T-type Ca2+ conductance [S/cm2] 0.0005 0.0005 0.0005 0.0005 EL (mV) −65 −65 −65 −65 ENa (mV) 50 50 50 50 EK (mV) −80 −80 −80 −80 cm: membrane capacity, Rm: membrane resistivity, Ra: intracellular resistivity, EL: equilibrium potential of the leak current, ENa: equilibrium potential of the sodium current, Ek: equilibrium potential of the K+ current. Table 1 Passive parameters and ionic conductance of channels for all compartments of pyramidal cells Table 1 Passive parameters and ionic conductance of channels for all compartments of id l ll Page 14 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Page 14 of 22 parameters of a bursting model of the CA3 pyramidal cell built by Lazarewicz et al. [41]. Responses of the pyramidal cell to current injections are shown in Figure 5A. We can see that the pyramidal cell generates bursts, which have been revealed to be the ability most pyramidal cells in the CA3 region have during the application of somatic current in- jections [4,5]. In addition, the firing pattern of this pyramidal cell is similar to the pattern measured in experiments [39]. As suggested by the CA3 models built earlier [24,31,36,37] and anatomical studies [21,38], each pyramidal cell receives somatic synaptic inhibition from the basket cell, distal apical inhibition from the OLM cell and medial dendritic excitation from the other pyr- amidal cell. It has been reported that the input from the DG projects proximally to the stratum lucidum, whereas the input from the EC projects distally to the stratum lacunosum-moleculare [36]. The input from the DG was relatively selective and projected to some of the CA3 pyramidal cells [36]. Pyramidal cell Thus, in our model, one pyramidal cell receives proximal excitation from the DG and both pyramidal cells receive distal excitation from the EC. Basket cell The simulated basket cell has one compartment [31,36,37] and contains leak current, sodium current, and delayed rectifier K+ current [31,53]. It is the same as the basket cell model built by Neymotin et al. [31]. It has been proven that basket cells are prevalent within the CA3 region [36], and they are fast-spiking interneurons [53]. With fast kinetics of the sodium current and delayed rectifier K+ current, the basket cell has the ability to fire repetitive spikes at high frequencies (Figure 5B). As a consequence, compared to the OLM cell, the basket cell affects pyramidal cells with a higher frequency. The basket cell inhibits the somata of both pyramidal cells directly and simultaneously, which help to control the synchrony of action potentials of pyramidal cells [46]. The parameters of ionic conductance used in the model are listed in Table 2. According to the CA3 models built earlier [36,37] and anatomical studies [38], the basket cell receives excitatory inputs from pyramidal cells and the inhibitory input from itself. OLM cell The simulated OLM cell has one compartment [31,36,37] and contains leak current, sodium current, delayed rectifier K+ current, Ca2+-activated K+ current, Ca2+ current, and hyperpolarization-activated current Ih [31,54]. The Ca2+-activated K+ current allows the cells to have long lasting inactivation after bursting [31], and as well as Ca2+ current, help the cells to produce spike-frequency adaptation [54]. The OLM cell also has hyperpolarization-activated current Ih for bursting [31]. The OLM cell projects to the dendrites of both pyramidal cells in the model, and affects the membrane potentials there. As glutamatergic receptors locate in the dendrites of pyramidal cells and their conductance is related to the local membrane potentials [47,48], the OLM cell controls the efficacy of glutamatergic receptors, which affects Mg2+-free-ACSF induced epileptiform activities. The parameters of ionic conductance used in the model are listed in Table 2, which are the same as the parameters used by Wang [54]. According to anatomical studies, the OLM cell receives excitatory inputs from pyramidal cells and the inhibitory input from the basket cell [37,38]. Page 15 of 22 Page 15 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Table 2 Passive parameters and ionic conductance of channels for all compartments of basket and OLM cells Mechanism Basket cell OLM cell Cm, μF/cm2 1 1 Rm, Ωcm2 60000 60000 Ra, Ωcm 200 200 Leak conductance [S/cm2] 0.000017 0.000017 Sodium conductance [S/cm2] 0.035 0.035 Delayed rectifier K+ conductance [S/cm2] 0.0009 0.0009 Ca2+-activated K+ conductance [S/cm2] – 0.01 Ca2+-activated [S/cm2] – 0.001 Ih conductance [S/cm2] – 0.15 EL (mV) −65 −65 ENa (mV) 66 55 EK (mV) −90 −90 ECa (mV) – 120 Eh (mV) – −40 cm: membrane capacity, Rm: membrane resistivity, Ra: intracellular resistivity, EL: equilibrium potential of the leak current, ENa: equilibrium potential of the sodium current, Ek: equilibrium potential of the K+ current, ECa: equilibrium potential of the Ca2+ current, Eh: equilibrium potential of the h current. Table 2 Passive parameters and ionic conductance of channels for all compartments of basket and OLM cells cm: membrane capacity, Rm: membrane resistivity, Ra: intracellular resistivity, EL: equilibrium potential of the leak current, ENa: equilibrium potential of the sodium current, Ek: equilibrium potential of the K+ current, ECa: equilibrium potential of the Ca2+ current, Eh: equilibrium potential of the h current. Synaptic properties In this model, AMPA, NMDA and GABAA receptors are considered, which are the commonest receptors in CA3 models [31,35]. As the network built here is small, AMPA and NMDA receptors are present in all excitatory connections, and GABAA receptors are present in all inhibitory connections. AMPA and GABAA receptors are modelled by a standard NEURON double-exponential mechanism [47,48]. The synaptic conductance gsyn(t) is given by: gsyn tð Þ ¼ c exp −t τ2 −exp −t τ1 ð1Þ ð1Þ where c is the weight, τ1 is the rising time constant, and τ2 is the falling time constant. The NMDA receptor conductance is dependent on the local membrane potential and the external Mg2+ concentration [31,47], and it is given by: where c is the weight, τ1 is the rising time constant, and τ2 is the falling time constant. The NMDA receptor conductance is dependent on the local membrane potential and the external Mg2+ concentration [31,47], and it is given by: gsyn tð Þ ¼ c exp −t τ2 −exp −t τ1 1 þ Mg 3:57 exp −0:062V tð Þ ð Þ ð2Þ ð2Þ where Mg is the external Mg2+ concentration in mM, V(t) is the local membrane potential in mV, and the other variables are as in eq. (1). AMPA and NMDA receptors have reversal potentials of 0 mV, while GABAA receptors have reversal potentials of −80 mV [31]. The parameters of synapses are present in Table 3, which are based on the model built by Neymotin et al. [31]. OLM cell cm: membrane capacity, Rm: membrane resistivity, Ra: intracellular resistivity, EL: equilibrium potential of the leak current, ENa: equilibrium potential of the sodium current, Ek: equilibrium potential of the K+ current, ECa: equilibrium potential of the Ca2+ current, Eh: equilibrium potential of the h current. Model inputs The model receives two extrahippocampal inputs from the DG and EC respectively. The input from the DG projects to the proximal dendrites of pyramidal cells, whereas Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Page 16 of 22 Table 3 Synaptic parameters Presynaptic Postsynaptic Receptor T1(ms) T2(ms) Conductance (μS) Pyramidal Pyramidal AMPA 0.05 15.3 0.00045 Pyramidal Pyramidal NMDA 15 150 0.00036 Pyramidal Basket AMPA 0.05 5.3 0.18 Pyramidal Basket NMDA 15 150 0.36 Pyramidal OLM AMPA 0.05 5.3 0.18 Pyramidal OLM AMPA 15 150 0.36 Basket Pyramidal GABAA 0.07 9.1 0.13 Basket Basket GABAA 0.07 9.1 0.65 Basket OLM GABAA 20 40 0.39 OLM Pyramidal GABAA 0.2 20 1.3 DG Pyramidal AMPA 0.05 5.3 0.0036 DG Pyramidal NMDA 15 150 0.0012 EC Pyramidal AMPA 0.05 5.3 0.00036 EC Pyramidal NMDA 15 150 0.000072 the input from the EC projects distally to the dendrites of pyramidal cells [55]. In the simulation, the input from the DG was relatively strong and selective, exciting only one of the pyramidal cells. However, the input from the EC was relatively weak and diffuse, exciting both pyramidal cells [21,36,55]. As suggested by the data from experiments, interictal discharges were induced by Mg2+-free-ACSF in hippocampal slices, and CA3 were considered to be the initiation place [16,18,19]. As hippocampal slices didn’t contain EC, we only considered the input from the DG, which followed a Possion process (λ = 1) in the model [31]. In combined EC-hippocampal slices, alternate interictal and ictal discharges were induced by Mg2+-free-ACSF, and epileptiform activities in the EC and DG were synchronized with each other [18,19]. An extral pyramidal cell was built, which received current injections and generated ictal-like and interictal-like discharges, to simulate the input signals from the DG and EC. AMPA and NMDA receptors were presented in all the connections, and the connection strength between the input from the DG and the model was larger than that between the input from the EC and the model. Pyramidal cell The somatic (s), proximal dendritic (pd), medium dendritic (md), and distal dendritic (dd) compartments obey the following current balance equations: c dVs dt ¼ −IL −INa −Ikdr −Ikc −IAHP −Ikm −ICaL −ICaN −ICaT −Isyn c dVpd dt ¼ −IL −INa −Ikdr −Ikc −Ikm −ICaL −ICaN −ICaT −Isyn c dVmd dt ¼ −IL −INa −Ikdr −Ikc −IAHP −Ikm −ICaN −ICaT −Isyn c dVdd dt ¼ −IL −INa −Ikdr −IAHP −Ikm −ICaN −ICaT −Isyn Where IL is the leak current, INa is the sodium current, Ikdr is the delayed rectifier K+ current, Ikc is the Ca2+ activated K+ current, IAHP is the AHP K+ current, Ikm is the Page 17 of 22 Page 17 of 22 Page 17 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 M-type K+ current, ICaL, ICaN, and ICaT are the L-, N- and T-type Ca2+ currents, and Isyn is the synaptic current. The conductance and reversal potential values for all ionic currents are listed in Table 1. M-type K+ current, ICaL, ICaN, and ICaT are the L-, N- and T-type Ca2+ currents, and Isyn is the synaptic current. The conductance and reversal potential values for all ionic currents are listed in Table 1. ICaN ¼ −gCaNm2hV 1−Ca2þ i Ca2þ o exp 2V F kT 1−exp 2V F kT αm ¼ 15:69 81:5−V ð Þ exp 81:5−V 10 −1 βm ¼ 0:29 exp −V 10:86 ICaT ¼ −gCaTm2hV 1−Ca2þ i Ca2þ o exp 2V F kT 1−exp 2V F kT Pyramidal cell The sodium current is described by [56]: INa ¼ gNam3h V−ENa ð Þ αm ¼ 0:32 13:1−V ð Þ exp 13:1−V 4 −1 βm ¼ 0:28 V−40:1 ð Þ exp V−40:1 5 −1 αh ¼ 0:128 exp 17−V 18 βh ¼ 4 1 þ exp 40−V 5 βm ¼ 0:28 V−40:1 ð Þ exp V−40:1 5 −1 αh ¼ 0:128 exp 17−V 18 βh ¼ 4 1 þ exp 40−V 5 The K+ currents are given by [39]: The K+ currents are given by [39]: Ikdr ¼ gkdrm3h V−Ek ð Þ αm ¼ 0:03 exp 2 V þ 32 ð Þ F RT βm ¼ 0:03 exp 3 V þ 32 ð Þ F RT αh ¼ 0:001 exp −2 V þ 61 ð Þ F RT βh ¼ 0:001 Ikm ¼ gkmm V−Ek ð Þ αm ¼ 0:006 exp 0:6 V þ 55 ð Þ F RT βm ¼ 0:06 exp −9:4 V þ 55 ð Þ F RT IAHP ¼ gAHPm V−Ek ð Þ αm ¼ 1:3 1013 Ca2þ 4 i βm ¼ 0:005 Ikc ¼ gkcm V−Ek ð Þ αm ¼ 0:28 Ca2þ i Ca2þ i þ 0:48 10−3 exp −1:68V F RT βm ¼ 0:48 1 þ Ca2þ i 0:13 10−6 exp −2V F RT Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 Page 18 of 22 http://www.tbiomed.com/content/11/1/14 Page 18 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 αm ¼ 0:19 19:88−V ð Þ exp 19:88−V 10 −1 βm ¼ 0:046 exp −V 20:73 αh ¼ 1:6 10−4 exp −V 48:4 βh ¼ 1 exp 39−V 10 þ 1 The Ca2+ currents are given by [57]: The Ca2+ currents are given by [57]: The Ca2+ currents are given by [57]: ICaL ¼ −gCaLm2V 1−Ca2þ i Ca2þ o exp 2V F kT 1−exp 2V F kT αm ¼ 15:69 81:5−V ð Þ exp 81:5−V 10 −1 βm ¼ 0:29 exp −V 10:86 ICaN ¼ −gCaNm2hV 1−Ca2þ i Ca2þ o exp 2V F kT 1−exp 2V F kT αm ¼ 0:19 19:88−V ð Þ exp 19:88−V 10 −1 βm ¼ 0:046 exp −V 20:73 αh ¼ 1:6 10−4 exp −V 48:4 βh ¼ 1 exp 39−V 10 þ 1 ICaT ¼ −gCaTm2hV 1−Ca2þ i Ca2þ o exp 2V F kT 1−exp 2V F kT αm ¼ 0:2 19:26−V ð Þ exp 19:26−V 10 −1 βm ¼ 0:009 exp −V 22:03 αh ¼ 1:0 10−6 exp −V 16:26 βh ¼ 1 exp 29:79−V 10 þ 1 αh ¼ 1:0 10−6 exp −V 16:26 Where [Ca2+]o = 2 mM, resting [Ca2+]i = 50 nM, and the dynamics of [Ca2+]i come from Migliore et al. [39]. In all the equations above, V is the membrane potential, k is the Boltzmann’s constant, F is the Faraday’s constant, T is the absolute temperature, R is the gas constant, [Ca2+]i is Page 19 of 22 Page 19 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 the intracellular Ca2+ concentration, and [Ca2+]o is the extracellular Ca2+ concentration. State variables m and h obey the following equations: the intracellular Ca2+ concentration, and [Ca2+]o is the extracellular Ca2+ concentration. State variables m and h obey the following equations: State variables m and h obey the following equations: dm dt ¼ αm 1−m ð Þ−βmm dh dt ¼ αh 1−h ð Þ−βhh dm dt ¼ αm 1−m ð Þ−βmm dh dt ¼ αh 1−h ð Þ−βhh dm dt ¼ αm 1−m ð Þ−βmm dh dt ¼ αh 1−h ð Þ−βhh Basket cell The somatic compartment obeys the following current balance equation: The somatic compartment obeys the following current balance equation: c dV dt ¼ −IL−INa−Ikdr−Isyn Where IL is the leak current, INa is the sodium current, Ikdr is the delayed rectifier K+ current, and Isyn is the synaptic current. The conductance and reversal potential values for all ionic currents are listed in Table 2. The model built here is the same as that built by Neymotin et al. [31]. The sodium current is described by: The sodium current is described by: INa ¼ gNam3h V−ENa ð Þ αm ¼ − 0:1 V þ 35 ð Þ exp −0:1 V þ 35 ð Þ ð Þ−1 βm ¼ 4 exp −V þ 60 18 αh ¼ 0:35 exp −V þ 58 20 βh ¼ 5 exp −0:1 V þ 28 ð Þ ð Þ þ 1 The delayed rectifier K+ current is given by: Ikdr ¼ gkdrm4 V −Ek ð Þ αm ¼ −0:05 V þ 34 ð Þ= exp −0:1 V þ 34 ð Þ ð Þ−1 ð Þ βm ¼ 0:625 exp −V þ 44 80 In all the equations above, V is the membrane potential, State variables m and h obey the following equations: dm dt ¼ αm 1 −m ð Þ −βmm dh dt ¼ αh 1 −h ð Þ −βhh dm dt ¼ αm 1 −m ð Þ −βmm dh dt ¼ αh 1 −h ð Þ −βhh Competing interests p g The authors declare that they have no competing interests. Authors’ contributions Authors’ contributions HR built the model and drafted the manuscript, YJS did the experiments, PMZ and QCL instructed the experiments, PMZ and PJL improved the model, PMZ improved the manuscript. All authors read and approved the final manuscript. Acknowledgments g This work was supported by the Key Basic Research Project of Science and Technology Commission of Shanghai (13DJ1400303), the Natural Science Foundation of Shanghai (12ZR1413800), the Shanghai Jiao Tong University Fund for Interdisciplinary Research for Medical Applications (YG2012ZD08), and the Seed Fund of Ren Ji Hospital (RJZZ13-005). OLM cell OLM cell The somatic compartment obeys the following current balance equation: The somatic compartment obeys the following current balance equation: c dV dt ¼ −IL −INa −Ikdr −Ih −ICa −Ikc −Isyn Page 20 of 22 Page 20 of 22 Page 20 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Where IL is the leak current, INa is the sodium current, Ikdr is the delayed rectifier K+ current, Ih is the hyperpolarization-activated current, ICa is the Ca2+ current, Ikc is the Ca2+ activated K+ current, and Isyn is the synaptic current. In this model, the sodium current INa and the delayed rectifier K+ current Ikdr are the same as in the basket cell. The conductance and reversal potential values for all ionic currents are listed in Table 2. The model built here is the same as that built by Wang [54]. The h current is described by: The h current is described by: Ih ¼ ghn V−Eh ð Þ n∞¼ 1= 1 þ exp V þ 80 10 τn ¼ 200 exp V þ 70 20 þ exp −V þ 70 20 þ 5 Ih ¼ ghn V−Eh ð Þ n∞¼ 1= 1 þ exp V þ 80 10 200 τn ¼ 200 exp V þ 70 20 þ exp −V þ 70 20 þ 5 n∞¼ αn αn þ βn τn ¼ 1 αn þ βn dn dt ¼ αn 1 −n ð Þ −βnn The Ca2+ current is given by: ICa ¼ gCam2 ∞V −ECa ð Þ m∞¼ 1= 1 þ exp −V þ 20 ð Þ 9 Where m is replaced by its steady-state m∞. The Ca2+ activated K+ current is given by: Where m is replaced by its steady-state m∞. The Ca2+ activated K+ current is given by: Where m is replaced by its steady-state m∞. Where m is replaced by its steady-state m∞. The Ca2+ activated K+ current is given by: Ikc ¼ gkc Ca2þ i Ca2þ i þ 30 V−Ek ð Þ d Ca2þ i dt ¼ −0:002ICa −Ca2þ i=80 In all the equations above, V is the membrane potential, [Ca2+]i is the intracellular Ca2+ concentration. Author details 1 Author details 1School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China. 2Department of Neurology, Ren Ji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai 200127, China. Page 21 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Received: 19 January 2014 Accepted: 17 March 2014 Published: 24 March 2014 References III of the entorhinal cortex in patients with temporal lobe epilepsy. Epilepsy Res 1993, 16:223–233. 10. Bartolomei F, Khalil M, Wendling F, Sontheimer A, Regis J, Ranjeva JP, Guye M, Chauvel P: Entorhinal cortex involvement in human mesial temporal lobe epilepsy: an electrophysiologic and volumetric study. Epilepsia 2005, 46:677–687. 11. de Curtis M, Jefferys JGR, Avoli M: Interictal Epileptiform Discharges in Partial Epilepsy: Complex Neurobiological Mechanisms Based on Experimental and Clinical Evidence. In Jasper’s Basic Mechanisms of the Epilepsies [Internet]; 2012. 12. Charpak S, Pare D, Llinas R: The entorhinal cortex entrains fast CA1 hippocampal oscillations in the anaesthetized guinea-pig: role of the monosynaptic component of the perforant path. Eur J Neurosci 1995, 7:1548–1557. 13. Kumar SS, Jin X, Buckmaster PS, Huguenard JR: Recurrent circuits in layer II of medial entorhinal cortex in a model of temporal lobe epilepsy. J Neurosci 2007, 27:1239–1246. 14. Nagao T, Alonso A, Avoli M: Epileptiform activity induced by pilocarpine in the rat hippocampal-entorhinal slice preparation. Neuroscience 1996, 72:399–408. slice preparation. Neuroscience 1996, 72:399–408. p p 15. Yang F, Gong XW, Gong HQ, Zhang PM, Liang PJ, Lu QC: Microelectrode array recording of excitability of low Mg2+ −induced acute hippocampal slices. Neural Regeneration Research 2010, 5:1548–1551. 16. Gong XW, Yang F, Liu JS, Lu QC, Gong HQ, Liang PJ, Zhang PM: Investigation of The Initiation Site and Propagation of Epileptiform Discharges in Hippocampal Slices Using Microelectrode Array. Progress Biochem Biophys 2010, 37:1240–1247. y 17. Li JB, Lu QC, Gong XW, Gong HQ, Zhang PM, Liang PJ: The Modulatory Effects of Exogenous Zn2+ on Propagation and Frequency of Epileptiform Discharges on Rat Hippocampal Slices. Acta Biophysica Sinica 2012, 2:119–130. 18. Shi YJ, Lu QC, Gong XW, Gong HQ, Zhang PM, Liang PJ: Comparative study of the characteristics of epileptiform discharges in the hippocampal slices and combined entorhinal cortex-hippocampal slices. Chinese J Clin (Electronic Edition) 2013, 7:5397–5400. 19. Shi YJ, Gong XW, Gong HQ, Liang PJ, Zhang PM, Lu QC: Effect of the entorhinal cortex on ictal discharges in low-Mg2+-induced epileptic hippocampal slice models. Neural Plasticity 2014. doi:10.1155/2014/205912. g p p pp p y 20. Jefferys JGR, Jiruska P, de Curtis M, Avoli M: Limbic Network Synchronization and Temporal Lobe Epilepsy. In Jasper’s Basic Mechanisms of the Epilepsies [Internet]. 2012. 21. Andersen P, Morris R, Amaral D, Bliss T, O’Keefe J: The hippocampus book. Oxford University Press; 2007. 21. References Andersen P, Morris R, Amaral D, Bliss T, O’Keefe J: The hippocam pp p y 22. Traub RD, Miles R, Buzsaki G: Computer simulation of carbachol-driven rhythmic population oscillations in the CA3 region of the in vitro rat hippocampus. J Physiol 1992, 451:653–672. 22. Traub RD, Miles R, Buzsaki G: Computer simulation of carbachol-driven rhythmic population oscillations in the CA3 region of the in vitro rat hippocampus J Physiol 1992 451 653 672 22. Traub RD, Miles R, Buzsaki G: Computer simulation of carbachol-driven rhyth CA3 region of the in vitro rat hippocampus. J Physiol 1992, 451:653–672. g pp p y 23. Traub RD, Miles R, Jefferys JG: Synaptic and intrinsic conductances shape picrotoxin-induced synchronized after-discharges in the guinea-pig hippocampal slice. J Physiol 1993, 461:525–547. y 24. Traub RD, Jefferys JG, Whittington MA: Enhanced NMDA conductance can account for epileptiform activity i d d b l M 2 i h hi l li J Ph i l 1994 478 379 393 24. Traub RD, Jefferys JG, Whittington MA: Enhanced NMDA conductance can account for epileptiform activity induced by low Mg2+ in the rat hippocampal slice. J Physiol 1994, 478:379–393. 25. Johannessen CU, Johannessen SI: Valproate: past, present, and future. CNS Drug Rev 2003, 9:199–216. 26. Bruckner C, Heinemann U: Effects of standard anticonvulsant drugs on different patterns of epileptiform discharges induced by 4-aminopyridine in combined entorhinal cortex-hippocampal slices. Brain Res 2000, 859:15–20. 27. D’Antuono M, Kohling R, Ricalzone S, Gotman J, Biagini G, Avoli M: Antiepileptic drugs abolish ictal but not interictal epileptiform discharges in vitro. Epilepsia 2010, 51:423–431. 28. Fueta Y, Avoli M: Effects of antiepileptic drugs on 4-aminopyridine-induced epileptiform activity in young and adult rat hippocampus. Epilepsy Res 1992, 12:207–215. 28. Fueta Y, Avoli M: Effects of antiepileptic drugs on 4-am adult rat hippocampus. Epilepsy Res 1992, 12:207–215. pp p p p y 29. Hines ML, Carnevale NT: The NEURON simulation environment. Neural Comput 1997, 9:1179–1209. 29. Hines ML, Carnevale NT: The NEURON simulation environme 30. Lopantsev V, Avoli M: Participation of GABAA-mediated inhibition in ictallike discharges in the rat entorhinal cortex. J Neurophysiol 1998, 79:352–360. p y 31. Neymotin SA, Lazarewicz MT, Sherif M, Contreras D, Finkel LH, Lytton WW: Ketamine disrupts theta modula of gamma in a computer model of hippocampus. J Neurosci 2011, 31:11733–11743. 31. References 1. Schmidt D, Loscher W: Drug resistance in epilepsy: putative neurobiologic and clinical mechanisms. Epilepsia 2005, 46:858–877. 2. Bertram EH: Temporal lobe epilepsy: where do the seizures really begin? Epilepsy Behav 2009, 14:32– ys JG: Models and mechanisms of experimental epilepsies. Epilepsia 3. Jefferys JG: Models and mechanisms of experimental epilepsies. Epilepsia 2003, 44:44–50. 4. McCormick DA, Contreras D: On the cellular and network bases of epileptic seizures. Annu Rev Physiol 2001, 63:815–846. 5. McNamara JO: Cellular and molecular basis of epilepsy. J Neurosci 1994, 14:3413–3425. McNamara JO: Cellular and molecular basis of epilepsy. J Neurosci 6. Avoli M, D’Antuono M, Louvel J, Kohling R, Biagini G, Pumain R, D’Arcangelo G, Tancredi V: Network and pharmacological mechanisms leading to epileptiform synchronization in the limbic system in vitro. Prog Neurobiol 2002, 68:167–207. 7. Jutila L, Ylinen A, Partanen K, Alafuzoff I, Mervaala E, Partanen J, Vapalahti M, Vainio P, Pitkanen A: MR volumetry of the entorhinal, perirhinal, and temporopolar cortices in drug-refractory temporal lobe epilepsy. AJNR Am J 7. Jutila L, Ylinen A, Partanen K, Alafuzoff I, Mervaala E, Partanen J, Vapalahti M, Vainio P, Pitkanen A: MR volumetry of the entorhinal, perirhinal, and temporopolar cortices in drug-refractory temporal lobe epilepsy. AJNR Am J Neuroradiol 2001, 22:1490–1501. 8. Du F, Eid T, Lothman EW, Kohler C, Schwarcz R: Preferential neuronal loss in layer III of the medial entorhinal 8. Du F, Eid T, Lothman EW, Kohler C, Schwarcz R: Preferential neuronal loss in layer III of the medial entorhina cortex in rat models of temporal lobe epilepsy. J Neurosci 1995, 15:6301–6313. cortex in rat models of temporal lobe epilepsy. J Neurosci 1995, n rat models of temporal lobe epilepsy. J Neurosci 1995, 15:6301–63 9. Du F, Whetsell WO Jr, Abou-Khalil B, Blumenkopf B, Lothman EW, Schwarcz R: Preferential neuronal loss in layer III f th t hi l t i ti t ith t l l b il E il R 1993 16 223 233 9. Du F, Whetsell WO Jr, Abou-Khalil B, Blumenkopf B, Lothman EW, Schwarcz R: Preferential neuronal loss in layer III of the entorhinal cortex in patients with temporal lobe epilepsy. Epilepsy Res 1993, 16:223–233. 9. Du F, Whetsell WO Jr, Abou Khalil B, Blumenkopf B, Lothman EW, Schwarcz R: Preferential neuronal loss in laye III of the entorhinal cortex in patients with temporal lobe epilepsy. Epilepsy Res 1993, 16:223–233. Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 33. Rutecki PA: Noradrenergic modulation of epileptiform activity in the hippocampus. Epilepsy Res 1995, 20:125–136. 34. Froriep UP, Kumar A, Cosandier-Rimele D, Haussler U, Kilias A, Haas CA, Egert U: Altered theta coupling between medial entorhinal cortex and dentate gyrus in temporal lobe epilepsy. Epilepsia 2012, 53:1937–1947. 35. Stanley DA, Talathi SS, Parekh MB, Cordiner DJ, Zhou J, Mareci TH, Ditto WL, Carney PR: Phase shift in the 24-hour rhythm of hippocampal EEG spiking activity in a rat model of temporal lobe epilepsy. J Neurophysiol 2013, 110:1070–1086. 36. Kunec S, Hasselmo ME, Kopell N: Encoding and retrieval in the CA3 region of the hippocampus: a model of theta-phase separation. J Neurophysiol 2005, 94:70–82. theta-phase separation. J Neurophysiol 2005, 94:70–82. 37. Tort ABL, Rotstein HG, Dugladze T, Gloveli T, Kopell NJ: On the formation of gamma-coherent cell assemblies by oriens lacunosum-moleculare interneurons in the hippocampus. Proc Natl Acad Sci USA 2007, 104:13490–13495. 38 Freund TF Buzsaki G: Interneurons of the hippocampus Hippocampus 1996 6:347 470 Freund TF, Buzsaki G: Interneurons of the hippocampus. Hippocam 39. Migliore M, Cook EP, Jaffe DB, Turner DA, Johnston D: Computer simulations of morphologically reconstructed CA3 hippocampal neurons. J Neurophysiol 1995, 73:1157–1168. 39. Migliore M, Cook EP, Jaffe DB, Turner DA, Johnston D: Computer CA3 hippocampal neurons. J Neurophysiol 1995, 73:1157–1168. y 0. Yaari Y, Beck H: “Epileptic neurons” in temporal lobe epilepsy. B 41. Lazarewicz MT, Migliore M, Ascoli GA: A new bursting model of CA3 pyramidal cell physiology su multiple locations for spike initiation. Biosystems 2002, 67:129–137. 41. Lazarewicz MT, Migliore M, Ascoli GA: A new bursting model of CA3 multiple locations for spike initiation. Biosystems 2002, 67:129–137. 41. Lazarewicz MT, Migliore M, Ascoli GA: A new bursting model of CA3 pyramidal cell physiology suggests multiple locations for spike initiation Biosystems 2002 67:129 137 2. Johannessen CU: Mechanisms of action of valproate: a commen 42. Johannessen CU: Mechanisms of action of valproate: a commentatory. Neurochem Int 2000, 37:103–110. 43. Franceschetti S, Hamon B, Heinemann U: The action of valproate on spontaneous epileptiform activity in the absence of synaptic transmission and on evoked changes in [Ca2+]o and [K+]o in the hippocampal slice. Brain Res 1986, 386:1–11. 44. Yamamoto R, Yanagita T, Kobayashi H, Yokoo H, Wada A: Up-regulation of sodium channel subunit mRNAs and their cell surface expression by antiepileptic valproic acid: activation of calcium channel and catecholamine secretion in adrenal chromaffin cells. Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 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 References Neymotin SA, Lazarewicz MT, Sherif M, Contreras D, Finkel LH, Lytton WW: Ketamine disrupts theta modulation f d l f h a computer model of hippocampus. J Neurosci 2011, 31:11733–11743 32. Jonas P, Major G, Sakmann B: Quantal components of unitary EPSCs at the mossy fibre synapse on CA3 id l ll f t hi J Ph i l 1993 472 615 663 32. Jonas P, Major G, Sakmann B: Quantal components of unitary EPS pyramidal cells of rat hippocampus. J Physiol 1993, 472:615–663. 32. Jonas P, Major G, Sakmann B: Quantal components of unitary EPSCs at the mossy fibre synapse on CA3 pyramidal cells of rat hippocampus. J Physiol 1993, 472:615–663. Page 22 of 22 Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 doi:10.1186/1742-4682-11-14 Cite this article as: Ren et al.: The role of the entorhinal cortex in epileptiform activities of the hippocampus. Theoretical Biology and Medical Modelling 2014 11:14. Ren et al. Theoretical Biology and Medical Modelling 2014, 11:14 http://www.tbiomed.com/content/11/1/14 J Neurochem 1997, 68:1655–1662. 45. Loscher W: Valproate: a reappraisal of its pharmacodynamic properties and mechanisms of action. Prog Neurobiol 1999, 58:31–59. 46. Parekh R: Interactions between oriens, basket and pyramidal cells during in vitro seizure-like events in CA1 of the rat hippocampus. Dissertation of Philosophy Doctor, Pennsylvania State University; 2010. 47. Baker JL, Perez-Rosello T, Migliore M, Barrionuevo G, Ascoli GA: A computer model of unitary responses from associational/commissural and perforant path synapses in hippocampal CA3 pyramidal cells. J Comput Neurosci 2011, 31:137–158. 48. Cutsuridis V, Cobb S, Graham BP: Encoding and retrieval in a model of the hippocampal CA1 microcircuit. Hippocampus 2010, 20:423–446. 49. Liu YH, Wang XJ: Spike-frequency adaptation of a generalized leaky integrate-and-fire model neuron. J Comput Neurosci 2001, 10:25–45. p 50. Lerche H, Jurkat-Rott K, Lehmann-Horn F: Ion channels and epilepsy. Am J Med Genet 2001, 106:146–159. 51. Blalock EM, Chen KC, Vanaman TC, Landfield PW, Slevin JT: Epilepsy-induced decrease of L-type Ca2+ channel activity and coordinate regulation of subunit mRNA in single neurons of rat hippocampal ‘zipper’ slices. Epilepsy Res 2001, 43:211–226. 52. Zamponi GW, Lory P, Perez-Reyes E: Role of voltage-gated calcium channels in epilepsy. Pflugers Arch 2010, 460:395–403. 53. Wang XJ, Buzsaki G: Gamma oscillation by synaptic inhibition in a hippocampal interneuronal network model. J Neurosci 1996, 16:6402–6413. 54. Wang XJ: Pacemaker neurons for the theta rhythm and their synchronization in the septohippocampal reciprocal loop. J Neurophysiol 2002, 87:889–900. 55. Witter MP: Organization of the entorhinal-hippocampal system: a review of current anatomical data. Hippocampus 1993, 3:33–44. 56. Traub RD, Wong RK, Miles R, Michelson H: A model of a CA3 hippocampal pyramidal neuron incorporating voltage-clamp data on intrinsic conductances. J Neurophysiol 1991, 66:635–650. 57. Jaffe DB, Ross WN, Lisman JE, Lasser-Ross N, Miyakawa H, Johnston D: A model for dendritic Ca2+ accumulation in hippocampal pyramidal neurons based on fluorescence imaging measurements. J Neurophysiol 1994, 71:1065–1077. 57. Jaffe DB, Ross WN, Lisman JE, Lasser-Ross N, Miyakawa H, Johnston D: A model for dendritic Ca2+ accumulation in hippocampal pyramidal neurons based on fluorescence imaging measurements. J Neurophysiol 1994, 71:1065–1077. doi:10.1186/1742-4682-11-14 Cite this article as: Ren et al.: The role of the entorhinal cortex in epileptiform activities of the hippocampus. Theoretical Biology and Medical Modelling 2014 11:14. doi:10.1186/1742-4682-11-14 Cite this article as: Ren et al.: The role of the entorhinal cortex in epileptiform activities of the hippocampus. Theoretical Biology and Medical Modelling 2014 11:14. 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 • 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 • Convenient online submission
https://openalex.org/W2759734524	https://www.frontiersin.org/articles/10.3389/fnins.2017.00532/pdf	English	null	Lateralization of Executive Function: Working Memory Advantage for Same Hemifield Stimuli in the Monkey	Frontiers in neuroscience	2,017	cc-by	8,726	Lateralization of Executive Function: Working Memory Advantage for Same Hemiﬁeld Stimuli in the Monkey Hua Tang 1, 2, Mitchell R. Riley 1 and Christos Constantinidis 1* 1 Department of Neurobiology & Anatomy, Wake Forest School of Medicine, Winston-Salem, NC, United States, 2 School of Life Science and Institute of Life Science, Nanchang University, Nanchang, China Working memory capacity, the amount of information that may be maintained in mind over a period of seconds, is extremely limited, to a handful of items. Some evidence exists that the number of visual items that may be maintained in working memory is independent for the two hemiﬁelds. To test this idea, we trained monkeys to perform visual working memory tasks that required maintenance in memory of the locations and/or shapes of 3–5 visual stimuli. We then tested whether systematic performance differences were present for stimuli concentrated in the same hemiﬁeld, vs. distributed across hemiﬁelds. We found little evidence to support the expectation that working memory capacity is independent in the two hemiﬁelds. Instead, when an advantage of stimulus arrangement was present, it involved multiple stimuli presented in the same hemiﬁeld. This conclusion was consistent across variations of the task, performance levels, and apparent strategies adopted by individual subjects. This result suggests that factors such as grouping that favor processing of stimuli in relative proximity may counteract the beneﬁts of independent processing in the two hemispheres. Our results reveal an important property of working memory and place constraints on models of working memory capacity. Edited by: Aldo Genovesio, Sapienza Università di Roma, Italy Edited by: Aldo Genovesio, Sapienza Università di Roma, Italy Edited by: Aldo Genovesio, Sapienza Università di Roma, Italy Reviewed by: Natasha Sigala, Brighton and Sussex Medical School, United Kingdom Suliann Ben Hamed, UMR5229 Institut des Sciences Cognitives Marc Jeannerod, France Correspondence: Christos Constantinidis cconstan@wakehealth.edu Correspondence: Christos Constantinidis cconstan@wakehealth.edu Specialty section: This article was submitted to Decision Neuroscience, a section of the journal Frontiers in Neuroscience INTRODUCTION Hemispheric specialization is an important principle of cortical processing, for vision and other sensory modalities (Essen and Zeki, 1978). The primary visual cortex of vertebrates displays hemispheric specialization so that information from one hemiﬁeld is represented by neurons located in the contralateral hemisphere. This contralateral bias is diminished in successive stages of processing along the cortical pathways, and neurons with bilateral or ipsilateral receptive ﬁelds are more frequent in the posterior parietal and inferior temporal cortices (Mountcastle et al., 1975; Desimone and Gross, 1979). Nonetheless, a number of psychophysical phenomena reveal perceptual advantages for stimuli presented in the same hemiﬁeld. For example, subjects are faster at detecting repeating visual stimuli, perceiving illusory contours, or judging whether stimuli are the same or diﬀerent when presented in the same hemiﬁeld (Pillow and Rubin, 2002; Butcher and Cavanagh, 2008; Hayes et al., 2010). Received: 14 April 2017 Accepted: 13 September 2017 Published: 26 September 2017 Received: 14 April 2017 Accepted: 13 September 2017 Published: 26 September 2017 Keywords: working memory, monkey, visual ﬁeld, cerebral hemisphere, visuospatial memory ORIGINAL RESEARCH published: 26 September 2017 doi: 10.3389/fnins.2017.00532 Behavioral Tasks The monkeys faced a computer monitor 60 cm away in a dark room with their head ﬁxed, as described in detail previously (Qi et al., 2010; Meyer et al., 2011). Eye position was sampled at 240 Hz, digitized, and recorded with an infrared eye position tracking system (model RK-716; ISCAN, Burlington, MA). The visual stimulus presentation and behavior monitoring were controlled by in-house software (Meyer and Constantinidis, 2005) implemented in the MATLAB computational environment (Mathworks, Natick, MA), using the Psychophysics Toolbox (Brainard, 1997). Monkeys were required to maintain their gaze on the ﬁxation target throughout a trial; breaks in ﬁxation aborted the trial. On the other hand, factors, such as grouping, Gestalt principles of proximity and connectedness, and the speciﬁc arrangement of stimuli may confer an advantage for groups of stimuli maintained in memory, over stimuli distributed between ﬁelds (Jiang et al., 2000; Xu, 2006; Peterson and Berryhill, 2013). It is not obvious, therefore, that multi-stimulus displays will provide a general advantage when distributed between hemiﬁelds over being concentrated in the same ﬁeld. Within- ﬁeld advantages are thought to occur due to processes, such as perceptual grouping, which likely originate in early visual cortex and are propagated along the visual pathways (Banich, 1998a,b; Weissman et al., 2000). All four monkeys were trained with spatial versions of the Match/Nonmatch task (Figures 1A,B). Additionally, one of the monkeys was trained with a feature Match/Nonmatch task (Figure 1C). The basic spatial Match/Nonmatch task (Figure 1A) required monkeys to remember the locations of multiple stimuli in a cue display and to determine if a second display was identical or not. For two monkeys (DA and CA) the trial started with the monkeys pulling the lever to initiate the trial, and keeping their eyes ﬁxated on a central ﬁxation target. After 1 s of stable ﬁxation, a cue display was presented for 0.5 s. The cue display consisted of 1–5 white squares, measuring 1.5◦of visual angle in size. Each square was displayed at 1 of 24 possible locations arranged on a circle at an eccentricity of 10◦of visual angle, with a 15◦angular separation between locations (Figure 1E). This was followed by a delay period of 1 s when only the ﬁxation target was visible. Citation: Tang H, Riley MR and Constantinidis C (2017) Lateralization of Executive Function: Working Memory Advantage for Same Hemiﬁeld Stimuli in the Monkey. Front. Neurosci. 11:532. doi: 10.3389/fnins.2017.00532 September 2017 \| Volume 11 \| Article 532 Frontiers in Neuroscience \| www.frontiersin.org Working Memory across Hemiﬁelds Tang et al. MATERIALS AND METHODS Among higher cognitive functions, language and visual- spatial representations are strongly lateralized in humans, so that their processing occurs predominantly by a single hemisphere (Corballis, 2012; Skeide and Friederici, 2016). Executive functions, including working memory, are not thought to be lateralized, nor has any advantage been associated with processing involving a single hemisphere. To the contrary, a bilateral ﬁeld advantage has been postulated for tasks with high computational complexity, which might beneﬁt from the processing power of two hemispheres (Leblanc-Sirois and Braun, 2014: Umemoto et al., 2010). Working memory and attention are notoriously subject to a processing bottleneck which limits how many items can be processed or maintained in mind at any point in time (Constantinidis and Klingberg, 2016). Splitting items between the left and right hemiﬁeld has been shown to confer an advantage when maintaining in memory multiple items, as parallel processing by the left and right hemisphere can expand the capacity of visual information processing (Delvenne, 2005; Delvenne et al., 2011; Hudson et al., 2012). In recent years, direct evidence has emerged that working memory capacity and spatial attention may operate independently in the two hemiﬁelds (Alvarez and Cavanagh, 2005; Alvarez et al., 2012; Stormer et al., 2014). Four male, rhesus monkeys (Macaca mulatta) weighing 7–13 kg were used in this study. Experiments were carried out in accordance with the recommendations of the U.S. Public Health Service Policy on Humane Care and Use of Laboratory Animals and the National Research Council’s Guide for the Care and Use of Laboratory Animals. The protocol was approved by the Wake Forest University Institutional Animal Care and Use Committee. Frontiers in Neuroscience \| www.frontiersin.org Behavioral Tasks Subsequently, a second display appeared with the same number of stimuli as the cue, either at identical locations (constituting a match), or with one item appearing at a diﬀerent location (constituting a nonmatch). The second display was presented for 0.5 s and after its oﬀset the animals were required to release the lever within 0.5 s if it was a match, or to continue holding for 0.5 s if it was a nonmatch. The monkeys received a liquid reward for a correct response. The trial was immediately aborted if the monkeys released the lever at any other time during the trial, or if the monkeys broke ﬁxation at any point prior to the lever release. Non-human primates are capable of mastering tasks requiring memory for multiple stimuli (Buschman et al., 2011; Heyselaar et al., 2011; Lara and Wallis, 2012), which allows for the neural basis of working memory capacity and its limitations to be investigated with neurophysiological means. Indeed, neurophysiological studies have begun to investigate the representation of multiple-stimulus information in memory by individual neurons (Warden and Miller, 2007; Buschman et al., 2011; Lara and Wallis, 2012). Experiments in monkeys also suggest that activity representing multiple items in memory declines to a greater extent when these appear in the same hemiﬁeld as opposed to diﬀerent hemiﬁelds, thus eroding the information representing the multiple items, in each hemiﬁeld (Buschman et al., 2011; Matsushima and Tanaka, 2014). However, monkeys may employ diﬀerent strategies to perform multi-item working memory tasks (Wittig et al., 2016). The other two monkeys (EL and NI) were trained in a variation of this spatial Match/Nonmatch task (Figure 1B). For these animals, a trial started with a 1.0 s ﬁxation interval. Then a cue display was presented for 0.5 s, containing 1–5 white squares, as in the basic task. After a delay period of 1.0 s, a second display appeared with the same number of stimuli. Two choice targets appeared subsequently at the bottom and top of the screen, one of which was green and the second blue, with their position pseudo- randomly interleaved between trials. The monkey was required We were motivated to investigate how general such bilateral beneﬁts are across working memory tasks in non-human primates, and across individual subjects. We therefore trained monkeys in variations of tasks that require memory for multiple stimuli and determined their behavior for displays containing stimuli in one or both hemiﬁelds. Behavioral Tasks The sequence of displays is the same as in (A), but two choice targets were presented at the end of a trial. The monkey was required to saccade to a green target if the two sequential displays matched each other, or to a blue target otherwise. (C) Schematic illustration of the feature Match/Nonmatch task. The monkey was now required to remember the shapes of stimuli appearing at different locations during the cue interval. During the second stimulus presentation, an identical display constituted a match, and the monkey was required to saccade to a green target. Alternatively, the same number of stimuli appeared, at the same locations, but one stimulus differed in shape from the cue display, and constituted a nonmatch. The monkey was then required to saccade to the blue target. (D) Stimulus set consisting of eight possible shapes in the feature Match/Nonmatch task. (E) The 24 possible locations where stimuli could appear in the spatial Match/Nonmatch task. to saccade to the green target if the two stimuli displays were identical, and to the blue target if they were not. completed trials. Displays were grouped based on the number of stimuli, and on the arrangement of stimuli in the same hemiﬁeld, or across hemiﬁelds. Analysis of performance levels across displays of diﬀerent numbers of stimuli was performed with non-parametric 1-way tests, the rank-sum (also known as Mann-Whitney) test, equivalent to the parametric t-test, and the Kruskal-Wallis test, equivalent to the one-way ANOVA test, and the non-parametric two-way Scheirer-Ray-Hare test, equivalent to the two-way ANOVA test. In all instances, the proportion of correct trials in one behavioral session (typically of 150–200 correct trials) was used as one observation. A regression model was also used, testing the dependence of performance on the number of stimuli in the display, the mean distance between stimuli, and whether they all appeared in the same side or not. The distance used for this analysis was deﬁned as the mean distance of all possible stimulus combinations in the display. All analysis of behavioral data was performed in the MATLAB environment, version R2012-2015a (Mathworks, Natick MA). One monkey was additionally trained in the shape Match/Nonmatch task (Figure 1C). In this task, the monkey was required to remember the shapes of stimuli appearing at multiple locations. The trial started with a 1.0 s ﬁxation interval and then a cue display was presented for 0.5 s. Behavioral Tasks September 2017 \| Volume 11 \| Article 532 2 Tang et al. Working Memory across Hemiﬁelds FIGURE 1 \| Working memory capacity tasks. (A) Successive frames illustrate the sequence of stimulus presentations in the spatial Match/Nonmatch task requiring a lever pull. The cue is presented and after a delay period, a match or nonmatch stimulus display appears. The monkeys were required to remember the locations of all the squares in the cue stimulus and to withhold the lever if an identical stimulus display appeared (match). If one of the squares appeared at a new location, then the display constituted a nonmatch and the monkey was required to release the lever. (B) Schematic illustration of the spatial Match/Nonmatch task requiring an eye movement to one of two choice targets. The sequence of displays is the same as in (A), but two choice targets were presented at the end of a trial. The monkey was required to saccade to a green target if the two sequential displays matched each other, or to a blue target otherwise. (C) Schematic illustration of the feature Match/Nonmatch task. The monkey was now required to remember the shapes of stimuli appearing at different locations during the cue interval. During the second stimulus presentation, an identical display constituted a match, and the monkey was required to saccade to a green target. Alternatively, the same number of stimuli appeared, at the same locations, but one stimulus differed in shape from the cue display, and constituted a nonmatch. The monkey was then required to saccade to the blue target. (D) Stimulus set consisting of eight possible shapes in the feature Match/Nonmatch task. (E) The 24 possible locations where stimuli could appear in the spatial Match/Nonmatch task. FIGURE 1 \| Working memory capacity tasks. (A) Successive frames illustrate the sequence of stimulus presentations in the spatial Match/Nonmatch task requiring a lever pull. The cue is presented and after a delay period, a match or nonmatch stimulus display appears. The monkeys were required to remember the locations of all the squares in the cue stimulus and to withhold the lever if an identical stimulus display appeared (match). If one of the squares appeared at a new location, then the display constituted a nonmatch and the monkey was required to release the lever. (B) Schematic illustration of the spatial Match/Nonmatch task requiring an eye movement to one of two choice targets. Behavioral Tasks The cue display comprised 1–5 white geometric shapes ﬁtting within a 2◦aperture, drawn from a set of 8 shapes (Figure 1D), which we have described in detail before (Meyer et al., 2007). These were displayed at 1 of 16 locations arranged on a circle at an eccentricity of 10◦of visual angle, with 22.5◦angular separation between locations. After a delay period of 1 s, a second display appeared with the same number of stimuli at identical locations. In the second display, either all shapes were identical, or one new shape substituted one of the shapes in the cue display. The monkey was required to saccade to the green choice target appearing at the end of the trial if the two stimuli displays were identical, and to the blue target if they were not. Frontiers in Neuroscience \| www.frontiersin.org Data Analysis The improvement represented a 12.5% increase in total performance for subject CA who achieved the lowest overall performance, a 3.9% increase for subject NI, a 3.5% increase for subject DA, and a 0.5% improvement for subject EL, who achieved > 90% performance for this type of stimuli. FIGURE 2 \| Behavioral performance in the working memory capacity task as a function of number of stimuli. Proportion of correct trials is plotted for four different subjects (EL, NI, DA, and CA) in the spatial task, and one subject (EL) in the shape task. Error bars represent mean ± SEM across daily sessions. FIGURE 2 \| Behavioral performance in the working memory capacity task as a function of number of stimuli. Proportion of correct trials is plotted for four different subjects (EL, NI, DA, and CA) in the spatial task, and one subject (EL) in the shape task. Error bars represent mean ± SEM across daily sessions. For displays with 4 stimuli, again the lowest performance was observed when stimuli appeared in both ﬁelds, than when they all appeared either in the left or right side. The eﬀect reached statistical signiﬁcance, evaluated with a Kruskal-Wallis test, for two animals CA (H = 17.33, df = 2, p = 1 × 10−4, n = 10 vs. 55 vs. 55 sessions for the three possible arrangements) and NI (H = 23.75, df = 2, p = 7 × 10−6, n = 54 vs. 84 vs. 84 sessions). For displays with 4 stimuli, again the lowest performance was observed when stimuli appeared in both ﬁelds, than when they all appeared either in the left or right side. The eﬀect reached statistical signiﬁcance, evaluated with a Kruskal-Wallis test, for two animals CA (H = 17.33, df = 2, p = 1 × 10−4, n = 10 vs. 55 vs. 55 sessions for the three possible arrangements) and NI (H = 23.75, df = 2, p = 7 × 10−6, n = 54 vs. 84 vs. 84 sessions). The beneﬁt represented an 18.4 and 11.0% improvement for stimuli in the same ﬁeld, over the average of the two split- stimulus arrangements. A similar trend, albeit without reaching signiﬁcance, was present for monkey DA (3.7% improvement, p = 0.063) and EL (1.7% improvement, p = 0.069). the two stimulus displays were the same or not. Data Analysis We trained monkeys to perform working memory tasks that required memory of multiple visual items. In a spatial Match/Nonmatch task (Figures 1A,B), four monkeys viewed a sample display with 1–5 white squares. After a delay period of 1 s, a second display appeared with the same number of stimuli, either at identical locations, or with one item appearing at a diﬀerent location. The subjects were required to judge whether Behavioral performance in the Match/Nonmatch task was determined in daily sessions, based on the proportion of correct responses for each stimulus display. Trials that were prematurely terminated, e.g. because of a break in ﬁxation, or due to release of the lever before the match/nonmatch stimulus was even displayed, were omitted from this analysis. Performance reﬂects only the proportion of correct and incorrect choices in We trained monkeys to perform working memory tasks that required memory of multiple visual items. In a spatial Match/Nonmatch task (Figures 1A,B), four monkeys viewed a sample display with 1–5 white squares. After a delay period of 1 s, a second display appeared with the same number of stimuli, either at identical locations, or with one item appearing at a diﬀerent location. The subjects were required to judge whether September 2017 \| Volume 11 \| Article 532 Frontiers in Neuroscience \| www.frontiersin.org 3 Working Memory across Hemiﬁelds Tang et al. FIGURE 2 \| Behavioral performance in the working memory capacity task as a function of number of stimuli. Proportion of correct trials is plotted for four different subjects (EL, NI, DA, and CA) in the spatial task, and one subject (EL) in the shape task. Error bars represent mean ± SEM across daily sessions. in the same hemiﬁeld (Figure 3A). For displays with 3 stimuli, a signiﬁcant diﬀerence was present for all four animals: subject CA (rank-sum test, p = 1.4 × 10−8, n = 55 sessions with 3+0 stimuli, 55 sessions with 2+1 stimuli), DA (rank-sum test, p = 0.045, n = 95 sessions with 3+0 stimuli, 96 sessions with 2+1 stimuli), NI (rank-sum test, p = 0.019, n = 72 sessions with 3+0 stimuli, 84 sessions with 2+1 stimuli), EL (rank-sum test, p = 0.041, n = 93 sessions with 3+0 stimuli, 115 sessions with 2+1 stimuli). Frontiers in Neuroscience \| www.frontiersin.org Same Hemiﬁeld vs. Different Hemiﬁelds Same Hemiﬁeld vs. Different Hemiﬁelds To determine if capacity was independent in the left and right hemiﬁelds, we analyzed performance for stimulus displays divided into diﬀerent groups depending on the number of stimuli appearing in diﬀerent hemiﬁelds. Performance generally declined as a function of stimulus number (Figure 2). We examined displays with at least 3 or more stimuli, for which suﬃcient numbers of errors were available from all monkeys. We then examined the spatial determinants that inﬂuenced performance across all possible 3-, 4-, and 5-stimulus conﬁgurations. Displays that contained exactly 3 stimuli were grouped into two groups termed “3+0” and “2+1” depending on whether all 3 stimuli appeared in one hemiﬁeld, or if 2 stimuli appeared in one hemiﬁeld, and 1 stimulus in the other. For 4-stimulus displays three groupings were possible: “4+0,” “3+1,” and “2+2.” For 5-stimulus displays also three groupings were possible: “5+0”, “4+1,” and “3+2.” As a control, we analyzed performance for stimulus displays in the upper and lower hemiﬁeld. Stimulus displays could be grouped in the exact same fashion for upper- and lower-ﬁeld stimuli, as well. Data Analysis We recorded performance in the task from four monkeys at diﬀerent stages of training, achieving diﬀerent overall levels of performance in the task (Figure 2). Generally, performance decreased as a function of number of stimuli that the monkeys had to remember. Across all conditions and numbers of stimuli, subject EL had an overall correct performance of 89% (n = 115 sessions, 24,317 total trials). Subject NI achieved an overall performance of 79% (n = 84 sessions, 19,835 trials). Subject DA achieved an overall performance of 66% correct (n = 96 sessions, 21,998 total trials). Subject CA’s overall performance was 56% (n = 55 sessions, 23,061 total trials). p p For displays with 5 stimuli a signiﬁcantly diﬀerent performance was also seen in two monkeys, with performance being lower when stimuli appeared in both hemiﬁelds than all in the same hemiﬁeld: subject DA (Kruskal-Wallis test, H = 8.25, df = 2, p = 0.016, n = 22 vs. 96 vs. 96 sessions) and subject NI (Kruskal-Wallis test, H = 8.65, df = 2, p = 0.013, n = 29 vs. 83 vs. 84 sessions). This improvement represented a 7.5 and 8.2% advantage in performance for stimuli in the same ﬁeld. A similar trend was observed in monkey CA with a 2.4% improvement for unilateral stimuli (p > 0.3). The only exception across all conditions was monkey EL, which exhibited a trend in the opposite direction, of a 2.4% beneﬁt in performance when 5 stimuli were split between two hemiﬁelds. However, the eﬀect size was minimal (Cohen’s d = 0.15 for one-side vs. split displays pooled together) and the diﬀerence did not reach statistical signiﬁcance (p > 0.8). In summary, the greatest performance advantages were observed when all stimuli appeared in the same hemiﬁeld, and this advantage was generally greater in subjects performing the task at lower levels. Upper vs. Lower Visual Field Displays were also more diﬃcult to remember when stimuli were split between the upper and lower hemiﬁeld, than when they appeared in only one of the two (Figure 3B). The eﬀect reached statistical signiﬁcance for subject DA in displays with 3 stimuli (rank-sum test, p = 0.006, n = 83 vs. 96 sessions), for subject NI in displays with 4 and 5 stimuli (4 stimuli: Kruskal-Wallis test, H = 14.87, df = 2, p = 6 × 10−4, n = 36 vs. 83 vs. 83 sessions; 5 stimuli: Kruskal-Wallis test, H = 16.82, df = 2, p = 2 × 10−4, n = 15 vs. 77 vs. 84 sessions) and for subject EL in displays with 3, 4, and 5 stimuli (3 stimuli: rank-sum test, p = 0.001, n = 91 vs. 115 sessions; 4 stimuli: Kruskal-Wallis test, H = 7.5, df = 2, Contrary to what we would expect if working memory capacity saturated independently in the two hemiﬁelds, when the stimulus groups were deﬁned based on arrangement of stimuli on the left and right ﬁeld, performance was poorer when stimuli appeared in both hemiﬁelds, than when they appeared September 2017 \| Volume 11 \| Article 532 4 Working Memory across Hemiﬁelds Tang et al. FIGURE 3 \| Behavioral performance in the spatial Match/Nonmatch task for different stimulus groups. (A) Proportion of trials that ended in a correct response is shown for different numbers of stimuli appearing in the left and right hemiﬁeld. Data from four subjects are shown in different colors: EL, NI, DA, and CA. (B) Stimulus groups were sorted between up and down hemiﬁelds. Error bars represent mean ± SEM. p < 0.05, p < 0.01, *p < 0.001, n.s., not signiﬁcant. FIGURE 3 \| Behavioral performance in the spatial Match/Nonmatch task for different stimulus groups. (A) Proportion of trials that ended in a correct response is shown for different numbers of stimuli appearing in the left and right hemiﬁeld. Data from four subjects are shown in different colors: EL, NI, DA, and CA. (B) Stimulus groups were sorted between up and down hemiﬁelds. Error bars represent mean ± SEM. p < 0.05, p < 0.01, *p < 0.001, n.s., not signiﬁcant. = 1, p > 0.05); subject NI (H = 1.501, df = 1, p > 0.05); subject DA (H = 2.424, df = 1, p > 0.1). Upper vs. Lower Visual Field The result conﬁrmed our expectation that working memory capacity is not strongly lateralized, so as to confer an advantage for stimuli appearing in one of the two hemiﬁelds. p = 0.024, n = 49 vs. 112 vs. 114 sessions; 5 stimuli: Kruskal- Wallis test, H = 17.9, df = 2, p = 1 × 10−4, n = 28 vs. 83 vs. 115 sessions). The only exceptions were subjects CA (Kruskal-Wallis test, H = 7.65, df = 2, p = 0.022, n = 10 vs. 55 vs. 55 sessions) and DA (Kruskal-Wallis test, H = 6.9, df = 2, p = 0.032, n = 46 vs. 96 vs. 96 sessions) in displays involving 4 stimuli, in which case, slightly lower performance was observed when all stimuli appeared in either the lower or upper ﬁeld. Stimulus Distance The similar results observed for left/right and upper/lower ﬁelds led us to suspect that the distance between stimuli may be the most important factor that determined the variability of behavioral performance across stimulus displays, rather than the speciﬁc arrangement of stimuli in these hemiﬁelds. To test the eﬀect of these variables we created a regression model that incorporated number of stimuli, average distance between stimuli and left/right side as independent variables: The results of this analysis indicate that dividing the hemiﬁeld in up and down halves generally produced performance advantages for stimuli appearing in the same hemiﬁeld, though this eﬀect was less consistent than dividing into left and right ﬁelds. This result conﬁrms that potential beneﬁts from independent processing of stimuli in two ﬁelds was rare, and in those occasions that an advantage was observed, it could not be attributed to independent left vs. right hemisphere processing, but rather appeared for stimuli distributed between the upper and lower ﬁeld (Figure 3B, middle). (1) P = β0 + β1NUM + β2DIST + β3SIDE + ε (1) Here P represents performance (percent correct responses in a session), NUM the total number of stimuli in the display, DIST the mean distance between stimuli calculated by averaging all possible angular distances between stimuli in the display, and SIDE a binary variable representing displays with stimuli either split between ﬁelds, or all appearing in the same (either left or right) ﬁeld. Frontiers in Neuroscience \| www.frontiersin.org Left vs. Right Visual Field We also examined whether there was an overall advantage for stimuli appearing speciﬁcally in the left or in the right hemisphere. The results reported above regarding unilateral displays were averaged from the left and right hemiﬁelds, and might have obscured an improved performance in one, concurrent with a diminished performance in the other ﬁeld. Data from three animals were suﬃcient to perform comparisons for displays with 3–5 stimuli, all appearing on the same side (Figures 5A–C). A Scheirer-Ray-Hare test with factors left/right side, and number of stimuli revealed no signiﬁcant main eﬀect for side in any of the subjects: subject EL (H = 2.804, df As expected, the coeﬃcient corresponding to stimulus number was negative (i.e., higher performance was observed for displays with fewer stimuli), and highly signiﬁcant for all monkeys (regression analysis, subject EL: p = 7.9 × 10−22, subject NI: p = 0.0197; subject DA: p = 0.0011, subject CA: p = 1.98 × 10−9). The eﬀects of distance and side did not prove to be uniform across monkeys, however. For subject EL, September 2017 \| Volume 11 \| Article 532 Frontiers in Neuroscience \| www.frontiersin.org 5 Working Memory across Hemiﬁelds Tang et al. we observed a signiﬁcant positive distance coeﬃcient (regression analysis, p = 0.015) suggesting that this monkey beneﬁtted from stimuli spread out. Once the distance variable was included in the model, however, no signiﬁcant eﬀect of whether the stimuli were concentrated or split between hemiﬁelds was present (regression analysis, p = 0.34). In other words, as long as the distance between stimuli was large, there was no further beneﬁt for stimuli appearing in diﬀerent hemiﬁelds. In contrast, monkeys NI and CA exhibited a signiﬁcantly negative distance coeﬃcient (regression analysis, p = 0.002 and p = 1.22 × 10−5, respectively) suggesting that for these animals, tighter grouping of stimuli oﬀered an advantage. A signiﬁcant beneﬁt of side was also present (p = 0.001 and p = 0.0057, respectively), in the direction of higher performance for stimuli in the same hemiﬁeld. Finally, subject DA exhibited a signiﬁcantly negative distance coeﬃcient (regression analysis, p = 0.02), suggesting that for this animal too, tighter grouping of stimuli oﬀered an advantage, though, once distance was accounted for, no signiﬁcant eﬀect of whether the stimuli were concentrated or split between hemiﬁelds was present (p = 0.54). were distributed between the two ﬁelds vs. present in the same ﬁeld. DISCUSSION Our study set out to test if working memory performance for displays of multiple stimuli is higher when stimuli are distributed across the left and right hemiﬁeld, as might be predicted if this ability were subserved relatively independently by the two cerebral hemispheres. We performed this behavioral analysis in advance of obtaining neurophysiological results from non-human primates, which will ultimately provide insights on the neural mechanisms behind working memory resources. Unexpectedly, our results indicated that this was not the case. In fact, performance was generally higher for displays that involved stimuli appearing in the same (left or right) hemiﬁeld, in the spatial version of the working memory task. In cases where a subject exhibited a preference for stimuli on separate sides of the visual ﬁeld, the advantage was not exclusive to the left- and right hemiﬁeld but it was present for lower and upper ﬁeld as well, and it could be accounted for by the relative distance between stimuli. We failed to observe an independent- hemiﬁeld beneﬁt in four diﬀerent monkeys, performing the task at diﬀerent levels of mastery, and appearing to rely on diﬀerent strategies. No qualitative diﬀerence was observed between left/right and up/down ﬁelds, either, which would have suggested independence of working memory capacity based on hemispheric specialization, or for a shape working memory task. There are a number of important limitations to our l i O l b i d i i i f i l We repeated the analysis of performance based on arrangement of stimuli across hemiﬁelds, using the same groups of displays as in the spatial task. There were no signiﬁcant diﬀerences for groups of stimulus displays, based on their appearance of either the left vs. right or the upper vs. lower hemiﬁeld (Figures 4A,B). To ensure that we had suﬃcient power to detect a potential diﬀerence between conditions in this experiment, we pooled performance from all possible bilateral displays (e.g., 1+3 and 2+2 arrangement in displays with 4 stimuli) and we performed a Scheirer-Ray-Hare test using the number of stimuli and the unilateral or bilateral arrangement of stimuli as factors (Figures 4C,D). When examining performance in left/right hemiﬁelds, we detected a signiﬁcant eﬀect of stimulus number (H = 9.097, df = 2, p = 1.7 × 10−4), but again no signiﬁcant eﬀect of unilateral/bilateral arrangement (H = 0.181, df = 1, p > 0.6). Shape Working Memory Task The results discussed so far were obtained from a task that only required memory for the spatial location of the stimuli. We considered that this eﬀect may lend itself to grouping of stimuli into patterns or other type of mental transformation of the stimulus display. Therefore, we trained one monkey to perform a working memory task requiring memory of both the location and shape of the stimuli (Figure 1C). The monkey now had to observe a display of diﬀerent stimuli, drawn from a set of eight white geometric shapes (Figure 1D). After a delay period, a second display was shown and the monkey had to determine if any of the shapes was diﬀerent or if the second display was identical to the ﬁrst. Subject EL achieved an overall performance of 85% correct trials (Figure 2) in this task (n = 34 sessions, 8,425 total trials). Left vs. Right Visual Field The average diﬀerence between these two conditions was 0.7% (81.2 and 81.9%, respectively). This advantage, however, was no greater than the diﬀerence in performance for stimuli appearing all in the upper or lower hemiﬁelds (79.7%) vs. distributed between upper and lower ﬁeld (81.0%), which also did not reach signiﬁcant diﬀerence (Scheirer-Ray-Hare test, H = 0.369, df = 1, p > 0.5 for eﬀect of unilateral/bilateral displays in Figure 4D). There was no signiﬁcant eﬀect of having all stimuli speciﬁcally in the left vs. the right side of the screen in this task, either (Scheirer-Ray-Hare test, H = 0.501, df = 1, p > 0.4 in Figure 5D). We also repeated the regression analysis described above for the analysis of performance in the shape working memory task. The results revealed similar preferences that monkey EL displayed for the spatial task as well. A slightly negative coeﬃcient was present for the distance variable, suggesting a small preference for spread out displays, however this eﬀect failed to reach signiﬁcance for the shape task (regression analysis, p = 0.085). Once the distance variable was included in the model, the coeﬃcient representing whether the stimuli were concentrated or split between hemiﬁelds was also not signiﬁcant (regression analysis, p = 0.1). In this case too, we conclude that working memory capacity does not improve when stimuli are distributed between the left and right side of the ﬁeld, as would be predicted if two independent capacities operated in the two cerebral hemispheres. These results reveal a considerable variability in strategies adopted by monkeys and stimulus arrangements that are easier to recall by each. In no instance however, independent hemiﬁeld processing accounted for an improvement in performance. DISCUSSION Still, we did observe an overall trend toward higher performance for displays where all stimuli g y There are a number of important limitations to our conclusions. Our results were obtained in variations of spatial- location and shape-working memory tasks, which represent a September 2017 \| Volume 11 \| Article 532 Frontiers in Neuroscience \| www.frontiersin.org 6 Working Memory across Hemiﬁelds Tang et al. FIGURE 4 \| Behavioral performance in the feature Match/Nonmatch task for different stimulus groups. (A) Proportion of correct trials are shown for stimulus groups sorted between left/right hemispheres. (B) Stimulus groups were sorted between up/down hemispheres. Error bars represent mean ± SEM, n.s. indicates not signiﬁcant. (C) Proportion of correct trials depicted in (A), now averaged across all bilateral and unilateral displays in the left and right ﬁeld, and plotted as a function of number of stimuli. (D) Proportion of correct trials depicted in (B), now averaged across all displays in the upper and lower ﬁeld, and plotted a function of number of stimuli. FIGURE 4 \| Behavioral performance in the feature Match/Nonmatch task for different stimulus groups. (A) Proportion of correct trials are shown for stimulus groups sorted between left/right hemispheres. (B) Stimulus groups were sorted between up/down hemispheres. Error bars represent mean ± SEM, n.s. indicates not signiﬁcant. (C) Proportion of correct trials depicted in (A), now averaged across all bilateral and unilateral displays in the left and right ﬁeld, and plotted as a function of number of stimuli. (D) Proportion of correct trials depicted in (B), now averaged across all displays in the upper and lower ﬁeld, and plotted a function of number of stimuli. Working memory is thought to be mediated by the persistent activity of neurons in the prefrontal cortex and other cortical areas, though alternative mechanisms have also been proposed in recent years (Stokes, 2015; Riley and Constantinidis, 2016). When multiple items are maintained in memory, then the activity of the subpopulation of neurons activated by each stimulus can be thought of as a “bump” (peak) in the network (Edin et al., 2009; Wimmer et al., 2014). Distributing the activity across separate populations of neurons, particularly in the two hemispheres, may in principle make the representation more robust, and less likely to be subject to interference, as is predicted by computational models (Compte et al., 2000). Lateralization Hemispheric specialization is well-known in humans for functions, such as language (localized predominantly in the left hemisphere) and visuo-spatial processing (in the right hemisphere). Less evidence of lateralization has been available in animal models, which do not possess language, though some traces of lateralization appear to be present in great apes (Corballis, 2012). Distributing stimuli between the two ﬁelds may therefore oﬀer an advantage, at least for demanding tasks, which might beneﬁt from the independent processing power of two hemispheres (Umemoto et al., 2010; Leblanc- Sirois and Braun, 2014). Indeed, imaging studies suggest that communication between hemispheres may increase with eﬀort, during performance of complex tasks (Davis and Cabeza, 2015). Other recent results, however, suggest that the organization of memory ﬁelds in the prefrontal cortex follows a quadrantic pattern of organization, with spatial working memory representations being biased by the vertical and horizontal meridians of the visual ﬁeld, and functional connectivity between neurons more rarely crossing quadrant boundaries, rather than left vs. right ﬁeld, speciﬁcally (Leavitt et al., 2017). Psychophysical studies in humans also conﬁrm quadrant-level interference eﬀects (Carlson et al., 2007). Our results are broadly consistent with these ﬁndings, as any behavioral eﬀects we observed between left and right ﬁelds tended to be present between up and down ﬁelds, as well. DISCUSSION Indeed, neurophysiological studies examining correlates of working memory for multiple stimuli reveal that activity representing multiple stimuli degrades faster, and information saturates with the presentation of more than one stimulus, when these appear in the same, left or right hemiﬁeld, than when stimuli are presented bilaterally (Buschman et al., 2011; Matsushima and Tanaka, 2014). A contralateral representation of stimuli is also favored early after stimulus appearance, whereas bilateral representations appear at later responses (Kadohisa et al., 2015). tiny fraction of tasks that can challenge the capacity of working memory. The stimulus displays we used are subject to grouping and Gestalt factors (discussed further below), which do not necessarily inﬂuence other types of visual working memory tasks. Additionally, we examined a small sample of non-human primates, recruited for the purposes of neurophysiological study, rather than behavioral analysis, per se. Nonetheless, our results provided clear examples of demanding working memory tasks that rely on visual working memory, and for which independent processing in the two hemispheres does not oﬀer an advantage. Neural Representation of Multiple Items in Working Memory In order to understand the factors behind any potential lateralization, it is instructive to consider in more detail the neural mechanisms maintaining memory of multiple stimuli. September 2017 \| Volume 11 \| Article 532 Frontiers in Neuroscience \| www.frontiersin.org 7 Working Memory across Hemiﬁelds Tang et al. FIGURE 5 \| Behavioral performance for stimuli appearing either in the left or the right hemisphere. (A–C) Proportion of correct trials from three monkeys (EL, NI, DA), in the spatial Match/Nonmatch task, for different numbers of stimuli, all displayed in either the left (green) or right (red) hemiﬁelds. (D) Proportion of correct trials for monkey EL in the shape task. et al., 2000). Spatial proximity (and other Gestalt principles, including connectedness, common region, and similarity) inﬂuence working memory performance (Peterson and Berryhill, 2013). Properties of nearby stimuli are better recalled from short- term memory (Xu, 2006). In that sense, stimuli appearing within the same hemiﬁeld and being subject to grouping principles may have an advantage over stimuli appearing at diﬀerent hemiﬁelds. Eﬃciencies may also be achieved by mentally transforming a multi-stimulus display, such as that used in our spatial working memory task into a polygon. In such an abstraction, higher performance may also be achieved when stimuli are nearby. In human studies, event-related potentials and BOLD fMRI activation elicited by visual displays amenable to grouping were diminished (i.e., require less resources for maintenance in memory) relative to displays of equal number of stimuli that cannot be grouped (Xu and Chun, 2007; Peterson et al., 2015). We should note however, that such an abstraction might be expected from animals that are able to master the task and strategically exploit stimulus grouping to perform it at a higher level. In our results, the greatest beneﬁts of stimulus proximity were observed for animals that performed at the lowest levels (Figure 3). We also saw no signiﬁcant advantage of separating stimuli across the left and right hemiﬁelds in the shape working memory task, which does not lend itself to an obvious transformation or grouping process of this kind. On the other hand, we saw no beneﬁt of grouping stimuli in this experiment, either. This ﬁnding may suggest that a potential advantage for same hemiﬁeld stimuli may only be present when no other task relevant stimulus features are available, although we should caution that our dataset was smaller in this experiment. FUNDING FIGURE 5 \| Behavioral performance for stimuli appearing either in the left or the right hemisphere. (A–C) Proportion of correct trials from three monkeys (EL, NI, DA), in the spatial Match/Nonmatch task, for different numbers of stimuli, all displayed in either the left (green) or right (red) hemiﬁelds. (D) Proportion of correct trials for monkey EL in the shape task. FIGURE 5 \| Behavioral performance for stimuli appearing either in the left or the right hemisphere. (A–C) Proportion of correct trials from three monkeys (EL, NI, DA), in the spatial Match/Nonmatch task, for different numbers of stimuli, all displayed in either the left (green) or right (red) hemiﬁelds. (D) Proportion of correct trials for monkey EL in the shape task. Research reported in this paper was supported by the National Eye Institute of the National Institutes of Health under award numbers R01 EY017077 and R01 EY016773 to CC; NIMH award F31 MH104012 to MR; and by the Tab Williams Family Endowment and Harry O’Parker Neurosciences Fund at the Wake Forest School of Medicine. Neural Representation of Multiple Items in Working Memory Taken together, our results argue that the proximity of stimuli maintained in memory provides advantages that counteract potential beneﬁts of independent hemispheric processing, at least in the context of some working memory tasks. FIGURE 5 \| Behavioral performance for stimuli appearing either in the left or the right hemisphere. (A–C) Proportion of correct trials from three monkeys (EL, NI, DA), in the spatial Match/Nonmatch task, for different numbers of stimuli, all displayed in either the left (green) or right (red) hemiﬁelds. (D) Proportion of correct trials for monkey EL in the shape task. AUTHOR CONTRIBUTIONS CC conceived and designed the research; HT and MR performed experiments; HT and CC analyzed data; HT and CC interpreted results of experiments; HT and CC drafted manuscript; HT, MR, and CC approved ﬁnal version of manuscript. REFERENCES Kadohisa, M., Kusunoki, M., Petrov, P., Sigala, N., Buckley, M. J., Gaﬀan, D., et al. (2015). Spatial and temporal distribution of visual information coding in lateral prefrontal cortex. Eur. J. Neurosci. 41, 89–96. doi: 10.1111/ejn.12754 Alvarez, G. A., and Cavanagh, P. (2005). Independent resources for attentional tracking in the left and right visual hemiﬁelds. Psychol. Sci. 16, 637–643. doi: 10.1111/j.1467-9280.2005.01587.x Lara, A. H., and Wallis, J. D. (2012). Capacity and precision in an animal model of visual short-term memory. J. Vis. 12:13. doi: 10.1167/12.3.13 Alvarez, G. A., Gill, J., and Cavanagh, P. (2012). Anatomical constraints on attention: hemiﬁeld independence is a signature of multifocal spatial selection. J. Vis. 12:9. doi: 10.1167/12.5.9 Leavitt, M. L., Pieper, F., Sachs, A. J., and Martinez-Trujillo, J. C. (2017). A quadrantic bias in prefrontal representation of visual-mnemonic space. Cereb Cortex. doi: 10.1093/cercor/bhx142. [Epub ahead of print]. Leblanc-Sirois, Y., and Braun, C. M. (2014). Intra and inter hemispheric dynamics revealed by reaction time in the Dimond paradigm: a quantitative review of the literature. Neuropsychologia 58, 1–13. doi: 10.1016/j.neuropsychologia.2014.03.012 Banich, M. T. (1998a). Integration of information between the cerebral hemispheres. Curr. Dir. Psychol. Sci. 7, 32–37. doi: 10.1111/1467-8721.ep11521844 Banich, M. T. (1998b). The missing link: the role of interhemispheric interaction in attentional processing. Brain Cogn. 36, 128–157. doi: 10.1006/brcg.19 97.0950 Matsushima, A., and Tanaka, M. (2014). Diﬀerent neuronal computations of spatial working memory for multiple locations within versus across visual hemiﬁelds. J. Neurosci. 34, 5621–5626. doi: 10.1523/JNEUROSCI.0295- 14.2014 Brainard, D. H. (1997). The psychophysics toolbox. Spat. Vis. 10, 433–46. doi: 10.1163/156856897X00357 Buschman, T. J., Siegel, M., Roy, J. E., and Miller, E. K. (2011). Neural substrates of cognitive capacity limitations. Proc. Natl. Acad. Sci. U.S.A. 108, 11252–11255. doi: 10.1073/pnas.1104666108 Meyer, T., and Constantinidis, C. (2005). A software solution for the control of visual behavioral experimentation. J. Neurosci. Methods 142, 27–34. doi: 10.1016/j.jneumeth.2004.07.009 Butcher, S. J., and Cavanagh, P. (2008). A unilateral ﬁeld advantage for detecting repeated elements. Percept. Psychophys. 70, 714–724. doi: 10.3758/PP.70. 4.714 Meyer, T., Qi, X. L., and Constantinidis, C. (2007). Persistent discharges in the prefrontal cortex of monkeys naive to working memory tasks. Cereb. Cortex 17(Suppl. 1), i70–i76. doi: 10.1093/cercor/bhm063 Carlson, T. A., Alvarez, G. A., and Cavanagh, P. (2007). Quadrantic deﬁcit reveals anatomical constraints on selection. Proc. Natl. Acad. Sci. U.S.A. 104, 13496–13500. doi: 10.1073/pnas.0702685104 Meyer, T., Qi, X. L., Stanford, T. R., and Constantinidis, C. (2011). REFERENCES Stimulus selectivity in dorsal and ventral prefrontal cortex after training in working memory tasks. J. Neurosci. 31, 6266–6276. doi: 10.1523/JNEUROSCI.6798-10.2011 Compte, A., Brunel, N., Goldman-Rakic, P. S., and Wang, X. J. (2000). Synaptic mechanisms and network dynamics underlying spatial working memory in a cortical network model. Cereb. Cortex 10, 910–923. doi: 10.1093/cercor/10.9.910 Mountcastle, V. B., Lynch, J. C., Georgopoulos, A., Sakata, H., and Acuna, C. (1975). Posterior parietal association cortex of the monkey: command functions for operations within extrapersonal space. J. Neurophysiol. 38, 871–908. Constantinidis, C., and Klingberg, T. (2016). The neuroscience of working memory capacity and training. Nat. Rev. Neurosci. 17, 438–449. doi: 10.1038/nrn. 2016.43 Peterson, D. J., and Berryhill, M. E. (2013). The Gestalt principle of similarity beneﬁts visual working memory. Psychon. Bull. Rev. 20, 1282–1289. doi: 10.3758/s13423-013-0460-x Corballis, M. C. (2012). Lateralization of the human brain. Prog. Brain Res. 195, 103–121. doi: 10.1016/B978-0-444-53860-4.00006-4 Peterson, D. J., Gozenman, F., Arciniega, H., and Berryhill, M. E. (2015). Contralateral delay activity tracks the inﬂuence of Gestalt grouping principles on active visual working memory representations. Atten. Percept. Psychophys. 77, 2270–2283. doi: 10.3758/s13414-015-0929-y Davis, S. W., and Cabeza, R. (2015). Cross-hemispheric collaboration and segregation associated with task diﬃculty as revealed by structural and functional connectivity. J. Neurosci. 35, 8191–8200. doi: 10.1523/JNEUROSCI.0464-15.2015 Pillow, J., and Rubin, N. (2002). Perceptual completion across the vertical meridian and the role of early visual cortex. Neuron 33, 805–813. doi: 10.1016/S0896-6273(02)00605-0 Delvenne, J. F. (2005). The capacity of visual short-term memory within and between hemiﬁelds. Cognition 96, B79–B88. doi: 10.1016/j.cognition.2004.12.007 Qi, X. L., Katsuki, F., Meyer, T., Rawley, J. B., Zhou, X., Douglas, K. L., et al. (2010). Comparison of neural activity related to working memory in primate dorsolateral prefrontal and posterior parietal cortex. Front. Syst. Neurosci. 4:12. doi: 10.3389/fnsys.2010.00012 Delvenne, J. F., Castronovo, J., Demeyere, N., and Humphreys, G. W. (2011). Bilateral ﬁeld advantage in visual enumeration. PLoS ONE 6:e17743. doi: 10.1371/journal.pone.0017743 Riley, M. R., and Constantinidis, C. (2016). Role of prefrontal persistent activity in working memory. Front. Syst. Neurosci. 9:181. doi: 10.3389/fnsys.2015.00181 j Desimone, R., and Gross, C. G. (1979). Visual areas in the temporal cortex of the macaque. Brain Res. 178, 363–380. doi: 10.1016/0006-8993(79)9 0699-1 Skeide, M. A., and Friederici, A. D. (2016). The ontogeny of the cortical language network. Nat. Rev. Neurosci. 17, 323–332. doi: 10.1038/nrn.2016.23 Edin, F., Klingberg, T., Johansson, P., McNab, F., Tegner, J., and Compte, A. (2009). ACKNOWLEDGMENTS There are other factors at work as well, that may counteract the beneﬁts of distributing stimuli between hemispheres. There is strong evidence that the spatial relations of stimuli in a display play an important role in visual short term memory (Jiang We wish to thank Xuelian Qi for her help with the design of the basic task, and Kathini Palaninathan for technical help. September 2017 \| Volume 11 \| Article 532 Frontiers in Neuroscience \| www.frontiersin.org 8 Working Memory across Hemiﬁelds Tang et al. REFERENCES Mechanism for top-down control of working memory capacity. Proc. Natl. Acad. Sci. U.S.A. 106, 6802–6807. doi: 10.1073/pnas.09018 94106 Stokes, M. G. (2015). ‘Activity-silent’ working memory in prefrontal cortex: a dynamic coding framework. Trends Cogn. Sci. 19, 394–405. doi: 10.1016/j.tics.2015.05.004 Stormer, V. S., Alvarez, G. A., and Cavanagh, P. (2014). Within- hemiﬁeld competition in early visual areas limits the ability to track multiple objects with attention. J. Neurosci. 34, 11526–11533. doi: 10.1523/JNEUROSCI.0980-14.2014 Essen, D. C., and Zeki, S. M. (1978). The topographic organization of rhesus monkey prestriate cortex. J. Physiol. 277, 193–226. doi: 10.1113/jphysiol.1978.sp012269 Umemoto, A., Drew, T., Ester, E. F., and Awh, E. (2010). A bilateral advantage for storage in visual working memory. Cognition 117, 69–79. doi: 10.1016/j.cognition.2010.07.001 Hayes, M. T., Swallow, K. M., and Jiang, Y. V. (2010). The unilateral ﬁeld advantage in repetition detection: eﬀects of perceptual grouping and task demands. Atten. Percept. Psychophys. 72, 583–590. doi: 10.3758/APP. 72.3.583 Warden, M. R., and Miller, E. K. (2007). The representation of multiple objects in prefrontal neuronal delay activity. Cereb. Cortex 17(Suppl. 1), i41–i50. doi: 10.1093/cercor/bhm070 Heyselaar, E., Johnston, K., and Pare, M. (2011). A change detection approach to study visual working memory of the macaque monkey. J. Vis. 11:11. doi: 10.1167/11.3.11 Weissman, D. H., Banich, M. T., and Puente, E. I. (2000). An unbalanced distribution of inputs across the hemispheres facilitates interhemispheric interaction. J. Int. Neuropsychol. Soc. 6, 313–321. doi: 10.1017/S1355617700633064 Hudson, C., Howe, P. D., and Little, D. R. (2012). Hemiﬁeld eﬀects in multiple identity tracking. PLoS ONE 7:e43796. doi: 10.1371/journal.pone.00 43796 Jiang, Y., Olson, I. R., and Chun, M. M. (2000). Organization of visual short-term memory. J. Exp. Psychol. Learn. Mem. Cogn. 26, 683–702. doi: 10.1037/0278-7393.26.3.683 Wimmer, K., Nykamp, D. Q., Constantinidis, C., and Compte, A. (2014). Bump attractor dynamics in prefrontal cortex explains behavioral precision in spatial working memory. Nat. Neurosci. 17, 431–439. doi: 10.1038/nn.3645 September 2017 \| Volume 11 \| Article 532 Frontiers in Neuroscience \| www.frontiersin.org 9 Tang et al. Working Memory across Hemiﬁelds 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. Copyright © 2017 Tang, Riley and Constantinidis. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). Xu, Y. (2006). Understanding the object beneﬁt in visual short-term memory: the roles of feature proximity and connectedness. Percept. Psychophys. 68, 815–828. doi: 10.3758/BF03193704 Wittig, J. H. Jr., Morgan, B., Masseau, E., and Richmond, B. J. (2016). Humans and monkeys use diﬀerent strategies to solve the same short- term memory tasks. Learn. Mem. 23, 644–647. doi: 10.1101/lm.0417 64.116 Xu, Y., and Chun, M. M. (2007). Visual grouping in human parietal cortex. Proc. Natl. Acad. Sci. U.S.A. 104, 18766–18771. doi: 10.1073/pnas.07056 18104 Frontiers in Neuroscience \| 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. Xu, Y., and Chun, M. M. (2007). Visual grouping in human parietal cortex. Proc. Natl. Acad. Sci. U.S.A. 104, 18766–18771. doi: 10.1073/pnas.07056 18104 September 2017 \| Volume 11 \| Article 532 Frontiers in Neuroscience \| www.frontiersin.org 10
https://openalex.org/W2747874418	https://digilib.phil.muni.cz/bitstream/handle/11222.digilib/136923/1_MusicologicaBrunensia_52-2017-1_3.pdf?sequence=1	English	null	Finding a way to tie technology, aesthetics and dramaturgy together in terms of experimental sound-based music	Musicologica Brunensia	2,017	cc-by-sa	7,137	Landy, Leigh Landy, Leigh Finding a way to tie technology, aesthetics and dramaturgy together in terms of experimental sound-based music Musicologica Brunensia. 2017, vol. 52, iss. 1, pp. 5-16 Musicologica Brunensia. 2017, vol. 52, iss. 1, pp. 5-16 Terms of use: Digital Library of the Faculty of Arts, Masaryk University provides access to digitized documents strictly for personal use, unless otherwise specified. Terms of use: Digital Library of the Faculty of Arts, Masaryk University provides access to digitized documents strictly for personal use, unless otherwise specified. Digital Library of the Faculty of Arts, Masaryk University digilib.phil.muni.cz Musicologica Brunensia 52 / 2017 / 1 DOI: 10.5817/MB2017–1-1 1 The talk was dedicated to my colleague and fellow keynote speaker, John Richards, as he celebrated his 50th birthday during the conference and therefore was with us, the conference attendees, not his family, on this special day. Abstract Not many years ago, I gave a keynote at the SPEEC Conference at Oxford University entitled, ‘music Technology, Music technology or Music Technology?’ (Contemporary Music Review. 32/5: 459–471, 2013) and have continued to investigate the subjects that arose in that talk both as a scholar and as a composer, issues that align strongly with many themes included in this conference’s Call for Papers. Suffice to say that some tension was discovered between the two words in that discussion. This conference’s food for thought keynote talk (modified in the form of the current article) fo- cuses on questions including: Where do we stand in terms of art for art’s sake in today’s world? How has this influenced our understanding of what aesthetics currently signifies? Who are our communities of listeners? And, with this in mind, what roles do or should communication and dramaturgy play in terms of music making? In consequence are the tensions between the words music and technology in any way getting resolved? As fellow keynote speaker, John Ri- chards1 and I are currently involved with writing a book dealing with many of these very issues, one of the book’s themes, sampling culture, will be used as this article’s case study. Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms of Experimental Sound-based Music Leigh Landy / llandy@dmu.ac.uk Music, Technology and Innovation Research Centre, De Montfort University, Leicester, UK Leigh Landy / llandy@dmu.ac.uk Music, Technology and Innovation Research Centre, De Montfort University, Leicester, UK Music, Technology and Innovation Research Centre, De Montfort University, Leicester, UK Keywords music technology, sound-based music, music aesthetics, music as a craft, music as a voca- tion, music as an art, sampling, dramaturgy in music 1 The talk was dedicated to my colleague and fellow keynote speaker, John Richards, as he celebrated his 50th birthday during the conference and therefore was with us, the conference attendees, not his family, on this special day. 5 5 Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... To begin As this text is based on an invited keynote talk, the following words will form a brief launch to the current article. I believe that it is quite important when offering a keynote talk not to just talk about one’s own work but instead to attempt to take some of my ideas and thoughts and combine them with the challenge that has been provided by the host. In fact Martin Flašar’s conference theme was perhaps easier to write than to respond to given the multiple forms of provocation implied in the questions posed. These provocative thoughts therefore present me with a good challenge. Thus, the fol- lowing text is in fact my reaction to many of the issues raised in the conference call com- bined with a personal statement, an attempt to offer those present and future readers food for thought and, perhaps logically, offer a few personal provocative views as part of this reaction. With this in mind the current text is more a statement seeking reaction than a report regarding a specific scholarly research project. The introduction involves walking through the abstract a little bit as it sets the theme clearly. Not many years ago I presented another keynote at Oxford University that was called ‘music Technology, Music technology or Music Technology?’ Reading it one notes its nuance, as the first one has a small m and a capital T, the second one has a capital M and a small t and the third contains two capitals. The one with two capitals is, in my view, ideal in my field of making music with any sounds given its interdisciplinary nature and tendency towards innovative approaches to both art and technology. However, I actually belong to category number two as I invite technology specialists to offer a helping hand when I need technology developed towards musical goals that I am unable to develop myself. Keywords Given my own area of expertise and what has been sketched above, the next step is to comment briefly on some of the points raised in the call and then add a few questions of my own that should aid in the article’s argument. Keywords More importantly, I want to state clearly that I definitely do not belong to the first one, which brings us right to the heart of my subject and equally to the challenge posed in the conference call. Why am I proud not to belong to the first one? Because for people who make music that is driven by technology, I would suggest that the per- centage of likelihood that a good solid piece of music is going to come out of such an approach is about 1%. Without a founding musical motivation that supports creativity using technology, it is just chance that leads towards a successful work of music. As Martin Flašar’s call offered some very interesting, yet challenging thoughts about how technology has almost taken over music, my preliminary contribution is to suggest that technology is ideally a tool supporting musical creativity, not a means leading towards creative goals. Yet one must be careful about talking about ‘good’ or ‘successful’ music, as these terms are relative, open to personal, community or cultural interpretation. We shall re- turn to the word, aesthetics fairly soon. The point here is not so much whether technol- ogy has taken over music as such. It is instead whether technology is there to serve musi- cal goals, which it does most of the time. And, of course, there is the often-encountered phenomenon of technology not working properly … There is also a third axis here not mentioned in that Oxford title, namely humani- ties – musicology in the current case – the field which best binds art with technology in terms of various forms of understanding, the field of this conference’s community of interest. In the area of sound-based musical creativity, a large percentage of the sub- ject’s musicologists consist of artistic practitioners like myself. These are people who are 6 Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... Leigh Landy g y Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... able to have both sides of their work inform each other (or all three if they are involved with technological development) using both sides of their brain as it were. They rein- force each other as the intellectual side is helping them to understand their music and their music is helping them to support or question ideas that they are investigating. 2 Sound-based music was defined as music in which the sound, that is material not normally associated with musical notes, is the unit measure of its content. Selected subjects from the conference call • Music as a craft, music as a vocation, music as art This first point immediately brings us very close to the small and capital M and T discussion. A friend once told me that when the arts separated from their original combined state in Western society, the artist (musician-dancer-actor, etc.) became an artisan, that is, a specialist of a given craft. For the purposes of our discussion, I would like to put it slightly differently. My wife, Evelyn Jamieson, is a choreographer who often makes the distinction between someone pos- sessing technique and possessing the ability to perform. Craft is perhaps more aligned with technique and creative imagination and communication more aligned with per- formance. Possessing technique or craft is undoubtedly important, but in my view, the ability to perform is even more vital. Optimising the two is ideal. Putting it differently, it is about determining one’s small and capital c’s and p’s in craft and performance. As far as vocation is concerned, in my field, there is virtually no one who can earn his or her bread from artistic endeavour alone. Therefore, optimisation is again relevant, such as combining artistic work with the requirements of a position at a university or conservatoire. • Aesthetic issues of ‘high’ versus ‘low’ music As the key discussion regarding aesthetics will be presented in a few paragraphs, I would like to make an important remark at this point regarding ‘high’ and ‘low’ music. In Landy 2007, I proposed that what I defined as sound-based music possessed its own paradigm, that is body of knowledge related to creative endeavour, reception and understanding.2 Within this paradigm, it was sug- gested that a good deal of this sound-based repertoire may have roots in forms of art or popular music, but the musician often either ignores or transcends this boundary. This work exists in its own field and is neither dictated by commercial means of pro- duction nor by the reasonably declining vector of social relevance that many forms of today’s new art music is facing. • Questions of rationality and irrationality in musical structure and Technologically gener ated / supported / mediated music As these points are not that pertinent to this discus- sion, suffice to say that the use of formal approaches to composition, another form 7 Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... 3 This is discussed more fully in John Richards’ article in this volume. Note: noise music does not need to be loud. So-called ‘lower case’ music is quiet, often subtle and open to a range of tastes. Selected subjects from the conference call of technological endeavour be it not necessarily involving electricity, works best when founded on a strong musical foundation. This is analogous to the remark above when discussing music Technology. I have often joked (only half seriously): if you can’t hear the algorithm, then why use it? There is much more to be said on this subject, but that is best saved for another publication. • Virtuosity as a product of industrial society This element of the call is a wonderful provo- cation. However, I would like to ‘abuse’ its presence by noting that technology is offer- ing a number of new forms of virtuosity today, such as people gaining a high level of mastery of a particular type of studio, of software, of a controller or other new device. Furthermore, virtuosity is being redefined in another way in our time. A six-year-old can become quite virtuosic with the software, Garage Band, due to its rather intuitive ap- proach to music making, just to name one example. This final point leads to one issue of my own that deserves to be raised This final point leads to one issue of my own that deserves to be raised I believe 8 Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... Leigh Landy Leigh Landy Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... that the answer is a definite ‘no’ and have been involved with research that provided rich statistics to back up this opinion. In relation to the above, the following questions will also serve this presentation’s argument • Where do we stand in terms of art for art’s sake [umění pro umění] in today’s world? I believe that the two-century old notion of art for art’s sake has had an enormous impact on our consideration and definition of aesthetics and I would like to share my view that this impact has been largely negative. We therefore must return to this. • How do these subjects influence our understanding of what aesthetics currently signifies? This question follows directly from the previous one. • How do these subjects influence our understanding of what aesthetics currently signifies? This question follows directly from the previous one. This final point leads to one issue of my own that deserves to be raised • Issues related to access and inclusivity (on participation and communities) What do I mean by this? In contemporary music, much more than in Janáček’s and Bach’s time, we have a small public in general. This is no secret to anybody. I have spent over 25 years writing about why this is and whether things have to be this way. I believe that when we are talk- ing about these issues about craft and everything else, if we do not talk about the lack of and, leading from that, the creation of a public, that is, the creation of interest in ap- preciating and making the music which is very dear to us through establishing a dynamic loop of communication, then we are not doing our work completely. This concentration of the socio-cultural side of our field is a crucial theme, yet one that is hardly investi- gated. Access is about reaching people. Inclusivity is about making our music open to broader groups of people. Participation is an obvious next step. And community is the heart of the matter, that is, a group of people who share an interest. I would therefore like to offer the following view at this point that many musicians do not think about their potential public, their community, when they make a piece. They make a piece for ‘anyone’ and hopefully someone will come and listen. I am suggesting here that this is not good enough. In other words, we need to think these things over and include the following questions in our music: why make music in a particular manner and for whom is it intended? Let me put it another way: where does such music fit in today’s society? Clearly noise music has, in general, a different audience than does acousmatic music. The latter has roots in high art whilst noise might be said to have roots in popular cul- ture, at least as far as dynamic levels and its venues are concerned. But this is a simplifi- cation. Noise music does have a reasonably dynamic community as does today’s hacking culture.3 Acousmatic music has been presented in a variety of contexts, but when pre- sented as art music, it indeed suffers from a similar lack of community beyond special- ists, as does contemporary instrumental music. Does this need to be the case? • Who are our communities of listeners? In other words, for whom is our music? Dramaturgy plays a role in terms of the 9 Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... issue of ‘for whom’ as it allows the musician to be able to share intention concepts and receive remarks related to a listener’s reception in order to see whether musical com- munication is actually taking place. It is a means of relating to an audience and learning about developing one’s own community, one’s own public. issue of ‘for whom’ as it allows the musician to be able to share intention concepts and receive remarks related to a listener’s reception in order to see whether musical com- munication is actually taking place. It is a means of relating to an audience and learning about developing one’s own community, one’s own public. • In consequence, are the tensions between the words music and technology in any way getting resolved? As this talk was presented at a musicology conference I am not going to spend a good deal of time on this question; however, if I were speaking to composers or people studying music technology I could speak for weeks addressing this last question. These people all want to use technology in their music; they want to develop technology for music but the question is: what is the musical advantage of all of this? In other words, the why and for whom are equally relevant here. Expressing the musical advantage or being able to envision the musical advantage of all of this is not always easy but it is pos- sible. That’s my profound belief. When technology is developed for its own sake without a clear view of (specific) musical application, we might use the phrase, technology for technology’s sake. With this in mind, the expression of the musical advantage regarding the use of a particular technology forms part of dramaturgy and the presence of drama- turgy infers that the small m and capital T represent an undesirable combination. Investigating the sonic musician in the 21st century, in particular sampling culture Fel- low keynote speaker, John Richards and I are currently writing a book together, a vision- ary one, all things going well. • Who are our communities of listeners? In other words, for whom is our music? • And, with this in mind, what roles do or should communication and dramaturgy play in terms of music making? We now arrive at the heart of my contribution, be it at the end of a lengthy introduction. These questions are based on an assumption, namely that art, and thus aesthetics, has something to do with communication. Many of the composers who taught me absolutely disagree with that idea. For them, music is a much more intel- lectual and formal discourse. Nihilistically put, your musical body consists solely of your head. The words, embodiment, emotion and the heart were completely absent in the teaching of composition that I received. I disagreed then and I disagree now. Perhaps I am very conservative in terms of these thoughts. In fact I reject the notion that aesthet- ics is solely to do with intellectual satisfaction or emotional satisfaction. I believe it is to do with the optimisation of both. The second term raised here is dramaturgy. When I was taught composition, and to a large extent when I was taught musicology, I learned about what type of artefact a com- poser (or I) was making and how the piece was made. I learned about the musical gram- mar of a given work. When studying performance there were equivalents. During those years of study, no one asked me the next two questions that follow from the how and the what of a work. As mentioned above, the third question is why? Why make this piece? This has to do with dramaturgy. It has to do with communication. My contemporaries did not think like this. They made abstract works – nothing wrong there – and were of the view that people could see and hear their abstract compositions and receive them in any way they found fitting. This is a little too simplistic for me actually. I would suggest that there exist particular types of reactions to abstract works. They are not identical to all people but there are these prototypes and there could be dramaturgical concepts supporting communication in such situations. I am not speaking of communication in terms of narrative; there are other means of navigating a work. The second question is for whom – for whom are we making our music? By re-emphasising these questions, I am attempting to underline their importance. • Who are our communities of listeners? In other words, for whom is our music? I sup- pose that if I asked the reader to come up with an example of early sampling you might think of Pierre Schaeffer’s musique concrète in which recordings were made of sounds from the real world and then using them as musical material. If we broaden this history a bit, we might go back to Dada which involved both Marcel Duchamp’s found objects, objets trouvés, and the technique of collage employing objects from daily life. We could go back even further, prior to recording technology, and find unexpected objects from the real world included as a musical instrument; however, this article is not intended to survey sampling history. Most of us think of sampling these days with respect to genres such as hip hop music. This is, of course, one of several genres in which samples are employed ranging from the most popular to the most experimental. Although aspects of sampling culture that are also related to hip hop are pertinent, we shall not focus on this genre and similar ones as they are primarily pitch- and note-based music, using sounds that are derived from notes that are intended to be heard as such. Scratching (a percus- sive sound) and more experimental approaches do take us into the world of timbre and non-note sounds. Our interest is more sound- or timbre-based forms of musical creativ- ity, an area that we believe is increasing in importance within today’s musical horizon. Neither of the book’s two approaches, at least in terms of more experimental varie- ties, lend themselves easily to either high art and low art. Of course there exist works in both areas that possess roots in popular music and most likely involve a regular pulse. However the sonic result will not necessarily suggest that the musicians are popular mu- sicians or were not art music composers. Similarly, there are pieces that have been made by people like myself trained in art/high/serious music, yet these works do not really fit comfortable in either category. Of course some do; the point here is: many do not. This ability to exist outside of the popular/art divide or, eventually, in an entirely different category of music making, allows us to look at classification in a different way. This is of major importance here because perhaps it suggests an exit from the notion that new music is only for a few people. • Who are our communities of listeners? In other words, for whom is our music? We come from two completely different aesthetic worlds but we believe that both are exemplary of many dynamic musical developments in this young century. In his case he specialises in what we call hacking or do-it-yourself culture, He calls his domain that of ‘dirty electronics’ and that is his focus of study in the book in progress. In my case, the focus is that of sampling culture, something reflecting in a diverse number of musical approaches in my own works for quite some time. Sampling is having a significant impact on our society in a many interesting ways, not solely re- stricted to music. The same can be said of hacking. With these two foci in mind, our book, provisionally titled The Music of Sounds and the Music of Things, investigates some important changes in the world of experimental arts practice. One of our discoveries is that there is less importance given in new musical pro- duction in recent years to finding new materials, developing new structures, new musical languages or even investigating new approaches to physical space all of which featured in the previous century, particularly in the last forty to fifty years of that century. Instead, our view is that the means of production – in sampling co-production, in DIY culture vari- ous means of making – and dissemination are not only becoming more prominent but also changing enormously rapidly in today’s experimental music world, which implies the forming of new communities of interest (which are related to new forms of dissemination) and new approaches to dramaturgy. I am aware that musicologists are a little hesitant to look forward because the field’s history is based on just that, history, investigating music that already has a trace. However, I believe that these two subjects lend themselves to both 10 Leigh Landy Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... traditional musicological treatment and investigation of their potential impact on music making, thus a more visionary approach to musicological discourse. On sampling As stated, my primary role in this book is to investigate sampling. We have known sampling of one form or another in music for quite a number of decades. • Who are our communities of listeners? In other words, for whom is our music? Furthermore, in the areas of sampling and hack- ing, but in particular the music of sampling, the material used is familiar to listeners. Therefore, a connection with the material that already exists offers listeners something to hold on to when listening to a work. Examples that are worthy of mention include the works of John Oswald, the Cana- dian artist who started the Plunderphonics movement, and his followers. Plunderphon- ics involves taking very short clips from known music and re-creating it. It is a halfway house between note-based and sound-based approaches that I have called music-based experimental music. It has a common history with hip hop, but it also involves more experimentation and, in Oswald’s case, humour. A second and very different type of ex- ample is soundscape composition. Here longer recordings are used, from nature, from cities, from anything related to real life and recomposed or, in extreme cases, left as is. 11 Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... Soundscape has virtually nothing in common with hip hop. A third area of potential relevance to sampling is that of sound art including the wide diversity of sound installa- tions. Many of these works use samples. Many of these works are also dependent on the context in which they are presented, thus offering another link to community (audience) and to dramaturgy. As sound art works normally do not have a beginning or an end, people can drop in and out as they wish. Installations in public spaces attract a public much larger than the normal contemporary music audience. In short there is an entire musical world in constant development using samples. The number of people who have had the experience of using samples is absolutely huge, not only in music, but also in writing, in video, in coding, etc. In the world of sampling, al- though this is not always the case, the notion of co-production, of remix is of relevance. The ‘work’ is but a form of a work and can be recomposed by the maker(s) and/or by others. How do such works sit with art to art’s sake? In another keynote talk, given in 2011, I came up with an antidote, or opposite to art for art’s sake, which I called art for life’s sake or, alternatively, art for goodness’ sake. 4 One of the questions raised after the talk concerned the listener’s understanding of musical grammars. In my response, I suggested that people who went to performances by Mozart or Beethoven at the time were aware of the grammars prior to their coming and thus could both identify the grammar (shared experience) and the novel approaches within a given piece. This can hardly be said of much contemporary music where structurally many pieces are reasonably unique and the shared experience can be found in the reception of virtuosity and abstraction or perhaps complexity. In sampling culture, the material and, ideally, the treatment of that material, offer experiential links that may not be as strong as that of traditional musical grammars, but at least allow the listener to have something to hold on to in terms of navigating a given work’s development in time. • Who are our communities of listeners? In other words, for whom is our music? The idea here is: if we take musical elements from daily life, which you can easily do with sonic material, and organise it imaginatively within an artistic context, you then already have something in common with your listener. If you do something artistic with that material that the listener can actually identify and attempt to follow in time, they indeed have something to hold on to in terms of navigating that work. Therefore, one might conclude at this point that when one works with sounds one is potentially better able to address communication, dramaturgy and other artistic as- pects related to communication than perhaps is the case regarding most more abstract instrumental works. In terms of dramaturgy, communication and aesthetics, it is my view that links with lived experience, e.g., the re-contextualisation of sounds and the art of re-composition, can lead not only to a new, experimental artwork, it can also lead towards a shared experience.4 In contrast, traditionally, someone wrote a work and then somebody else stole it, or somebody wrote variations on a piece by someone else. In the world of sampling, something exists, somebody remixes it, somebody re-composes it, etc. The sampled work is a living thing, it is a flexible thing and, following sampling culture, often nobody owns a given piece. In terms of participation, therefore, we have not only composers, we have transformers and we work with co-composers also called sequential composers. Sampling culture opens up creativity allowing people to join together and play with the same sounds and with the same sequences. This is a pronounced anti-high- art attitude. In high art: this is my piece, I have the rights, I have the publisher, I have the Supraphon recording and that’s it. You can buy the work or you can hear it live or dur- 12 Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... Leigh Landy Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... ing a broadcast or, more recently, online if the rights issues have been dealt with. This is quite different, the opposite in fact, from sampling culture. In terms of communities, inclusivity and the ‘for whom factor’ tend to be largely built into sampling culture. I find that both interesting and noteworthy. • Who are our communities of listeners? In other words, for whom is our music? And it goes without saying that there is increased accessibility because a piece possibly has many more people involved with it from its early stages. These musicians do not necessarily call themselves innovative composers in the high art sense; they just enjoy the act of organising sampled sounds. With this in mind, perhaps the notion of composition is becoming redefined, at least for some people in our areas of focus.5 Music is not isolated from the other arts as sampling can be applied to just about eve- rything. And certainly sampling in video culture is enormous. There are many stories of videos that have gone viral and have been changed, that is, remixed thousands of times. (See, for example, the history of ‘TechnoViking’.) Sampling (remix, mash-up) culture is offering opportunities to many people from all sorts of backgrounds. It is offering new musical challenges and influencing current concepts of musical practice. Before moving on, there are a couple of aspects related to sampling culture that de- serve attention. First of all, there is the question of so-called appropriation. One takes a sample from somewhere. How does that person treat the sample? You can treat a sam- ple of a recording of, say, a politician you do not like in a very negative way, For example, it would be very easy for me to ‘grab’ a very short recording of Donald Trump and put it into a musical context and make everybody laugh. In this case I am not treating him with respect. I do not want to treat him with respect; I am being clear about this. This could be seen as Brechtian in the sense of political art and satire. However appropriation can become more complex. There are cases in sampling where people hear something exotic such as music from China and use it because it is exotic. However, they are not necessar- ily treating the sounds with respect, as these sounds perhaps have a particular meaning in China that are not being taken into consideration. What is the musician’s attitude in terms of appropriation? This is, I believe, a subject for consideration within both musi- cal and musicological studies if we are going to engage with this field. The second, crucial aspect that is worthy of attention is to do with legality or illegal- ity and is much more difficult. 5 A second question posed after the talk focused on this notion of redefinition. The reply was indirect as composition is being redefined in a particular way within hacking culture, an area where the expression ‘in- strument as composition’ is often used (that is, that making of a particular instrument serves the subsequent music-making process). More pertinent to the current discussion is the fact that the roles of composer and performer have become fuzzy within many forms of experimental music over the years. In the case of sampling culture, co-composition and ownership represent key subjects related to the dynamic development of the tra- ditional métier of composition. 6 I have been informed that the Czech Radio holds all copyright and can provide samples from their broadcasts for artists to reuse. This is quite exceptional. I am therefore pleased to be making my first legal 6 I have been informed that the Czech Radio holds all copyright and can provide samples from their broadcasts for artists to reuse. This is quite exceptional. I am therefore pleased to be making my first legal • Who are our communities of listeners? In other words, for whom is our music? In fact legality issues associated with sampling, at least currently, might be called its Achilles heel. There are two aspects regarding this subject to be raised. I have composed a series of pieces in which all of the material comes from one nation’s radio stations. In many countries this means that the vast majority of the samples that I use are illegally being re-composed in my work.6 This being the 13 g y Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... case, why do I go ahead and compose such works? Beyond the satisfaction of seeing a nation’s public react to its media being tilted ever so slightly and being presented as an art work, I am taking a small risk to let the world know that current sampling leg- islation is absurd. To be honest, I believe this largely holds true in hip hop culture as well. You cannot use the sound of a radio station’s logo. You cannot use the sound of the advertisements and certainly not much of the music. In the UK the sound of Big Ben is copyright protected. This means it is theoretically impossible to really do any- thing involving sampling with radio broadcasts. I go ahead and do it because I believe that it is a question of time before the law changes as the law protecting most of these sounds is completely ridiculous. One of the consequences, bringing me to my second point regarding legality, is the following. A sampling artist within popular music may become quite well known, because the whole thing about popular music is about being popular, being famous and maybe even being rich. In our world of experimental music it is most unusual that a musician wants to become a celebrity, famous or rich. I would even go so far to say that in both sampling and in DIY cultures, we are not looking for more Boulez’s and Stockhausen’s. We are looking for interesting, exciting musicians. Whether their music outlives them is not that relevant. One French video artist and former collaborator, Michel Jaffrennou, once said to me, l’immortalité est morte, thus immortality is dead. I cannot say whether this is now true or not in both high art or in the new cultures being presented here. radio piece in the coming year entitled Mezihlas – Přeshlas – Nahlas or Radio – Voice – Overs in English using recordings from a number of Czech state broadcasters. • Who are our communities of listeners? In other words, for whom is our music? Regardless, I believe that it is a question we need to think about, especially with so much art being produced these days with so few opportunities to share it, the Internet being one clear exception. The question is instead: what happens with the sampled works and who are the major figures? I cannot provide you with a lengthy list of names of really interesting sampling artists because many work ‘underground’ due to the fact that much of their work is illegal. Thus we find ourselves in an awkward situation here – very interesting, perhaps unique in musical history. Again, I believe that these subjects are worthy of further musicological attention in the future. If musicologists can spend months researching how composer X stole from composer Y, why can they not investi- gate sample-based works themselves, the thorny issues of sampling and the position of sampled works in today’s culture? [At this point in the keynote, the author presented excerpts from three of his works to demonstrate the breadth of experimentation with sampling. The most traditional in- volved samples taken from recordings made on the Chinese ocarina, the xūn, played by a master musician. The recordings were re-composed into a sample-based composition, a form of music-based music in the sense of Oswald. The second work utilised texts by Gertrude Stein that were recomposed for live and recorded speaking voices (the same voice as the performer). The third piece was one of the above-mentioned radio works.] Without getting into analysis, two things separate these works from most contemporary art music. 1) If I make a mistake in performing the Stein work even the untrained lis- tener will know. I have often heard performances of Xenakis and Boulez works and been 14 Leigh Landy Leigh Landy Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... told by the performers that they had made hundreds of mistakes and yet they received enthusiastic applause in the end, as they apparently had been able to communicate the gist of the work. 2) I do not believe that these works need to be destined for contempo- rary music’s micro-public. Given their connection with lived experience, I believe that many people can enjoy any of these experimental works. • Who are our communities of listeners? In other words, for whom is our music? In fact I have discovered that, without grasping all of the language in the radio piece or the Stein performance or even without the background of Chinese performance traditions regarding the xūn, many people can discover things to hold on to and discover a means of enjoying such works. It then has to do with the level of shared experience (communities), communication and comprehension alongside the level of emotional enjoyment. I therefore again sug- gest that musicology should open the door to investigating the intention-reception loop of such experimental outings and discover why this music is, in the end, much more accessible than one might imagine, that is, accessible to people in general and to new communities that evolve from their common interest regarding the sampled material and means of production. To conclude In tackling the challenges presented in the conference call regarding the roles of technology in today’s musical cultures and combining these with the areas relat- ed to aesthetics found in the title and the sampling culture case study, I have attempted to demonstrate that: • Innovative music including sound-based music is a subject of study that is of potential importance to musicians and musicologists alike. It is also open to a larger public, group of communities, than most might imagine. As new means of dissemination evolve, these communities will evolve, intersect and evolve further. • Innovative music including sound-based music is a subject of study that is of potential importance to musicians and musicologists alike. It is also open to a larger public, group of communities, than most might imagine. As new means of dissemination evolve, these communities will evolve, intersect and evolve further. • After a period when I was a student and a young composer when everything had to be different than anything that came before it, a period in which neo-atonal music avoided consonance, most new works contained an inaudible structure and John Cage introduced the music of no rules, thus a period of antithesis, a period of synthesis inevi- tably was to follow. Sampling culture, due to its interest in recycling sounds, is a form of music that allows for both innovative and experiential connections, taking music beyond antithesis. • Who are our communities of listeners? In other words, for whom is our music? • After a period when I was a student and a young composer when everything had to be different than anything that came before it, a period in which neo-atonal music avoided consonance, most new works contained an inaudible structure and John Cage introduced the music of no rules, thus a period of antithesis, a period of synthesis inevi- tably was to follow. Sampling culture, due to its interest in recycling sounds, is a form of music that allows for both innovative and experiential connections, taking music beyond antithesis. • With this in mind, aesthetics becomes a notion of optimising intellectual understand- ing with forms of visceral communication. In general both are based upon relating the musical act of listening to previous experience, also known as knowledge. Communica- tion is therefore integrated into the aesthetic experience. Communities of interest will possess a shared knowledge allowing for this communication to take place. All of this serves experimental music that, often in the past, has avoided this type of discourse. • In terms of the conference theme, the key contribution here is for technology to serve musical goals. The proposal is, in the end, that those involved with technology work 15 g y Finding a Way to Tie Technology, Aesthetics and Dramaturgy Together in Terms ... closely with musicians (often the same person) and, ideally, with musicologists to holisti- cally demonstrate the musical relevance of a technological or a music-technological de- velopment. Aesthetics and dramaturgy can be of invaluable help in terms of supporting this point. This three-way form of communication – between musicians, technologists and musicologists – is valuable for the betterment of technology serving music and for (innovative forms of) music itself. LANDY, Leigh. Understanding the Art of Sound Organization. Cambridge, Mass.: MIT Press, 2007. Bibliography 16 16
https://openalex.org/W2763122548	https://europepmc.org/articles/pmc5634792?pdf=render	English	null	Extensive Core Microbiome in Drone-Captured Whale Blow Supports a Framework for Health Monitoring	MSystems	2,017	cc-by	10,711	RESEARCH ARTICLE Host-Microbe Biology crossm Editor J. Gregory Caporaso, Northern Arizona University Copyright © 2017 Apprill et al. This is an open- access article distributed under the terms of the Creative Commons Attribution 4.0 International license. Address correspondence to Amy Apprill, aapprill@whoi.edu. A.A. and C.A.M. contributed equally to this work. IMPORTANCE The conservation and management of large whales rely in part upon health monitoring of individuals and populations, and methods generally necessitate invasive sampling. Here, we used a small, unmanned hexacopter drone to noninva- sively ﬂy above humpback whales from two populations, capture their exhaled breath (blow), and examine the associated microbiome. In the ﬁrst extensive exami- nation of the large-whale blow microbiome, we present surprising results about the discovery of a large core microbiome that was shared across individual whales from geographically separated populations in two ocean basins. We suggest that this core microbiome, in addition to other microbiome characteristics, could be a useful fea- ture for health monitoring of large whales worldwide. msystems.asm.org 1 September/October 2017 Volume 2 Issue 5 e00119-17 Extensive Core Microbiome in Drone- Captured Whale Blow Supports a Framework for Health Monitoring Gregory Caporaso, Northern Arizona University Copyright © 2017 Apprill et al. This is an open- access article distributed under the terms of the Creative Commons Attribution 4.0 International license. Address correspondence to Amy Apprill, aapprill@whoi.edu. A.A. and C.A.M. contributed equally to this work. Extensive core microbiome in drone- captured whale blow supports a framework for health monitoring Received 5 September 2017 Accepted 13 September 2017 Published 10 October 2017 Citation Apprill A, Miller CA, Moore MJ, Durban JW, Fearnbach H, Barrett-Lennard LG. 2017. Extensive core microbiome in drone-captured whale blow supports a framework for health monitoring. mSystems 2:e00119-17. https://doi .org/10.1128/mSystems.00119-17. Editor J. Gregory Caporaso, Northern Arizona University Extensive Core Microbiome in Drone- Captured Whale Blow Supports a Framework for Health Monitoring Amy Apprill,a Carolyn A. Miller,a Michael J. Moore,b John W. Durban,c Holly Fearnbach,d Lance G. Barrett-Lennarde,f Amy Apprill,a Carolyn A. Miller,a Michael J. Moore,b John W. Durban,c Holly Fearnbach,d Lance G. Barrett-Lennarde,f Department of Marine Chemistry and Geochemistry, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USAa; Biology Department, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts, USAb; Marine Mammal and Turtle Division, Southwest Fisheries Science Center, National Marine Fisheries Service, NOAA, La Jolla, California, USAc; SR3 SeaLife Response, Rehabilitation, and Research, Mukilteo, Washington, USAd; Coastal Ocean Research Institute, Vancouver Aquarium, Vancouver, BC, Canadae; Zoology Department, University of British Columbia, Vancouver, BC, Canadaf ABSTRACT The pulmonary system is a common site for bacterial infections in ceta- ceans, but very little is known about their respiratory microbiome. We used a small, unmanned hexacopter to collect exhaled breath condensate (blow) from two geo- graphically distinct populations of apparently healthy humpback whales (Megaptera novaeangliae), sampled in the Massachusetts coastal waters off Cape Cod (n 17) and coastal waters around Vancouver Island (n 9). Bacterial and archaeal small- subunit rRNA genes were ampliﬁed and sequenced from blow samples, including many of sparse volume, as well as seawater and other controls, to characterize the associated microbial community. The blow microbiomes were distinct from the sea- water microbiomes and included 25 phylogenetically diverse bacteria common to all sampled whales. This core assemblage comprised on average 36% of the microbiome, making it one of the more consistent animal microbiomes studied to date. The closest phylogenetic relatives of 20 of these core microbes were previously detected in marine mammals, suggesting that this core microbiome assemblage is specialized for marine mammals and may indicate a healthy, noninfected pulmonary system. Pathogen screen- ing was conducted on the microbiomes at the genus level, which showed that all blow and few seawater microbiomes contained relatives of bacterial pathogens; no known cetacean respiratory pathogens were detected in the blow. Overall, the discovery of a shared large core microbiome in humpback whales is an important advancement for health and disease monitoring of this species and of other large whales. Received 5 September 2017 Accepted 13 September 2017 Published 10 October 2017 Citation Apprill A, Miller CA, Moore MJ, Durban JW, Fearnbach H, Barrett-Lennard LG. 2017. Extensive core microbiome in drone-captured whale blow supports a framework for health monitoring. mSystems 2:e00119-17. https://doi .org/10.1128/mSystems.00119-17. Editor J. KEYWORDS SSU rRNA gene, bacteria, drone, humpback whale, microbiome A number of large whale populations are listed as endangered or critically endan- gered (1), and their conservation and management greatly depend on understand- ing the relationship between anthropogenic disturbances and health (2–5). The pul- msystems.asm.org 1 September/October 2017 Volume 2 Issue 5 e00119-17 Apprill et al. monary system is a common site of cetacean infection and disease (6, 7), and yet cases are diagnosed mostly in live-stranded or dead whales (8) and rarely in the wild. In humans, exhaled breath is used to test for bacterial and fungal infections of the lower respiratory tract (9). Therefore, examining the microorganisms in the exhaled breath of whales, commonly referred to as blow, may serve as an important indicator of whale respiratory health, including the identiﬁcation of potential bacterial, fungal, and viral respiratory pathogens. Advances in aerial drone technology offer new opportunities for studying the health of whales remotely and noninvasively. For example, the exhaled breath of large whales can be sampled with a small aerial drone (10), given that whales exhale large volumes of air when they surface and usually breathe several times during a surfacing interval. The exhaled breath contains mucus and moisture that, when released into the com- parably cooler external air, condense to form a visible mass of vapor, which can be collected. Often blow is collected by approaching the whale in a small boat and holding an ~7-m pole with a collection plate above the blow hole (11, 12), requiring a skilled team and presenting safety risks to both the researchers and the whale. However, as mentioned above, blow was successfully collected by ﬂying a remotely operated drone through the visible mass of vapor, thus offering a less invasive and safer platform for blow collections (10). Knowledge of respiratory-associated microorganisms in cetaceans is limited. Several studies examined the diversity of respiratory-associated bacteria in captive and wild bottlenose dolphins (Tursiops truncatus) and provided preliminary evidence that dol- phins host a core group of bacteria associated with the respiratory system (13–15). A cultivation-based study of blow from killer whales (Orcinus orca) identiﬁed pathogenic and antibiotic-resistant bacteria and fungi (16), which may be presenting health risks to the whales. The only study of blow-associated microorganisms in baleen whales was conducted using taxonomic screening for speciﬁc bacteria, leaving a number of ques- tions remaining about the broader diversity of the blow microbiome (10). KEYWORDS SSU rRNA gene, bacteria, drone, humpback whale, microbiome Given this limited knowledge about the large-whale respiratory microbiome and the possible implications of using the microbiome for health and pathogen monitoring, a broader understanding of the large-whale blow microbiome is needed. In this study, we sought to determine if drone-captured blow microbiomes of large whales could be used to remotely monitor the respiratory health of large whales. Speciﬁcally, we sought to characterize the microbiomes associated with drone- collected samples of blow to assess commonalities and differences between the blow microbiomes of individual whales and identify the presence of potential pathogens. To address these goals, a small, unmanned hexacopter drone (10) (Fig. 1a) was used to collect blow samples from humpback whales (Megaptera novaeangliae) from Race Point Channel, north of Cape Cod, MA, and coastal waters surrounding Vancouver Island, in both British Columbia and Washington State. We examined the blow microbiomes, as well as those associated with the surface seawater, by sequencing and comparing partial small-subunit (SSU) rRNA genes of bacteria and archaea. These analyses revealed that blow microbiomes are distinct from seawater and contain an extensive network of consistent core microbiome members whose absence could serve as an important framework for health monitoring. RESULTS Blow microbiomes are similar between individuals and distinct from surface seawater. By ﬂying a small, remotely operated hexacopter drone 2 to 4 m above humpback whales (Fig. 1b), blow was collected from 17 whales off Cape Cod and nine whales near Vancouver Island (see Table S1 in the supplemental material). Partial bacterial and archaeal SSU rRNA genes were ampliﬁed from DNA extractions of the blow samples, including replicate blows from eight animals; environmental controls (a nonblow ﬂight and nine replicate surface seawater samples from around Vancouver Island; sampling equipment was not available for Cape Cod seawater); and technical controls (DNA extraction of sterile swabs and PCR blanks) and sequenced, resulting in msystems.asm.org 2 September/October 2017 Volume 2 Issue 5 e00119-17 Drone-Captured Whale Blow Microbiome a b FIG 1 (a) Photograph of the APH-22 hexacopter launching for ﬂight, with a petri dish atop and a 96-well PCR plate attached on a forward arm for whale blow sampling. (b) Photograph of the hexacopter collecting blow from a humpback whale off Cape Cod. Photographs courtesy of the authors. a FIG 1 (a) Photograph of the APH-22 hexacopter launching for ﬂight, with a petri dish atop and a 96-well PCR plate attached on a forward arm for whale blow sampling. (b) Photograph of the hexacopter collecting blow from a humpback whale off Cape Cod. Photographs courtesy of the authors. 2,337 to 180,141 sequences for the noncontrol samples. Minimum entropy decompo- sition (MED) (17), a sensitive method for partitioning sequences into operational taxonomic units (here referred to as MEDs or nodes), identiﬁed 616 MEDs for the entire data set. A cluster dendrogram analysis of Bray-Curtis dissimilarity (18) of the MEDs showed that community compositions of the technical control samples were similar to each other but different from the majority of blow samples (Fig. 2). Five sparse-volume (volume observed in the ﬁeld) blow samples (WA_A_F06, H_C_a, H_A_a, H_K, and H_B_b) clustered with the technical controls, indicating that the volume of these samples likely was so low that they only reﬂected the background microbial signal of technical contaminants such as laboratory reagents. Thus, these ﬁve low-volume samples were removed from the data set and are not included in the results presented below. Although two blow samples (BC_A_b_mix and BC_A_B_unk) from the same whale were more similar to the surface seawater than to the other humpback blow samples (Fig. RESULTS 2), seawater microorganisms may be incorporated into the blow; hence, these blow samples were included in all further analyses. The compositions of the humpback blow microbiotas were signiﬁcantly different from those of the microbiotas of surface seawater (permutational multivariate analysis of variance [PERMANOVA], F 61.364, P 0.001). Although the compositions of the humpback whale blow microb- 2,337 to 180,141 sequences for the noncontrol samples. Minimum entropy decompo- sition (MED) (17), a sensitive method for partitioning sequences into operational taxonomic units (here referred to as MEDs or nodes), identiﬁed 616 MEDs for the entire data set. A cluster dendrogram analysis of Bray-Curtis dissimilarity (18) of the MEDs showed that community compositions of the technical control samples were similar to each other but different from the majority of blow samples (Fig. 2). Five sparse-volume (volume observed in the ﬁeld) blow samples (WA_A_F06, H_C_a, H_A_a, H_K, and H_B_b) clustered with the technical controls, indicating that the volume of these samples likely was so low that they only reﬂected the background microbial signal of technical contaminants such as laboratory reagents. Thus, these ﬁve low-volume samples were removed from the data set and are not included in the results presented below. Although two blow samples (BC_A_b_mix and BC_A_B_unk) from the same whale were more similar to the surface seawater than to the other humpback blow samples (Fig. 2), seawater microorganisms may be incorporated into the blow; hence, these blow samples were included in all further analyses. The compositions of the humpback blow microbiotas were signiﬁcantly different from those of the microbiotas of surface seawater (permutational multivariate analysis of variance [PERMANOVA], F 61.364, P 0.001). Although the compositions of the humpback whale blow microb- msystems.asm.org 3 September/October 2017 Volume 2 Issue 5 e00119-17 Apprill et al. msystems.asm.org 4 RESULTS Data from replicate collections are shown for the Vancouver Island blow and seawater samples with multiple symbols per sample name. 0.05). The technical controls were used only to assess the microbial signal of blow and seawater samples, and while they appear in Fig. 2, they are not represented in any further analyses but do serve as a constant source for assessing contamination (see below description of core microbiome). Diverse assemblage of microorganisms found in humpback blow. The hump- back whale blow samples contained a diverse assemblage of microorganisms, in terms of both richness and phylogeny. In whale blow, the observed number of MEDs, which is comparable to a ﬁne-resolution species richness index, ranged from 164 to 515, with an average of 321 (Fig. 3a). The number of MEDs in surface seawater samples generally fell into this range as well, suggesting that the blow and surface seawater support a similarly rich community of cells (Fig. 3a). There was considerably more consistency between the numbers of observed MEDs within replicate samples in the seawater microbiome than for the Vancouver Island humpback whale blow samples (the only blow samples that were replicated), which could be related to inconsistencies in whale blows, volume of blow collection, or sequencing depth (Fig. 3a). Simpson’s index of diversity (19), which also considers evenness, generally ranged above 0.90, indicating high microbial diversity in the samples. The Simpson index was comparable for blow and seawater samples from Vancouver Island but was more variable for the Cape Cod blow samples (Fig. 3b). A phylogenetically diverse assemblage of sequences that spanned 15 phyla of Bacteria and two phyla of Archaea was identiﬁed in the humpback whale blow microbiomes. Several classes were shared with the seawater samples, including Gam- maproteobacteria, Flavobacteriia, and Alphaproteobacteria (Fig. 4). However, the whale blow samples from both locations harbored classes not common in the surface seawater, such as Actinobacteria, Bacilli, Clostridia, Fusobacteriia, Bacteroidia, Acidimicro- biia, Epsilonproteobacteria, Deltaproteobacteria, Erysipelotrichia, and Mollicutes (Fig. 4). Extensive core microbiome in whale blow with relatedness to other marine mammals. Twenty-six MEDs were present in all humpback whale blow samples. However, one of these MEDs, 6038, identiﬁed as Bacillus, was considered a technical contaminant and not a member of the blow microbiome because it was found in all technical control samples and is a common contaminant in laboratory reagents (20). RESULTS iotas were 50 to 90% similar to each other, the microbiotas of the blow samples collected off Cape Cod were nevertheless signiﬁcantly different from those collected around Vancouver Island (PERMANOVA, F 5.8224, P 0.001), and neither ﬁnding was impacted by the factor of sequencing depth (PERMANOVA with pairwise tests, P PCR_S PCR_BCWA DNA_BCWA1 DNA_BCWA3 WA_A_F06 Flight_S DNA_S H_C_a H_A_a H_K DNA_BCWA2 H_B_b BC_SW_A_b BC_A_b_mix BC_A_b_unk WA_SW_C_a WA_SW_C_b WA_SW_A_a WA_SW_B_a WA_SW_A_b WA_SW_B_b BC_SW_B_b BC_SW_B_b BC_SW_C_b BC_SW_F_b BC_SW_E_a BC_SW_E_b BC_SW_F_a BC_SW_C_a BC_SW_D_a BC_SW_D_b BC_SW_A_a H_A_b H_D WA_C_d WA_C_p BC_E_p BC_C_d BC_C_p H_M H_C_b H_O H_J BC_A_a BC_F_p H_G H_B_a H_H H_F H_N H_I H_E H_L H_Q H_P WA_A_F08 WA_B_p WA_B_d BC_E_d BC_F_d BC_D_bp BC_D_bd BC_D_ap BC_D_ad BC_B_d BC_B_p 100 80 60 40 20 0 Humpback blow, Cape Cod Humpback blow, Vancouver Island Seawater Flight control Technical control Similarity Humpback blow Seawater Controls FIG 2 Comparison of humpback blow, surface seawater, and ﬂight and technical control samples using a cluster dendrogram of bacterial and archaeal SSU rRNA genes grouped using minimum entropy decomposition (17) and compared using Bray-Curtis dissimilarity (18). The categories “controls,” “seawater,” and “humpback blow” were inferred based on the clustering patterns and sample types. September/October 2017 Volume 2 Issue 5 e00119-17 msyste 60 Controls Humpback blow FIG 2 Comparison of humpback blow, surface seawater, and ﬂight and technical control samples using a cluster dendrogram of bacterial and archaeal SSU rRNA genes grouped using minimum entropy decomposition (17) and compared using Bray-Curtis dissimilarity (18). The categories “controls,” “seawater,” and “humpback blow” were inferred based on the clustering patterns and sample types. RESULTS iotas were 50 to 90% similar to each other, the microbiotas of the blow samples collected off Cape Cod were nevertheless signiﬁcantly different from those collected around Vancouver Island (PERMANOVA, F 5.8224, P 0.001), and neither ﬁnding was impacted by the factor of sequencing depth (PERMANOVA with pairwise tests, P msystems.asm.org 4 September/October 2017 Volume 2 Issue 5 e00119-17 Drone-Captured Whale Blow Microbiome 200 300 400 500 0.84 0.88 0.92 0.96 H_D H_A H_J H_G H_M H_C H_O H_H H_L H_B H_N H_F H_I H_P H_Q H_E BC_C BC_E BC_D BC_B WA_C WA_B BC_A WA_A BC_F BC_SW_A BC_SW_F BC_SW_D WA_SW_B WA_SW_A BC_SW_E BC_SW_C WA_SW_C BC_SW_B Cape Cod humpback blow Cape Cod humpback blow Vancouver Island humpback blow Vancouver Island humpback blow Vancouver Island seawater Vancouver Island seawater Observed MEDs Simpson’s Index a b Whale or seawater sample identification FIG 3 Diversity of whale blow and seawater samples from minimum entropy decomposition (MED) node groupings (17), including observed number of MEDs, a relative estimate of species richness between samples (a), and Simpson’s index (19), a relative estimate of diversity and evenness (b). Data from replicate collections are shown for the Vancouver Island blow and seawater samples with multiple symbols per sample name. 200 300 400 500 0.84 0.88 0.92 0.96 H_D H_A H_J H_G H_M H_C H_O H_H H_L H_B H_N H_F H_I H_P H_Q H_E BC_C BC_E BC_D BC_B WA_C WA_B BC_A WA_A BC_F BC_SW_A BC_SW_F BC_SW_D WA_SW_B WA_SW_A BC_SW_E BC_SW_C WA_SW_C BC_SW_B Cape Cod humpback blow Cape Cod humpback blow Vancouver Island humpback blow Vancouver Island humpback blow Vancouver Island seawater Vancouver Island seawater Observed MEDs Simpson’s Index a b Whale or seawater sample identification FIG 3 Diversity of whale blow and seawater samples from minimum entropy decomposition (MED) node groupings (17), including observed number of MEDs, a relative estimate of species richness between samples (a), and Simpson’s index (19), a relative estimate of diversity and evenness (b). Data from replicate collections are shown for the Vancouver Island blow and seawater samples with multiple symbols per sample name FIG 3 Diversity of whale blow and seawater samples from minimum entropy decomposition (MED) node groupings (17), including observed number of MEDs, a relative estimate of species richness between samples (a), and Simpson’s index (19), a relative estimate of diversity and evenness (b). RESULTS The remaining 25 MEDs common to all blow samples were considered “core” members of the humpback whale blow microbiome. These core microbiome members spanned September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 5 Apprill et al. 0 25 50 75 100 H_H H_P H_Q H_L H_F H_B_a H_E H_J H_M H_I H_O H_G H_D H_N H_A_b H_C_b BC_C_p BC_E_d BC_F_p BC_C_d BC_E_p BC_F_d BC_B_d BC_D_bp BC_D_ap BC_B_p BC_D_ad BC_D_bd BC_A_a BC_SW_A_a BC_SW_A_b BC_A_b_unk BC_SW_F_a BC_SW_C_a BC_SW_B_a BC_SW_D_a BC_SW_E_a BC_SW_D_b BC_SW_E_b BC_SW_F_b BC_SW_C_b BC_SW_B_b Class Percentage of SSU rRNA gene sequences Gammaproteobacteria Flavobacteriia Alphaproteobacteria Actinobacteria Bacilli Clostridia Fusobacteriia Cyanobacteria Bacteroidia Acidimicrobiia Marine Group I (Archaea) Betaproteobacteria Epsilonproteobacteria Thermoplasmata Deltaproteobacteria Erysipelotrichia Verrucomicrobiae Sphingobacteriia Mollicutes Other Cape Cod humpback blow Vancouver Island humpback blow Vancouver Island seawater WA_C_d WA_B_d WA_B_p WA_A_F08 WA_C_p BC_A_b_mix WA_SW_A_a WA_SW_A_b WA_SW_C_b WA_SW_C_a WA_SW_B_a WA_SW_B_b FIG 4 Overview of the phylogeny of the bacteria and archaea associated with the humpback whale blow and seawater samples on a class level based on partial SSU rRNA gene sequences. 0 25 50 75 100 Class Percentage of SSU rRNA gene sequences Gammaproteobacteria Flavobacteriia Alphaproteobacteria Actinobacteria Bacilli Clostridia Fusobacteriia Cyanobacteria Bacteroidia Acidimicrobiia Marine Group I (Archaea) Betaproteobacteria Epsilonproteobacteria Thermoplasmata Deltaproteobacteria Erysipelotrichia Verrucomicrobiae Sphingobacteriia Mollicutes Other 0 H_H H_P H_Q H_L H_F H_B_a H_E H_J H_M H_I H_O H_G H_D H_N H_A_b H_C_b BC_C_p BC_E_d BC_F_p BC_C_d BC_E_p BC_F_d BC_B_d BC_D_bp BC_D_ap BC_B_p BC_D_ad BC_D_bd BC_A_a BC_SW_A_a BC_SW_A_b BC_A_b_unk BC_SW_F_a BC_SW_C_a BC_SW_B_a BC_SW_D_a BC_SW_E_a BC_SW_D_b BC_SW_E_b BC_SW_F_b BC_SW_C_b BC_SW_B_b Cape Cod humpback blow Vancouver Island humpback blow Vancouver Island seawater WA_C_d WA_B_d WA_B_p WA_A_F08 WA_C_p BC_A_b_mix WA_SW_A_a WA_SW_A_b WA_SW_C_b WA_SW_C_a WA_SW_B_a WA_SW_B_b FIG 4 Overview of the phylogeny of the bacteria and archaea associated with the humpback whale blow and seawater samples on a class level based on partial SSU rRNA gene sequences. Vancouver Island seawater FIG 4 Overview of the phylogeny of the bacteria and archaea associated with the humpback whale blow and seawater samples on a class level based on partial SSU rRNA gene sequences. seven phyla or classes (for the Proteobacteria) and ranged in relative abundance from 0.01 to 18% of the total community in each blow sample (Fig. 5). Collectively, the 25 core microbiome members comprised 36.0% (standard deviation [SD], 10.5%) of the total humpback blow sequences from whales residing in both geographic locations. RESULTS To identify samples that might be suitable for respiratory pathogen screening, the taxonomy of the blow MEDs was screened at the level of genus against a custom pathogen database September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 6 Drone-Captured Whale Blow Microbiome Homology with sequences from marine mammals Homology with sequences from seawater Homology with sequences from terrestrial mammals 20 0 5 10 15 MED6786 MED6836 MED857 MED8466 MED678 MED680 MED7942 MED5795 MED8352 MED8398 MED5986 MED8097 MED1312 MED6327 MED6423 MED6459 MED9495 MED2426 MED7254 MED7186 MED7188 MED9687 MED9388 MED3986 MED7357 Percentage of total SSU rRNA gene sequences in humpback blow Actinobacteria Bacteroidetes Firmicutes Gammaproteobacteria Alphaproteo Epsilonproteo Fusobacteria Corynebacterium Corynebacterium “Candidatus Actinomarina” Zimmermannella Leucobacter Bacteroidetes VC2.1 Bac22 Tenacibaculum Porphyromonas Guggenheimella Helcococcus Helcococcus Dielma Cardiobacteraceae Cardiobacteraceae Oceanospirillaceae Oceanospirillaceae Moraxella Psychrobacter Moraxella Moraxella Moraxella Rhodovulum SAR11, Surface 1 Arcobacter Oceanivirga FIG 5 Box plots displaying the median, minimum, maximum, and ﬁrst and third quartiles of the percentage of the 25 members of the core microbiome detected in all humpback whale blow samples, obtained from minimum entropy decomposition (17) nodes of bacterial and archaeal SSU rRNA genes. The phyla or subphyla (for Proteobacteria) of each MED are listed below the box plot, the most detailed level of taxonomy available from the SILVA database (v.123) (52) is shown above, and the colors refer to sequences with homology to those previously recovered from marine mammals (blue), terrestrial mammals (green), or seawater (red). The color of MED 7188 is blue. FIG 5 Box plots displaying the median, minimum, maximum, and ﬁrst and third quartiles of the percentage of the 25 members of the core microbiome detected in all humpback whale blow samples, obtained from minimum entropy decomposition (17) nodes of bacterial and archaeal SSU rRNA genes. The phyla or subphyla (for Proteobacteria) of each MED are listed below the box plot, the most detailed level of taxonomy available from the SILVA database (v.123) (52) is shown above, and the colors refer to sequences with homology to those previously recovered from marine mammals (blue), terrestrial mammals (green), or seawater (red). The color of MED 7188 is blue. that included human and animal pathogens recognized by the American Biological Safety Association, as well as previously identiﬁed and potential pathogens from studies of marine mammals (7, 21–27) (Table S2). RESULTS Phylogenetic analysis of the core microbiota using ARB and the SILVA database (v.128) revealed that 20 of the 25 core members were most closely related to microbial sequences recovered from other species of marine mammals, most commonly from the mouths and blowholes of bottlenose dolphins (Fig. 5; Table 1). Core members with homology to other marine mammal-associated microbes that were represented at mean abundances of 1% or greater in all humpback blow samples were Corynebacte- rium MED6786 (bottlenose dolphin forestomach), Tenacibaculum MED7942 (bottlenose dolphin blowhole), Porphyromonas MED5795 (bottlenose dolphin mouth and hump- back whale skin), Cardiobacteriaceae MED1312 (bottlenose dolphin blowhole), Oceano- spirillaceae MED6423 and MED6459 (bottlenose dolphin forestomach and blowhole), Moraxella MED9495 and MED7186 (bottlenose dolphin mouth), Psychrobacter MED2426 (bottlenose dolphin mouth and blowhole), and Arcobacter MED3986 (bottlenose dol- phin blowhole and forestomach). The nine other core MEDs with homology to se- quences recovered from marine mammals were present at mean abundances less than 1% (Fig. 5; Table 1). Additionally, two of the core members, Corynebacterium MED6836 and Zimmermannella MED8466, shared homology to sequences previously identiﬁed in terrestrial mammals (Fig. 5; Table 1). Last, three core members present in the humpback blow shared homology to sequences commonly recovered from seawater: “Candidatus Actinomarina” MED857; the Roseobacter-afﬁliated Rhodovulum 9687; and SAR11, Sur- face 1 clade 9833. Indeed, these three seawater-afﬁliated MEDs were well represented in the seawater samples, each present at average abundances ranging from 4.3 to 5.1%. No cetacean respiratory pathogens detected in humpback blow. RESULTS One hundred ﬁfteen MEDs from whale blow and seawater spanning 31 genera had a genus-level phylogenetic afﬁliation with pathogens listed in the database (Fig. 6). The pathogen relatives identiﬁed in the whale blow were numerous and distinct from those present in the seawater samples (Fig. 6). Of the potential pathogens identiﬁed in the blow samples, 10 genera were previously identiﬁed in other species of marine mammals, and of those, Corynebacterium was the only genus also represented in the core microbiome of humpback whale blow (Table 2). Speciﬁcally, two uncharacterized Corynebacterium species were previously cultivated from the spleen, blood, and lymph nodes of bottlenose dolphins following mortality (21), although it should be noted that this genus harbors many nonpathogens that often associate with healthy humans (28, 29). Of the potential pathogens identiﬁed in the humpback blow samples, none were known cetacean respiratory pathogens. September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 7 Apprill et al. TABLE 1 List of core MED nodes of SSU rRNA gene sequences with taxonomic afﬁliations and description of environment where the most similar sequences were recovered MED Taxonomic afﬁliation Environment of most similar sequences (GenBank identiﬁer) 6786 Corynebacterium Bottlenose dolphin forestomach (JQ192966) 6836 Corynebacterium Horse uterus (CP011546) 857 “Candidatus Actinomarina” Seawater next to bottlenose dolphin (JQ195517) 8466 Zimmermannella Human skin (GQ043066) 678 Leucobacter Bottlenose dolphin blowhole (FJ959933) 680 Bacteroidetes VC2.1 Bac22 Bottlenose dolphin mouth (JQ210604) 7942 Tenacibaculum Bottlenose dolphin blowhole (FJ959464) 5795 Porphyromonas Bottlenose dolphin mouth (KC259428) and humpback whale skin (GU202009) 8352 Guggenheimella Bottlenose dolphin mouth (JQ208689) 8398 Helcococcus Bottlenose dolphin mouth (FJ959814) 5986 Helcococcus Sea lion rectum (JQ208548) 8097 Dielma Bottlenose dolphin mouth (JQ209430) 1312 Cardiobacteriaceae Bottlenose dolphin blowhole (FJ960054) 6327 Cardiobacteriaceae Bottlenose dolphin mouth (KC260696) 6423 Oceanospirillaceae Bottlenose dolphin forestomach (JQ194233) and blowhole (FJ959835) 6459 Oceanospirillaceae Bottlenose dolphin forestomach (JQ193528) and blowhole (FJ959835) 9495 Moraxella Bottlenose dolphin mouth (JQ216648) 2426 Psychrobacter Bottlenose dolphin blowhole (FJ960065) and mouth (KC260479) 7254 Moraxella Bottlenose dolphin mouth (JQ216648) 7186 Moraxella Bottlenose dolphin mouth (JQ216648) 7188 Moraxella Bottlenose dolphin mouth (JQ216648) 9687 Rhodovulum Sub-Antarctic seawater (AY697867) 9388 SAR11, Surface 1 clade Gulf of Mexico seawater (KU578707) 3986 Arcobacter Bottlenose dolphin blowhole (FJ959747) and forestomach (JQ194125) 7357 Oceanivirga Bottlenose dolphin forestomach (JQ193505) and mouth (KC260320) strates that the microbial communities in blow and surface seawater are different, indicating that whale blow is not just aerosolized seawater. Although both baleen and toothed whale blows were previously compared to seawater (10, 16), this was the ﬁrst study to apply more comprehensive microbial diversity analyses to whale blow. Indeed, there were some surface seawater-associated bacteria within the humpback blow examined in our study, but this is not surprising because seawater may remain in the upper tract between breaths and enter and leave the blowhole cavity and the upper nasal tract of a surfacing whale during the inhalation phase (M. Moore, unpublished observations), to the point where these samples could be described as seawater lavages of the upper respiratory tract seeded with condensed exhalation. Three hump- back blow core MEDs, including bacteria from the globally abundant SAR11 clade (30), were well represented in the seawater samples, at relative abundances of 4 to 5%, suggesting that the most abundant cells in the seawater are likely the cells making their way into the upper respiratory tract. September/October 2017 Volume 2 Issue 5 e00119-17 DISCUSSION In the ﬁrst examination of whale blow microbial diversity, we show that blow from humpback whales supports a diverse and rich community of microorganisms with a number of features that may be useful for monitoring health. First, our study demon- September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 7 Genus Vancouver Island seawater Vancouver Island whale blow Cape Cod whale blow FIG 6 Heat map of the distribution and relative abundance of the minimum entropy decomposition nodes (MEDs) (17), obtained from bacterial and archaeal SSU rRNA genes, from humpback whale blow and surface seawater samples that are related to potential pathogens. The MEDs are classiﬁed to genus level, and the line below a genus name indicates membership within the same genus. FIG 6 Heat map of the distribution and relative abundance of the minimum entropy decomposition nodes (MEDs) (17), obtained from bacterial and archaeal SSU rRNA genes, from humpback whale blow and surface seawater samples that are related to potential pathogens. The MEDs are classiﬁed to genus level, and the line below a genus name indicates membership within the same genus. from different populations residing in distinct ocean basins. Deﬁning core microorgan- isms in a host environment is helpful because the persistence of core members between individual animals suggests that these core microbes may be beneﬁcial for the host (35, 36). Core microbiomes are generally identiﬁed by applying criteria such as sequences or taxonomic groups that are present in 30 to 80% of hosts, and only rarely is 100% membership within all hosts used as the deﬁning criterion (35). Here, we used the 100% host membership criterion at a highly discriminative species-type level of phylogenetic similarity (as determined by MED) and identiﬁed 25 core microbiome members, which to our knowledge is an unprecedented number of core members at this discriminative scale for any marine or terrestrial host microbiome. The most similar ﬁnding using the 100% core membership criterion is for the human gut and hands, which host 18 and 5 core microbial members, respectively (37, 38). While core mem- bership in the human respiratory tract has not been investigated extensively, studies suggest high variability in the lower respiratory tract microbiome between individuals (39). Stable and persistent core microbiomes with low interindividual variability, as observed here for the humpback whales, suggest that the microbiome and/or host may receive beneﬁts from the presence of this collective group of cells, such as nutritional or immune beneﬁts. However, examining blow microbiomes from unhealthy or dis- eased animals will be necessary to understand if the core microbiome does indeed change with health state. Blow samples from populations of large whales with healthy from different populations residing in distinct ocean basins. Furthermore, the remaining 22 core microbiome sequences were most closely related to marine and terrestrial mammals, suggesting that these cells were, indeed, coming from the whales, not the seawater. The exhaled breath of most mammals is believed to comprise cells that originate from both the mouth and the nasopharynx because they are anatomically connected (31–33). Because the nasopharynx of cetaceans is not connected to the mouth, cetacean breath passes through only the respiratory tract; therefore, the breath microbes originating from the cetacean are from the respiratory tract and almost certainly do not include oral microbes. Two core members of the blow microbiome were previously detected on humpback whale skin, Porphyromonas (MED5795) and Psychrobacter (MED2426) (34). It is possible that these whale skin-associated microbes reside on the epithelium of the blowhole and become aerosolized with the force of the whale’s exhalation. Thus, in addition to seawater-associated microbes, a mixture of pulmonary bacteria and mi- crobes associated with the epithelial cells of the blowholes likely comprise the blow microbiome in cetaceans. The second and possibly most useful feature of the humpback whale blow micro- biome for health monitoring is that it contains a surprisingly high number of core microbiome members shared by all individuals, despite our samples being collected September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 8 Drone-Captured Whale Blow Microbiome Veillonella Mycoplasma Vibrio Psychrobacter Pseudomonas Pseudoalteromonas Moraxella Haemophilus Escherichia-Shigella Balneatrix Actinobacillus Acinetobacter Fusobacterium Chryseobacterium Capnocytophaga Helicobacter Campylobacter Arcobacter Peptostreptococcus Helcococcus Neisseria Porphyromonas Bacteroides Streptococcus Staphylococcus Listeria Lactococcus Lactobacillus Bacillus Sphingomonas Corynebacterium Arthrobacter H_A_b H_B_a H_C_b H_D H_E H_F H_G H_H H_I H_J H_L H_M H_N H_O H_P H_Q BC_A_a BC_A_b1 BC_A_b2 BC_B_d BC_B_p BC_C_d BC_C_p BC_D_ad BC_D_ap BC_D_bd BC_D_bp BC_E_d BC_E_p BC_F_d BC_F_p WA_A WA_B_d WA_B_p WA_C_d WA_C_p BC_SW_A_a BC_SW_A_b BC_SW_B_a BC_SW_B_b BC_SW_C_a BC_SW_C_b BC_SW_D_a BC_SW_D_b BC_SW_E_a BC_SW_E_b BC_SW_F_a BC_SW_F_b WA_SW_A_a WA_SW_A_b WA_SW_B_a WA_SW_B_b WA_SW_C_a WA_SW_C_b Genus 0.01 1.00 Abundance Cape Cod whale blow Vancouver Island whale blow Vancouver Island seawater FIG 6 Heat map of the distribution and relative abundance of the minimum entropy decomposition nodes (MEDs) (17), obtained from bacterial and archaeal SSU rRNA genes, from humpback whale blow and surface seawater samples that are related to potential pathogens. The MEDs are classiﬁed to genus level, and the line below a genus name indicates membership within the same genus. September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 9 Apprill et al. TABLE 2 List of potential pathogens identiﬁed in the humpback whale blow according to genus-level identity, as well as descriptions of recognized pathogens and putative marine mammal pathogens ogens identiﬁed in the humpback whale blow according to genus-level identity, as well as descriptions of ative marine mammal pathogens g p g p p g Genus MED(s) in whale blowa Putative marine mammal pathogen (reference[s]) No. of recognized pathogens in genus Acinetobacter 9761, 8888, 5164, 8857, 5165, 1070, 9765, 1067 Acinetobacter lwofﬁi (21) NAb Actinobacillus 6581 NA 9 Arcobacter 8192, 3986, 8223, 8193, 8194, 3989, 8222 NA 2 Arthrobacter 100 Arthrobacter cumminsii (16) NA Bacillus 6010 NA 4 Bacteroides 6758 NA 14 Balneatrix 4335 NA 1 Campylobacter 3645, 3641, 3642, 3651, 80 NA 11 Capnocytophaga 6603 NA 6 Chryseobacterium 7829 NA 4 Corynebacterium 6786, 6788, 6836, 8673, 8671, 6797, 8652, 8654, 3514 Corynebacterium spp. (21) 33 Escherichia-Shigella 5544 Escherichia coli (21) 0 Fusobacterium 35 Fusobacterium varium (21) 13 Haemophilus 9046, 9047, 9045, 5437 NA 18 Helcococcus 4258, 8395, 8398, 8377, 261, 5988, 8396, 1947, 1949, 257, 8397,8378, 5986, 260 NA 2 Helicobacter 3988 NA 12 Lactobacillus 4629, 4625, 4645, 322 NA 3 Lactococcus 2216 NA 1 Listeria 5860 NA 2 Moraxella 7227, 8609 NA 9 Mycoplasma 7640, 3008, 7642 Mycoplasma sp. (16) 63 Neisseria 9019, 9020 NA 11 Peptostreptococcus 8129, 8130 NA 10 Porphyromonas 9273, 9078, 9274, 5800, 9077, 9112, 5796, 9114, 5799, 5795, 5797, 9109, 9080, 5486 NA 12 Pseudoalteromonas 5116, 5119 NA 1 Pseudomonas 9740, 5159, 9739, 5873 Pseudomonas aeruginosa (7) 9 Psychrobacter 8626, 2428, 2426, 7254, 2065, 8585, 4878, 8627, 8584, 2062, 2441,4879, 7255, 8610, 9480, 9482, 8582, 7157, 4874 NA 1 Sphingomonas 231 NA 2 Staphylococcus 6011, 6041, 6038 Staphylococcus aureus (7, 16), Staphylococcus cohnii (16), Staphylococcus epidermidis (16, 21), Staphylococcus warneri (16), Staphylococcus sp. (16), Staphylococcus delphini (23) 13 Streptococcus 7098, 3720, 7097, 2227, 2217, 2228 Alpha-hemolytic Streptococcus (16), Streptococcus zooepidemicus (7), Streptococcus group D (21) 27 Veillonella 1666, 1668 NA 1 Vibrio 2695 Vibrio anguillarum (16), Vibrio alginolyticus (16), Vibrio wodanis (16), Vibrio sp. (21) 15 aMED nodes (17) present in the core microbiome are in bold Streptococcus 7098, 3720, 7097, 2227, 2217, 2228 Veillonella 1666, 1668 Vibrio 2695 aMED nodes (17) present in the core microbiome are in bold. bNA, not applicable. and unhealthy individuals, such as the North Atlantic right whales, may be particularly useful for this comparison (40). Deﬁning core microorgan- isms in a host environment is helpful because the persistence of core members between individual animals suggests that these core microbes may be beneﬁcial for the host (35, 36). Core microbiomes are generally identiﬁed by applying criteria such as sequences or taxonomic groups that are present in 30 to 80% of hosts, and only rarely is 100% membership within all hosts used as the deﬁning criterion (35). Here, we used the 100% host membership criterion at a highly discriminative species-type level of phylogenetic similarity (as determined by MED) and identiﬁed 25 core microbiome members, which to our knowledge is an unprecedented number of core members at this discriminative scale for any marine or terrestrial host microbiome. The most similar ﬁnding using the 100% core membership criterion is for the human gut and hands, which host 18 and 5 core microbial members, respectively (37, 38). While core mem- bership in the human respiratory tract has not been investigated extensively, studies suggest high variability in the lower respiratory tract microbiome between individuals (39). Stable and persistent core microbiomes with low interindividual variability, as observed here for the humpback whales, suggest that the microbiome and/or host may receive beneﬁts from the presence of this collective group of cells, such as nutritional or immune beneﬁts. However, examining blow microbiomes from unhealthy or dis- eased animals will be necessary to understand if the core microbiome does indeed change with health state. Blow samples from populations of large whales with healthy September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 9 msystems.asm.org 10 Another reason that the presence of a core microbiome in whale blow could be a useful feature for monitoring health is that the absence of core members and/or the presence of atypical microbes in the blow could suggest an alteration in the growth environment or immune response of the microbes, as is typical for pulmonary infec- tions and diseases (33). Analyzing microbiomes associated with blow from humpback whales with known pulmonary conditions would greatly advance our understanding about the stability of the core microbiome. However, these conditions are challenging to diagnose in the wild. Instead, examining the blow microbiomes of whales with msystems.asm.org 10 msystems.asm.org 10 September/October 2017 Volume 2 Issue 5 e00119-17 Drone-Captured Whale Blow Microbiome ﬁeld-diagnosable conditions, such as poor body condition detectable by drone-enabled photogrammetry (41), will help advance our understanding of the variability of the core microbiome in relation to health. The third feature of the blow microbiome of humpback whales that is important for health monitoring is that the microbiomes can be coarsely screened for pathogens using the amplicon sequencing approach. We chose to screen for relatives of patho- gens, instead of just species-level identity to known pathogens, because cetacean pathogens are not well described, and the pulmonary and blowhole system of ceta- ceans is unique compared to other mammals. As knowledge of speciﬁc cetacean respiratory pathogens increases, a stricter sequence similarity approach could be applied. Until that time, analyzing blow microbiomes for potential pathogens using amplicon sequencing provides a means to screen for samples for further analysis of pathogens using quantitative PCR (qPCR) or metagenomics. Additionally, this amplicon- based method provides a means to understand the other possibly mutualistic or commensal members of the respiratory microbiome that would not be available if analysis were limited to pathogen detection. In this study, no known marine mammal respiratory pathogens were identiﬁed for further veriﬁcation using qPCR or metag- enomics. Recently, Raverty et al. suggested that potentially pathogenic bacteria and fungi, including some exhibiting resistance to antimicrobial agents, identiﬁed in killer whale blow and surface seawater may pose a threat to the endangered killer whale population (16). Large whales, including humpbacks, frequently reside in coastal areas of high anthropogenic inﬂuence, including wastewater and sewage dispersal (42, 43), and therefore, the probability of whales being exposed to pathogens is high. September/October 2017 Volume 2 Issue 5 e00119-17 Whales also exhibit behaviors, such as surfacing in close proximity to each other or feeding cooperatively (44, 45), which could facilitate the spread of pathogens between indi- viduals. Thus, screening for potential pathogens in the blow microbiome is a particu- larly useful endeavor for large whales. A fourth useful feature of the blow microbiome is that the richness, diversity, and sequence homology of microorganisms detected in whale blow were on par with those previously identiﬁed in the blowholes of bottlenose dolphins (13–15), the only com- parable microbiome data sets (i.e., broad taxonomic microbiome surveys that used cultivation-independent approaches). This feature suggests that blow microbiome monitoring criteria for one species may be applicable to other cetaceans. Although the criteria for sequence delineation differed between this study (which used MEDs) and the previous dolphin studies (which used 97% sequence similarity), all studies exam- ined surface seawater and found similar microbiome richness patterns between the marine mammal and seawater environments, thus making the data sets comparable on this generalized level. To further advance the blow microbiome as a health monitoring tool for large whales, several features require consideration and future development. While drones are safer and less invasive to the whale than pole sampling, they can result in low-volume samples. Indeed, several low-volume samples collected from this study were found to be similar in composition to samples from ﬂight and technical controls. Contaminating DNA is frequently identiﬁed in low-volume samples (20, 46), including those targeting low microbial biomass from human lungs (39), and the present study suggests that ﬂight and technical controls are necessary to ensure that biological samples have sufﬁcient volume and microbial biomass for accurate comparisons. Also, the differences in richness and community composition that were observed in the replicate blow samples were probably related to inconsistencies in whale blows or blow collections and signify the importance of repeated sampling. In addition to these recommendations, we acknowledge that this study presents a limited description of the microbiome by focusing on bacteria and archaea. Protists, viruses, and fungi can also be important indicators of respiratory illnesses (26, 47), and detection of these organisms may be achievable on some of the higher-volume samples using a metagenomics-based sequencing approach. Efforts should also be made to examine the growth environment for the blow-associated microorganisms, including character- msystems.asm.org 11 September/October 2017 Volume 2 Issue 5 e00119-17 Apprill et al. ization of blow chemistry, temperature, and salinity, which could inform conditions for microbial cell growth, enhance cultivation efforts, and ultimately allow us to better evaluate the beneﬁcial, commensal, or pathogenic potential of the microbial cells residing in the whale’s respiratory tract. Last, this and a previous study (10) used drones to collect blow-associated microorganisms from large whales, where this approach appears to be broadly applicable. These methods may also be applicable to smaller cetaceans, but this still needs to be examined in the context of the ability of small unmanned aerial systems (UAS) to operate near small cetaceans unobtrusively. Conclusions. This study demonstrates that remote sampling via drones provides a noninvasive means to collect whale blow for microbiome analysis. Using this technique, we showed that different humpback whales and populations harbor similar microbial communities in their exhaled blow, including the presence of a large number of speciﬁc core bacteria shared between all individuals. The persistence of these core members in apparently healthy individuals suggests that they may be indicative of a healthy, noninfected pulmonary system, and their presence or absence could be informative for health monitoring of humpback whales and possibly other large whales. MATERIALS AND METHODS Sample collection. Exhaled breath condensate (blow) was collected from humpback whales (Megaptera novaeangliae) in the Race Point Channel, north of Cape Cod, MA (n 17), during July 2015 and in two locations around Vancouver Island: Johnstone Strait and surrounding channels, British Columbia (n 6 whales with replicate samples for all animals), in August 2016 and off San Juan Island, Washington State (n 3 whales with replicate samples for 2 animals), in September 2016. Samples were collected using a small, unmanned hexacopter drone (APH-22; Aerial Imaging Solutions, Old Lyme, CT) operated by a pilot and copilot team from a vessel (48) (Fig. 1). Collection surfaces differed between the ﬂights and locations. For Cape Cod, the hexacopter’s dome and one to three 96-well PCR plates ﬁxed to the struts of the hexacopter were used as the sterile collection surfaces. For Vancouver Island, one forward-facing sterile, 96-well PCR plate; the dome of the hexacopter; and, for some whales, a 150-mm- diameter sterile petri dish afﬁxed to the top of the dome (Fig. 1a) were used. Prior to ﬂight, the hexacopter’s propellers, arms, struts, and dome were sterilized with 95% isopropanol. The hexacopter then was ﬂown 2 to 4 m or more above the blowhole, and once the whale exhaled, the hexacopter was returned to the boat so that the sample could be processed. Using sterile technique, the blow was swabbed from the collection surfaces using sterile cotton-tipped swabs or ﬂocked swabs (Copan Diagnostics, Inc., Murrieta, CA). Each swab then was placed in a sterile 2-ml cryovial, frozen in a liquid nitrogen vapor shipper, and transferred to 80°C until processing. In one case, a sample had a large-enough volume to pipette and thus was pipetted directly into a 2-ml cryovial prior to freezing. Sampled whales were photographed for identiﬁcation purposes using standard methods to identify duplicate samples from the same whale (49). As a control, the hexacopter was ﬂown with sterile PCR plates attached for the same ﬂight duration, altitude over the water, and distance from the boat as had been done when collecting actual blow samples. Upon landing, a sterile cotton-tipped swab was used to wipe the PCR plates and was placed in a sterile 2-ml cryovial, frozen, and stored in the same manner as the blow samples. September/October 2017 Volume 2 Issue 5 e00119-17 SUPPLEMENTAL MATERIAL Supplemental material for this article may be found at https://doi.org/10.1128/ mSystems.00119-17. Supplemental material for this article may be found at https://doi.org/10.1128/ mSystems.00119-17. TABLE S1, XLSX ﬁle, 0.01 MB. TABLE S1, XLSX ﬁle, 0.01 MB. TABLE S2, XLSX ﬁle, 0.04 MB. DATA SET S1, TXT ﬁle, 0.2 MB. MATERIALS AND METHODS Based on this high similarity to the technical controls, these samples were considered to have been of such low volume that they only reﬂected the background microbial signal of the lab reagents. Therefore, these ﬁve samples were removed from the remaining analyses. Differences in microbial community compositions between whale populations and between seawater and whale blow were tested on Bray-Curtis dissimilarity (18) using PERMANOVA (v7.0.9; Primer-E, Auckland, New Zealand) with replicate samples treated as a random effect and sequencing depth also tested as a factor using the categories of 50,000 and 50,000 sequences. The phyloseq package (55) in R was used to examine alpha diversity of the humpback blow and seawater MEDs, without repeated subsampling, including richness and Simpson’s index of diversity (19). R also was used to determine the core MEDs (MEDs present in all high-quality samples of humpback whale blow). Taxonomic afﬁliations of the core microbiome and most-homologous sequence were determined using ARB with the most recent nonredundant SILVA database (v.128) at the time of analysis. The ggplot2 package (56) in R was used to construct the taxonomic stacked bar and box plot ﬁgures. Pathogen database and screening. A custom pathogen database of phylogenetic afﬁliations was constructed to screen the humpback whale blow microbiome sequences for the presence of potential pathogens at the genus level. The database included putative pathogens of any marine mammal body site identiﬁed from a compilation made by Raverty and colleagues (16) and from other published studies (see Table S2 in the supplemental material). To account for any bacteria that have not yet been identiﬁed as marine mammal pathogens, the database also included any human and animal bacterial pathogens recognized by the American Biological Safety Association (899 pathogens). To ensure that the humpback blow and seawater MEDs were resolved to the most descriptive taxonomic level offered by the SILVA database, the representative sequences were assigned taxonomic identity using both the k-nearest neighbor algorithm in mothur with the SILVA rRNA sequence database customized for humpback whale skin as mentioned above and the SINA Alignment Service v1.2.11 (52). Data availability. Sequence data from this study are available at NCBI under BioProject accession no. PRJNA401637. Representative MED sequences are available in fasta format in Data Set S1 in the supplemental material. MATERIALS AND METHODS After quantiﬁcation on a Qubit 2.0 ﬂuorometer with a double-stranded DNA (dsDNA) high- sensitivity assay kit (Invitrogen Corp., Carlsbad, CA), the PCR products were pooled into two libraries of equal concentrations. Amplicons were sequenced over two 2- by 250-bp MiSeq (Vancouver Island) and NanoSeq (Cape Cod) formats (Illumina, San Diego, CA) at the University of Illinois W. M. Keck Center for Comparative and Functional Genomics. Sequence data processing. Raw sequences (6,358,878) were assembled, denoised, and quality ﬁltered using mothur v.1.36.1 (51). Speciﬁcally, barcoded primers were removed, sequence reads were joined, sequences were trimmed to 255 bp, and ambiguous base pair calls were removed, resulting in sequences with an average of 253 bp. The sequencing error rate was calculated as 0.0015771% from the mock community samples. Sequences from technical and ﬂight control samples did not meet quality control criteria. However, due to the low-volume nature of the blow samples, these samples were initially included in the analysis (Fig. 2). Sequences were classiﬁed using a k-nearest neighbor consensus algorithm in mothur with the SILVA rRNA sequence database (v.123) (52, 53), and those that were identiﬁed as chloroplasts (233,253 sequences, from the seawater samples) and unknown (355 sequences) were removed from the data set. Chimeric sequences found by UCHIME (54) within mothur also were removed, and the cumulative outcome of these quality control measures resulted in 4,903,825 sequences (22.9% loss of sequences). Minimum entropy decomposition (MED) (17) was used to bin sequences to homogeneous operational taxonomic units (here referred to as MEDs) on the entire data set, 4,903,825 sequences, with a minimum substantive abundance (M) set to 490 to reduce the impact of noise, resulting in 616 nodes and a further reduction of the data to 4,183,042 reads. Taxonomy was assigned to sequences representing each MED using the k-nearest neighbor consensus algorithm in mothur with the nonredundant SILVA rRNA sequence database v.123 that was customized for humpback whale skin for an unrelated study by adding Tenacibaculum and Psychrobacter hypervariable region IV sequences from the partial (shorter-read) version of the same database. Microbial community analysis. Using the Primer software (v7.0.9; Primer-E, Auckland, New Zealand), Bray-Curtis dissimilarity (18) was calculated from nonrareﬁed, square-root-transformed relative abun- dances of the MED nodes and compared using a single linkage clustering algorithm. The resulting dendrogram showed that ﬁve sparse-volume blow samples clustered with the technical control samples (DNA isolation, PCR, and mock community controls). MATERIALS AND METHODS To sample surface seawater microbes, 1 liter surface seawater was collected at 0.25-m depth from the same general area as the blow collections around Vancouver Island and ﬁltered through an 0.22-m Supor membrane ﬁlter (Millipore, Boston, MA) using a peristaltic pump. Each ﬁlter was placed in a sterile 2-ml cryovial, frozen in a liquid nitrogen vapor shipper, and transferred to 80°C until processing. Two replicate seawater samples were collected per sampling site. g Sample preparation, PCR ampliﬁcation, and sequencing. Nucleic acids were isolated from the swabs, 50 l of the pipetted sample, or the ﬁlters using the PowerBioﬁlm DNA isolation kit (Mo Bio Laboratories, Inc., Carlsbad, CA). Barcoded 515FY and 806RB (50, 57) primers, utilized by the Earth Microbiome Project, were used to amplify the V4 region of the SSU rRNA gene in triplicate 25-l PCR mixtures per sample on a Bio-Rad Thermocycler (Hercules, CA) as follows: an initial denaturation step at 95°C for 2 min; 30 to 38 cycles (blow samples) or 20 to 25 cycles (seawater samples) of 95°C for 20 s, 55°C for 15 s, and 72°C for 5 min; and an extension step at 72°C for 10 min. Each PCR mixture contained 1 ng of DNA, 200 nM barcoded primers, GoTaq Flexi DNA polymerase, GoTaq Flexi 5 colorless buffer, 2.5 mM MgCl2, and 200 M deoxynucleoside triphosphate (dNTP) mix (Promega, Madison, WI). Products of the triplicate reactions were combined for each sample, screened on a 1% agarose–Tris-borate-EDTA (TBE) gel using HyperLadder (50 bp; Bioline USA Inc., Taunton, MA) to conﬁrm amplicon size, and puriﬁed either with a 1.5% agarose-TBE gel using the MinElute gel extraction kit (Qiagen, Valencia, CA) or without the gel extraction using a Wizard PCR cleanup system (Promega, Madison, WI). A low-concentration microbial mock community with equimolar rRNA operon counts (obtained through BEI Resources, NIAID, NIH, as part of the Human Microbiome Project; genomic DNA from microbial mock community B [even, low concentration], v5.1L, for 16S rRNA gene sequencing, HM-782D) was ampliﬁed and sequenced with the samples to test for sequencing errors. To test for reagent contamination from the DNA isolation kit, 50 l sterile water or a dry swab was included in each batch of isolations. Each PCR run included sterile water as a negative control, for which ampliﬁcation was not detected via gel electrophoresis for any of msystems.asm.org 12 Drone-Captured Whale Blow Microbiome the runs. September/October 2017 Volume 2 Issue 5 e00119-17 REFERENCES 1. Thomas PO, Reeves RR, Brownell RL. 2016. Status of the world’s baleen whales. Mar Mamm Sci 32:682–734. https://doi.org/10.1111/mms.12281. layer in the eastern north Paciﬁc. Sci Rep 7:394. https://doi.org/10.1038/ s41598-017-00457-5. 17. Eren AM, Morrison HG, Lescault PJ, Reveillaud J, Vineis JH, Sogin ML. 2015. Minimum entropy decomposition: unsupervised oligotyping for sensitive partitioning of high-throughput marker gene sequences. ISME J 9:968–979. https://doi.org/10.1038/ismej.2014.195. 2. Kraus SD, Brown MW, Caswell H, Clark CW, Fujiwara M, Hamilton PK, Kenney RD, Knowlton AR, Landry S, Mayo CA, McLellan WA, Moore MJ, Nowacek DP, Pabst DA, Read AJ, Rolland RM. 2005. North Atlantic right whales in crisis. Science 309:561–562. https://doi.org/10.1126/science .1111200. J 9:968–979. https://doi.org/10.1038/ismej.2014.195. 18. Bray JR, Curtis JT. 1957. An ordination of the upland forest communities of southern Wisconsin. Ecol Monogr 27:325–349. https://doi.org/10 .2307/1942268. 3. Clapham PJ, Young SB, Brownell RL. 1999. Baleen whales: conservation issues and the status of the most endangered populations. Mamm Rev 29:37–62. https://doi.org/10.1046/j.1365-2907.1999.00035.x. 19. Simpson EH. 1949. Measurement of diversity. Nature 163:688. https:// doi.org/10.1038/163688a0. 4. Fleming AH, Clark CT, Calambokidis J, Barlow J. 2016. Humpback whale diets respond to variance in ocean climate and ecosystem conditions in the California Current. Glob Chang Biol 22:1214–1224. https://doi.org/ 10.1111/gcb.13171. 20. Salter SJ, Cox MJ, Turek EM, Calus ST, Cookson WO, Moffatt MF, Turner P, Parkhill J, Loman NJ, Walker AW. 2014. Reagent and laboratory contamination can critically impact sequence-based microbiome analy- ses. BMC Biol 12:87. https://doi.org/10.1186/s12915-014-0087-z. 5. Knowlton AR, Hamilton PK, Marx MK, Pettis HM, Kraus SD. 2012. Moni- toring North Atlantic right whale Eubalaena glacialis entanglement rates: a 30 yr retrospective. Mar Ecol Prog Ser 466:293–302. https://doi.org/10 .3354/meps09923. 21. Venn-Watson S, Smith CR, Jensen ED. 2008. Primary bacterial pathogens in bottlenose dolphins Tursiops truncatus: needles in haystacks of com- mensal and environmental microbes. Dis Aquat Organ 79:87–93. https:// doi.org/10.3354/dao01895. 6. Sweeney JC, Ridgway SH. 1975. Common diseases of small cetaceans. J Am Vet Med Assoc 167:533–540. 22. Foster G, Ross HM, Malnick H, Willems A, Garcia P, Reid RJ, Collins MD. 1996. Actinobacillus delphinicola sp. nov., a new member of the family Pasteurellaceae Pohl (1979) 1981 isolated from sea mammals. Int J Syst Bacteriol 46:648–652. https://doi.org/10.1099/00207713-46-3-648. 7. Venn-Watson S, Daniels R, Smith C. 2012. Thirty year retrospective evaluation of pneumonia in a bottlenose dolphin Tursiops truncatus population. Dis Aquat Organ 99:237–242. https://doi.org/10.3354/ dao02471. eriol 46:648–652. https://doi.org/10.1099/00207713-46-3-648. 23. Varaldo PE, Kilpper-Balz R, Biavasco F, Satta G, Schleifer KH. 1988. Staphylococcus delphini sp. REFERENCES nov., a coagulase-positive species isolated from dolphins. Int J Syst Bacteriol 38:436–439. https://doi.org/10.1099/ 00207713-38-4-436. 8. Fernández A, Esperón F, Herraéz P, de Los Monteros AE, Clavel C, Bernabé A, Sánchez-Vizcaino JM, Verborgh P, DeStephanis R, Toledano F, Bayón A. 2008. Morbillivirus and pilot whale deaths, Mediterranean Sea. Emerg Infect Dis 14:792–794. https://doi.org/10.3201/eid1405.070948. 24. Bernardelli A, Bastida R, Loureiro J, Michelis H, Romano MI, Cataldi A, Costa E. 1996. Tuberculosis in sea lions and fur seals from the south- western Atlantic coast. Rev Sci Tech 15:985–1005. https://doi.org/10 .20506/rst.15.3.963. 9. Chambers ST, Scott-Thomas A, Epton M. 2012. Developments in novel breath tests for bacterial and fungal pulmonary infection. Curr Opin Pulm Med 18:228–232. https://doi.org/10.1097/MCP.0b013e328351f98b. 25. Bogomolni AL, Gast RJ, Ellis JC, Dennett M, Pugliares KR, Lentell BJ, Moore MJ. 2008. Victims or vectors: a survey of marine vertebrate zoonoses from coastal waters of the northwest Atlantic. Dis Aquat Organ 81:13–38. https://doi.org/10.3354/dao01936. 10. Acevedo-Whitehouse K, Rocha-Gosselin A, Gendron D. 2010. A novel non-invasive tool for disease surveillance of free-ranging whales and its relevance to conservation programs. Anim Conserv 13:217–225. https:// doi.org/10.1111/j.1469-1795.2009.00326.x. 26. Higgins R. 2000. Bacteria and fungi of marine mammals: a review. Can Vet J 41:105–116. 11. Hogg CJ, Rogers TL, Shorter A, Barton K, Miller PJO, Nowacek D. 2009. Determination of steroid hormones in whale blow: it is possible. Mar Mamm Sci 25:605–618. https://doi.org/10.1111/j.1748-7692.2008.00277.x. 27. Cusick PK, Bullock BC. 1973. Ulcerative dermatitis and pneumonia asso- ciated with Aeromonas hydrophila infection in the bottle-nosed dolphin. J Am Vet Med Assoc 163:578–579. 12. Hunt KE, Rolland RM, Kraus SD. 2014. Detection of steroid and thyroid hormones via immunoassay of North Atlantic right whale (Eubalaena glacialis) respiratory vapor. Mar Mamm Sci 30:796–809. https://doi.org/ 10.1111/mms.12073. 28. Costello EK, Lauber CL, Hamady M, Fierer N, Gordon JI, Knight R. 2009. Bacterial community variation in human body habitats across space and time. Science 326:1694–1697. https://doi.org/10.1126/science.1177486. 13. Johnson WR, Torralba M, Fair PA, Bossart GD, Nelson KE, Morris PJ. 2009. Novel diversity of bacterial communities associated with bottlenose dolphin upper respiratory tracts. Environ Microbiol Rep 1:555–562. https://doi.org/10.1111/j.1758-2229.2009.00080.x. 29. The Human Microbiome Project Consortium. 2012. Structure, function and diversity of the healthy human microbiome. Nature 486:207–214. https://doi.org/10.1038/nature11234. 30. Morris RM, Rappé MS, Connon SA, Vergin KL, Siebold WA, Carlson CA, Giovannoni SJ. 2002. SAR11 clade dominates ocean surface bacterio- plankton communities. Nature 420:806–810. https://doi.org/10.1038/ nature01240. 14. ACKNOWLEDGMENTS We thank Hannah Moore, William Durban, and Carla Crossman for ﬁeld assistance and Chris Wright and the University of Illinois W. M. Keck Center for Comparative and Functional Genomics for sequencing. Wayne Perryman provided advice on unmanned aircraft operations, and Don LeRoi constructed attachments for collection surfaces on msystems.asm.org 13 September/October 2017 Volume 2 Issue 5 e00119-17 Apprill et al. the hexacopter. A research vessel for ﬁeld collection of samples around Vancouver Island was provided by the Vancouver Aquarium Marine Science Centre. Research approaches by both the boat and the hexacopter were authorized by NMFS permits 19091 and 17355 in U.S. waters and a research license issued by Fisheries and Oceans Canada (2014-5 SARA-327) in Canadian waters. Funding for sample analysis was provided through a grant to A.A., M.J.M., and J.W.D. from the Ocean Life Institute of the Woods Hole Oceanographic Institution. Attach- ments for collection surfaces on the hexacopter were constructed with funding support from NOAA’s UAS Program. September/October 2017 Volume 2 Issue 5 e00119-17 Drone-Captured Whale Blow Microbiome 34. Apprill A, Robbins J, Eren AM, Pack AA, Reveillaud J, Mattila D, Moore M, Niemeyer M, Moore KMT, Mincer TJ. 2014. Humpback whale populations share a core skin bacterial community: towards a health index for marine mammals? PLoS One 9:e90785. https://doi.org/10.1371/journal.pone .0090785. 1984. Behavior of bowhead whales, Balaena mysticetus, summering in the Beaufort Sea: surfacing, respiration, and dive characteristics. Can J Zool 62:1910–1921. https://doi.org/10.1139/z84-281. 1984. Behavior of bowhead whales, Balaena mysticetus, summering in the Beaufort Sea: surfacing, respiration, and dive characteristics. Can J Zool 62:1910–1921. https://doi.org/10.1139/z84-281. 46. Kim D, Hofstaedter CE, Zhao C, Mattei L, Tanes C, Clarke E, Lauder A, Sherrill-Mix S, Chehoud C, Kelsen J, Conrad M, Collman RG, Baldassano R, Bushman FD, Bittinger K. 2017. Optimizing methods and dodging pitfalls in microbiome research. Microbiome 5:52. https://doi.org/10 .1186/s40168-017-0267-5. 35. Hernandez-Agreda A, Gates RD, Ainsworth TD. 2017. Deﬁning the core microbiome in corals. Trends Microbiol 25:125–140. https://doi.org/10 .1016/j.tim.2016.11.003. 47. Nollens HH, Wellehan JF, Saliki JT, Caseltine SL, Jensen ED, Van Bonn W, Venn-Watson S. 2008. Characterization of a parainﬂuenza virus isolated from a bottlenose dolphin (Tursiops truncatus). Vet Microbiol 128: 231–242. https://doi.org/10.1016/j.vetmic.2007.10.005. 36. Astudillo-García C, Bell JJ, Webster NS, Glasl B, Jompa J, Montoya JM, Taylor MW. 2017. Evaluating the core microbiota in complex communities: a systematic investigation. Environ Microbiol 19:1450–1462. https://doi.org/ 10.1111/1462-2920.13647. 37. Qin J, Li R, Raes J, Arumugam M, Burgdorf KS, Manichanh C, Nielsen T, Pons N, Levenez F, Yamada T, Mende DR, Li J, Xu J, Li S, Li D, Cao J, Wang B, Liang H, Zheng H, Xie Y, Tap J, Lepage P, Bertalan M, Batto JM, Hansen T, Le Paslier D, Linneberg A, Nielsen HB, Pelletier E, Renault P, Sicheritz- Ponten T, Turner K, Zhu H, Yu C, Li S, Jian M, Zhou Y, Li Y, Zhang X, Li S. 2010. A human gut microbial gene catalogue established by metag- enomic sequencing. Nature 464:59 – 65. https://doi.org/10.1038/ nature08821. 48. Durban JW, Fearnbach H, Barrett-Lennard LG, Perryman WL, LeRoi DJ. 2015. Photogrammetry of killer whales using a small hexacopter launched at sea. J Unmanned Veh Syst 3:1–21. https://doi.org/10.1139/ juvs-2014-0011. 49. Mizroch S. 2007. Whale photo-identiﬁcation: digital photography tech- niques, using EXIF data, and photo editing tools. NOAA, Washington, DC. https://www.afsc.noaa.gov/nmml/pdf/NMMLDigitalPhotoProtocol.pdf. 50. Apprill A, McNally SP, Parsons R, Weber L. 2015. Minor revision to V4 region SSU rRNA 806R gene primer greatly increases detection of SAR11 bacterioplankton. Aquat Microb Ecol 75:129–137. https://doi.org/10 .3354/ame01753. 38. Fierer N, Hamady M, Lauber CL, Knight R. 2008. The inﬂuence of sex, handedness, and washing on the diversity of hand surface bacteria. Proc Natl Acad Sci U S A 105:17994–17999. https://doi.org/10.1073/pnas .0807920105. 51. Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH, Robinson CJ, Sahl JW, Stres B, Thallinger GG, Van Horn DJ, Weber CF. 2009. Introducing mothur: open- source, platform-independent, community-supported software for de- scribing and comparing microbial communities. Appl Environ Microbiol 75:7537–7541. https://doi.org/10.1128/AEM.01541-09. 39. Charlson ES, Bittinger K, Haas AR, Fitzgerald AS, Frank I, Yadav A, Bushman FD, Collman RG. 2011. Topographical continuity of bacterial populations in the healthy human respiratory tract. Am J Respir Crit Care Med 184:957–963. https://doi.org/10.1164/rccm.201104-0655OC. 40. Rolland RM, Schick RS, Pettis HM, Knowlton AR, Hamilton PK, Clark JS, Kraus SD. 2016. Health of North Atlantic right whales Eubalaena glacialis over three decades: from individual health to demographic and popu- lation health trends. Mar Ecol Prog Ser 542:265–282. https://doi.org/10 .3354/meps11547. 52. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glöckner FO. 2013. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 41: D590–D596. https://doi.org/10.1093/nar/gks1219. 53. Pruesse E, Quast C, Knittel K, Fuchs BM, Ludwig W, Peplies J, Glöckner FO. 2007. SILVA: a comprehensive online resource for quality checked and aligned ribosomal RNA sequence data compatible with ARB. Nucleic Acids Res 35:7188–7196. https://doi.org/10.1093/nar/gkm864. 41. Durban JW, Moore MJ, Chiang G, Hickmott LS, Bocconcelli A, Howes G, Bahamonde PA, Perryman WL, LeRoi DJ. 2016. Photogrammetry of blue whales with an unmanned hexacopter. Mar Mamm Sci 32:1510–1515. https://doi.org/10.1111/mms.12328. g g 54. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. 2011. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27:2194–2200. https://doi.org/10.1093/bioinformatics/btr381. 42. Elfes CT, VanBlaricom GR, Boyd D, Calambokidis J, Clapham PJ, Pearce RW, Robbins J, Salinas JC, Straley JM, Wade PR, Krahn MM. 2010. Geo- graphic variation of persistent organic pollutant levels in humpback whale (Megaptera novaeangliae) feeding areas of the North Paciﬁc and North Atlantic. Environ Toxicol Chem 29:824–834. https://doi.org/10 .1002/etc.110. 55. McMurdie PJ, Holmes S. 2013. phyloseq: an R package for reproducible interactive analysis and graphics of microbiome census data. PLoS One 8:e61217. https://doi.org/10.1371/journal.pone.0061217. 43. Kraus SD, Rolland R. 2007. The urban whale: North Atlantic right whales at the crossroads. REFERENCES Bik EM, Costello EK, Switzer AD, Callahan BJ, Holmes SP, Wells RS, Carlin KP, Jensen ED, Venn-Watson S, Relman DA. 2016. Marine mammals harbor unique microbiotas shaped by and yet distinct from the sea. Nat Commun 7:10516. https://doi.org/10.1038/ncomms10516. 31. Dickson RP, Huffnagle GB. 2015. The lung microbiome: new principles for respiratory bacteriology in health and disease. PLoS Pathog 11: e1004923. https://doi.org/10.1371/journal.ppat.1004923. 15. Lima N, Rogers T, Acevedo-Whitehouse K, Brown MV. 2012. Temporal stability and species speciﬁcity in bacteria associated with the bottle- nose dolphins respiratory system. Environ Microbiol Rep 4:89–96. https://doi.org/10.1111/j.1758-2229.2011.00306.x. 32. Dickson RP, Erb-Downward JR, Huffnagle GB. 2013. The role of the bacterial microbiome in lung disease. Expert Rev Respir Med 7:245–257. https://doi.org/10.1586/ers.13.24. 16. Raverty SA, Rhodes LD, Zabek E, Eshghi A, Cameron CE, Hanson MB, Schroeder JP. 2017. Respiratory microbiome of endangered southern resident killer whales and microbiota of surrounding sea surface micro- p g 33. Beck JM, Young VB, Huffnagle GB. 2012. The microbiome of the lung. Transl Res 160:258–266. https://doi.org/10.1016/j.trsl.2012.02.005. September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 14 msystems.asm.org 14 September/October 2017 Volume 2 Issue 5 e00119-17 Harvard University Press, Cambridge, MA. 56. Wickham H. 2009. ggplot2: elegant graphics for data analysis. Springer Science and Business Media, New York, NY. 57. Parada A, Needham DM, Fuhrman JA. 2015. Every base matters: assess- ing small subunit rRNA primers for marine microbiomes with mock communities, time-series and global ﬁeld samples. Environ Microbiol 18:1403–1414. https://doi.org/10.1111/1462-2920.13023. 44. Hain JW, Carter R, Kraus D, Mayo A, Winni E. 1982. Feeding behavior of the humpback whale, Megaptera novaeangliae, in the western North Atlantic. Fish Bull 80:259–268. 45. Würsig B, Dorsey EM, Fraker MA, Payne RS, Richardson WJ, Wells RS. September/October 2017 Volume 2 Issue 5 e00119-17 msystems.asm.org 15
https://openalex.org/W2737711039	https://revistasinvestigacion.unmsm.edu.pe/index.php/Arqueo/article/download/12198/10907, http://repositorio.usil.edu.pe/bitstreams/d3789975-3ed3-4bb2-a72f-a87f5c0c974b/download	es		EL PERIODO FORMATIVO EN LOS ANDES SEPTENTRIONALES Y SUS RELACIONES CON LOS ANDES CENTRALES	Arqueología y sociedad	2,014	cc-by	11,044	Arqueología y Sociedad Nº 27, 2014: 111-130 ISSN: 0254-8062 Recibido: enero de 2014 Aceptado: mayo de 2014 El pEriodo formativo En los andEs sEptEntrionalEs y sus rElacionEs con los andEs cEntralEs Christian Mesía Montenegro Ministerio de Cultura christian.mesia@gmail.com RESUMEN El presente artículo realiza una descripción analítica del periodo Formativo ecuatoriano (4500-300 ANE) teniendo como base comparativa los procesos descritos y conocidos para el Formativo peruano (1800-300 ANE), estableciendo los mecanismos a partir de los cuales se dieron procesos de transmisión cultural así como los productos derivados de estos procesos, los cuales reflejan en el registro arqueológico (fundamentalmente cerámico) la intensidad de los contactos entre ambas regiones, fundamentalmente durante el apogeo de la sociedad Chorrera en el norte y Chavín- Cupisnique en el sur. PALABRAS CLAVE: Ecuador, Perú, Formativo, Cupisnique, Chavín, Valdivia, Chorrera, Andes. ABSTRACT This paper reviews the Ecuadorian Formative (4500-300 ANE) in relationship with its counterpart in the Central Andes (1800-300 ANE), trying to establish the mechanisms that operated processes of cultural transmission and their outcomes, which reflects in the archaeological record (basically ceramics) the level of interaction between these two areas, especially during the height of Chorrera, Cupisnique and Chavín societies. KEYWORDS: Ecuador, Perú, Formative, Chavín, Valdivia, Chorrera, Andes. Los procesos sociales ocurridos en el área central andina, ejercieron influencia y la vez se vieron influenciados por los desarrollos nucleados que se dieron en varios sectores de los Andes Septentrionales (Lumbreras 1981). Tradicionalmente los Andes septentrionales han sido observados como áreas intermedias en las cuales los procesos de complejización o de innovación cultural se dieron a partir de influencias exógenas provenientes de las áreas nucleares. En el caso del territorio que actualmente ocupa el Ecuador, se puede observar ciertamente que la innovación cultural se da a partir de contactos con los Andes centrales y con Mesoamérica, pero al mismo tiempo, existe procesos de independientes, por lo que podemos hipotetizar que el desarrollo cultural se dio a partir de innovación cultural y transmisión cultural. El estudio del período Formativo (4500-300 ANE) del Ecuador, nos puede ayudar a disgregar aquellos elementos que se dieron partir de transmisión cultural (proveniente de áreas nucleares) y aquellos que se dieron a través de procesos independientes de innovación cultural. 111 Arqueología y Sociedad 27, 2014: 111-130 En ese sentido, la intención del presente artículo es no solamente la de brindar una visión general del período Formativo en territorio ecuatoriano, con énfasis en los desarrollos costeños1, sino además de identificar aquellos elementos compartidos con formaciones contemporáneas en los Andes centrales, a fin de dar un espectro regional a la naturaleza del registro arqueológico en ambas áreas culturales. Es importante precisar la necesidad de realizar estudios comparativos entre estas dos regiones debido a —como se argumentará en el presente trabajo— las similitudes existentes entre los materiales e iconografías de ambas regiones las cuales se manifestaron a partir de un intenso tráfico de ideas y/o poblaciones. Cabe decir que estas relaciones no son exclusivas al período formativo, sino más bien continuaron después del mismo como ya lo han indicado otros autores (Guffroy 2004; Hocquenghem 1991; Izumi y Terada 1966). I. TIEMPO Y ESPACIO Es preciso delimitar los límites espaciales y temporales, para lo cual como ya se ha mencionado, incidiremos no solamente en los procesos sociales formativos de la costa de los Andes septentrionales, sino también en aquellas regiones en las cuales se pueda observar relaciones y/o influencia de los Andes centrales. En términos temporales, existe un consenso entre diversos investigadores, en definir los límites cronológicos del Formativo ecuatoriano a partir del limite terminus a quo de Valdivia y terminus ad quem de Chorrera, (Lippi 2003; Zeidler 2003) por lo que para propósitos del presente artículo el Formativo ecuatoriano se enmarca entre los 4500–300 ANE (Tabla 1). Durante este espacio temporal, tres desarrollos sociales caracterizaron fundamentalmente la costa y parte de la sierra ecuatoriana, Valdivia, Machalilla y Chorrera. Indudablemente que estos desarrollos no fueron los únicos en este gran lapso de tiempo, por lo que se harán importantes menciones a otras regiones del Ecuador. II. PRIMEROS AGRICULTORES Y CERAMISTAS La sociedad Valdivia ha merecido la atención de diversos investigadores debido a lo temprano de su datación y a las características de este desarrollo social. El antecedente inmediato a Valdivia es Las Vegas, caracterizada por la ausencia de cerámica pero por la presencia de los primeros cultígenos domesticados en la región (Stothert 1985). Valdivia, en términos generales se caracteriza por tener fechados muy tempranos de pequeños asentamientos aldeanos asociados a evidencias de fitolitos de maíz, lo que sugiere el consumo de maíz desde épocas muy tempranas (Bonavía 2008; Pearsall 2003; Van der Merwe, Lee-Thorp, y Raymond 1993; Zeidler 2008). A pesar de que existen controversias que merecen ser aclaradas (Bonavía 2008; Van der Merwe, Lee-Thorp y Raymond 1993)2, para propósito del presente artículo, asumiremos que la sociedad Valdivia fue responsable del cultivo de maíz desde los 4500 ANE; lo que no queda del todo claro es el tipo de maíz cultivado o si es producto de un proceso de innovación (Zevallos et al. 1977) o de transmisión cultural (Bonavía 2008). Las evidencias a la mano indican que el maíz procedente de San Pablo podría tratarse de la especie kcello Ecuatoriano el cual de acuerdo a Lathrap podría ser una especie derivada del Nat-Tel mexicano (Lathrap, Collier y Chandra 112 1 Se incide en la costa ecuatoriana ya que es una zona extensamente estudiada y que guarda mayor relación con los Andes centrales. 2 Algunos restos de fitolitos ubicados en contextos Valdivia tempranos han sido datados en 7000-5500 ANE (Bonavía 2008; Van der Merwe, N., J. Lee-Thorp, y S. Raymond 1993). Zeidler y Pearsall afirman que estos restos en realidad deben de tener una antigüedad que va desde los 4500 ANE (consistente con lo que se estima para la Fase Valdivia I). Bonavía es escéptico con respecto a la presencia de maíz durante Valdivia, argumenta que la falta de evidencia de caries en restos óseos de la época (comparativamente con Machalilla en donde si hay abundancia de esta patología) así como una errónea identificación de especies (Bonavía 2008) mientras que los análisis de isótopos realizados por Van der Merwe y Tschauner en restos óseos de Valdivia Temprano indican una dieta basada en recursos de floresta y de río y no precisamente basada en el maíz. (Van der Merwe y Tschauner, 1999). Las Vegas Valdivia Temprano Arcaico Machalilla Chorrera Chorrera/ Tabuchilla Chorrera/ Tabuchilla Medio Tardío Chorrera Chorrera/ Engoroy Chorrera/ Engoroy Centro Cotocollao Alausi Chaullabamba La Cerro Chimba Narrío I Sierra Norte Catamayo B Catamayo A Catamayo C Catamayo D Sur PreUpano Pastaza B Pastaza A Upano I Mayo Chinchipe Mayo Chinchipe Oriente Centro Sur Paiján Huaca Prieta PreCupisnique Cupisnique La Chira La Florida PreChavinoide Chavinoidec Cerro Cupisnsique Santa Ana Cupisnique Transitorio Huayco Salinar Andes Centrales Costa Norte Centro Bahía de Paracas Disco Verde Puerto Nuevo Karwa Paracas Cavernas Paracas Necrópolis Sur Huarás Conchucos Pandanche Ídolo Kotosh Chavín Sajarapatq Centro Cívico Mito Tutishcaynio Temprano Tutishcaynio Tardío Shakimu Oriente HuaYarillaga nacocha Higueras WairaLa Banda jirca Urabarroide Kuntur- JanaWasi barroide Sierra Sierra Norte a Elaborado por el autor. b La periodificación está en referencia a la clasificación cronológica del formativo ecuatoriano. c Los términos pre- chavinoide y chavinoide siguen la clasificación de Hermilio Rosas para la cerámica de influencia Chavín recuperada en Ancón. 10000 4500 1300 1500 300 Periodob Andes Septentrionales Costa Norte Centro Sur TABLA I CUADRO CRONOLÓGICO DEL FORMATIVO EN LOS ANDES SEPTENTRIONALES Y CENTRALES Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales 113 Arqueología y Sociedad 27, 2014: 111-130 1975), ante lo cual el maíz en el Ecuador aparecería como producto de un fenómeno de transmisión cultural, posición similar a la de Deborah Pearsall quien afirma en base a análisis de ADN que el maíz en los Andes proviene del Zea mays subsp. Parviglumis (conocido también como Balsas Teosinte) el cual es endémico en el sur de Mexico (Pearsall 2008). Otro indicador de la presencia del maíz es la decoración de vasijas Valdivia 3 (2800 ANE) con aplicados que han sido interpretados como representaciones de granos de maíz (Lathrap, Collier y Chandra 1975; Pearsall 2008). De cualquier modo la agricultura temprana no aparece exclusivamente relacionada al maíz, sino también a otros productos como curcubitas, lagenarias, maní y achira (Pearsall 2003). Paralelamente a este desarrollo agrícola, se da la producción de cerámica la cual al parecer ingresa al Ecuador a través de un proceso de transmisión cultural procedente de Colombia, específicamente del complejo San Jacinto el cual ha sido fechado en aproximadamente 6000 ANE, es decir 1500 años antes de la presencia de cerámica en el actual territorio ecuatoriano (Meggers 2005; Oyuela-Caycedo y Bonzani 2005; Pratt 1999). La cerámica Valdivia en sus inicios se caracteriza por presentar tres formas elementales: vasijas con cuellos altos, bordes cóncavos y cuerpos esféricos, las cuales en su mayoría presentan el cuello decorado con líneas incisas transversales cruzadas (Evans y Meggers 1957; Meggers, Evans y Estrada 1965), desarrollándose vasijas de bordes evertidos y cuerpo globulares con líneas en zigzag horizontales incisas en estado de cuero blando bajo el borde; borde y cuerpos delimitados por incisiones ovoidales verticales. Aparecen durante la fase 3 cuencos con bordes almenados y representaciones excisas de rostros al parecer humanos sobre la superficie de los mismos (Fig. 1). Aunque también existen cuencos altos, de cuerpos globulares y bordes redondeados, con decoración incisa en cuero duro, representando sobre la mitad superior rostros flanqueados por lo que parecen ser las extremidades superiores del individuo, es necesario indicar que a diferencia de los rostros de cuencos cortos, la boca del individuo presenta un achurado en forma de tablero de ajedrez (Fig. 2.) Figura 1. Cuenco Valdivia, de borde almenado, con decoración excisa sobre el cuerpo del mismo representando rostros al parecer humanos dispuestos a modo de paneles, separados por una línea vertical excisa. Presenta pequeños soportes sobre la base del mismo. Museo Casa del Alabado. 114 Figura 2. Cuenco Valdivia, alto globular de borde circular, superificie con engobe rojo con diseños incisos en estado de cuero duro sobre la mitad superior. Se observan rostros flanqueados por lo que parecen ser las extremidades superiores del individuo. La parte inferior de la vasija presenta bandas achuradas verticales que convergen hacia el centro de la base. Museo de Arte Precolombino Casa del Alabado. Alrededor de los 2800 ANE, se empiezan a producir figulinas femeninas de cerámica, conocidas como «Venus» (Fig. 3), las cuales en Real Alto se han hallado en contextos primarios en el montículo del «Osario» así como en los pisos de viviendas a lo largo de la secuencia (Marcos 2005). De acuerdo a Marcos «de esta manera se pudieron no sólo asociar las figurinas a ritos funerarios y de curación, sino que se pudo determinar que las curaciones que involucraban figurinas se dieron al interior de las viviendas» (Marcos 2005: 105). Estas figulinas generalmente representan una mujer con un elaborado tocado, en posición vertical, ya sea en estado de gravidez, tocando un instrumento musical de viento o con un infante en brazos. Algunas otras interpretaciones podrían apuntar a una serie de amuletos relacionados con la Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales Figura 3. Figulina Valdivia de un personaje femenino con tocado elaborado, al parecer tocando un instrumento de viento. Museo de Arte Precolombino Casa del Alabado. fertilidad humana. Siguiendo en esta línea de pensamiento se podría afirmar que la importancia de estos objetos están relacionados con la necesidad de contar con el suficiente recurso humano en una época en la cual las actividades productivas requerían de un nivel demográfico adecuado para llevar a cabo estas tareas que iban más allá de la caza o extracción marina. Tradicionalmente se ha clasificado a Valdivia como una sociedad eminentemente marina con una dependencia marginal hacia la horticultura, debido fundamentalmente a que los sitios que sirvieron para la identificación de esta sociedad se encuentran en la línea costera de las provincias de Manabi y Guayas. Investigaciones llevadas a cabo en sitios de tierra adentro como Loma Alta, Colimes y San Lorenzo del Mate, han ayudado a revaluar el rol de la economía marina en Valdivia y a retomar la posición de Lathrap quien argumentó que Valdivia fue principalmente un desarrollo de la floresta tropical con un patrón de asentamiento basado en el poblamiento de zonas de valle medio cuyo origen se debe a los patrones de dispersión proveniente del poblamiento amazónico (Lathrap 1970), es decir un poblamiento que desde la selva se extiende hacia la costa y no viceversa. Parte de este planteamiento se basa en la distribución de vasijas carenadas tempranas con tendencia a presentar las áreas carenadas y ángulos basales decorados (Lathrap 1963).3 Siguiendo lo planteado por Lathrap, los sitios costeros vendrían a ser asentamientos especializados en la explotación de recursos marinos, por lo que es de esperar una progresiva adaptación a diversos nichos ecológicos, de este modo el medio ambiente estaría ejerciendo una presión selectiva en las diversas manifestaciones culturales dadas en ambientes particulares. De acuerdo a diversas investigaciones en Manabi y Guayas, se observa una progresiva adaptación al medio así como una paulatina jerarquización social «es muy claro que Valdivia representa una sociedad dinámica con un progresivo incremento demográfico y de jerarquización social y desigualdad en estatus a través del tiempo» (Zeidler 2008: 464). Esta aseveración está fundamentada en sitios como Real Alto, en donde alrededor de los 4500 ANE se observa una modesta aldea con disposición en forma de U, mientras que alrededor de los 1400 ANE se tiene un complejo arquitectónico monumental con una sectorización ceremonial y doméstica claramente definida (y al interior del área domestica, se observan elementos distintivos de estatus) (Marcos 1988); por otra parte sitios como San Isidro en el valle de Jama y La Emerenciana en el Oro son ejemplos de arquitectura monumental no antes vista en el desarrollo Valdivia (Zeidler 1994b, 2008; Zeidler e Isaacson 2003). La concentración de poder en las villas nucleadas de Valdivia sirvió como germen en el desarrollo de desigualdades a través del tiempo, a partir de la transmisión hereditaria de privilegios y jerarquías lo cual eventualmente culminó en la construcción de centros ceremoniales como los antes mencionadas, asociados a asentamientos domésticos de élite. En un lapso aproximado de 2600 años, se da una completa transformación social en la costa del Ecuador (provincias del Oro, Los Ríos, Guayas y Manabí) originando una progresiva y completa modificación de las relaciones de poder en la región. 3 Vasijas de este tipo se encuentran en Cotocollao y La Chimba en la sierra norte ecuatoriana. 115 Arqueología y Sociedad 27, 2014: 111-130 III. DESPUÉS DE LOS VOLCANES Los procesos evolutivos se dan a partir de progresiones o modificaciones puntuadas. Estas últimas en muchos casos se deben a eventos dramáticos que obligan a los organismos sociales a buscar salidas innovativas a fin de poder reproducirse en el tiempo. Podría decirse que en una estructura social jerarquizada, las élites son en realidad quienes están en la obligación de buscar aquellos mecanismos que les permitan seguir manteniendo posiciones privilegiadas. En el caso de las élites Valdivia, al parecer no tuvieron suerte en encontrar alternativas que les permita mantener la estructura configurada en 2600 años, frente a eventos catastróficos medioambientales. Alrededor de los 1900 ANE, se dan erupciones volcánicas al norte de los Andes septentrionales, no habiéndose identificado aún el lugar preciso de estos fenómenos (Zeidler e Isaacson 2003). Estos eventos fueron de gran intensidad y perturbaron los modos de vida de las poblaciones de centro y occidente del Ecuador. La extensión de dichas erupciones alcanzó a las provincias de Esmeraldas, El Oro, Guayas, Los Ríos e incluso la provincia de Pichincha en la cual se encuentra la ciudad de Quito (Issacson 1994; Zeidler e Isaacson 2003). Estas erupciones alteraron los modos de vida de las poblaciones ahora identificadas como parte de la tradición cultural Valdivia, debilitando no solamente elementos relacionados con el sistema económico y actividades productivas sino causando una gran inestabilidad política que derivó en el surgimiento de una nueva tradición cultural, conocida por los arqueólogos como Machalilla (Estrada 1958; Meggers, Evans y Estrada 1965). Esta tradición se extendió por un área geográfica menor a la de Valdivia, presente de modo continuo entre las secciones sur de la provincia de Manabi y Guayas, mostrando una discontinuidad entre las secciones norte de Manabi y sur de Esmeraldas (Zeidler 2008). La cerámica Machalilla se origina a partir de las fases tardía y terminal (fases 7-8) de Valdivia (Figs. 9-12) , cuya transición se puede observar en los sitios de San Lorenzo de Mate, al sur de la provincia del Guayas (Cruz y Holm 1982; Marcos 1989) y La Emerenciana en la provincia del Oro (Staller 2001a). Aparecen del mismo modo botellas con asa estribo (Lathrap, Collier y Chandra 1975) las cuales al parecer llegaron a la costa proveniente de la cuenca del río Mayo, a pocos kilómetros de la Cordillera del Cóndor (Valdez 2008) siendo características de los desarrollos Cupisnique y Chavín (Larco 1948; Lumbreras 1993; Tello 1960). Por otra parte la cerámica Machalilla tiene una fuerte influencia en la cerámica temprana del norte peruano, a juzgar los diseños de la cerámica Ídolo, Pandanche y Huaca Loma Temprano (Burger 1992; Kaulicke 1975; Kaulicke 1994; Morales 1993; Onuki 1995). Esta influencia podría remontarse desde la época Valdivia a juzgar por las formas de las vasijas de doble cuerpo de la fase Tutishcainyo Temprano, cuyas formas son muy similares a las vasijas Valdivia 7, así como a los aplicados circulares y en bandas. Similarmente Lathrap ha sugerido que Machalilla ejerce una fuerte influencia en la fase Tutishcainyo Tardio (Lathrap 1963). Adicionalmente, la incidencia de vasijas carenadas y achurados zonales es característica de la fase Kotosh-Wairajirca la cual está muy relacionada con Tutishcainyo (Figs. 4-6).4 La cerámica Machalilla se caracteriza por presentar cuencos carenados, vasijas reducidas con un fuerte bruñido, con finas líneas incisas pintadas post cocción, generalmente de color blanco, práctica que es igualmente común en la cerámica inicial del Formativo peruano. Otras características incluyen el uso de engobe blanco y líneas rojas pintadas y/o incisas, platos con pedestal, así como el achurado y puntueado zonal los cuales aparecen también en la fase Tutishcainyo Temprano del Yarinacocha (Estrada 1958; Lathrap 1963, 1970). Por otra parte, se continua la producción de figulinas pero de morfologia marcadamente diferente a las llamadas «Venus» de Valdivia. Las figulinas Machalilla son de manufactura cruda y estilizadas, con ojos en forma de granos de café y narices protuberantes, y mayormente con una línea de perforaciones bajo el labio inferior, a lo largo de la cabeza y a lo largo de los lóbulos auditivos las cuales de acuerdo a Lathrap pudo servir para insertar algún tipo de adorno (Lathrap, Collier y Chandra 1975). Sobre el significado de estas figulinas, se podría extender el uso antes descrito para las figulinas Valdivia. 116 4 Nótese la similitud de diseño entre las figuras 5 y 6. Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales Figura 4. Botella de doble cuerpo. En el cuerpo superior se observa una línea incisa bajo el borde de la vasija, así como círculos incisos dispuestos en pares, de forma vertical sobre la superficie del mismo. Museo de Arte Precolombino Casa del Alabado. Figura 5. Cuenco Valdivia de cuerpo semiesférico, presenta una banda con incisión achurada diagonal por debajo del labio. En la zona central del cuerpo se observan diseños rectangulares rellenos con incisiones achuradas verticales. Museo de Arte Precolombino Casa del Alabado. Figura 6. Cuenco Kotosh-Wairajirca, de paredes convergentes. Decorado con motivos de rombos y trapecios achurados en forma de tablero de ajedrez. Museo Nacional de Arqueología, Antropología e Historia del Perú. En las zonas de influencia Machalilla, destaca la ausencia de centros cívicos ceremoniales o montículos (tal como se observa en las postrimerías del desarrollo Valdivia), en su lugar se observan aldeas a lo largo de zonas ribereñas de menos de 0,5 ha (Zeidler 1994b; Zeidler e Isaacson 2003) o extensas villas nucleadas (Schwarz y Raymond 1996). Este último tipo de sitios es más frecuente en el litoral, sobre colinas frente al mar, en donde se observa una economía altamente especializada (Zeidler e Isaacson 2003). La continuidad estilística ceremográfica no es la única, sino al parecer Machilla prosiguió con una economía diversificada característica de Valdivia, donde no solamente se dependía fuertemente de los recursos marinos, sino también de la agricultura (fuertemente orientada hacia el cultivo de maíz y achira (Pearsall 2003) y de la caza de animales terrestres como el venado (Stahl 2003). La organización social de Machalilla, al parecer no llegó hasta los niveles de jerarquización alcanzados por Valdivia en sus fases 7 y 8. Existe una marcada ausencia de complejos monumentales, lo que nos invita a pensar en la ausencia de grandes proyectos corporativos y por ende de una autoridad centralizada capaz de movilizar grandes grupos humanos. Dicha ausencia obedece a la falta de instituciones corporativas y de líderes o autoridades encargadas de dirigir proyectos de mediana y gran envergadura, posiblemente debido a la inhabilidad de las autoridades Valdivia de preservar sus condiciones de poder en una nueva configuración social como la encontrada en Machalilla. A pesar de la influencia existente de formas cerámicas Valdivia en la alfarería Machalilla, esta influencia al parecer no fue más allá de experiencias acumuladas en cuanto a estrategias de producción de alimentos, cambiando elementos tradicionales —por ejemplo figulinas— por nuevas formas que no rememoran en nada a aquellas manifestaciones Valdivia de poder y autoridad (como se ha indicado las figulinas Valdivia estuvieron ligados a rituales de fertilidad representando figuras de poder). La ausencia de cuerpos centralizados de poder (manifestados en centros ceremoniales) sugiere relaciones sociales de producción menos elaboradas que en Valdivia, con mecanismos de acumulación de excedentes que posiblemente apuntaron a la redistribución antes que a la centralización de recursos en segmentos específicos. Si bien es cierto el sistema de cargo puede ser uno de los mecanismos de negociación capaces de movilizar grandes grupos, la ausencia de grandes estructuras nos invita a pesar que la población no alcanzó un nivel de cohesión lo suficientemente adecuado como para generar este tipo de proyectos, no en sus tramos iniciales ni finales, ya que es de esperar una reconfiguración social 117 Arqueología y Sociedad 27, 2014: 111-130 y política intensa luego de un evento traumático como los desastres naturales citados en el párrafo anterior. En promedio, Machalilla tuvo una duración de casi un milenio, y la ausencia de estos centros sugiere la existencia de una sociedad plenamente conservadora, a diferencia de Valdivia, en la cual las transformaciones tecnológicas, estilísticas y sociales son bastante obvias en el análisis del registro arqueológico, ¿podríamos entonces hablar de mecanismos sociales que sancionaban la monopolio de excedentes en grupos específicos? ¿A través de estos supuestos mecanismos se promovió la redistribución por encima de la acumulación? Quedan estas preguntas abiertas, siendo necesario indicar que comparativamente es más lo que se ha investigado de Valdivia con respecto a Machalilla por lo que el efecto conservador pudiera deberse a una ausencia de datos por investigaciones insuficientes. IV. CHORRERA Chorrera fue definida a partir de los trabajos de Evans y Meggers en el sitio del mismo nombre (Evans y Meggers 1954) y por los de Bushnell en el sitio de La Carolina en la península de Santa Elena (Bushnell 1951). Su rango de extensión al parecer cubrió tanto la costa como la sierra ecuatoriana y a pesar de la escasez de datos arqueológicos, es considerada por algunos investigadores como un horizonte cultural en donde se reconoce una cierta unidad en estilo y manufactura como producto de una manifestación cultural general pero con la existencia de diversas variaciones regionales (Bischof 1975; Cummins 2003; Lathrap, Collier y Chandra 1975; Simmons 1970). La utilización del término «horizonte» para Chorrera, es hasta cierto punto similar a su uso al referirse a Chavín en los Andes centrales, debido a que los elementos regionales no han sido adecuadamente entendidos, restándose la importancia que los mismo tienen, privilegiándose a similitudes por encima de diferencias. En términos generales, lo que ha llamado mayormente la atención de investigadores ha sido el alto desarrollo logrado en la producción alfarera Chorrera (Cummins 2003; Lathrap, Collier y Chandra 1975). La cerámica se caracteriza por vasijas silbadoras, figulinas en la misma vena que las Valdivia y Machalilla pero de mejor factura tecnológica y artística. La cerámica utilitaria Chorrera es de buena cocción con las paredes delgadas, con superficies muy bien pulidas con engobe rojo o ahumadas (Fig. 7) y se prosigue con la tradición de vasijas carenadas. La decoración es sumamente elaborada en comparación con la cerámica Valdivia o Machalilla. Uno de los puntos importantes en relación con los Andes centrales es la semejanza entre las vasijas de asa estribo Chorrera con las fases Cupisnique, Cupisnique Transitorio y en menor medida Cupisnique Santa Ana establecidas por Rafael Larco, tanto en las formas zoomorfas y fitomorfas, como en las asas gruesas de pico corto (Larco 1948). Al respecto, ya Jijón y Camaño al observar fragmentos Chorrera procedentes de la Hacienda La Compañía, sobre el río Babahoyo (en la provincia de Los Ríos), había indicado que los fragmentos procedían de una cultura relacionada con Chavín y su variante costeña Cupisnique (Holm 2001; Jijón y Camaño 1945). La cronología relativa de Chorrera no es lo suficientemente adecuada hasta el momento, a pesar de intentos muy localizados, de establecer secuencias ceramográficas. (Bischof 1975; Evans y Meggers 1982; Simmons 1970; Zeidler y Sutliff 1994). Por razones estilísticas, se le ha denominado a la variante Chorrera de las provincias de Guayas y El Oro como Engoroy (Lathrap, Collier y Chandra 1975) y es 118 Figura 7. Botella silbato fitomorfa Chorrera. Decoración con engobe crema y rojo, con protuberancias y depresiones. Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales en esta variante en la cual los trabajos de secuencia han sido mayores (Bischof subdivide Engoroy en seis fases agrupadas en tres períodos a saber: Engoroy Temprano, Engoroy Medio y Engoroy Tardío) encapsuladas entre los 900-300/100 ANE (Bischof 1982). Es necesario considerar que las secuencias del sur no pueden ser aplicadas in strictu sensu en el norte debido a las particulares variaciones estilísticas entre las regiones (a pesar del uso del término «horizonte»), lo cual es explicable no solamente por variaciones regionales, por lo limitado de las muestras a partir de las cuales se han elaborado las secuencias hasta el momento. Por el momento, un ejercicio interesante desde el punto de vista metodológico, sería el de tomar como marco de referencia la secuencia estilística Cupisnique definida por Larco, hasta que mejores datos estén disponibles. Es innegable la semejanza entre la cerámica Chorrera y la Cupisnique considerando que de acuerdo a las cronologías disponibles, ambas fueron contemporáneas en un milenio como promedio. La existencia de cerámica Chorrera con formas semejantes a Cupisnique, nos sugiere un intenso fenómeno de transmisión cultural entre ambos desarrollos.5 Como se ha mencionado, las características principales de la cerámica Chorrera ya están presentes en las fases Cupisnique, Cupisnique Transitorio y Cupisnique Santa Ana (Figs. 8 y 9). Las botellas de la fase Kunturwasi, recuperadas por Onuki en contextos funerarios en el sitio del mismo nombre, presentan semejanza con el Transitorio de Larco y están datadas entre los 900-500 ANE, por lo que se reforzaría el argumento de influencia Chorrera en Cupisnique como ya lo había sugerido Jijón y Camaño, Holm y Lathrap anteriormente (Holm 2001; Jijón y Camaño 1945; Lathrap, Collier y Chandra 1975; Onuki 1995). Tal como lo indica Cummins, la esfera en la cual estos objetos fueron movilizados en el área andina permanece en el sistema de intercambio de la que formaron parte, antes que los contextos en los cuales los objetos fueron producidos (Cummins 2003). Muchos de los objetos de los períodos Cupisnique, Cupisnique Transitorio formaron parte de la red de intercambio de ofrendas que se dio durante el Formativo Medio peruano (1500-900 ANE) y Formativo Tardío peruano (900-500 ANE), tal y como se observa en Chavín de Huántar (Fig. 10), Kuntur Wasi, Huaca Partida (Ihekara y Shibata 2005; Kaulicke 2010; Lumbreras 1993; Mesia 2007; Onuki 1995), y si bien estos estilos son distinguibles en asociaciones a contextos funerarios y de arquitectura monumental en el Perú, no tenemos mucha suerte con respecto a Chorrera en el Ecuador y no es posible asociarla hasta el momento a sitios complejos como los del Perú. La cerámica Chorrera comparte incluso con Moche el afán realista por reducir el mundo natural en un objeto (Figs. 11 y 12). En Chorrera la técnica decorativa es variada, incluyéndose la incisión fina, el rocker stamping (Fig. 13), el ahumado, la pintura iridiscente y la combinación de colores en superficies continuas (Evans y Meggers 1982). De realizarse este ejercicio metodológico, es necesario trabajar con muestras del sur de la provincia de Guayas y de la provincia del Oro, ya que es el área más cercana al Perú y el estilo Chorrera presenta ciertas variaciones de sur a norte. Otra característica importante de la cerámica es la presencia de figulinas (Fig. 14). Cummins indica que existen figulinas Chorrera de más de 40 cm (Cummins 2003) mientras que Zeidler precisa que existen ejemplos de hasta 80 cm (Zeidler 2008). Figulinas de este tamaño recuerdan al títere compuesto que Burger halló en el relleno asociado a la escalinata trasera del centro cívico ceremonial de Mina Perdida en el valle de Lurín, costa central peruana (Burger y Salazar 1998). Si bien este espécimen estuvo elaborado a partir de mate y cerámica cruda, objetos de este tipo pudieron servir como oficiantes invitados en ceremonias específicas frente a audiencias de diverso tamaño. De acuerdo a Lathrap, la figulina de Curayacu excavada por Engel, podría ser un derivado de la tradición de figulinas Chorrera (Lathrap, Collier y Chandra 1975), lo que podría ser extendido a figulinas encontradas en Ancón y la costa norte peruana (Fig. 15), cuyas similitudes con sus contrapartes norteñas son merecedoras de resaltar. Del mismo modo, destaca la presencia de platos tetrápodos y pentápodos, cuyos 5 Es necesario mencionar que los términos «Cupisnique» y «Chorrera» comparten cierto grado de incertidumbre ya que no queda aún claro si se refieren a unidades políticas, estilo, tecnología, etc. Para efectos del presente artículo ambos términos se refieren a formas y modos de elaborar cerámica asociadas a una población en particular, independientemente del tamaño de la misma. 119 Arqueología y Sociedad 27, 2014: 111-130 Figura 8. Botella antropomorfa Chorrera. Individuo arqueado por la espalda. Gollete directo, borde evertido y asa simple. Museo de Arte Precolombino Casa del Alabado. Figura 9. Botella zoomorfa Chorrera. Mono en posición sedente. Gollete directo, labio engrosado y asa simple. Museo de Arte Precolombino Casa del Alabado. Figura 11. Cerámica escultórica Chorrera. Cabeza de ciervo. Museo de Arte Precolombino Casa del Alabado. 120 Figura 13. Botella antropomorfa Chorrera. Personaje femenino, recostado sobre su lado derecho. Cuerpo decorado mediante la técnica del rocker stamping. Museo de Arte Precolombino Casa del Alabado. Figura 10. Botella de asa estribo, estilo Wacheqsa, encontrada en la Galería de las Ofrendas en Chavín de Huántar. Museo de Antropología y Arqueología de la Universidad Nacional Mayor de San Marcos. Figura 12. Botella zoomorfa Chorrera. Ave sobre cuerpo carenado. Gollete directo, borde vertido y asa simple. Museo de Arte Precolombino Casa del Alabado. Figura 14. Figulina antropomorfa Chorrera. Personaje femenino en posición vertical, porta orejeras y un elaborado tocado delineado a partir de detalles incisos y de alto relieve. Museo Casa del Alabado. Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales Figura 15. Figulina antropomorfa Ancón. Personaje femenino en posición vertical. Cabello simulado a partir de incisiones, lleva una vincha a modo de cerquillo, decorada con líneas verticales, brazos en relieve pegados al cuerpo. Museo Nacional de Arqueología, Antropología e Historia del Perú. soportes en algunos casos presentan formas fitomorfas (Fig. 16). Por otra parte, llama la atención la presencia de botellas con asa puente, las cuales son características de las fases Cavernas (500–200 ANE) y Necrópolis (200 ANE-100 NE) en la costa sur peruana (Figs. 17 y 18), así como las formas de cuerpo semiglobular carenado que en algunos casos imita a las formas de lagenarias, características de Paracas Necrópolis. La similitud en el asa puente y forma de cuerpo entre vasijas Chorrera y Paracas ya había sido notada por Lathrap, quien especuló que la naturaleza de estas similitudes se debió a contacto marino entre las costas de Ica y del sur ecuatoriano, dada la ausencia de estos rasgos en la costa norte peruana (Lathrap, Collier y Chandra 1975). En términos de sistema de asentamientos, la información disponible no es precisamente abundante. En ese sentido vale la pena destacar los trabajos de Zeidler en el valle de Jama. A través de una prospección sistemática de 785 km² en el Valle de Jama, al norte de la provincia de Manabí, ha identificado 33 sitios Chorrera, distribuidos en su gran mayoría en los pisos de valle, mientras que un pequeño porcentaje se encuentra en zonas elevadas. (Zeidler 1994a, 1994b) Al parecer la densidad poblacional localizada en el piso del valle, originó la movilización de pequeños grupos hacia las zonas más altas, lográndose una economía mixta a partir de la explotación de diversos nichos ecológicos(Zeidler 1994b) . De acuerdo a Stall, la economía Chorrera al parecer estuvo fundamentada en el maíz, frejol, calabaza y achira, el consumo del armadillo, diversas variedades de ciervos, patos, y algunas variedades de roedores (Stahl 2003) En el valle de Jama, destaca el sitio de San Isidro, como el exponente más importante de arquitectura monumental Chorrera, el cual fue construido sobre los cimientos de una plataforma Valdivia 7 (en la cual hay una notable ausencia de elementos Machalilla) (Zeidler 1994b). Es interesante notar que al interior del Valle de Jama, el sitio de San Isidro destaca como la cabeza de una jerarquía de asentamientos que se extiende tanto hacia las zonas bajas y altas del valle, lo que le lleva Zeidler a argumentar por la existencia de una jefatura estratificada, o una jefatura compleja en el sentido clásico Figura 16. Plato fitomorfo Chorrera. Base irregular calada evertida, presenta cinco patas bulbares puntiagudas, imitando rizomas o tubérculos. Museo de Arte Precolombino Casa del Alabado. Figura 17. Botella zoomorfa Chorrera. Representación de strombus modelado el cual presenta un asa puente en la parte superior. Museo de Arte Precolombino Casa del Alabado. Figura 18. Botella fitomorfa Chorrera. Representación de calabaza con asa puente, superficie del cuerpo con engobe rojo. Museo de Arte Precolombino Casa del Alabado. 121 Arqueología y Sociedad 27, 2014: 111-130 Figura 19. Cuenco Mayo-Chinchipe. Presenta diseños incisos sobre la superficie, entre los que destaca la representación de rostros al parecer humanos, similares a los existentes en la cerámica valdivia. Museo de Arte Precolombino Casa del Alabado. de la definición (Flannery 1972). La existencia de edificios monumentales de similar magnitud en Salango podría llevar a pensar en un fenómeno similar, lo que nos hace preguntarnos: ¿Se trata de una red de asentamientos integrada a partir de diversas cabezas? ¿Podemos hablar de un sistema de peer polities? (Renfrew 1986). De tener respuestas afirmativas, la configuración de asentamientos sería similar a la que se observa durante el período Formativo en los Andes centrales (Burger 2008; Mesía 2000; Rick 2005) Si erupciones volcánicas fueron responsables de la crisis económica y social de la sociedad Valdivia, la erupción del Pululahua alrededor del 476 ANE (Zeidler e Isaacson 2003), fue responsable de similar situación en la sociedad Chorrera, con lo que concluye uno de los episodios más fascinantes de la prehistoria ecuatoriana. V. AL ESTE DE LA COSTA: EL FORMATIVO SERRANO 122 Una gran porción de la sierra ecuatoriana permanece aún sin investigar, debido a lo agreste del territorio y también a la existencia de gruesos depósitos volcánicos que dificultan las investigaciones. Si bien en la costa encontramos marcados desarrollos de alcance regional, en la sierra la situación es diferente, aislándose conjuntos arqueológicos antes que culturas plenamente establecidas y diferenciadas. A continuación mencionaré algunos elementos destacados dentro del Formativo serrano, los cuales gracias a los datos disponibles nos permiten tener un panorama más claro sobre los procesos sociales que se desarrollaron en la sierra ecuatoriana durante el período Formativo. Cotocollao fue una aldea formativa, identificada por Villaba (Villalba 1988) y fechada entre los 1800-400 ANE, dividida en cuatro fases cerámicas (Lippi 2003). Los datos de los trabajos de Villaba proceden de excavaciones de rescate arqueológico, en los cuales no se han registrado evidencias de arquitectura monumental. La cerámica Cotocollao incluye vasijas carenadas, botellas de asa estribo, decoración punteada, incisa y engobes rojos, los cuales son elementos característicos de Machalilla y Chorrera (cierto, las formas carenadas se originan en Valdivia, pero dadas las distancias cronológicas, lo lógico es pensar en Machalilla como influencia directa). Del mismo modo, el trabajo en morteros de piedra guarda relación con artefactos costeños, específicamente con los del sitio Chorrera de San Isidro en el valle de Jama, a pesar de no presentarse elementos zoomorfos en Cotocollao (Zeidler 1988; Zeidler y Sutliff 1994). La economía de esta villa estuvo basada en el consumo de maiz, achira, oca, papa, y quinua (Pearsall 2003) y de ciervos, llamas y cuyes (Stahl 2003). Sitios con formas cerámicas identificadas en Cotocollao, se han registrado en los cercanos valles de Tumbaco y los Chillos (Zeidler 2008), pero no se han identificado sitios monumentales. ¿Podríamos hablar de un sistema de villas interconectadas a partir del comercio? Esta y otro tipo de preguntas podrán ser respondidas a partir de mayores investigaciones arqueológicas. Otro sitio que merece ser mencionado es el de La Chimba, el cual se encuentra a 55 km de Quito y alcanza una extensión de 12 ha. De acuerdo a la evidencia paleobotánica, el sitio comparte similitudes con Cotocollao, consumiendo maíz, papa, oca, quinua y frejoles. La cerámica de la Chimba guardaría relación con la de Cotocollao, fundamentalmente en lo que respecta a las ollas carenadas y decoración puntuada y aplicada (Bruhns 2003). La Chimba mantuvo un comercio intenso con la costa e incluso Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales la selva, dada la presencia de madreperlas, spondylus, strombus, vasijas Chorrera, cerámica de oriente (Cosanga) e incluso coca a juzgar por la decoración en algunas vasijas (Athens 1995). El sitio de Tulipe, ubicado en las cercanías de Quito, presenta una fase denominada Nueva Era, la cual se caracteriza por presentar vasijas relacionadas estrechamente con Chorrera, antes que a sitos serranos más cercanos como Cotocollao y La Chimba. Esta fase se encontró íntegramente cubierta por depósitos volcánicos, producto de la erupción del Pululahua, la cual fue responsable de la desestabilización que puso fin a Chorrera alrededor de los 400 ANE (Isaacson 1987) Challuabamba, ubicado en el valle de Tomebamba en la sierra sur ecuatoriana, es un extenso sitio de 70 ha, el cual presenta una ocupación que va desde los 2000 hasta los 1400 ANE y el inicio de la ocupación en este sitio es contemporáneo con la fase 7 de Valdivia en la costa. Grieder segrega la ocupación en cuatro fases (Períodos I, II, III y IV) e identifica tres alfares a lo largo de la misma (Grieder, T., J. Farmer, A. Carrillo y B. Jones 2002). Dichos alfares presentan lo que Bruhms ha identificado como cáscara de huevo, por sus paredes excesivamente delgadas y que son indicadores importantes de la tecnología alfarera de la sierra sur ecuatoriana. Al parecer, si bien no hay evidencias de arquitectura monumental, Challuabamba fue un importante centro de intercambio con la costa, a juzgar por la presencia de fauna marina en el registro arqueológico (Grieder, T., J. Farmer, A. Carrillo y B. Jones 2002). De similar importancia es la región de Loja, en el área de Catamayo, el cual presenta siete sitios del Formativo (Guffroy 1987, 1989, 2004) cuyos límites cronológicos son similares a los planteados por Grieder en Challuabamba (Grieder, T., J. Farmer, A. Carrillo y B. Jones 2002). Guffroy presenta una secuencia de cuatro fases dividida en Catamayo, A, B, C y D. A lo largo de esta secuencia se observan marcadas influencias costeñas entre las que destacan incisiones anchas, bandas aplicadas, y punteados zonales. Es importante la mención referente a botellas de formas similares a Cupisnique y/o Chavín en la fase Catamayo D (Guffroy 2004), fase cuyos límites están enmarcados entre los 500-300 ANE, espacio temporal en el cual Chavín no funcionaba como un centro ceremonial de importancia pan andina (Mesia 2007) y el Cupisnique Santa Ana era popular en costa norte dando paso al estilo Salinar. En oriente, el trabajo de Francisco Valdez, ha revelado una tradición cultural desconocida la cual ha denominado como Mayo-Chinchipe, la cual se encuentra en el sitio de Santa Ana-La Florida, en la cuenca del río Mayo (afluente del río Marañón), en las cercanías de la frontera con el Perú (Valdez 2007, 2008). El sitio en mención al parecer se trata de un centro ceremonial compuesto por un círculo de piedra de 40 m de diámetro, en cuyo interior se encontraron seis estructuras cuadrangulares dispuestas simétricamente a ambos lados del círculo. El sitio se encuentra sobre una terraza sostenida por un conjunto de muros de contención. Sobre la terraza se agrupó un conjunto de recintos en disposición circular, en cuyo centro se ubicó un fogón, con una ofrenda asociada de cuencos de piedra y cuentas de piedra; al parecer existió un entierro asociado a estas ofrendas, pero lamentablemente se ubicaron tan solo fragmentos de un cráneo (Valdez 2007, 2008). Los cuencos de piedra guardan similitudes cercanas con las ofrendas excavadas por Pedro Rojas Ponce en Huayurco, en la amazonía peruana (Rojas 1969). Otro entierro fue ubicado en las inmediaciones, asociado a fragmentos de strombus, vasijas carenadas hemiesféricas, cuatro vasijas con asa estribo, cuencos de piedra finamente pulidos, cuentas de turquesa y pseudo malaquita entre otros objetos. Este contexto presenta una fechado al 66,7% de 2141-2031ANE,6 por lo que se constituiría en el fechado más antiguo de contextos asociados a asas estribo en territorio ecuatoriano y sobrepasaría en antigüedad a las botellas de asa estribo de Machalilla. El resto de las formas cerámicas encontradas en Santa Ana-La Florida guardan relación con la fase Catamayo A y Valdivia 3-8 (Guffroy 1987; Hill 1974), lo cual es consistente con los 14 fechados 6 Este fechado fue calibrado por el autor del presente artículo, utilizando el software OxCal v.4.1.7, utilizando la curva IntCal 09, sobre la base del fechado Beta-197176 (3700 +40 AP), publicado por Valdez (Valdez 2008: 880, Tabla 43.1). Este y el resto de fechados procedentes de este sitio que se mencionen en el presente artículo han seguido similar tratamiento. Los cuencos de Huayurco son considerablemente más tardíos que los de Santa Ana-La Florida. 123 Arqueología y Sociedad 27, 2014: 111-130 publicados para el sitio7 (Valdez 2008). Los detalles de algunos cuencos de piedra hallados en el contexto funerario antes mencionado, guardan relación con diseños encontrados sobre soportes textiles en Huaca Prieta y La Galgada (específicamente las serpientes bicéfalas estilizadas) por lo que habría que pensar sobre el rol que tuvo esta cultura del oriente ecuatoriano en las sociedades del Arcaico Tardío de la costa peruana y posiblemente con las del oriente peruano. Es interesante notar también que los famosos mates de Huaca Prieta muestran diseños que podrían relacionarse con aquellos de las fases 3-5 de la cerámica Valdivia (Bischof 1999; Lathrap 1974), por lo que el sitio de Huaca Prieta podría haber tenido estrechos contactos con la costa y oriente ecuatorianos. Especial atención merece el cuenco Mayo-Chinchipe del Museo de Arte Precolombino Casa del Alabado (Foto 18), el cual presenta en su superficie rostros similares a aquellos que se encuentran sobre la superficie de cuencos Valdivia de las fases 3-5 y del mate de las caras procedente de Huaca Prieta. VI. DISCUSIÓN El desarrollo del Formativo en el Ecuador es bastante extenso en términos cronológicos, tiene una extensión de alrededor de 4300 años (4500-300 ANE) en relación con el Formativo en el Perú que presenta una extensión aproximada de 1800 años (2000-200 ANE). La larga maduración cultural de Valdivia (4500-1500 ANE), pudo contribuir a la ocurrencia de una explosión cultural en el Perú durante el Formativo, lo que sirvió para que los desarrollos tempranos del Formativo peruano marchen en parte sobre la experiencia ecuatoriana. Esta afirmación puede ser juzgada por su orientación difusionista, la cual no pretendemos ocultar, pero es necesario indicar que las palabras en parte deben de ser incluidas en el razonamiento. El Arcaico Tardío peruano se caracteriza por presentar un fuerte desarrollo de la arquitectura monumental, el cual no se refleja en las mismas dimensiones o densidades territoriales en el Formativo Ecuatoriano. Si la arqueología trabaja en función de evidencias materiales, se puede indicar en base a las evidencias materiales existentes que el Arcaico peruano no influyó fuertemente en el Formativo Ecuatoriano, pero si en el Formativo de los Andes centrales. La experiencia acumulada en organización social y arquitectura monumental fue igual de importante en el desarrollo del Formativo peruano. Y del mismo modo, es importante reconocer la importancia de la experiencia acumulada en los Andes septentrionales, la cual llegó a los Andes centrales, directa o indirectamente, la misma que pudo regresar, procesada y reinterpretada, luego de su paso por los Andes centrales. La cerámica temprana de Pandanche, Huaca Loma y Kuntur Wasi, tiene estrecha relación con formas y decoraciones Valdivia y Machalilla, mientras que en Tutishcainyo, Cueva de las Lechuzas y Kotosh se observa igualmente influencia Valdivia y Machalilla según lo argumentado en el presente artículo. Del mismo modo, como acabamos de mencionar el oriente ecuatoriano probablemente mantuvo contactos con la costa y sierras peruanas a juzgar por los diseños encontrados en los textiles de Huaca Prieta y La Galgada y probablemente haya ejercido influencia en la producción de morteros del Arcaico Tardío de los Andes centrales, pudiéndose dar esta relación de modo opuesto, lo cual tendría que corroborarse arqueológicamente. En Valdivia se observa, un largo proceso evolutivo de complejidad social, teniendo a Real Alto, el sitio en el cual se observa desde Valdivia 1 hasta Valdivia 7, un proceso de cambio en la configuración social del espacio, desde una aldea hasta un centro ceremonial y el poblamiento de la costa ecuatoriana de centros ceremoniales contemporáneos para las fases 7 y 8 de Valdivia. La influencia de Valdivia asimismo alcanza a la sierra ecuatoriana, en sitios como Cotocollao, La Chimba, Challuabamba, Loja entre otros. Lamentablemente desastres naturales impidieron observar hasta donde hubiera llegado el avance de esta sociedad. 124 7 La ocupación más temprana del sitio presenta un fechado (Beta-197175) al 59,1% de 2931-2883 ANE y está asociado a un piso de ocupación. El fechado más tardío al 62,7% (Beta-181459) da un valor de 1316-973 ANE. Podría descartarse el Beta-188267 por tratarse de un outlier en la secuencia de fechados, el cual al 49,6% da un margen de 317-207 ANE, sin embargo mayores investigaciones podrían llenar este espacio cronológico. Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales Machalilla es relativamente modesta en términos de circunscripción espacial y de desarrollo tecnológico y hasta social, aunque habría que investigar en detalle la naturaleza de los contactos con Santa Ana-La Florida, ya que las botellas asa estribo llegan a la costa a partir de las evidencias tempranas de este sitio. Habría que considerar hasta qué punto es certera la hipótesis del arribo de las botellas asa estribo desde la costa exclusivamente a la luz de las evidencias de Santa Ana-La Florida. En Machalilla no se observan centros monumentales y la cerámica no alcanza la sofistificación de las fases 7 y 8 de Valdivia. Chorrera, alcanza un alto nivel tecnológico en la producción de cerámica, lo que llevó a Estrada a afirmar que Chorrera es un fundamento prehistórico de la unidad nacional ecuatoriana (Estrada 1958). Ciertamente, la cerámica Chorrera destaca por su variedad de formas, representaciones realistas de la naturaleza y técnicas de decoración, las cuales guardan mucha similitud con la cerámica Cupisnique de la costa norte peruana (especialmente con las fases Cupisnique y Cupisnique Transitorio). La naturaleza de la relación de Chorrera con sociedades contemporáneas del Formativo peruano podría deberse a transacciones comerciales antes que a relaciones religiosas o de adherencia a un sistema de creencias, ya que no se observan (más allá de algunos elementos aislados citados por Lathrap) elementos que formen parte del sistema iconográfico Chavín–Cupisnique, el cual se extendió por los Andes centrales durante el Formativo Tardío peruano. Chorrera se extendió por casi toda la costa ecuatoriana, teniendo control sobre la isla de Salango, conocida por presentar Spondylus princeps de manera endémica en sus aguas, por lo que se puede esperar una relación comercial entre la costa ecuatoriana y la costa norte peruana a partir del intercambio de este producto como ya ha sido sugerido por otros autores (Marcos 2005). Sin embargo la relación Chorrera–Cupisnique no ha sido examinada a fondo por lo que es necesario investigar esta relación a partir del examen de secuencias cerámicas y contextos arqueológicos. Lo mismo se podría decir de una supuesta relación Chorrera– Paracas, sugerida por Lathrap en base a las formas de botellas Cavernas y Necrópolis (Lathrap 1974). De cualquier modo, tanto Valdivia como Machalilla y Chorrera ejercieron influencia directa e indirecta en el resto del territorio ecuatoriano y contribuyeron a los desarrollos que se dieron en el Formativo peruano, influencia que pudo darse del mismo modo en dirección sur a norte. Los trabajos en Pechiche y Garbanzal realizados en Tumbes, en el extremo norte peruano, han identificado una fase relacionada con elementos que —de acuerdo a los autores— recuerdan vagamente a Chavín (o Cupisnique) pero al mismo tiempo con influencias ecuatorianas (Izumi y Terada 1966). Se trata de la fase Pechiche, ubicada en el sitio del mismo nombre en la ribera sur del río Tumbes, cuyas medianas estadísticas de fechados calibrados nos indican un rango de 1064–918 ANE,8 los cuales son contemporáneos con la primera etapa de la fase Chorrera/Engoroy en la costa sur ecuatoriana y Cupisnique en la costa norte peruana (Tabla I). Staller es de la misma opinión, afirmando que: «[Pechiche] muestra claras afinidades estilísticas con los complejos del lado ecuatoriano de la frontera» (Staller 2001b). Trabajos recientes en los sitios de Uña de Gato y el Porvenir, en la cuenca del río Tumbes nos revelan una ocupación que se inicia en el período Arcaico Tardío de los Andes centrales hasta aproximadamente los 700 ANE, algo desconocido hasta algunos años, lo cual servirá para darnos más luces respecto a la interacción entre el norte peruano y el sur ecuatoriano (Moore 2010). Cuando Burger se refiere a la «esfera de influencia Chavín» (Burger 1992), coloca a Chiclayo como área límite de esta esfera en la costa norte peruana y a Cajamarca en la sierra norte como límites de esta esfera. Del mismo modo Hocquenghem se refiere a un área cultural centroandina, cuyos límites estarían al sur del río Olmos en Lambayeque (Hocquenghem 1991). Si bien estos límites culturales 8 Los códigos de los fechados son N-82 860+110; N-83, 910+120; N-72, 810+150 y N-75, 785+120. Los fechados han sido calibrados utilizando el software OxCal v.4.1.7, utilizando la curva IntCal 09. Las medianas de cada uno de los fechados son N-82, 999; N-83, 1064; N-72, 956 y N-75, 918. Los rangos de cada fechado al 68,2% son los siguientes N-82, 1114-840; N-83, 1212-902; N-72, 1189-792; N-75, 1053-792. Por otra parte Moore coloca a Pechiche entre los 2000-1000 ANE (Moore, 2010). 125 Arqueología y Sociedad 27, 2014: 111-130 pudieron haber existido en el pasado, los mismos no fueron impedimentos para que los pueblos de la costa norte y oriente de los Andes centrales entablaran contactos con los pueblos costeños y selváticos de los Andes septentrionales a juzgar por la evidencia discutida en el presente trabajo. Ciertamente, las fronteras de los estados nación contemporáneos, tienden a sesgar la visión en conjunto de un contexto cultural bastante amplio y extendido que fue alimentado por procesos de innovación y transmisión cultural bastante complejos que derivaron en los procesos sociales resumidos en el presente artículo. Una crítica similar formula Guffroy, quien rebate los argumentos de Hocquenghem y Burger, indicando que las fronteras culturales precisadas por ambos autores parecen estar influidas por un determinismo ecológico severo (Guffroy 2004), de acuerdo a lo observado en la región de Loja. Como se ha mencionado, la fase Catamayo D presenta rasgos de los Andes centrales bastante marcados como ollas sin cuello, cuencos biselados y carenados, cuellos de botella con reborde, los cuales ciertamente constituyen elementos presentes en la generalidad Janabarroide (Burger 1984, 1992; Guffroy 2004). Sin embargo, también están presentes elementos Cupisnique como la decoración delimitada por líneas incisas, la cual también aparece en Kuntur Wasi, Pacopampa, Huacaloma y otros sitios contemporáneos. La presencia de elementos procedentes de la costa ecuatoriana (Chorrera) con otros del Formativo Tardío de los Andes centrales, nos indican la presencia de un fuerte tráfico cultural entre Loja, la costa del Ecuador, la costa del Perú y posiblemente la sierra central del Perú. La explicación más razonable es apuntar a la existencia de ejes paralelos e incluso transversales entre estas regiones. La ausencia de elementos ecuatorianos en Chavín de Huántar sugiere que la probable llegada de elementos tradicionalmente reconocidos como Chavín a la sierra sur ecuatoriana se debió a transmisión indirecta, probablemente teniendo a centros de la sierra cajamarquina o piurana como intermediarios. De cualquier modo, concuerdo con Guffroy en afirmar que es necesario pensar en espacios abiertos antes que en rígidas fronteras ecológicas al referirnos al norte de los Andes centrales y al sur de los Andes septentrionales. En tal sentido, una mejor comprensión del Formativo peruano se logrará cuando investigadores peruanos analicen y comprendan los desarrollos ecuatorianos y del mismo modo, es necesario para la arqueología formativa del Ecuador, integrar a la discusión los procesos sociales del Formativo peruano. Agradecimientos A los editores de Arqueología y Sociedad, al anónimo revisor que tuvo la paciencia de revisar el presente trabajo y hacer comentarios atingentes en pos de la mejora del mismo. Al Museo de Arte Precolombino Casa del Alabado y al Museo Nacional de Arqueología, Antropología e Historia del Perú por el acceso a sus colecciones. BIBLIOGRAFÍA 126 ATHENS, S. 1995 «Relaciones interregionales prehistóricas en el norte de los Andes: evidencia del sitio La Chimba en el Ecuador septentrional». En: C. Gnecco (ed.) Perspectivas Regionales en la Arqueología del Suroccidente de Colombia y Norte del Ecuador. pp. 3-29. Popayán: Editorial Universidad del Cauca. BISCHOF, H. 1975 La fase de Engoroy: períodos, cronología, y relaciones. Bonn: Bonner Amerikanistische Studien 3. 1982 «La Fase Engoroy: Períodos, cronología, y relaciones. I» Primer Simposio de Correlaciones Antropológicas Andino–Mesoamericano 1982, pp. 135-176. 1999 «Los mates tallados de Huaca Prieta: ¿evidencias del arte Valdivia en el arcaico centroandino?» Boletín de Arqueología PUCP 3: 85-120. BONAVÍA, D. 2008 El Maíz, su origen, su domesticación y el rol que ha cumplido en el desarrollo de la cultura. Lima: Universidad San Martín de Porres. Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales BRUHNS, K. O. 2003 «Social and cultural development in the Ecuadorian highlands and eastern lowlands during the Formative». En: S. Raymond y R. Burger (eds.) Archaeology of Formative Ecuador. pp. 125-174. Washington, D.C.: Dumbarton Oaks Research Library and Collection. BURGER, R. 1984 The Prehistoric Occupation at Chavín de Huántar, Perú. Vol. 14. University of Calfornia Publications, Anthropology. Berkeley: University of Berkeley. 1992 Chavin and the origins of Andean civilization. London: Thames and Hudson. 2008 «Chavin de Huantar and its Sphere of Influence». En: H. Silverman y W. Isbell (eds.) Handbook of South American Archeology, pp. 681-706. New York. BURGER, R. y L. SALAZAR 1998 A sacred effigy from Mina Perdida and the unseen ceremonies of the Peruvian Formative RES: Anthropology and Aesthetics: 85-95. BUSHNELL, G. 1951 The Archaeology of the Santa Elena Peninsula in South-west Ecuador. Cambridge: Cambridge University Press. CRUZ, F. y O. HOLM 1982 «Un informe muy preliminar: San Lorenzo del Mate, provincia del Guayas.» Guayaquil: Museo Antropológico del Banco Central del Ecuador, Guayaquil. CUMMINS, T. 2003 «Nature as culture’s representation: a change of focus in Late Formative iconography». En: S. Raymond y R. Burger (eds.) Archaeology of Formative Ecuador, pp. 423-464. Washington, D.C.: Dumbarton Oaks Research Library and Collection. ESTRADA, E. 1958 Las culturas pre-clásicas, formativas o arcaicas del Ecuador. Guayaquil: Museo Arqueológico Víctor Emilio Estrada. EVANS, C. y B. MEGGERS 1954 «Informe preliminar sobre las investigaciones arqueológicas realizadas en la cuenca del Guayas, Ecuador». Cuadernos de Historia y Arqueología 12: 307-336. 1957 «Formative Period cultures of the Guayas basin, coastal Ecuador». American Antiquity: 235-247. 1982 «Técnicas decorativas diagnósticas y variantes regionales Chorrera, un análisis preliminar». En: J. Marcos y P. Norton (eds.) Primer Simposio de Correlaciones Antropológicas Andino-Mesoamericano. pp. 121-133. Guayaquil: Escuela Superior Politécnica del Litoral. FLANNERY, K. 1972 «The Cultural Evolution of Civilizations». Annual Review of Ecology and Systematics: 399-426. GRIEDER, T., J. FARMER, A. CARRILLO y B. JONES 2002 «Art and prestige among noble houses of the equatorial Andes». En: H. Silverman y W. Isbell (eds.) Andean Archaeology II: Art, Landscape and Society. pp. 157-177. New York: Kluwer Academic/Plenum Publishers. GUFFROY, J. 1987 «La Periode Formative». En: J. Guffroy (ed.) Loja Préhispanique, Recherches Archéologiques dans les Andes Méridionales de l’Equateur. pp. 61-139. Paris: Editions Recherches sur les Civilisations. 1989 «Las tradiciones culturales de Catamayo en el ámbito Formativo Andino». En: S. Moreno (ed.) Antropología del Ecuador: Memorias del Primer Simposio Europeo sobre Antropología del Ecuador. Quito: Abya-Yala. 2004 Catamayo Precolombino: Investigaciones Arqueologicas en la Provincia de Loja (Ecuador). Paris: IRD Éditions. HILL, B. 1974 «A new chronology of the Valdivia ceramic complex from the coastal zone of Guayas province, Ecuador». Ñawpa Pacha 12: 1-32. 127 Arqueología y Sociedad 27, 2014: 111-130 128 HOCQUENGHEM, A.-M. 1991 «Frontera entre Áreas Culturales Nor y Centro Andinas en los valles de la costa del extremo norte peruano». Boletín del Instituto Francés de Estudios Andinos 20: 309-348. HOLM, O. 2001 «La Cultura Chorrera». En: K. Stothert (ed.) Lanzas silbadoras y otras contribuciones de Olaf Holm en el estudio del pasado ecuatoriano, vol. I. pp. 161-168. Guayaquil: Banco Central del Ecuador, Museo Antropológico y de Arte Contemporáneo de Guayaquil. IHEKARA, H. y K. SHIBATA 2005 «Festines e integración social en el Período Formativo: nuevas evidencias de Cerro Blanco, valle bajo de Nepeña». Boletín de Arqueología PUCP: 123-160. ISAACSON, J. S. 1987 Volcanic Activity and Human Occupation of the Northern Andes: The Application of Tephrostratigraphic Techniques to the Problem of Human Settlement in the Western Montaña during the Ecuadorian Formative. Ph.D. dissertation. Departament of Anthropology. University of Illinois at Urbana-Champaign. University Microfilms. Ann Arbor. 1994 «Volcanic sediments in archaeological contexts from western Ecuador». En: D. Pearsall y J. Zeidler (eds.) Regional Archaeology in Northern Manabí, Ecuador, Volume 1: Environment, Cultural Chronology, and Prehistoric Subsistence in the Jama River Valley. pp. 131-140. Quito: University of Pittsburgh Memoirs in Latin American Archaeology. Ediciones Librimundi. IZUMI, S. y K. TERADA 1966 Andes 3: Excavtions at Pechiche and Garbanzal, Tumbes Valley, Peru. Tokio: Kadokawa Publishing Co. JIJóN Y CAMAñO, J. 1945 Antropología Prehispánica del Ecuador. Quito: La Prensa Católica. KAULICKE, P. 1975 Pandanche. Un Caso del Formativo de los Andes de Cajamarca. Lima: Seminario de Hisoria Rural Andina. Universidad nacional Mayor de San Marcos (UNMSM). 1994 Los origenes de la civilizacion andina. Vol. I. Lima: Editorial Brasa. 2010 Las cronologías del formativo. Lima: Pontificia Universidad Católica del Perú (PUCP). LARCO, R. 1948 Cronología Arqueológica del Norte del Perú. Buenos Aires: Sociedad Geográfica Americana. LATHRAP, D. 1963 «Possible Affiliations of the Machalilla Complex of Coastal Ecuador». American Antiquity 29: 239241. 1970 The Upper Amazon. London: Thames and Hudson. 1974 «The moist tropics, the arid lands, and the appearance of great art styles in the New World». En: M. King y R. Taylor (eds.) Art and Environment in Native America. Texas: Texas Tech University. LATHRAP, D., D. COLLIER y H. CHANDRA 1975 Ancient Ecuador: Culture, Clay, and Creativity, 3000-300 BC. Chicago: Field Museum of Natural History. LIPPI, R. 2003 «Appendix B: Formative Period chronology for the northern and central highlands of Ecuador». En: S. Raymond y R. Burger (eds.) Archaeology of Formative Ecuador. pp. 529-537. Washington, D.C.: Dumbarton Oaks Research Library and Collection. LUMBRERAS, L. 1981 Arqueología de la América Andina. Lima: Milla Batres. 1993 Chavin de Huantar, excavaciones en la Galeria de las Ofrendas. Mainz: KAVA. MARCOS, J. 1988 Real Alto: La Historia de Un Centro Ceremonial. Guayaquil: Biblioteca Ecuatoriana de Arqueología 4 y 5. Escuela Superior Politécnica del Litoral. 1989 «Informe arqueológico». En: J. Marcos y S. Alvarez (eds.) Proyecto San Lorenzo del Mate. Guayaquil: Fundación Pedro Vicente Maldonado. Christian Mesía / El periodo Formativo en los Andes septentrionales y sus relaciones con los Andes centrales 2005 Los Pueblos Navegantes del Ecuador Prehispánico. Quito: Abya-Yala. MEGGERS, B. 2005 «The subversive significance of transpacific contact». New England Antiquities Research Association (NEARA) Journal 22-30. MEGGERS, B., C. EVANS y E. ESTRADA 1965 Early Formative Period of Coastal Ecuador: The Valdivia and Machalilla Phases. Smithsonian Contributions to Anthropology. Washington, D.C. MESIA, C. 2007 Intrasite spatial organization at Chavín de Huántar during the Andean Formative: three-dimensional modeling, stratigraphy and ceramics. Ph.D Dissertation, Stanford University. 2000 «Anchucaya: aproximación teórica sobre un Complejo en «U» en el Valle Medio del río Lurín». Arqueológicas 24 : 45-52. MOORE, J. D. 2010 «Architecture, settlement, and Formative Developments in the Equatorial Andes: New Discoveries in the Department of Tumbes». Latin American Antiquity 21: 147-172. MORALES, D. 1993 Historia Arquelógica del Perú.. Vol. I. Compendio Histórico del Perú. Lima: Editorial Milla Batres. ONUKI, Y. 1995 Kuntur Wasi y Cerro Blanco, dos sitios del Formativo en el norte del Peru. Tokyo: Hokusen-Sha. OYUELA-CAYCEDO, A. y R. BONZANI 2005 San Jacinto 1, an historical ecological approach to an archaic site in Colombia. Alabama: The University of Alabama Press. PEARSALL, D. 2003 «Plant food resources of the Ecuadorian Formative: an overview and comparison to the Central Andes». En: S. Raymond y R. Burger (eds.) Archaeology of Formative Ecuador. pp. 213-257. Washington, D.C.: Dumbarton Oaks Research Library and Collection. 2008 «Plant Domestication and the Shift to Agriculture in the Andes». En: H. Silverman y W. Isbell (eds.) Handbook of South American Archaeology. pp. 105-120. New York: Springer. PRATT, J. A. 1999 «Determining the function of one of the new world’s earliest pottery assemblages: the case of San Jacinto, Colombia». Latin American Antiquity 10:71-85. RENFREW, C. 1986 «Introduction: peer polity interaction and sociopolitical change». En: C. Renfrew y J. Cherry (eds.) Peer polity interaction and sociopolitical change. pp. 1-18. Cambridge: Cambridge University Press. RICK, J. 2005 «The Evolution of Authority and Power at Chavin de Huantar, Peru». En: D. O. a. C. A. C. Kevin J. Vaughn (ed.) Foundations of Power in the Prehispanic Andes, vol. 14. pp. 71-89. Los Angeles: American Anthropological Association. ROJAS, P. 1969 «La ‘Huaca’ Huayurco, Jaén, Cajamarca». Boletín del seminario de Arqueología. Instituto Riva Agüero: 48-56. SCHWARZ, F. y S. RAYMOND 1996 «Formative settlement patterns in the Valdivia valley. SW Coastal Ecuador». Journal of field archaeology 10: 205-224. SIMMONS, M. 1970 The ceramic sequence from La Carolina, Santa Elena Peninsula, Ecuador. Ph.D, Arizona. STAHL, P. 2003 «The zoarchaeological record from Formative Ecuador». En: S. Raymond y R. Burger (eds.) Archaeology of Formative Ecuador. pp. 175-191. Washington, D.C: Dumbarton Oaks Research Library and Collection. 129 Arqueología y Sociedad 27, 2014: 111-130 STALLER, J. 2001a «Explorando el contexto y significado del intercambio a larga distancia en un paisaje sagrado: el caso del Formativo ecuatoriano». En: M. Funes (ed.) Cuadernos de Historia y Arqueología 1951-2001. pp. 56-101. Guayaquil: Casa de la Cultura Ecuatoriana, Núcleo del Guayas. 2001b «Reassessing the Developmental and Chronological Relationships of the Formative of Coastal Ecuador». Journal of World Prehistory 15: 193-256. STOTHERT, K. 1985 «The preceramic Las Vegas culture of coastal Ecuador». American Antiquity 50: 613-637. TELLO, J. C. 1960 Chavin Cultura Matriz de la Civilizacion Andina. Lima: UNMSM. VALDEZ, F. 2007 «Mayo Chinchipe: la puerta entreabierta». En: D. Klein y I. Cruz (eds.) Ecuador: el secreto del arte precolombino. pp. 321-349. Milan: 5 Continentes. 2008 «Inter-zonal Relationships in Ecuador». En: H. Silverman y W. Isbell (eds.) Handbook of South American Archaeology. pp. 865-888. New York: Springer. VAN DER MERWE, N., J. LEE-THORP y S. RAYMOND 1993 «Light, stable isotopes and the subsistence of Formative cultures at Valdivia». En: J. Lambert y G. Gruppe (eds.) Prehistoric human bone. pp. 63-97. Berlin: Springer. VILLALBA, M. 1988 Cotocollao: Una Aldea Formative del Valle de Quito. Miscelánea Antropológica Ecuatoriana. Quito: Museos del Banco Central del Ecuador. ZEIDLER, J. 1988 «Feline imagery, stone mortars, and Formative Period interaction spheres in the northern Andean area». Journal of Latin American Lore 14: 243-283. 1994a «Archaeological testing in the lower Jama valley». En: D. Pearsall y J. Zeidler (eds.) Regional Archaeology in Northern Manabí, Ecuador, Volume 1: Environment, Cultural Chronology, and Prehistoric Subsistence in the Jama River Valley. pp. 71-98. Quito: University of Pittsburgh Memoirs in Latin American Archaeology Ediciones Libri Mundi. 1994b «Archaeological testing in the middle Jama Valley». En: D. Pearsall y J. Zeidler (eds.) Regional Archaeology in Northern Manabí, Ecuador, Volume 1: Environment, Cultural Chronology, and Prehistoric Subsistence in the Jama River Valley. pp. 71-98. Quito: University of Pittsburgh Memoirs in Latin American Archaeology. Ediciones Libri Mundi. 2003 «Appendix A: Formative Period chronology for the coast and western lowlands of Ecuador». En: S. Raymond y R. Burger (eds.) Archaeology of Formative Ecuador. pp. 487–527. Washington, D.C.: Dumbarton Oaks Research Library and Collection. 2008 «The Ecuadorian Formative». En: H. Silverman y W. Isbell (eds.) Handbook of South American Archaeology. pp. 459-488. New York: Springer. ZEIDLER, J. y J. S. ISAACSON 2003 «Settlement process and historical contingency in the western Ecuadorian Formative». En: S. Raymond y R. Burger (eds.) Archaeology of Formative Ecuador. pp. 69-123. Washington, D.C.: Dumbarton Oaks Research Library and Collection. ZEIDLER, J. y M. SUTLIFF 1994 «Definition of ceramic complexes and cultural occupation in the Jama valley». En: Regional Archaeology in Northern Manabí, Ecuador, Volume 1: Environment, Cultural Chronology, and Prehistoric Subsistence in the Jama River Valley. Editado por D. Pearsall y J. Zeidler, pp. 111-130. Quito: University of Pittsburgh Memoirs in Latin American Archaeology Ediciones Libri Mundi. ZEVALLOS, C., W. GALINAT, D. LATHRAP, E. LENG, J. MARCOS y K. KLUMPP 1977 «The San Pablo corn kernel and its friends». Science 196:385-389. 130
https://openalex.org/W3177036176	https://bmcmusculoskeletdisord.biomedcentral.com/track/pdf/10.1186/s12891-021-04435-x	English	null	Slipped capital femoral epiphysis: an epidemiological Nationwide study in Italy from 2001 to 2015	BMC musculoskeletal disorders	2,021	cc-by	6,159	Open Access Abstract Background: Slipped capital femoral epiphysis (epiphysiolysis of the femoral head, SCFE) is the most common pediatric hip disease in 10–14 years old children. The most used procedure to correct a stable form of SCFE is in situ pinning. Instead, the proper treatment for unstable forms is controversial. The first purpose of this study was to estimate annual admissions for SCFE in Italian patients from 2001 to 2015, basing on the hospitalization reports. The second aim was to assess the difference between regions regarding SCFE procedures. Lastly, a statistical prediction of the volume of SCFE procedures performed in Italy based on data from 2001 to 2015 was performed. Methods: Data of this study were collected from the National Hospital Discharge Reports (SDO) reported at the Italian Ministry of Health regarding the years of this paper. The yearly number of hospital admission for SCFE, the percentage of males and females, the average age, days of hospitalization, primary diagnoses and primary procedures in the whole Italian population were calculated using descriptive statistical analyses. Results: From 2001 to 2015, 4893 hospitalizations for SCFE were recorded in Italy, with a mean incidence of 2.9 (cases/100.000 inhabitants). The majority of patients treated by SCFE were males (70.6%). Conclusion: National health statistics for SCFE are attractive for an international audience, as different approaches to screening are reported between countries. These differences allow comparing outcomes internationally. Moreover, sharing national statistics and correlating those to other countries protocols, could be helpful to compare outcomes for different procedures internationally. However, further studies are required to understand the specific reasons for regional variation for SCFE procedures in Italy. Level of evidence: III Slipped capital femoral epiphysis: an epidemiological Nationwide study in Italy from 2001 to 2015 Umile Giuseppe Longo1* , Rocco Papalia1, Sergio De Salvatore1, Laura Ruzzini2, Vincenzo Candela1, Ilaria Piergentili1, Leonardo Oggiano2, Pier Francesco Costici2 and Vincenzo Denaro1 Keywords: Epiphysiolysis, Slipped capital femoral epiphysis, SCFE, In situ fixation, Dunn procedure, Triplane proximal osteotomy, Epidemiology, Surgery, Young © 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. * Correspondence: g.longo@unicampus.it Correspondence: g.longo@unicampus.it 1Department of Orthopedic and Trauma Surgery, Campus Bio-Medico University of Rome, Rome, Italy Full list of author information is available at the end of the article Correspondence: g.longo@unicampus.it 1Department of Orthopedic and Trauma Surgery, Campus Bio-Medico University of Rome, Rome, Italy Full list of author information is available at the end of the article Level of evidence: III Keywords: Epiphysiolysis, Slipped capital femoral epiphysis, SCFE, In situ fixation, Dunn procedure, Triplane proximal osteotomy, Epidemiology, Surgery, Young (2021) 22:570 (2021) 22:570 Longo et al. BMC Musculoskeletal Disorders https://doi.org/10.1186/s12891-021-04435-x Background hospital admission occurring in Italy, both from public and private institutions. In Italy, the regional authorities are responsible for organizing and supervising healthcare services delivered through local structures (public or pri- vate). Data on the healthcare services are collected by hospitals and periodically sent to the Ministry of Health [16]. These data are anonymous and described the pa- tient’s age, sex, residence, the region of hospital admis- sion, days of stays, diagnoses and procedures [17]. Population data were obtained from the National Insti- tute for Statistics (ISTAT) for each year [18]. Epiphysio- lysis was defined by the following International Classification of Diseases, Ninth Revision, Clinical Modi- fication (ICD-9-CM) with the diagnosis code: 732.2. Since SCFE procedures for patients over 19 years were only 263 over the 15-year study period, the study was re- stricted to the patients with 0–19 years of age to avoid underestimation. g Slipped Capital Femoral Epiphysis (also known as Epi- physiolysis of the femoral head, SCFE) is the most com- mon pediatric hip disease that affects patients 10–14 years old [1, 2]. SCFE is defined by posterior and inferior displacement (through the epiphyseal plate) of the prox- imal femoral epiphysis with the metaphysis. Each year in the USA, approximately 10.8 cases per 100.000 children of SCFE occurs [3, 4], and 18–50% are bilateral [5]. A timely diagnosis is challenging due to the relative fre- quency of the disease and the lack of significant symp- toms [6]. Pain along with the adductor muscle (groin pull is uncommon in adolescents) is the most frequent symptom. The aetiology is multifactorial (endocrine dis- eases, hypogonadism, panhypopituitarism, growth spurts), but pediatric obesity represents the most rele- vant risk factor [7, 8]. SCFE is classified (using radiog- raphy and clinics) as stable or unstable forms based on the stability of the femoral physis and the capability to weight-bearing [9]. The former is treated using in situ closed screw fixation; instead, open reduction and fix- ation are usually adopted for unstable forms [10]. How- ever, the proper treatment for unstable forms is still debated [11–13]. The major problem of this disease is the rapidity of diagnosis and the timing of surgery. The decision between conservative or operative treatment could also be influenced by geographical factors [14]. The prevalence of SCFE surgery in Europe is not fully defined, and only Sweden reported the nationwide inci- dence of this disease [15]. © 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. Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Page 2 of 9 Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Page 2 of 9 Macro-areas of Italy y Italy is divided into three macro-areas: North, Center and South-and-islands. The former includes four regions in the Western part (Aosta Valley, Liguria, Lombardy and Piedmont) and five in the Eastern part (Autono- mous Province of Trento, Autonomous Province of Bol- zano, Friuli-Venezia Giulia, Emilia-Romagna and Veneto). The Center counts four countries (Lazio, Mar- che, Tuscany and Umbria). The latter part of Italy in- cludes five regions (Apulia, Basilicata, Calabria, Fig. 1 Macro-areas of Italy are the North, the Centre and the South. (The figure was made using R software version i368 4.0.3) This study was conducted from 2001 to 2015, based on official data source as hospitalization records. The principal purpose is to evaluate the yearly number of SCFE surgeries in Italy. The second purpose is to as- sess geographical variation in hospitalization for SCFE between three macro-areas of Italy (North, Center and South). Finally, a statistical projection of the vol- ume of SCFE procedures in the next 5 years was performed. Background National health statistics for SCFE are attractive for an international audience, as dif- ferent approaches to screening are reported between countries (type of screening, method of classification, mean age at the time of screening, diagnosis and subse- quent treatment protocols). These differences allow comparing outcomes internationally. Moreover, sharing national statistics and correlating those to other coun- tries protocols, could be helpful to compare outcomes for different procedures internationally. Methods Data of this study were collected from the National Hos- pital Discharge Reports (SDO) reported at the Italian Ministry of Health regarding the years of this paper (2001–2015). These reports provided data concerning all Fig. 1 Macro-areas of Italy are the North, the Centre and the South. (The figure was made using R software version i368 4.0.3) Fig. 1 Macro-areas of Italy are the North, the Centre and the South. (The figure was made using R software version i368 4.0.3) Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Page 3 of 9 Fig. 2 Number of SCFE procedures from 2001 to 2015 Fig. 2 Number of SCFE procedures from 2001 to 2015 Campania and Molise) and the islands (Sardinia and Si- cily) (Fig. 1). 2.9 procedures for every 100,000 Italian inhabitants (0– 19 years old). From 2001 to 2015, the incidence of oper- ations slight decreased from 2.8 to 2.3 per 100,000 person-years 0–19 years old without statistically signifi- cant result (p = 0.5) (Table 1). Statistics Table 1 Incidence of SCFE procedures per 100′000 resident from 2001 to 2015 Table 1 Incidence of SCFE procedures per 100′000 resident from 2001 to 2015 Years Frequency Percent Incidence 2001 309 6.3 2.8 2002 348 7.1 3.2 2003 355 7.3 3.2 2004 308 6.3 2.8 2005 299 6.1 2.7 2006 322 6.6 2.9 2007 344 7.0 3.0 2008 334 6.8 2.9 2009 327 6.7 2.9 2010 280 5.7 2.4 2011 349 7.1 3.1 2012 349 7.1 3.1 2013 333 6.8 2.9 2014 375 7.7 3.3 2015 261 5.3 2.3 Total 4893 100.0 2.9 The authors used a descriptive statistical analysis were estimate the yearly number of SCFE procedures, the per- centage of males and females, the average age, days of hospitalization, primary diagnoses and primary proce- dures in Italy. The annual adult population size obtained from ISTAT, a statutory electronic national population register, was used to calculate the incidence rates [19]. The incidence was based on the size of the whole Italian population of patients under 19 years old. To find statis- tical differences between years or sex, the linear regres- sion analysis and the Mann-Whitney U Test were used, as applicable. The Exponential Smoothing (ETS) algo- rithm without seasonality was used to assess the forecast model. The Statistical Package for Social Sciences (SPSS) version 26 was used for this data analysis. Tables, graphs and forecast were performed using Excel (Microsoft) software. Results Population During the study period, 4893 SCFE procedures were performed in Italy (Fig. 2), representing an incidence of Table 1 Incidence of SCFE procedures per 100′000 resident from 2001 to 2015 Years Frequency Percent Incidence 2001 309 6.3 2.8 2002 348 7.1 3.2 2003 355 7.3 3.2 2004 308 6.3 2.8 2005 299 6.1 2.7 2006 322 6.6 2.9 2007 344 7.0 3.0 2008 334 6.8 2.9 2009 327 6.7 2.9 2010 280 5.7 2.4 2011 349 7.1 3.1 2012 349 7.1 3.1 2013 333 6.8 2.9 2014 375 7.7 3.3 2015 261 5.3 2.3 Total 4893 100.0 2.9 The authors used a descriptive statistical analysis were estimate the yearly number of SCFE procedures, the per- centage of males and females, the average age, days of hospitalization, primary diagnoses and primary proce- dures in Italy. The annual adult population size obtained from ISTAT, a statutory electronic national population register, was used to calculate the incidence rates [19]. Population The authors divided the patients for macro-area of domicile (North, Centre and South). The patients were divided into two groups (named “regional” and “extra-re- gional surgeries”) to identify the presence of possible mi- gratory flux. The former includes patients that were treated in the same macro-area of their domicile. In- stead, the extra-regional group included patients who migrated from their domicile and were treated in other macro-areas. Over the study period, the 10–14-year age group re- ported the highest incidence of surgeries (Fig. 3). Males represented the majority of patients, both in total and over the years (females 29.4% and males 70.6%) (Fig. 4). No statistically significant differences in sex trend during the years were found (p = 0.4). g y From 2001 to 2015, the average age of patients was 12.55 ± 2.2. During the entire period, the average age of males was always higher than in females (mean age of Statistics The incidence was based on the size of the whole Italian population of patients under 19 years old. To find statis- tical differences between years or sex, the linear regres- sion analysis and the Mann-Whitney U Test were used, as applicable. The Exponential Smoothing (ETS) algo- rithm without seasonality was used to assess the forecast model. The Statistical Package for Social Sciences (SPSS) version 26 was used for this data analysis. Tables, graphs and forecast were performed using Excel (Microsoft) software. Results Population During the study period, 4893 SCFE procedures were performed in Italy (Fig. 2), representing an incidence of Table 1 Incidence of SCFE procedures per 100′000 resident from 2001 to 2015 Years Frequency Percent Incidence 2001 309 6.3 2.8 2002 348 7.1 3.2 2003 355 7.3 3.2 2004 308 6.3 2.8 2005 299 6.1 2.7 2006 322 6.6 2.9 2007 344 7.0 3.0 2008 334 6.8 2.9 2009 327 6.7 2.9 2010 280 5.7 2.4 2011 349 7.1 3.1 2012 349 7.1 3.1 2013 333 6.8 2.9 2014 375 7.7 3.3 2015 261 5.3 2.3 Total 4893 100.0 2.9 The authors used a descriptive statistical analysis were estimate the yearly number of SCFE procedures, the per- centage of males and females, the average age, days of hospitalization, primary diagnoses and primary proce- dures in Italy. The annual adult population size obtained from ISTAT, a statutory electronic national population register, was used to calculate the incidence rates [19]. The incidence was based on the size of the whole Italian population of patients under 19 years old. To find statis- tical differences between years or sex, the linear regres- sion analysis and the Mann-Whitney U Test were used, as applicable. The Exponential Smoothing (ETS) algo- rithm without seasonality was used to assess the forecast model. The Statistical Package for Social Sciences (SPSS) version 26 was used for this data analysis. Tables, graphs and forecast were performed using Excel (Microsoft) software. Population The highest rate of extra-regional surgeries was recorded for patients that moved from the South to the North (29.7%) and from Population Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Page 4 of 9 females 11.5 ± 2 and mean age of males 13 ± 2; p < 0.001) (Fig. 5). Days of hospitalizations The mean length of days of hospitalization was 4.97 days (range 1–62 days). Males had more days of admission than females (M: 4.97 and F: 4.95 days; p = 0.8). A gen- eral trend of decrease in days of hospitalization in both groups was observed (p = 0.001) (Fig. 6). Patients aged from 5 to 9 had more days of hospitalization on average. Differentiating by sex, males with a higher number of days of hospitalization are be- tween 10 and 14 years old, while women between 5 and 9 years old (Fig. 7). Region of admission and migratory flow Regarding the regional distribution, 2422 cases of SCFE procedures were performed in the North (49.5%), 868 (17.7%) in the Center, and 1603 (32.8%) in the South. Of 4893 SCFE procedures performed in Italy during the study period, data on the patient’s domicile was not available for 20 patients; therefore, only 4873 procedures were included in the analysis. From 2001 to 2015, 1491 patients (30.6%) lived in the North, 823 patients (16.9%) in the Center, and 2559 patients (52.5%) in the South. Regional surgeries were 99.1% in the North, 77.9% in the Center and 61.9% in the South. The highest rate of extra-regional surgeries was recorded for patients that moved from the South to the North (29.7%) and from Fig. 3 Number of SCFE procedures by age group Fig. 3 Number of SCFE procedures by age group Fig. 3 Number of SCFE procedures by age group Region of admission and migratory flow females 11.5 ± 2 and mean age of males 13 ± 2; p < 0.001) (Fig. 5). Regarding the regional distribution, 2422 cases of SCFE procedures were performed in the North (49.5%), 868 (17.7%) in the Center, and 1603 (32.8%) in the South. Of 4893 SCFE procedures performed in Italy during the study period, data on the patient’s domicile was not available for 20 patients; therefore, only 4873 procedures were included in the analysis. From 2001 to 2015, 1491 patients (30.6%) lived in the North, 823 patients (16.9%) in the Center, and 2559 patients (52.5%) in the South. Regional surgeries were 99.1% in the North, 77.9% in the Center and 61.9% in the South. Procedure performed and admission diagnosis codes Procedure performed and admission diagnosis codes During the 15-year study period, the main primary diag- noses were “Nontraumatic slipped upper femoral epiphysis” (84.1%), “Aftercare involving internal fixation device” (3.8%), “Other orthopaedic aftercare” (3.6%) and “Encounter for removal of internal fixation device” (2.4%). The main primary procedures performed are re- ported in Fig. 8. Days of hospitalizations The mean length of days of hospitalization was 4.97 days (range 1–62 days). Males had more days of admission than females (M: 4.97 and F: 4.95 days; p = 0.8). A gen- eral trend of decrease in days of hospitalization in both groups was observed (p = 0.001) (Fig. 6). Patients aged from 5 to 9 had more days of hospitalization on average. Differentiating by sex, males with a higher number of days of hospitalization are be- tween 10 and 14 years old, while women between 5 and 9 years old (Fig. 7). Fig. 4 Percentage of SCFE procedures hospital admission divided by year and sex Fig. 4 Percentage of SCFE procedures hospital admission divided by year and sex Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Page 5 of 9 Fig. 5 Average age over the years by sex Fig. 5 Average age over the years by sex the Center to the North (21.0%). All the data are re- ported in Table 2. Discussion Epiphysiolysis of the femoral head is the most common hip disease in the pediatric population [13, 20], with an incidence of 0.2–0.3 per 100,000 children aged 10–14 years [1]. The aetiology of SCFE is multifactorial and in- cludes endocrine disorders, growth spurs and obesity [2, 4, 7, 21–23]. History of trauma to the hip is uncommon [9]. The most common symptoms are pain and limping localized to the hip, groin, thigh or knee [24, 25]. A pre- cise and rapid diagnosis is challenging due to the differ- ential causes of hip pain in young patients [11]. Apophyseal avulsion fracture or apophysitis of the ante- rosuperior and anteroinferior iliac spine; septic arthritis and adductor muscle strain need to be excluded in these patients [4, 26]. Moreover, also transient synovitis, frac- tures and Legg-Calvè-Perthes should present similar symptoms. However, these conditions are uncommon in the SCFE age group [4, 26]. A delayed diagnosis could avoid short and long-term complications as avascular Projection model BMC Musculoskeletal Disorders (2021) 22:570 Page 7 of 9 Fig. 8 Main primary procedures for SCFE from 2001 to 2015 Fig. 8 Main primary procedures for SCFE from 2001 to 2015 treatment for stable SCFE is in situ pinning with a single screw, performed regardless of the timing of presenta- tion [36]. The unstable SCFE is related to a higher risk of osteonecrosis (20–50% of cases) [37–39], but the proper treatment and the timing associated with the lowest risk of AVN are still debated [11, 36].. The tech- nique described by Parsch and colleagues (open capsu- lotomy and partial reduction) seems to be the most promising, reporting a low rate of AVN [40]. Moreover, the modified Dunn procedure historically reported satis- factory outcomes with a low rate of necrosis, but it is widely influenced by the surgeon’s technique and skills [36]. an AVN incidence of 0 and 47% for stable and unstable SCFE forms, respectively. However, recent literature re- ported an incidence of AVN between 0 and 3.3% for stable forms and 23.9% for unstable forms [32]. [ ] SCFE is a relevant disease in the pediatric population and deserves to be known by clinicians. The main diag- nosis code used for this analysis was 732.2 (Epiphysioly- sis). The main procedures performed for SCFE were the “Open Reduction Of Separated Epiphysis, Femur” (23.4%), followed by “Limb shortening Procedures, Femur” (8.3%). In Italy, from 2001 to 2015, the mean in- cidence of hospital admission for SCFE was 2.9 for every 100,000 Italian inhabitants 0–19 years old. The majority of patients were males of the 10–14 years age group, in line with the Swedish results, as reported by Herngren et al. [15]. Males reported a higher mean age compared to females (p < 0.001). The highest number of patients treated was domiciled in the South of Italy (n = 2559), followed by the North (n = 1491) and the Center (n = 823). Otherwise, the highest number of procedures were The most relevant complications of SCFE are AVN; degenerative osteoarthritis; acute loss of cartilages known as chondrolysis (reported after SCFE surgery or in untreated SCFE); femoroacetabular impingement [3, 12, 41–43]. The rate of AVN varies in the literature, but it is usually more frequent in unstable SCFE compared to stable forms [36]. Loder and colleagues [9] reported Fig. Projection model The projection model indicates a stable volume of hospitalization for SCFE procedures (Fig. 9). The projec- tion model showed that the demand for SCFE proce- dures hospital admissions was estimated to remain unchanged from 261 in 2015 to 278 by 2025. Fig. 6 Average days of hospitalization during the years Fig. 6 Average days of hospitalization during the years Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Page 6 of 9 Fig. 7 Average days of hospitalization by age groups and sex Fig. 7 Average days of hospitalization by age groups and sex The initial step in the treatment of SCFE is to place the patient on non-weight bearing crutches or in a wheel- chair [31]. It is mandatory to prevent slip progression and the insurgence of complications [32]. A closed re- duction should not be attempted because it can result in AVN caused by the restricted blood supply to the epiphysis [33, 34]. Some authors recommend the prophylactic treatment of the contralateral hip, but there is no consensus concerning this topic [35]. There is a lack of high-quality literature on SCFE surgical manage- ment. However, based on the current literature, the best necrosis (AVN) of the femoral head and hip osteoarth- ritis, respectively [27]. Symptom’s duration is used to classify SCFE in acute, acute on chronic and chronic forms. If symptoms present within 3 weeks, it is consid- ered acute; instead, after 3 weeks, it is chronic [28]. The former is the most dangerous because related to a higher rate of AVN. Loder and colleagues [9] classified the sta- bility of the physis based on the patient’s capacity to bear weight, with or without crutches. Moreover, it is possible to evaluate SCFE by radiographical parameters using the Wilson and Southwick methods [9, 29, 30]. Table 2 Analysis of migratory flows by macro-region 2001–2015 Analysis of migratory flows by macro-region 2001–2015 Macroregion of residence Macroregion of hospitalization Frequency Percent Unspecified Center 5 25.0 North 10 50.0 South 5 25.0 Total 20 100.0 Center Center 641 77.9 North 173 21.0 South 9 1.1 Total 823 100.0 North Center 8 0.5 North 1478 99.1 South 5 0.3 Total 1491 100.0 South Center 214 8.4 North 761 29.7 South 1584 61.9 Total 2559 100.0 Table 2 Analysis of migratory flows by macro-region 2001–2015 Analysis of migratory flows by macro-region 2001–2015 Table 2 Analysis of migratory flows by macro-region 2001–2015 Center Longo et al. Projection model 9 The projection model showed a stable volume of hospitalizations for SCFE procedures in the next 5 years Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Page 8 of 9 Page 8 of 9 Page 8 of 9 performed in the North (n = 2422) and the South (n = 1603). The highest number of “extra-regional surgeries” were patients from the South that migrated to the North or the Center. Instead, patients from the North and the Center tended to be hospitalized in their macro-region of domicile. Moreover, patients from the South reported a higher rate of diagnosis of epiphysiolysis compared to the North and the South. A significant decrease in days of hospitalization during the study period was found, but further studies are required to identify possible explana- tions. The forecast model showed that the demand for SCFE hospital admissions was estimated to remain un- changed from 2015 to 2025. Author details 1 1Department of Orthopedic and Trauma Surgery, Campus Bio-Medico University of Rome, Rome, Italy. 2Department of Surgery, Orthopedic Unit, Bambino Gesù Children’s Hospital, Rome, Italy. Received: 6 December 2020 Accepted: 6 June 2021 Availability of data and materials To our knowledge, the only study on the SCFE surgery trend was performed by Herngren et al. [15]. However, only patients aged from 9 to 15 years were included, while a broader range of age was analysed in the present paper (0–19 years old). Moreover, the study period con- sidered by Herngren et al. [15] was shorter compared to the present study (from 2007 to 2013 vs 2001 to 2015, respectively). The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. The access to the database is on request. All data were obtained by the Direzione Generale della Programmazione Sanitaria—Banca Dati SDO of the Italian Ministry of Health. Funding This research received no external funding. References O T 1. Otani T, Kawaguchi Y, Marumo K. Diagnosis and treatment of slipped capital femoral epiphysis: recent trends to note. J Orthop Sci Off J Jpn Orthop Assoc. 2018;23:220–8. 1. Otani T, Kawaguchi Y, Marumo K. Diagnosis and treatment of slipped capital femoral epiphysis: recent trends to note. J Orthop Sci Off J Jpn Orthop Assoc. 2018;23:220–8. 1. Otani T, Kawaguchi Y, Marumo K. Diagnosis and treatment of slipped capital femoral epiphysis: recent trends to note. J Orthop Sci Off J Jpn Orthop Assoc. 2018;23:220–8. 2. Gholve PA, Cameron DB, Millis MB. Slipped capital femoral epiphysis update. Curr Opin Pediatr. 2009;21(1):39–45. https://doi.org/10.1097/MOP.0b013e32 8320acea. 2. Gholve PA, Cameron DB, Millis MB. Slipped capital femoral epiphysis update. Curr Opin Pediatr. 2009;21(1):39–45. https://doi.org/10.1097/MOP.0b013e32 8320acea. 3. Lehmann CL, Arons RR, Loder RT, Vitale MG. The epidemiology of slipped capital femoral epiphysis: an update. J Pediatr Orthop. 2006;26(3):286–90. https://doi.org/10.1097/01.bpo.0000217718.10728.70. 3. Lehmann CL, Arons RR, Loder RT, Vitale MG. The epidemiology of slipped capital femoral epiphysis: an update. J Pediatr Orthop. 2006;26(3):286–90. https://doi.org/10.1097/01.bpo.0000217718.10728.70. 4. Peck DM, Voss LM, Voss TT. Slipped capital femoral epiphysis: diagnosis and management. Am Fam Physician. 2017;95(12):779–84. Acknowledgements We thank the Direzione Generale della Programmazione Sanitaria—Banca Dati SDO of the Italian Ministry of Health for the support in providing data for this research. Ethics approval and consent to participate The Institutional Review Board of Campus Bio-Medico University of Rome ruled that no formal ethics approval was required in this particular case. The access to the database is on request. All data were obtained by the Direzione Generale della Programmazione Sanitaria—Banca Dati SDO of the Italian Ministry of Health. Authors’ contributions Conceptualization, U.G.L. and V.D.; Data curation, S.D.S., V.C. and L.R.; Formal analysis, S.D.S. and I.P..; Methodology, S.D.S., L.R.; Software, I.P. and S.D.S..; Supervision, L.O. and V.D.; Validation, P.F.C..; Visualization, L.O., R.P.., and V.D.; Writing—original draft, U.G.L., V.C. and S.D.S..; Writing—review and editing, U.G.L., S.D.S., R.P.., and L.R. All authors have read and agreed to the published version of the manuscript. Conceptualization, U.G.L. and V.D.; Data curation, S.D.S., V.C. and L.R.; Formal analysis, S.D.S. and I.P..; Methodology, S.D.S., L.R.; Software, I.P. and S.D.S..; Supervision, L.O. and V.D.; Validation, P.F.C..; Visualization, L.O., R.P.., and V.D.; Writing—original draft, U.G.L., V.C. and S.D.S..; Writing—review and editing, U.G.L., S.D.S., R.P.., and L.R. All authors have read and agreed to the published version of the manuscript. Competing interests UGL i b f h UGL is a member of the Editorial Board of BMC Musculoskeletal Disorders. The remaining authors declare that they have no conflict of interest. 8. Villani C, Mantegna N, Chiozzi F, De Stefanis P, Persiani P. Epiphysiolysis of the hip: relationship between etiopathogenesis and hormone status. Chir Organi Mov. 2000;85:409–12. Limitations This study is based on administrative data from different hospitals and macro-regions. The International Classifi- cation of Diseases 9 (ICD-9) was used for all the proce- dures reported. Otherwise, with the ICD-9, it was possible to use different codes for the same surgical pro- cedure. This heterogeneity of codification could lead to an underestimation of our results. Secondly, the database has not been subject to internal validation. Moreover, we found 16.1% of “Removal of Implanted Devices from Bone, Femur” within the procedures included. This data could lead to an overestimation of our results concern- ing the migratory flux and the total amount of proce- dures performed. Patients over 19 years old (n = 263) were excluded to avoid underestimation. Moreover, it was impossible to distinguish monoliteral vs bilateral screw fixation because ICD-9 did not fully code it. Lastly, this is a database study, and therefore it is not possible to define specific reasons for migratory flux. Further studies are required to define this trend precisely. 6. Krauspe R. Epiphyseolysis capitis femoris. Orthopade. 2019;48:643. 7. Reichelt A, Rütt A. Etiology of epiphysiolysis (epiphysiolisthesis) of the femur head. Arch Orthop Unfallchir. 1969;67(1):28–38. https://doi.org/10.1007/ BF00417137. 9. Loder RT, Richards BS, Shapiro PS, Reznick LR, Aronson DD. Acute slipped capital femoral epiphysis: the importance of physeal stability. J Bone Joint Surg Am. 1993;75(8):1134–40. https://doi.org/10.2106/00004623-199308000- 00002. Open reduction and smooth Kirschner wire fixation for unstable slipped capital femoral epiphysis. J Pediatr Orthop. 2009;29(1):1–8. https://doi.org/10.1097/BPO.0b013e31818f0ea3. 17. Longo UG, Salvatore G, Locher J, Ruzzini L, Candela V, Berton A, et al. Epidemiology of Paediatric shoulder dislocation: a Nationwide study in Italy from 2001 to 2014. Int J Environ Res Public Health. 2020;17(8). https://doi. org/10.3390/ijerph17082834. 41. Boero S, Brunenghi GM, Carbone M, Stella G, Calevo MG. Pinning in slipped capital femoral epiphysis: long-term follow-up study. J Pediatr Orthop Part B. 2003;12:372–9. 41. Boero S, Brunenghi GM, Carbone M, Stella G, Calevo MG. Pinning in slipped capital femoral epiphysis: long-term follow-up study. J Pediatr Orthop Part B. 2003;12:372–9. 18. Longo UG, Salvatore G, Rizzello G, Berton A, Ciuffreda M, Candela V, et al. The burden of rotator cuff surgery in Italy: a nationwide registry study. Arch Orthop Trauma Surg. 2017;137(2):217–24. https://doi.org/10.1007/s00402-01 6-2610-x. 42. Ziebarth K, Steppacher SD, Siebenrock KA. The modified Dunn procedure to treat severe slipped capital femoral epiphysis. Orthopade. 2019;48(8):668–76. https://doi.org/10.1007/s00132-019-03774-x. 42. Ziebarth K, Steppacher SD, Siebenrock KA. The modified Dunn procedure to treat severe slipped capital femoral epiphysis. Orthopade. 2019;48(8):668–76. https://doi.org/10.1007/s00132-019-03774-x. 43. Soni JF, Valenza WR, Uliana CS. Surgical treatment of femoroacetabular impingement after slipped capital femoral epiphysis. Curr Opin Pediatr. 2018;30(1):93–9. https://doi.org/10.1097/MOP.0000000000000565. 19. Longo UG, Papalia R, De Salvatore S, Ruzzini L, Piergentili I, Oggiano L, et al. Trends in hospitalisation of Subtalar Joint Arthroereisis in Italy from 2009 to 2016. Foot Ankle Surg Off J Eur Soc Foot Ankle Surg 2021. 20. Dobbe AM, Gibbons PJ. Common paediatric conditions of the lower limb. J Paediatr Child Health. 2017;53(11):1077–85. https://doi.org/10.1111/jpc.13 756. Surg Am. 1993;75(8):1134–40. https://doi.org/10.2106/00004623-199308000- 00002. 32. Naseem H, Chatterji S, Tsang K, Hakimi M, Chytas A, Alshryda S. Treatment of stable slipped capital femoral epiphysis: systematic review and exploratory patient level analysis. J Orthop Traumatol Off J Ital Soc Orthop Traumatol. 2017;18(4):379–94. https://doi.org/10.1007/s10195-017-0469-4. 10. Aprato A, Conti A, Bertolo F, Massè A. Slipped capital femoral epiphysis: current management strategies. Orthop Res Rev. 2019;11:47–54. https://doi. org/10.2147/ORR.S166735. Traumatol. 2017;18(4):379–94. https://doi.org/10.1007/s10195-017- 33. Bartoszewski T, Mazurek T, Siwicka K, Samul G. Results of the treatment of the distal femoral epiphysiolysis. Chir Narzadow Ruchu Ortop Pol. 2007; 72(4):253–7. 11. Sucato DJ. Approach to the hip for SCFE: the north American perspective. J Pediatr Orthop. 2018;38(Suppl 1):S5–12. https://doi.org/10.1097/BPO. 0000000000001183. 34. Surgery for slipped capital femoral epiphysis in adolescents - PubMed. https://pubmed.ncbi.nlm.nih.gov/24397949/. Accessed 13 Aug 2020. 12. Lykissas MG, McCarthy JJ. Should all unstable slipped capital femoral epiphysis be treated open? J Pediatr Orthop. 2013;33(Suppl 1):S92–8. https://doi.org/10.1097/BPO.0b013e318274f716. 35. Lim YJ, Lam KS, Lee EH. Review of the management outcome of slipped capital femoral epiphysis and the role of prophylactic contra-lateral pinning re-examined. Ann Acad Med Singap. 2008;37:184–7. 13. Mathew SE, Larson AN. Natural History of Slipped Capital Femoral Epiphysis. J Pediatr Orthop. 2019;39(Issue 6, Supplement 1 Suppl 1):S23–7. 36. Daley E, Zaltz I. Strategies to avoid osteonecrosis in unstable slipped capital femoral epiphysis: a critical analysis review. JBJS Rev. 2019;7(4):e7. https:// doi.org/10.2106/JBJS.RVW.18.00129. 14. Loder RT, Schneble CA. Seasonal variation in slipped capital femoral epiphysis: new findings using a National Children’s hospital database. J Pediatr Orthop. 2019;39(1):e44–9. https://doi.org/10.1097/BPO. 0000000000001074. 37. McPartland TG, Sankar WN, Kim Y-J, Millis MB. Patients with unstable slipped capital femoral epiphysis have antecedent symptoms. Clin Orthop. 2013; 471(7):2132–6. https://doi.org/10.1007/s11999-013-3042-3. 15. Herngren B, Stenmarker M, Vavruch L, Hagglund G. Slipped capital femoral epiphysis: a population-based study. BMC Musculoskelet Disord. 2017;18(1): 304. https://doi.org/10.1186/s12891-017-1665-3. 38. Aronsson DD, Loder RT. Treatment of the unstable (acute) slipped capital femoral epiphysis. Clin Orthop. 1996:99–110. 39. Zaltz I, Baca G, Clohisy JC. Unstable SCFE: review of treatment modalities and prevalence of osteonecrosis. Clin Orthop. 2013;471(7):2192–8. https:// doi.org/10.1007/s11999-012-2765-x. 39. Zaltz I, Baca G, Clohisy JC. Unstable SCFE: review of treatment modalities and prevalence of osteonecrosis. Clin Orthop. 2013;471(7):2192–8. https:// doi.org/10.1007/s11999-012-2765-x. 16. Skrami E, Carle F, Villani S, Borrelli P, Zambon A, Corrao G, et al. Availability of real-world data in Italy: a tool to navigate regional healthcare utilization databases. Int J Environ Res Public Health. 2019;17(1). https://doi.org/10.33 90/ijerph17010008. 40. Parsch K, Weller S, Parsch D. Conclusions 5. Riad J, Bajelidze G, Gabos PG. Bilateral slipped capital femoral epiphysis: predictive factors for contralateral slip. J Pediatr Orthop. 2007;27(4):411–4. https://doi.org/10.1097/01.bpb.0000271325.33739.86. 5. Riad J, Bajelidze G, Gabos PG. Bilateral slipped capital femoral epiphysis: predictive factors for contralateral slip. J Pediatr Orthop. 2007;27(4):411–4. https://doi.org/10.1097/01.bpb.0000271325.33739.86. The incidence of surgery for SCFE in Italy is 2.9 cases/ 100,000 inhabitants of the same age group (from 2001 to 2015). The higher rate of hospitalization for SCFE was recorded in the South and the North. Epidemiological studies are helpful to understand the national variation of a specific surgical procedure and compare them with other countries. However, further studies are required to understand the specific reasons for regional variation for SCFE procedures in Italy. 6. Krauspe R. Epiphyseolysis capitis femoris. Orthopade. 2019;48:643. 7. Reichelt A, Rütt A. Etiology of epiphysiolysis (epiphysiolisthesis) of the femur head. Arch Orthop Unfallchir. 1969;67(1):28–38. https://doi.org/10.1007/ BF00417137. 9. Loder RT, Richards BS, Shapiro PS, Reznick LR, Aronson DD. Acute slipped capital femoral epiphysis: the importance of physeal stability. J Bone Joint Page 9 of 9 Longo et al. BMC Musculoskeletal Disorders (2021) 22:570 Page 9 of 9 Page 9 of 9 Surg Am. 1993;75(8):1134–40. https://doi.org/10.2106/00004623-199308000- 00002. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 21. Georgiadis AG, Zaltz I. Slipped capital femoral epiphysis: how to evaluate with a review and update of treatment. Pediatr Clin N Am. 2014;61(6):1119– 35. https://doi.org/10.1016/j.pcl.2014.08.001. g j 22. Hellmich HJ, Krieg AH. Slipped capital femoral epiphysis-etiology and pathogenesis. Orthopade. 2019;48(8):644–50. https://doi.org/10.1007/s00132- 019-03743-4. 23. Perry DC, Metcalfe D, Lane S, Turner S. Childhood obesity and slipped capital femoral epiphysis. Pediatrics. 2018;142(5):e20181067. https://doi.org/1 0.1542/peds.2018-1067. 24. Reynolds RA. Diagnosis and treatment of slipped capital femoral epiphysis. Curr Opin Pediatr. 1999;11(1):80–3. https://doi.org/10.1097/00008480-199902 000-00016. 25. Castillo C, Mendez M. Slipped capital femoral epiphysis: a review for pediatricians. Pediatr Ann. 2018;47(9):e377–80. https://doi.org/10.3928/193 82359-20180730-01. 26. Karkenny AJ, Tauberg BM, Otsuka NY. Pediatric hip disorders: slipped capital femoral epiphysis and Legg-Calvé-Perthes disease. Pediatr Rev. 2018;39(9): 454–63. https://doi.org/10.1542/pir.2017-0197. 27. Millis MB. SCFE diagnosis delayed among patients with referred pain. J Pediatr. 2019;212:242. https://doi.org/10.1016/j.jpeds.2019.07.010. 28. Bittersohl D, Bittersohl B, Westhoff B, Krauspe R. Slipped capital femoral epiphysis: clinical presentation, diagnostic procedure and classification. Orthopade. 2019;48(8):651–8. https://doi.org/10.1007/s00132-019-03767-w. 29. Jarrett DY, Matheney T, Kleinman PK. Imaging SCFE: diagnosis, treatment and complications. Pediatr Radiol. 2013;43(Suppl 1):S71–82. 30. Jones CE, Cooper AP, Doucette J, Buchan LL, Wilson DR, Mulpuri K, et al. Southwick angle measurements and SCFE slip severity classifications are affected by frog-lateral positioning. Skelet Radiol. 2018;47(1):79–84. https:// doi.org/10.1007/s00256-017-2761-z. 31. Rossiter DJ, Ahluwalia A, Vo P, Mapara R. The limping child: a systematic approach to assessment and management. Br J Hosp Med Lond Engl 2005. 2018;79:C150–3. 31. Rossiter DJ, Ahluwalia A, Vo P, Mapara R. The limping child: a systematic approach to assessment and management. Br J Hosp Med Lond Engl 2005. 2018;79:C150–3.
https://openalex.org/W3129658875	https://www.nature.com/articles/s41598-021-83892-9.pdf	English	null	Exploring the conservation of Alzheimer-related pathways between H. sapiens and C. elegans: a network alignment approach	Scientific reports	2,021	cc-by	9,103	Exploring the conservation of Alzheimer‑related pathways between H. sapiens and C. elegans: a network alignment approach Avgi E. Apostolakou1,3, Xhuliana K. Sula1,3, Katerina C. Nastou1,2, Georgia I. Nasi1 & Alzheimer disease (AD) is a neurodegenerative disorder with an –as of yet– unclear etiology and pathogenesis. Research to unveil disease processes underlying AD often relies on the use of neurodegenerative disease model organisms, such as Caenorhabditis elegans. This study sought to identify biological pathways implicated in AD that are conserved in Homo sapiens and C. elegans. Protein–protein interaction networks were assembled for amyloid precursor protein (APP) and Tau in H. sapiens—two proteins whose aggregation is a hallmark in AD—and their orthologs APL-1 and PTL-1 for C. elegans. Global network alignment was used to compare these networks and determine similar, likely conserved, network regions. This comparison revealed that two prominent pathways, the APP- processing and the Tau-phosphorylation pathways, are highly conserved in both organisms. While the majority of interactions between proteins in those pathways are known to be associated with AD in human, they remain unexamined in C. elegans, signifying the need for their further investigation. In this work, we have highlighted conserved interactions related to AD in humans and have identified specific proteins that can act as targets for experimental studies in C. elegans, aiming to uncover the underlying mechanisms of AD. Abbreviations AD Alzheimer disease APP Amyloid precursor protein PPIs Protein-protein interactions GNA Global network alignment Abbreviations AD Alzheimer disease APP Amyloid precursor protein PPIs Protein-protein interactions GNA Global network alignment Alzheimer disease (AD) is a chronic, progressive, neurodegenerative disorder, characterized clinically by a grad- ual impact on mental and cognitive functions, affecting a person’s ability to perform common daily activities1. It is the most common cause of dementia and is estimated that by 2050 the number of patients with this disease could exceed 100 million worldwide if no cure is discovered2. The pathological hallmarks of AD are amyloid plaques and neurofibrillary tangles; amyloid plaques are created by the extracellular deposition of fibrils consisting of abnormally folded Aβ peptide—a cleavage product of amyloid precursor protein (APP)3—while neurofibrillary tangles consist mainly of intracellular hyperphosphorylated twisted filaments of the microtubule-associated Tau protein (Tau hereafter)1,4. Moreover, both proteins have the ability to form fibrils extracellularly in vivo, and are thus characterized as amyloid fibril proteins by the International Society of Amyloidosis5. www.nature.com/scientificreports www.nature.com/scientificreports www.nature.com/scientificreports/ like mouse, pig and non-human primates, can offer great insight for the study of neurodegenerative diseases, and have extensively used as models to study the disease7–9. However, in many instances, in vivo experimentation in mammalian model organisms is complex, time- and resource-consuming, therefore other more distant organisms can be used to alleviate these issued. Prominent among them is the nematode Caenorhabditis elegans, which is often used as it provides an attractive model for neurodegenerative diseases, due to its distinctive features (see review by Alexander et al.10). Most importantly, many human genes (ca. 60–80%) have an ortholog—genes in different species that evolved from a common ancestor gene and possibly retain the same function11—in the genome of C. elegans12. Even though, C. elegans has a gene encoding for a protein similar to human Tau (PTL-1)13 the APP-related gene (APL-1) does not contain the Aβ peptide sequence. Moreover, C. elegans does not possess an ortholog for β-secretase, and thus has no β-secretase activity10. For this reason, many transgenic C. elegans strains that express human Aβ and Tau sequences in specific cell types have been created10,13 and are used as models for the study of AD and other neurodegenerative diseases. like mouse, pig and non-human primates, can offer great insight for the study of neurodegenerative diseases, and have extensively used as models to study the disease7–9. However, in many instances, in vivo experimentation in mammalian model organisms is complex, time- and resource-consuming, therefore other more distant organisms can be used to alleviate these issued. Prominent among them is the nematode Caenorhabditis elegans, which is often used as it provides an attractive model for neurodegenerative diseases, due to its distinctive features (see review by Alexander et al.10). Most importantly, many human genes (ca. 60–80%) have an ortholog—genes in different species that evolved from a common ancestor gene and possibly retain the same function11—in the genome of C. elegans12. Even though, C. elegans has a gene encoding for a protein similar to human Tau (PTL-1)13 the APP-related gene (APL-1) does not contain the Aβ peptide sequence. Moreover, C. elegans does not possess an ortholog for β-secretase, and thus has no β-secretase activity10. For this reason, many transgenic C. elegans strains that express human Aβ and Tau sequences in specific cell types have been created10,13 and are used as models for the study of AD and other neurodegenerative diseases. Methods i Protein and interaction datasets. Three network datasets were created for this study, each consisting of a H. sapiens PPI network and a C. elegans PPI network. Every dataset differs in the method of data collection and the purpose it serves. Cross-species network comparison for each pair of networks in the three datasets was achieved by using one or more network alignment algorithms. APP and Tau network from the amyloid interactome. The first human network we used in this study was iso- lated from the Amyloid Interactome23. The Amyloid Interactome is a PPI network of amyloidogenic proteins and their experimentally verified interaction partners. APP, Tau, their interaction partners, as well as any interactions between them, were extracted from this network. The interactions from the Amyloid Interactome are experimen- tally validated and extracted from the IntAct database24. For the C. elegans protein dataset, predicted orthologs of the aforementioned human proteins were recovered using OrthoList225, a compendium of predicted C. elegans- human orthologs. The predicted orthologs were further verified through WormBase26, a curated database about the genetics, genomics and biology of C. elegans. An attempt was initially made to collect interactions between the predicted C. elegans orthologs from IntAct, in a manner similar to the Amyloid Interactome, however, only a very small number of such interactions were available. Instead, the STRING database27—a database of known and predicted PPIs—was selected, since it integrates data from various sources and covers a vast number of organisms. STRING primarily uses 4 types of evidence—co-expression, experiments, databases and text min- ing—that it also propagates according to homology; all evidence contribute to a confidence score provided for each PPI. A cutoff interaction score of 0.7 was selected representing interaction partners of high confidence. This dataset was used to evaluate the network alignment algorithms and determine the best parameters that would be used for the subsequent network alignments. APL‑1 and PTL‑1 network from STRING. Afterwards, the interaction partners of APL-1 and PTL-1—the C. elegans orthologs of APP and Tau—were extracted from STRING. A cutoff interaction score of 0.7 was selected representing interaction partners and interactions of high confidence. The predicted human orthologs of those interaction partners were found and constitute the human protein dataset for this pair of networks. For the human network an interaction score of 0.9, representative of the highest confidence level in STRING, was utilized28. www.nature.com/scientificreports/ y g Protein–protein interactions (PPIs) govern most biological processes and investigating them in the context of human diseases is essential to fully comprehend their underlying mechanisms. Networks of PPIs are powerful tools used to conceptualize models of molecular interactions in various biological systems14,15. One of the main benefits of PPI networks is that they allow the conversion of a wealth of raw data into reasonably structured visual representations. Nowadays, due to high-throughput techniques, the growth of available PPI data is exponential. Utilization of this data has allowed the study of PPI networks for entire organisms (e.g. C. elegans16) or for specific diseases(e.g. AD17). Moreover, the availability of organism-wide PPI networks has made cross-species network comparisons possible. A popular method for comparing networks is network alignment, which, analogously to genome alignment, aims at mapping the nodes of two or more networks, and thereby, determining topologically and functionally similar regions18,19. Conserved network regions can be used to transfer biologically relevant information between humans and model organisms18,20. Previous studies21,22 have used PPI networks to explore AD mechanisms using the model organism C. elegans. To our awareness this is the first effort to employ network alignment for the transfer of knowledge between C. elegans and human for the study of this disease.h g g g y The goal of this work was the in silico construction and comparison of Alzheimer-related protein–protein interaction networks in humans and C. elegans. Our main aim was the discovery of common biological pathways that are conserved in both organisms and are potentially implicated in AD. Study of such pathways will guide experimental studies on the model organism C. elegans, and help in the elucidation of mechanisms involved in the pathogenesis of AD. Exploring the conservation of Alzheimer‑related pathways between H. sapiens and C. elegans: a network alignment approach Avgi E. Apostolakou1,3, Xhuliana K. Sula1,3, Katerina C. Nastou1,2, Georgia I. Nasi1 & Therefore, it comes to no surprise that the scientific community has placed much emphasis on understanding the role of these proteins in AD onset.h p The study of age-related diseases, such as AD, in humans is practically impossible, due to both the prohibi- tive nature of an observational study in a human’s lifespan and the serious ethical issues raised on performing experimental studies in humans6. Consequently, researchers use model organisms in order to gain insight into the molecular mechanisms underlying these diseases. Animals with close evolutionary relationships to humans, 1Section of Cell Biology and Biophysics, Department of Biology, National and Kapodistrian University of Athens, 15701 Panepistimiopolis, Athens, Greece. 2Present address: Novo Nordisk Foundation Center for Protein Research, University of Copenhagen, Copenhagen, Denmark. 3These authors contributed equally: Avgi E. Apostolakou and Xhuliana K. Sula. *email: veconom@biol.uoa.gr \| https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 www.nature.com/scientificreports/ Results and discussion In the course of evolution ortholog proteins occasionally maintain the same function across species, as is often reflected by their sequence similarity39. Likewise, PPIs between orthologs can be conserved across organisms and these interactions are referred to as interologs. The aim of this work was to explore both of these concepts in the context of comparing homologous PPI networks in H. sapiens and C. elegans to uncover conserved pathways between these organisms that could be exploited for the study of AD. Three pairs of networks were assembled and compared via GNA (Supplementary Table 5). APP and Tau network from the amyloid interactome. The Amyloid Interactome23 reported 86 inter- action partners for APP and Tau. This resulted in the human network extracted from it, to consist of 88 proteins and 216 PPIs (Fig. 1). Of the 88 human proteins, 61 were matched to 73 (68 unique) predicted protein orthologs in C. elegans; ortholog assignment criteria and mapping can be found in Supplementary File 2 and Supple- mentary Table 1. A query of IntAct using these 68 proteins returned only 6 interactions amongst the C. elegans proteins. Failure to construct a network with data from IntAct, led to the use of STRING instead, and the final C. elegans network used in this study consists of 51 proteins and 69 PPIs (Fig. 1).h g y p g The aforementioned pair of networks was used as a “gold-standard” for evaluating the network alignment algorithms. The performance of the GNA algorithms was assessed based on their ability to both align nodes according to the mapping of predicted orthologs and also identify interologs. Manual alignment of the networks according to predicted ortholog mapping revealed the existence of 22 interologs. An endeavor was made to align the networks with the algorithms relying exclusively on network topological information. This resulted in complete failure of the algorithms to align the two networks, with both MAGNA++ and NETAL unable to correctly align any nodes and with CytoGEDEVO making only 3 correct node alignments (Table 1). In an effort to optimize the alignments, biological information was introduced to MAGNA++ and CytoGEDEVO; NETAL on the contrary does not allow the inclusion of such information. The combination of topological and biologi- cal information was sufficient to allow the two algorithms to successfully align the two networks, as shown in Table 1. Methods i Furthermore, interactions without experimental validation were filtered out at this stage, in order to increase the reliability of collected data. Top 100 interaction partners for APP and Tau & APL‑1 and PTL‑1 from STRING. The final dataset is the least biased and therefore was the one primarily used to transfer biological information between the two organisms. The 100 interaction partners of APL-1 and PTL-1 with the best confidence score were collected from STRING, to create the C. elegans network. A cutoff interaction score of 0.5 was used for interactions between these proteins, to collect at least 100 additional proteins. Using the same approach, the top 100 interaction partners of APP and Tau were collected from STRING to create the human network. In this case a higher interaction cutoff (0.9), https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| www.nature.com/scientificreports/ allowed the retrieval of the top 100 interaction partners and the need to reduce it did not emerge. Once again, interactions without experimental validation were filtered out to increase data reliability. allowed the retrieval of the top 100 interaction partners and the need to reduce it did not emerge. Once again, interactions without experimental validation were filtered out to increase data reliability. Network visualization, alignment and comparison. All collected data was visualized as PPI net- works using Cytoscape 3.7.229. Cytoscape is a freely available platform for biological network visualization and analysis that provides a vast array of applications for specialized functions. Two such applications used in the current study are the stringApp30, which allowed the direct import of networks from the STRING database into Cytoscape and Omics Visualizer31, that was used for visualization purposes in this work. Additionally, Cytoscape.js32 was used to create interactive networks, available via a web interface at (http://thalis.biol.uoa.gr/ celegans_human_AD/), where detailed information about the proteins in the aforementioned networks can be retrieved. Detailed description of the functionalities offered by the web application is available at Supplementary File 1. Comparison between human and C. elegans networks was done using Global Network Alignment (GNA), which aims to locate similarities across entire networks. An array of tools is available to perform GNA, the major- ity of which are based on a cost function for node similarity and attempt to locate the alignment with a maximum node similarity score33. Three algorithms for GNA, namely MAGNA++ 34, CytoGEDEVO35 and NETAL36, were selected and tested to determine the best performing. Methods i All three algorithms produce one-to-one node alignments, where each node in one network is uniquely mapped to a single node in the other network. q y pp g MAGNA++ uses a genetic algorithm to simulate a population of alignments that evolves in time, until the alignment can no longer be improved. A major advantage of MAGNA++ lies in its effort to maximize edge con- servation during the alignment process. CytoGEDEVO is a Cytoscape application that employs the GEDEVO algorithm for network alignment. GEDEVO is based on the graph edit distance between two networks, and attempts to transform one network into the other by applying the minimum number of edge additions and removals. NETAL uses a greedy search algorithm to progressively align node pairs with the best score. An advantage of this tool is that topological information is being renewed during the search for the best alignment. g g g g g To optimize each comparison, GNA traditionally uses biological information on top of topological parameters37. MAGNA++ and CytoGEDEVO, but not NETAL, accept node similarity inputs allowing for the use of biological information to achieve a better network alignment. Protein sequence similarity was calculated and used as a biological information input, where applicable (Supplementary Table 1). To this end, pairwise sequence alignment was done for every human protein against all C. elegans proteins in the dataset using the Needleman-Wunsch algorithm38,which aims at finding the best alignment across the entire protein length. The protein sequence similarity was determined as the percentage of aligned residues that matched, i.e. have similar physicochemical properties. Evaluation of the algorithms was done based on their ability to correctly align the nodes of the networks according to their predicted orthologs. Details regarding the evaluation of the algorithms and a flowchart with the step-by-step procedure followed for the creation and alignment of the “Top 100 interac‑ tion partners” networks are available in Supplementary File 2. Results and discussion MAGNA++ outperformed CytoGEDEVO, both in correctly aligning predicted ortholog pairs and in identifying interologs (Table 1). More detailed information on the evaluation of the GNA algorithms is available in Supplementary File 2. Ultimately, MAGNA++ correctly aligned 33 out of 51 node pairs according to their pre- dicted ortholog mapping as shown in Fig. 1. Due to some proteins mapping to multiple predicted orthologs, and since GNA algorithms work by aligning unique pairs of proteins, it was inherently impossible for MAGNA++ to https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| www.nature.com/scientificreports/ Figure 1. Alignment of human APP and Tau network from Amyloid Interactome and homologous C. eleg Figure 1. Alignment of human APP and Tau network from Amyloid Interactome and homologous C. elegans network by MAGNA++ with combination of biological and topological information, containing both correct and incorrect alignments. The human network (top) and the C. elegans network (bottom) are displayed with the layout of nodes corresponding to the predicted ortholog mapping. Human proteins with no predicted orthologs in the C. elegans network are isolated on the upper side of the human network. Two dotted black lines show the positioning of ortholog pairs APP & APL-1 and Tau & PTL-1 that were not correctly aligned; APP and Tau were not mapped to any C. elegans protein, while APL-1 and PTL-1 were aligned with RNF32 and PRAM1, respectively. All node alignments are available in Supplementary Table 3. Groups of nodes surrounded by dotted lines represent proteins mapped to the same predicted ortholog. These protein nodes have been colored accordingly, to make it easier to detect. Detailed information about the proteins in these networks can be retrieved via the web application (http://thalis.biol.uoa.gr/celegans_human_AD/). Figure 1. Alignment of human APP and Tau network from Amyloid Interactome and homologous C. elegans network by MAGNA++ with combination of biological and topological information, containing both correct and incorrect alignments. The human network (top) and the C. elegans network (bottom) are displayed with the layout of nodes corresponding to the predicted ortholog mapping. Human proteins with no predicted orthologs in the C. elegans network are isolated on the upper side of the human network. Two dotted black lines show the positioning of ortholog pairs APP & APL-1 and Tau & PTL-1 that were not correctly aligned; APP and Tau were not mapped to any C. elegans protein, while APL-1 and PTL-1 were aligned with RNF32 and PRAM1, respectively. Results and discussion MAGNA++ succeeded in correctly aligning the majority of predicted orthologs (48 out of 55 pairs), thus providing further validation of its ability to accurately align pairs of networks with orthologous proteins. Top 100 interaction partners for APP and Tau & APL‑1 and PTL‑1 from STRING. The two afore- mentioned datasets relied on mapping predicted ortholog proteins across the two organisms, a procedure that inevitably introduced bias. Conversely, the “Top 100 interaction partners” dataset is the least biased dataset, and was therefore the one primarily used to transfer biological information between the two organisms. The human network had 102 proteins and 164 PPIs, while the C. elegans network had 102 proteins and 202 PPIs (Fig. 2b). Because PPIs without experimental validation were excluded, a number of proteins in each network are left with no PPIs. Specifically, interaction data was available for 91 human proteins and 82 C. elegans proteins. Alignment of these networks resulted in 82 node aligned pairs and a unified network of 74 aligned protein pairs and 152 conserved interactions (Fig. 2b). g During initial investigation of these networks, the network proteins of both organisms were functionally annotated by searching the literature for information; emphasis was placed on all AD related processes. The vast majority of human proteins in the dataset were found to be AD-associated, thus supporting the relevance of the network in regard to the disease under study. Every association was supported by at least one publication as shown in Supplementary Table 4. Additionally, the evidence provided for the interactions were collected from STRING and the evidence sourced from C. elegans, i.e. not transferred from another organism, were extracted (Supplementary Table 2 and Supplementary File 3).i ( pp y pp y ) Next, focus was applied to the identification of interactions conserved in both human and C. elegans. Com- mon elements between this dataset’s unified network and the unified network of the “APL-1 and PTL-1 network from STRING” dataset were extracted. In total 10 pairs of human and C. elegans proteins were commonly aligned in both approaches, forming the Common network made up of 10 nodes and 9 edges (Fig. 2c). Aside from being commonly aligned, these protein pairs are also predicted orthologs, thereby confirming the conserved interactions between them as interologs. Most of these conserved interactions are linked to post-translational modifications on APP or Tau in human, while information for their role in C. Results and discussion Alignment based on topological information Alignment based on topological and biological information MAGNA++ CytoGEDEVO NETAL MAGNA++ CytoGEDEVO Correct node alignments 0 3 0 33 24 Incorrect node alignments 51 48 51 18 27 Aligned interactions 56 46 40 31 39 Interologs identified 0 0 0 15 12 Table 1. GNA algorithm evaluation results. Three GNA algorithms were evaluated based on their ability to correctly align predicted ortholog proteins (51) and identify interologs (22). Bold is used to indicate the best value for each metric. All algorithms failed to align the networks using topological information exclusively and combination with biological information proved necessary for the correct alignment of protein orthologs. Alignment based on topological information Alignment based on topological and biological information MAGNA++ CytoGEDEVO NETAL MAGNA++ CytoGEDEVO Correct node alignments 0 3 0 33 24 Incorrect node alignments 51 48 51 18 27 Aligned interactions 56 46 40 31 39 Interologs identified 0 0 0 15 12 Alignment based on topological information Alignment based on topological and biological information MAGNA++ CytoGEDEVO NETAL MAGNA++ CytoGEDEVO Correct node alignments 0 3 0 33 24 Incorrect node alignments 51 48 51 18 27 Aligned interactions 56 46 40 31 39 Interologs identified 0 0 0 15 12 Alignment based on topological information Alignment based on topological and biological information MAGNA++ CytoGEDEVO NETAL MAGNA++ CytoGEDEVO Correct node alignments 0 3 0 33 24 Incorrect node alignments 51 48 51 18 27 Aligned interactions 56 46 40 31 39 Interologs identified 0 0 0 15 12 Table 1. GNA algorithm evaluation results. Three GNA algorithms were evaluated based on their ability to correctly align predicted ortholog proteins (51) and identify interologs (22). Bold is used to indicate the best value for each metric. All algorithms failed to align the networks using topological information exclusively and combination with biological information proved necessary for the correct alignment of protein orthologs. their predicted human orthologs. PPIs were collected from STRING for these proteins, resulting in the human network comprising of 127 proteins and 382 PPIs with experimental evidence (Fig. 2a). Alignment of the afore- mentioned networks, with the previously established procedure, returned 61 node aligned pairs and a unified network of 55 pairs of H. sapiens and C. elegans proteins and 92 interologs (Fig. 2a). Results and discussion All node alignments are available in Supplementary Table 3. Groups of nodes surrounded by dotted lines represent proteins mapped to the same predicted ortholog. These protein nodes have been colored accordingly, to make it easier to detect. Detailed information about the proteins in these networks can be retrieved via the web application (http://thalis.biol.uoa.gr/celegans_human_AD/). Figure 1. Alignment of human APP and Tau network from Amyloid Interactome and homologous C. elegans network by MAGNA++ with combination of biological and topological information, containing both correct and incorrect alignments. The human network (top) and the C. elegans network (bottom) are displayed with the layout of nodes corresponding to the predicted ortholog mapping. Human proteins with no predicted orthologs in the C. elegans network are isolated on the upper side of the human network. Two dotted black lines show the positioning of ortholog pairs APP & APL-1 and Tau & PTL-1 that were not correctly aligned; APP and Tau were not mapped to any C. elegans protein, while APL-1 and PTL-1 were aligned with RNF32 and PRAM1, respectively. All node alignments are available in Supplementary Table 3. Groups of nodes surrounded by dotted lines represent proteins mapped to the same predicted ortholog. These protein nodes have been colored accordingly, to make it easier to detect. Detailed information about the proteins in these networks can be retrieved via the web application (http://thalis.biol.uoa.gr/celegans_human_AD/). achieve a “perfect” alignment. Based on these results, all subsequent network alignments were performed using MAGNA++ with a combination of topological and biological information. achieve a “perfect” alignment. Based on these results, all subsequent network alignments were performed using MAGNA++ with a combination of topological and biological information. APL‑1 and PTL‑1 network from STRING. Next, the C. elegans network of APL-1, PTL-1 and their inter- action partners was extracted from STRING, using the stringApp30, and consisted of 61 proteins and 136 PPIs with experimental evidence (Fig. 2a). In a manner similar to above, the C. elegans proteins were mapped to https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| www.nature.com/scientificreports/ Table 1. GNA algorithm evaluation results. Three GNA algorithms were evaluated based on their ability to correctly align predicted ortholog proteins (51) and identify interologs (22). Bold is used to indicate the best value for each metric. All algorithms failed to align the networks using topological information exclusively and combination with biological information proved necessary for the correct alignment of protein orthologs. Results and discussion Below these networks is the unified network resulting from their alignment. (c) The common elements of the two unified networks (pairs of proteins and their PPIs aligned similarly in both datasets). Omics Visualizer31 was used to apply colors on the network’s nodes, with the green half representing the C. elegans protein and the blue half representing the human protein. Each half is labeled with their corresponding gene symbol (gene symbol for Tau is MAPT). All human interactions were annotated with at least one supporting publication (Supplementary Table 4); edges with underlined labels indicate annotations directly associated with AD. On the contrary, only one C. elegans interaction could be annotated as AD related based on literature evidence. Figure 2. Alignments for the “APL-1and PTL-1 network from STRING” and the “Top 100 interaction partners” datasets and their common elements. (a) The C. elegans APL-1 and PTL-1 network from STRING (green) and the homologous human network (blue), with the unified network resulting from their alignment below. (b) The Top 100 interaction partners of APL-1 and PTL-1 for the C. elegans network (green) and of APP and Tau for the human network (blue). Below these networks is the unified network resulting from their alignment. (c) The common elements of the two unified networks (pairs of proteins and their PPIs aligned similarly in both datasets). Omics Visualizer31 was used to apply colors on the network’s nodes, with the green half representing the C. elegans protein and the blue half representing the human protein. Each half is labeled with their corresponding gene symbol (gene symbol for Tau is MAPT). All human interactions were annotated with at least one supporting publication (Supplementary Table 4); edges with underlined labels indicate annotations directly associated with AD. On the contrary, only one C. elegans interaction could be annotated as AD related based on literature evidence. are responsible for the catalytic activity of the γ-secretase complex46. Cleavage of APP by β-secretase followed by γ-secretase, without any α-secretase activity, results in the release of Aβ, according to the amyloidogenic pathway of APP processing47. Thus, catalysis by the γ-secretase complex is crucial both for the production of Aβ and the determination of its length and in turn its propensity to aggregation46. The catalytic subunits of γ-secretase, PSEN1 and PSEN2, were aligned to C. elegans presenilin orthologs HOP-1 and SEL-12, respectively. Results and discussion elegans was available for only one interaction (Fig. 2c). Limited C. elegans evidence were available for 5 of these interactions, the interactions of APL-1 with proteins ULA-1, APH-2 and CDK-5, and the interactions of PTL-1 with CDK-5 and Y71H2AM.20 (Supplementary Table 2 and Supplementary File 3). Functional associations: the APP processing pathway. Taking a closer look at the Common net‑ work in Fig. 2c revealed several interesting associations between the network’s proteins. Located in the Common network and interacting with APP are two proteins involved in APP processing, namely disintegrin and metal- loproteinase domain-containing protein 10 (ADAM10) and nicastrin (NCSTN). ADAM10 is an α-secretase that proteolytically cleaves APP as part of the non-amyloidogenic APP processing pathway, leading to the release of sAPPα, an important peptide for neuronal function, while also prohibiting the production of Aβ40. ADAM10 participates in many important processes, including brain function and development, and is therefore consid- ered a viable drug target; however caution is necessary due to the existence of multiple ADAM10 substrates41. Network alignment paired ADAM10 to its ortholog SUP-17, one of two C. elegans proteins with α-secretase activity with the other protein being ADM-4. C. elegans APL-1, like APP, is cleaved by an α-secretase resulting in the release of sAPL-142. Following α- or β-secretase cleavage, APP is further cleaved by γ-secretase, a complex with four main components: presenilins 1 (PSEN1) and 2 (PSEN2), anterior pharynx-defective-1 (APH1A or APH1B) and nicastrin (NCSTN)43. Nicastrin is thought to act as a receptor binding to APP and other γ-secretase substrates44. This protein was aligned to its ortholog APH-2, a protein that also participates in the γ-secretase complex of C. elegans, functioning similarly to the human nicastrin45. p g g y Investigation of the network of “Top 100 interaction partners” revealed additional elements of the APP pro- essing pathway in the unified network (Fig. 3). These are presenilin 1 (PSEN1) and presenilin 2 (PSEN2) that https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| www.nature.com/scientificreports/ om/scientificreports/ Figure 2. Alignments for the “APL-1and PTL-1 network from STRING” and the “Top 100 interaction partners” datasets and their common elements. (a) The C. elegans APL-1 and PTL-1 network from STRING (green) and the homologous human network (blue), with the unified network resulting from their alignment below. (b) The Top 100 interaction partners of APL-1 and PTL-1 for the C. elegans network (green) and of APP and Tau for the human network (blue). Results and discussion HOP-1 and SEL-12 are parts of the γ-secretase complex, just like their human counterparts, with evidence suggesting that SEL-12, directly or indirectly, regulates the activity of APL-142. Overall, most of the APP processing machinery, and its corresponding orthologs, arose as important common elements during this work and were determined to indeed be conserved in the model organism. A notable exception is β-secretase (BACE1) that is necessary for the production of Aβ peptide and is absent in C. elegans, explaining why Aβ—and not APP—transgenes of this organism are used to model for AD48. https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ Figure 3. Conservation of the APP processing pathway in H. sapiens and C. elegans. Shown here are elements of the APP processing pathway found in the Common Network, extended with relevant proteins and interactions from the “Top 100 interaction partners” dataset. Elements associated with α- and γ-secretase activity are conserved in both organisms, while β-secretase is missing from C. elegans. Even though all interactions that take place in this pathway have been experimentally investigated in human, information related to AD could be retrieved only for two C. elegans interactions. Figure 3. Conservation of the APP processing pathway in H. sapiens and C. elegans. Shown here are elements of the APP processing pathway found in the Common Network, extended with relevant proteins and interactions from the “Top 100 interaction partners” dataset. Elements associated with α- and γ-secretase activity are conserved in both organisms, while β-secretase is missing from C. elegans. Even though all interactions that take place in this pathway have been experimentally investigated in human, information related to AD could be retrieved only for two C. elegans interactions. For the interaction between APL-1 and SUP-17, as well as for the interactions between APH-2 and proteins SEL-12 and HOP-1, an abundance of evidence was available that was not transferred from homology (Supplemen- tary File 3, Figure S14). On the contrary, most of the evidence for the interactions between APL-1 and proteins SEL-12, HOP-1 and APH-2, are transferred from homology (Supplementary Table 2 and Supplementary File 3, Figure S14). Given the central role of the γ-secretase complex in APP processing in humans, these three proteins are promising targets to experimentally study their interactions between APL-1 and the C. elegans γ-secretase complex or to study their role in transgenic C. elegans, models that express the human protein APP. Results and discussion Functional associations: phosphorylation pathways in AD. Aside from APP processing, many pro- teins recovered from the network were involved in phosphorylation (Fig. 4). One of these is cyclin-dependent kinase 5 (CDK5), a prominent kinase in AD, whose aberrant action is associated with pathological processes, including the formation of amyloid plaques and neurofibrillary tangles in AD49. CDK5 phosphorylates APP50, promoting the production of Aβ peptide51, and mediates Tau hyperphosphorylation, leading to its dissociation from microtubules and its aggregation52. Furthermore, Aβ peptide was shown to interact with phosphorylated Tau protein in AD brains, potentially causing neuronal damage53, while another study reported an affinity of Aβ peptide for non-phosphorylated Tau, triggering its phosphorylation and promoting its aggregation54. Modifica- tion of Tau by kinases can however be reversed through the action of phosphatases55. Primarily responsible for dephosphorylation of Tau is protein phosphatase 2A (PP2A), that is reportedly downregulated in AD, thereby contributing to the hyperphosphorylation of Tau56. Located in the Common network is the PP2A activator, PPP2R4, which interacts with Tau. In C. elegans the predicted orthologs of the aforementioned human proteins also interact, however no additional information is available concerning these interactions. Additionally, a num- ber of kinases that phosphorylate Tau were found in the human network of the “Top 100 interaction partners” https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ Figure 4. AD associated phosphorylation-related proteins conserved in H. sapiens and C. elegans. Shown here are elements related to phosphorylation of APP and Tau found in the Common Network, extended with relevant proteins and interactions from the “Top 100 interaction partners” dataset. The interactions appear to be conserved in both organisms, but no information was available for the C. elegans interactions. The gene symbol for Tau is MAPT. Figure 4. AD associated phosphorylation-related proteins conserved in H. sapiens and C. elegans. Shown here are elements related to phosphorylation of APP and Tau found in the Common Network, extended with relevant proteins and interactions from the “Top 100 interaction partners” dataset. The interactions appear to be conserved in both organisms, but no information was available for the C. elegans interactions. The gene symbol for Tau is MAPT. Figure 4. AD associated phosphorylation-related proteins conserved in H. sapiens and C. elegans. Shown here are elements related to phosphorylation of APP and Tau found in the Common Network, extended with relevant proteins and interactions from the “Top 100 interaction partners” dataset. Conclusions l AD is a complex disease still under study that involves an array of proteins and biological pathways. Two critical players in AD are Aβ and Tau, the main components of amyloid plaques and neurofibrillary tangles, respectively. Unveiling the pathological events leading to the development of AD is a prerequisite for the discovery of the effective prevention and treatment of AD. To this end research often relies in the experimentation on model organisms. In this study we attempted to bridge the gap between AD-related processes in H. sapiens and the model organism C. elegans. PPI networks were constructed for the human proteins APP and Tau and for their respective C. elegans orthologs APL-1 and PTL-1. To achieve this we used network alignment in order to iden- tify proteins and interactions conserved in both organisms that are potentially implicated in AD. Amongst the most prominent pathways, that emerged through this analysis, were the APP processing pathway and the Tau phosphorylation pathway. We extensively reviewed these, critical for AD, pathways and it was revealed that they are highly conserved between the two organisms. Most importantly, we discovered that while the majority of interactions involved in these pathways have been studied in human and associated with AD, almost no infor- mation was available about them in C. elegans. Additionally, for some of these interactions the evidence in C. elegans was very limited indicating the need for more experiments to expand the known C. elegans interactome. We therefore managed to showcase significant targets for the study of AD that have been as of yet unexplored in C. elegans. In regard to APP processing, candidates for experimental study are the interactions between the members of the C. elegans γ-secretase complex, SEL-12, APH-2 and HOP-1, as well as their respective interac- tions with APL-1. Also, emerging as a target for exploring the neuroprotective role of the human α-secretase responsible for the non-amyloidogenic processing of APP, was its ortholog SUP-17 and its interaction with APL-1. Additionally, several C. elegans kinases are prime subjects for investigation, namely, KIN-1, KIN-2, R90.1 and F47F2.1 that interact with PTL-1 and CDK-5, a kinase that interacts with both APL-1 and PTL-1. These as well as other pathways can be further investigated to identify more critical proteins for experimental validation in C. elegans. Results and discussion of PTL-1 and further studies focusing on the role of this process in C. elegans and how it relates to AD-related processes in humans is necessary. Conclusions l To enable this functionality, we have created a web application that allows detailed browsing of the networks presented in this work (http://thalis.biol.uoa.gr/celegans_human_AD/). In conclusion, we revealed promising AD-related targets for study in C. elegans and provided a framework for the transfer of knowledge between H. sapiens and C. elegans through the computational study of PPI networks. References 1. Lane, C. A., Hardy, J. & Schott, J. M. Alzheimer’s disease. Eur. J. Neurol. 25, 59–70. https://doi.org/10.1111/ene.13439 (2018). 2. Brookmeyer, R., Johnson, E., Ziegler-Graham, K. & Arrighi, H. M. Forecasting the global burden of Alzheimer’s disease. Alzheimers Dement. 3, 186–191. https://doi.org/10.1016/j.jalz.2007.04.381 (2007). 1. Lane, C. A., Hardy, J. & Schott, J. M. Alzheimers disease. Eur. J. Neurol. 25, 59 70. https://doi.org/10.1111/ene.13439 (2018). 2. Brookmeyer, R., Johnson, E., Ziegler-Graham, K. & Arrighi, H. M. Forecasting the global burden of Alzheimer’s disease. Alzheimer Dement. 3, 186–191. https://doi.org/10.1016/j.jalz.2007.04.381 (2007). p g j j 3. Zhang, Y. W., Thompson, R., Zhang, H. & Xu, H. APP processing in Alzheimer’s disease. Mol. Brain. 4, 3. https://doi org/10.1186/1756-6606-4-3 (2011). org/10.1186/1756-6606-4-3 (2011). 4. Binder, L. I., Guillozet-Bongaarts, A. L., Garcia-Sierra, F. & Berry, R. W. Tau, tangles, and Alzheimer’s disease. Biochim. Biophys. Acta 1739, 216–223. https://doi.org/10.1016/j.bbadis.2004.08.014 (2005). g 4. Binder, L. I., Guillozet-Bongaarts, A. L., Garcia-Sierra, F. & Berry, R. W. Tau, tangles, and Alzheimer’s disease. Biochim. Biophys Acta 1739, 216–223. https://doi.org/10.1016/j.bbadis.2004.08.014 (2005). p g j 5. Benson, M. D. et al. Amyloid nomenclature 2018: recommendations by the International Society of Amyloidosis (ISA) nomen- clature committee. Amyloid 25, 215–219. https://doi.org/10.1080/13506129.2018.1549825 (2018). 6. Mitchell, S. J., Scheibye-Knudsen, M., Longo, D. L. & de Cabo, R. Animal models of aging research: implications for human aging and age-related diseases. Annu. Rev. Animal Biosci. 3, 283–303. https://doi.org/10.1146/annurev-animal-022114-110829 (2015). g g 7. Janus, C. & Welzl, H. Mouse models of neurodegenerative diseases: criteria and general methodology. Methods Mol. Biol. 602 323–345. https://doi.org/10.1007/978-1-60761-058-8_19 (2010).i p g 8. Holm, I. E., Alstrup, A. K. & Luo, Y. Genetically modified pig models for neurodegenerative disorders. J. Pathol. 238, 267–287 https://doi.org/10.1002/path.4654 (2016). p g p 9. Emborg, M. E. Nonhuman Primate Models of Neurodegenerative Disorders. ILAR J. 58, 190–201. https://doi.org/10.1093/ilar/ ilx021 (2017). 10. Alexander, A. G., Marfil, V. & Li, C. Use of Caenorhabditis elegans as a model to study Alzheimer’s disease and other neurodegen- erative diseases. Front. Genet. 5, 279. https://doi.org/10.3389/fgene.2014.00279 (2014). p g g 11. Fitch, W. M. Distinguishing homologous from analogous proteins. Syst. Zool.. 19, 99–113 (1970).i 12. Corsi, A. K., Wightman, B. & Chalfie, M. A transparent window into biology: a primer on caenorhabditis elegans. Genetics 200, 387–407. https://doi.org/10.1534/genetics.115.176099 (2015). 13. Pir, G. J., Choudhary, B. & Mandelkow, E. Caenorhabditis elegans models of tauopathy. FASEB J 31, 5137–5148. https://doi. org/10.1096/fj.201701007 (2017). g fj 4. Liu, W., Wu, A., Pellegrini, M. Received: 1 October 2020; Accepted: 11 January 2021 Results and discussion The interactions appear to be conserved in both organisms, but no information was available for the C. elegans interactions. The gene symbol for Tau is MAPT. dataset and were aligned to C. elegans kinases. Glycogen synthase kinase-type 3β (GSK3B) is crucial for neu- rodevelopment and –like CDK5 –it is a main Tau kinase that contributes to the evident hyperphosphorylation of Tau in AD57. Another Tau kinase is Ca2+/calmodulin (CaM)-dependent protein kinase II (CaMKII)58 that is important for synaptic function and is reportedly dysregulated in AD59. Lastly, p38 mitogen-activated protein kinases (MAPKs), MAPK12 and MAPK13, were found to interact with Tau; p38 MAPK mediates proinflamma- tory signaling, is upregulated in AD and is associated with both Aβ and Tau pathology60. y g g p g β p gy Evidence in C. elegans was available for the interaction of APL-1 and CDK-5, as well as for the interactions between PTL-1 and proteins CDK-5, F47F2.1, KIN-1, KIN-2 and Y71H2AM.20 was mainly based on homology (Supplementary Table 2L and Supplementary File 3, Figure S15). In the case of APL-1 and CDK-5, consider- ing the central role phosphorylation by CDK5 has in the processing of APP61, targeted studies should focus on exploring the phosphorylation potential of the C. elegans ortholog or the lack thereof. More specifically, any conserved phosphorylation sites found would help further elucidate the functional role of APP and would be vital to explain the differences in the aggregation mechanisms in the two organisms. Phosphorylation also has a crucial role in the aggregation of Tau and given the relatively recent shift of focus towards Tau-related drug discovery for AD62, Tau kinases and phosphatases are subjects of intense study63. Despite the importance of Tau, its C. elegans ortholog PTL-1 remains understudied with very few available data regarding interactions with other proteins not transferred from homology. This is evident in the results of this study, as only one PTL-1 phospho- rylation–related interaction has evidence originated in experiments using C. elegans. CDK-5, F47F2.1, KIN-1 and KIN-2 are all kinases aligned to human kinases known to phosphorylate Tau, similarly with Y71H2AM.20 aligned to a human Tau phosphatase. It is therefore possible that these proteins play a role in the phosphorylation https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| www.nature.com/scientificreports/ of PTL-1 and further studies focusing on the role of this process in C. elegans and how it relates to AD-related processes in humans is necessary. www.nature.com/scientificreports/ MAGNA++: maximizing accuracy in global network alignment via both node and edge conservation. Bioinformatics 31, 2409–2411. https://doi.org/10.1093/bioinformatics/btv161 (2015). f p g 35. Malek, M., Ibragimov, R., Albrecht, M. & Baumbach, J. CytoGEDEVO-global alignment of biological networks with cytoscape. Bioinformatics 32, 1259–1261. https://doi.org/10.1093/bioinformatics/btv732 (2016). 35. Malek, M., Ibragimov, R., Albrecht, M. & Baumbach, J. CytoGEDEVO-global alignment of b Bioinformatics 32, 1259–1261. https://doi.org/10.1093/bioinformatics/btv732 (2016). f p g 36. Neyshabur, B., Khadem, A., Hashemifar, S. & Arab, S. S. NETAL: a new graph-based method for global alignment of protein-protein interaction networks. Bioinformatics 29, 1654–1662. https://doi.org/10.1093/bioinformatics/btt202 (2013). f p g 37. Sharan, R. & Ideker, T. Modeling cellular machinery through biological network comparison. Nat. Biotechnol. 24, 427–433. ://doi.org/10.1038/nbt1196 (2006). g 8. Needleman, S. B. & Wunsch, C. D. A general method applicable to the search for similarities in the amino acid sequence of two proteins. J. Mol. Biol. 48, 443–453. https://doi.org/10.1016/0022-2836(70)90057-4 (1970). 39. Dolinski, K. & Botstein, D. Orthology and functional conservation in eukaryotes. Annu. Rev. Genet. 41, 465–507. https://doi. org/10.1146/annurev.genet.40.110405.090439 (2007). g g 0. Asai, M. et al. Putative function of ADAM9, ADAM10, and ADAM17 as APP alpha-secretase. Biochem. Biophys. Res. Commun 301, 231–235. https://doi.org/10.1016/s0006-291x(02)02999-6 (2003). p g 1. Endres, K. & Deller, T. Regulation of alpha-secretase ADAM10 in vitro and in vivo: genetic, epigenetic, and protein-based mecha nisms. Front. Mol. Neurosci 10, 56. https://doi.org/10.3389/fnmol.2017.00056 (2017). g 42. Hornsten, A. et al. APL-1, a Caenorhabditis elegans protein related to the human beta-amyloid precursor protein, is essential for viability. Proc. Natl. Acad. Sci. U S A 104, 1971–1976. https://doi.org/10.1073/pnas.0603997104 (2007).h 3. Takasugi, N. et al. The role of presenilin cofactors in the gamma-secretase complex. Nature 422, 438–441. https://doi.org/10.1038/ nature01506 (2003). 4. Shah, S. et al. Nicastrin functions as a gamma-secretase-substrate receptor. Cell 122, 435–447. https://doi.org/10.1016/j cell.2005.05.022 (2005). ( ) 5. Levitan, D., Yu, G., St George, H. P. & Goutte, C. APH-2/nicastrin functions in LIN-12/Notch signaling in the Caenorhabditis elegans somatic gonad. Dev. Biol. 240, 654–661. https://doi.org/10.1006/dbio.2001.0486 (2001). , , , , g , , g g elegans somatic gonad. Dev. Biol. 240, 654–661. https://doi.org/10.1006/dbio.2001.0486 (2001). 46 De Strooper B Iwatsubo T & Wolfe M S Presenilins and gamma secretase: structure function and role in Alzheimer Disease 6. De Strooper, B., Iwatsubo, T. & Wolfe, M. S. Presenilins and gamma-secretase: structure, function, and role in Alzheimer Disease Cold Spring Harb. Perspect. Med. 2, a006304. https://doi.org/10.1101/cshperspect.a006304 (2012). 7. O’Brien, R. J. & Wong, P. C. References & Wang, X. Integrative analysis of human protein, function and disease networks. Sci. Rep. 5, 14344 https://doi.org/10.1038/srep14344 (2015).i 15. Kikuchi, M. et al. Identification of unstable network modules reveals disease modules associated with the progression of Alzheimer’s disease. PLoS ONE 8, e76162. https://doi.org/10.1371/journal.pone.0076162 (2013). 16. Li, S. et al. A map of the interactome network of the metazoan C. elegans. Science 303, 540–543. https://doi.org/10.1126/scien ce.1091403 (2004). 7. Chen, J. Y., Shen, C. & Sivachenko, A. Y. Mining Alzheimer disease relevant proteins from integrated protein interactome data Pacfic Symp. Biocomput. 50, 367–378 (2006).h i 18. Faisal, F. E., Meng, L., Crawford, J. & Milenkovic, T. The post-genomic era of biological network alignment. EURASIP J. Bioinform. Syst. Biol.. 2015, 3. https://doi.org/10.1186/s13637-015-0022-9 (2015). y g 19. Kelley, B. P. et al. Conserved pathways within bacteria and yeast as revealed by global protein network alignment. Proc. Natl. Acad. Sci. U S A 100, 11394–11399. https://doi.org/10.1073/pnas.1534710100 (2003). p g p ( ) 0. Qian, X. & Yoon, B. J. Comparative analysis of protein interaction networks reveals that conserved pathways are susceptible to HIV-1 interception. BMC Bioinf. 12, 19. https://doi.org/10.1186/1471-2105-12-S1-S19 (2011). https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| www.nature.com/scientificreports/ www.nature.com/scientificreports/ 21. Mukherjee, S. et al. Systems biology approach to late-onset Alzheimer’s disease genome-wide association study identifies novel candidate genes validated using brain expression data and Caenorhabditis elegans experiments. Alzheimers Dement. 13, 1133–1142. https://doi.org/10.1016/j.jalz.2017.01.016 (2017). p g j j ( ) 2. Godini, R., Pocock, R. & Fallahi, H. Caenorhabditis elegans hub genes that respond to amyloid beta are homologs of genes involved in human Alzheimer’s disease. PLoS ONE 14, e0219486. https://doi.org/10.1371/journal.pone.0219486 (2019).h p g j p 3. Biza, K. V. et al. The amyloid interactome: exploring protein aggregation. PLoS ONE 12, e0173163. https://doi.org/10.1371/journ al.pone.0173163 (2017).h p 4. Orchard, S. et al. The MIntAct project–IntAct as a common curation platform for 11 molecular interaction databases. Nucleic Acids Res. 42, D358-363. https://doi.org/10.1093/nar/gkt1115 (2014). p g g 5. Kim, W., Underwood, R. S., Greenwald, I. & Shaye, D. D. OrthoList 2: a new comparative genomic analysis of human and caeno- rhabditis elegans genes. Genetics 210, 445–461. https://doi.org/10.1534/genetics.118.301307 (2018). g g g g 6. Lee, R. Y. N. et al. WormBase 2017: molting into a new stage. Nucleic Acids Res. 46, D869–D874. https://doi.org/10.1093/nar/gkx99 8 (2018). ( ) 27. Szklarczyk, D. et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47, D607–D613. https://doi.org/10.1093/nar/gky1131 (2019).h g p p g g y 8. Szklarczyk, D. et al. The STRING database in 2017: quality-controlled protein-protein association networks, made broadly acces- sible. Nucleic Acids Res. 45, D362–D368. https://doi.org/10.1093/nar/gkw937 (2017).t g g 29. Shannon, P. et al. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504. https://doi.org/10.1101/gr.1239303 (2003). p g g 0. Doncheva, N. T., Morris, J. H., Gorodkin, J. & Jensen, L. J. Cytoscape stringapp: network analysis and visualization of proteomics data. J. Proteome Res. 18, 623–632. https://doi.org/10.1021/acs.jproteome.8b00702 (2019). p g jp 1. Legeay, M., Doncheva, N. T., Morris, J. H. & Jensen, L. J. Visualize omics data on networks with omics visualizer a cytoscape. Appl Res. 9, 157. https://doi.org/10.12688/f1000research.22280.1 (2020). p g 32. Franz, M. et al. Cytoscape.js: a graph theory library for visualisation and analysis. Bioinformatics 32, 309–311. https://doi. org/10.1093/bioinformatics/btv557 (2016). 33. Guzzi, P. H. & Milenkovic, T. Survey of local and global biological network alignment: the need to reconcile the two sides of the same coin. Brief Bioinform. 19, 472–481. https://doi.org/10.1093/bib/bbw132 (2018). f f p g ( ) 4. Vijayan, V., Saraph, V. & Milenkovic, T. Acknowledgementshi g This research has been co-financed by the European Union and Greek national funds through the Operational Program "Competitiveness, Entrepreneurship and Innovation", under the call "RESEARCH-CREATE-INNO- VATE" (Project Code:T1EDK-00353). www.nature.com/scientificreports/ www.nature.com/scientificreports/ 60. Kheiri, G., Dolatshahi, M., Rahmani, F. & Rezaei, N. Role of p38/MAPKs in Alzheimer’s disease: implications for amyloid beta toxicity targeted therapy. Rev. Neurosci. 30, 9–30. https://doi.org/10.1515/revneuro-2018-0008 (2018). y g py p g 61. Liu, F. et al. Regulation of amyloid precursor protein (APP) phosphorylation and processing by p35/Cdk5 and p25/Cdk5. F Lett. 547, 193–196. https://doi.org/10.1016/s0014-5793(03)00714-2 (2003). , p g ( ) ( ) 2. Brunden, K. R., Trojanowski, J. Q. & Lee, V. M. Advances in tau-focused drug discovery for Alzheimer’s disease and related tauopathies. Nat. Rev. Drug Discov. 8, 783–793. https://doi.org/10.1038/nrd2959 (2009). 3. Martin, L. et al. Tau protein kinases: involvement in Alzheimer’s disease. Ageing Res. Rev. 12, 289–309. https://doi.org/10.1016/j arr.2012.06.003 (2013). 63. Martin, L. et al. Tau protein kinases: involvement in Alzheimer’s disease. Ageing Res. Rev. 12, 289–309. https://doi.org/10.1016/j. arr.2012.06.003 (2013). Author contributions tudy design: K.C.N., V.A.I.; Conceptualization: A.E.A., K.C.N., V.A.I.; Methodology: A.E.A., K.C.N.; Software A.E.A., X.K.S., K.C.N.; Visualization: A.E.A., X.K.S., K.C.N.; Data curation: A.E.A., X.K.S.; Formal analysis Study design: K.C.N., V.A.I.; Conceptualization: A.E.A., K.C.N., V.A.I.; Methodology: A.E.A., K.C.N.; Software: A.E.A., X.K.S., K.C.N.; Visualization: A.E.A., X.K.S., K.C.N.; Data curation: A.E.A., X.K.S.; Formal analysis: A.E.A., X.K.S, K.C.N., G.I.N.; Writing-original draft: A.E.A.; Writing-review and editing: A.E.A., K.C.N., G.I.N., Study design: K.C.N., V.A.I.; Conceptualization: A.E.A., K.C.N., V.A.I.; Methodology: A.E.A., K.C.N.; Software: A.E.A., X.K.S., K.C.N.; Visualization: A.E.A., X.K.S., K.C.N.; Data curation: A.E.A., X.K.S.; Formal analysis: A.E.A., X.K.S, K.C.N., G.I.N.; Writing-original draft: A.E.A.; Writing-review and editing: A.E.A., K.C.N., G.I.N., X.K.S., V.A.I.; Funding acquisition: V.A.I.; Supervision: V.A.I.. A.E.A., X.K.S., K.C.N.; Visualization: A.E.A., X.K.S., K.C.N.; Data curation: A.E.A., X.K.S.; Formal analysis: A.E.A., X.K.S, K.C.N., G.I.N.; Writing-original draft: A.E.A.; Writing-review and editing: A.E.A., K.C.N., G.I.N., X.K.S., V.A.I.; Funding acquisition: V.A.I.; Supervision: V.A.I.. A.E.A., X.K.S., K.C.N.; Visualization: A.E.A., X.K.S., K.C.N.; Data curation: A.E.A., X.K.S.; Formal analysis: A.E.A., X.K.S, K.C.N., G.I.N.; Writing-original draft: A.E.A.; Writing-review and editing: A.E.A., K.C.N., G.I.N., X.K.S., V.A.I.; Funding acquisition: V.A.I.; Supervision: V.A.I.. Competing interests h p g The authors declare no competing interests. www.nature.com/scientificreports/ Amyloid precursor protein processing and Alzheimer’s disease. Annu. Rev. Neurosci. 34, 185–204 https://doi.org/10.1146/annurev-neuro-061010-113613 (2011). p g 8. Link, C. D. C. elegans models of age-associated neurodegenerative diseases: lessons from transgenic worm models of Alzheimer’s disease. Exp. Gerontol. 41, 1007–1013. https://doi.org/10.1016/j.exger.2006.06.059 (2006).h p p g j g 49. Liu, S. L. et al. The role of Cdk5 in Alzheimer’s Disease. Mol. Neurobiol. 53, 4328–4342. https://doi.org/10.1007/s12035-015-9369-x (2016).i 0. Iijima, K. et al. Neuron-specific phosphorylation of Alzheimer’s beta-amyloid precursor protein by cyclin-dependent kinase 5. J Neurochem. 75, 1085–1091. https://doi.org/10.1046/j.1471-4159.2000.0751085.x (2000). 1. Lee, M. S. et al. APP processing is regulated by cytoplasmic phosphorylation. J. Cell Biol. 163, 83–95. https://doi.org/10.1083 jcb.200301115 (2003). j 52. Patrick, G. N. et al. Conversion of p35 to p25 deregulates Cdk5 activity and promotes neurodegeneration. Nature 402, 615–622. https://doi.org/10.1038/45159 (1999). p g 53. Manczak, M. & Reddy, P. H. Abnormal interaction of oligomeric amyloid-beta with phosphorylated tau: implications to synaptic dysfunction and neuronal damage. J. Alzheimers Dis. 36, 285–295. https://doi.org/10.3233/JAD-130275 (2013). 54. Guo, J. P., Arai, T., Miklossy, J. & McGeer, P. L. Abeta and tau form soluble complexes that may promote self aggregation of both into the insoluble forms observed in Alzheimer’s disease. Proc. Natl. Acad. Sci. U S A 103, 1953–1958. https://doi.org/10.1073/ pnas.0509386103 (2006). p 5. Martin, L. et al. Tau protein phosphatases in Alzheimer’s disease: the leading role of PP2A. Ageing Res. Rev. 12, 39–49. https://doi org/10.1016/j.arr.2012.06.008 (2013). 56. Liu, F., Grundke-Iqbal, I., Iqbal, K. & Gong, C. X. Contributions of protein phosphatases PP1, PP2A, PP2B and PP5 to the regula- tion of tau phosphorylation. Eur. J. Neurosci. 22, 1942–1950. https://doi.org/10.1111/j.1460-9568.2005.04391.x (2005). l h d h h l f h b b l 57. Flaherty, D. B., Soria, J. P., Tomasiewicz, H. G. & Wood, J. G. Phosphorylation of human tau protein by microtubule-associ- ated kinases: GSK3beta and cdk5 are key participants. J. Neurosci. Res. 62, 463–472. https://doi.org/10.1002/1097-4547(20001 101)62:3%3c463::AID-JNR16%3e3.0.CO;2-7 (2000). 58. Yamamoto, H. et al. Phosphorylation of tau at serine 416 by Ca2+/calmodulin-dependent protein kinase II in neuronal soma in brain. J. Neurochem. 94, 1438–1447. https://doi.org/10.1111/j.1471-4159.2005.03307.x (2005). p g j 9. Ghosh, A. & Giese, K. P. Calcium/calmodulin-dependent kinase II and Alzheimer’s disease. Mol. Brain. 8, 78. https://doi org/10.1186/s13041-015-0166-2 (2015). https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \| © The Author(s) 2021 Additional informationh Additional information Supplementary Information The online version contains supplementary material available at https://doi. org/10.1038/s41598-021-83892-9. Correspondence and requests for materials should be addressed to V.A.I. Reprints and permissions information is available at www.nature.com/reprints. 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) 2021 https://doi.org/10.1038/s41598-021-83892-9 Scientific Reports \| (2021) 11:4572 \|
https://openalex.org/W1920289683	https://www.research-collection.ethz.ch/bitstream/20.500.11850/105370/2/document%2891%29.pdf	English	null	Surgical management of hallux valgus and hallux rigidus: an email survey among Swiss orthopaedic surgeons regarding their current practice	BMC musculoskeletal disorders	2,015	cc-by	5,503	ETH Library RESEARCH ARTICLE Open Access * Correspondence: lukas.iselin@usb.ch 1Department of Orthopaedics and Traumatology, University Hospital Basel, Spitalstrasse 21, CH-3041 Basel, Switzerland Full list of author information is available at the end of the article Author(s): Originally published in: BMC Musculoskeletal Disorders 16, https://doi.org/10.1186/s12891-015-0751-7 This page was generated automatically upon download from the ETH Zurich Research Collection. For more information, please consult the Terms of use. Iselin et al. BMC Musculoskeletal Disorders (2015) 16:292 DOI 10.1186/s12891-015-0751-7 Surgical management of hallux valgus and hallux rigidus: an email survey among Swiss orthopaedic surgeons regarding their current practice Lukas Daniel Iselin1, Georg Klammer2, Norman Espinoza3, Panagiotis D. Symeonidis4, David Iselin5 and Peter Stavrou6 Correspondence: lukas.iselin@usb.ch 1Department of Orthopaedics and Traumatology, University Hospital Basel, Spitalstrasse 21, CH-3041 Basel, Switzerland Full list of author information is available at the end of the article © 2015 Iselin 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. © 2015 Iselin 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. Abstract Background: Various clinical and radiological criteria have been suggested to choose one of the numerous techniques in surgical treatment of hallux valgus and rigidus. We hypothesized that the surgeons' professional background will influence that choice depending on specialization, age, type and institution of training as well as his orthopaedic cultural orientation. Since Switzerland is characterized by regional languages (the most important being German and French), we were interested to learn if the linguistic differences had an influence on the orientation of the surgeons towards e.g. Anglo-American or French surgical traditions and/or sources of literature on the subject. Methods: A survey was e-mailed to all members of the Swiss Orthopaedic Society (SGOT-SSOT). Questions were asked regarding respondents’ demographics as well as their preferred treatment for 3 separate cases of (1) moderate and (2) severe hallux valgus and (3) hallux rigidus. The responses were collected and statistically analyzed. Results: Two hundred thirty of 322 respondents completed the survey(response rate 46 %). as they perform foot surgery on a regular base; 39 % were members of the Swiss Orthopaedic Foot and Ankle Society (SFAS). Selected surgical treatments differed as follows: in joint sparing procedures older and busier surgeons were more likely to use Chevron osteotomies, however more than 50 % preferred a Scarf-type of osteotomy. Along the so-called "Rösti-Graben" separating the French from the German speaking part of Switzerland no significant difference was found in the choice of operation technique. heless the fact being a member of SFAS showed significant differences in technical choice in case Conclusions: There are significant associations between the surgeons’ age, expertise and training and their preferred operative intervention. Considerable differences in the surgical management were found in the practice of the general orthopaedic surgeons 72 and the foot and ankle specialists. The cultural background and training is not mirroring the classical Swiss east west discrepancy. Despite the large number of surgical options available for hallux valgus, only a small number were preferred by the majority of surgeons. Keywords: Bunions, Foot surgery techniques, Forefoot, Toe, Midfoot, Survey © 2015 Iselin 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. Methods A survey was electronically mailed out to members of the Swiss Society of Orthopaedic Surgery and Traumatology (Schweizer Gesellschaft für Orthopädie und Traumatolo- gie, SGOT) including fellows and registrars in the ortho- paedic training program. Participants completed the survey questionary online via a dedicated website which collected and collated the responses. Transla- tions into the three main languages (German, French and Italian) were available. intermetatarsal angle 11°) and a congruent joint without evidence of joint space narrowing. Case 2 (Fig. 2) showed a patient with a severe hallux valgus deformity with significant lateral displacement of the sesamoids (hallux valgus angle 44°, intermetatarsal angle 17°) an incongruent joint and some narrowing of the joint space. Case 3 (Fig. 3a/b) illustrated a patient suffering of advanced degenerative osteoarthritis of the first metatar- sophalangeal joint with significant dorsal osteophyte formation. and Italian) were available. The first question of the survey was if the respondent performed foot and ankle surgery; a positive response allowed them to complete the remainder of the survey. Three separate cases were presented in the survey. Expecting a higher response rate the X-rays illustrating the cases were accompanied with only brief information on patient history in order to minimize the time needed for the completion of the survey. In Figs. 1, 2, 3 the X- rays with corresponding texts (translated in English) as given in the survey are depicted. The first question of the survey was if the respondent performed foot and ankle surgery; a positive response allowed them to complete the remainder of the survey. The questions to the cases asked the participant to state which surgical technique with what type of fixation they would choose and for cases 1–2 if an additional dis- tal soft-tissue release would be performed. Answers were selected as multiple choice options including one for alternative solutions which the participant could state in detail (Table 1). Three separate cases were presented in the survey. Expecting a higher response rate the X-rays illustrating the cases were accompanied with only brief information on patient history in order to minimize the time needed for the completion of the survey. In Figs. 1, 2, 3 the X- rays with corresponding texts (translated in English) as given in the survey are depicted. Case 1 (Fig. Background Fig. 1 Survey case 1: Dorsoplantar weight-bearing radiograph of a patient’s right foot. History of complaints related to her hallux valgus deformity since one year, seeking surgical treatment after conservative measures had failed Hallux valgus and hallux rigidus are common conditions for which numerous operative interventions have been described in the literature [1–4]. Various clinical and radiological criteria have been used to guide the choice of surgical technique [5–7]. The surgeons’ professional background may influence that choice, depending on surgeons’ specialization, age, type and institution of training as well as their orthopaedic cultural orientation [8–20]. In a survey performed with the members of the Australian Orthopaedic Association the surgeon’s mem- bership to the Australian Foot and Ankle Association and age influenced the choice of treatment most. Youn- ger surgeons with a selective foot and ankle training tend to do more Scarf osteotomies in mild to moderate cases and metatarsaophalangeal (MTP)-I fusions in severe Hallux valgus or rigidus. Furthermore a trend to less joint replacements is visible [21]. Switzerland is divided into several distinct cultural and linguistic regions that were formed through the variable influences of the surrounding empires (French, German, Austrian and Italian) over time. An influence of the lan- guage difference on literature search, decision-making and practice due to membership of the surgeon in e.g. French or Anglo-American professional organization could not a priori be excluded. Thus we presented the questions of the Australian sur- vey to Swiss orthopaedic surgeons aiming to identify fac- tors that influenced their choice of treatment with special emphasis on that demographic peculiarity [21]. Fig. 1 Survey case 1: Dorsoplantar weight-bearing radiograph of a patient’s right foot. History of complaints related to her hallux valgus deformity since one year, seeking surgical treatment after conservative measures had failed Abstract 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. Iselin et al. BMC Musculoskeletal Disorders (2015) 16:292 Page 2 of 7 Methods 1) described a patient with a moderate hallux valgus deformity (hallux valgus angle 34°; In order to correlate the chosen treatments to demo- graphic data the following data was collected from each Page 3 of 7 Iselin et al. BMC Musculoskeletal Disorders (2015) 16:292 Fig. 3 a, b Survey Case 3: Oblique and lateral views of a patients left foot. Painful hallux rigidus since one year. Conservative measures have failed and the patients seeks surgical treatment Fig. 2 Survey Case 2: Dorsoplantar weight-bearing radiograph of a patient’s right foot. Complaints related to her hallux valgus deformity lasting since 2 years. She favours surgical treatment as conservative measures had failed Fig. 3 a, b Survey Case 3: Oblique and lateral views of a patients left foot. Painful hallux rigidus since one year. Conservative measures have failed and the patients seeks surgical treatment Fig. 2 Survey Case 2: Dorsoplantar weight-bearing radiograph of a patient’s right foot. Complaints related to her hallux valgus deformity lasting since 2 years. She favours surgical treatment as conservative measures had failed participant of the survey: number of surgical cases on the foot and ankle treated per year; most commonly used language (German, French, Italian, English); princi- pal Swiss region of medical training (primary German speaking-, French speaking-, Italian speaking region or combinations thereof); type of institution of practice (University hospital; public hospital; private practice or mix of private and public); age and membership in the SFAS (Swiss Foot and Ankle Society). Fig. 3 a, b Survey Case 3: Oblique and lateral views of a patients left foot. Painful hallux rigidus since one year. Conservative measures have failed and the patients seeks surgical treatment The responses were collated and then organised into an appropriate format for transfer to a statistical programme, SAS version 9.2 (SAS Institute Inc, Cary, NC, USA). Stat- istical analysis was performed to obtain percentages of all of the responses and chi squared tests were undertaken to investigate for significant statistical relationships between responses and demographic variables. Results The survey was mailed to 654 recipients according the list of members provided by the SGOT with an overall response rate of 46 % (322 responses). Seventy-one per- cent (230) of these stated to perform surgery on the foot and ankle, while 29 % did not (74). There was no need for an ethical approval as the sur- vey did not concern direct patient data according to our institutional review board (directed by the Head of the Orthopaedic Department of the University Hopital Basel, Prof M. Jakob) and the local ethical committee (EKNZ). The study was completed by medical profes- sionals. See questionnaire here (Additional file 1). So the response rate of the survey was actually 35 %. Demographic factors as well as participants case loads, type of training and practice are highlighted in Table 2. In summary all were board certified orthopaedic sur- geons with forty percent of the respondents with a Page 4 of 7 Iselin et al. BMC Musculoskeletal Disorders (2015) 16:292 Table 1 Demographic Information Survey recipients 654 Respondents 322/654 (=overall response rate 46 %) Board certified orthopaedic surgeons 322/322 (100 %) Trainees 0/322 (0 %) Performing Foot & Ankle Surgery 230/322 (71 %) = response rate 35 %) SFAS-Membersa 90/322 (28 %) Survey participants 230 Age groups (years of age) 31–40 35/230 (15 %) 41–50 91/230 (39 %) 51–60 85/230 (38 %) 61–70 18/230 (8 %) Case load (forefoot cases per year) 0–10 8/230 (4 %) < 25 58/230 (25 %) 25–50 71/230 (31 %) > 50 93/230 (40 %) Type of practice Public hospital 97/230 (43 %) Private practice 104/230 (45 %) University hospital 14/230 (6 %) Other institution 13/230 (6 %) Orthopaedic Training Fellowship 52/230 (22 %) Orthopaedic Training Area German part 190/230 (82 %) French part 24/230 (10 %) Italian part 0/230 (0 %) Mix German-French 12/230 (5 %) Mix German-Italian 2/230 (1 %) Mix french-Italian 2/230 (1 %) Language/Region of Practice German-Speaking 183/230 (79 %) French-Speaking 38/230 (17 %) Italian-Speaking 5/230 (2 %) Other 4/230 (2 %) aSFAS = Swiss Foot & Ankle Society; 2 % did not report on membership Table 1 Demographic Information 78 %) and was more likely to be preferred by members of SFAS (60 vs. 25 %, p <0.001). See details in Table 2. In case two (severe hallux valgus) a Lapidus procedure was the most commonly preferred (31 %). Results It was more likely to be performed by members of the SFAS (53 vs. 26 %, p = 0.003) and by those who were less than 50 years old (47 vs. 12 %, p <0.001). 21 % preferred first metatar- sophalangeal joint fusion with 80 % of those choosing a plate and screw construct for their fixation. See details in Table 3. Case 3 (first MTPJ arthritis), first MTPJ fusion was the treatment of choice for the majority of respondents (78 %) (Table 3) with 27 % preferring a plate and screw construct for fixation of the fusion and 60 % choosing screw fixation alone. Joint replacement arthroplasty was preferred by 2.2 % of respondents with a statistically sig- nificant percentage of them being in a practice which was 100 % private. Cheilectomy was chosen by 8 % and was more likely to be undertaken by those greater than 50 years old (12 vs. 1 %, p = 0.002). See details in Table 4. The cultural background analysis did not show any differences regarding the choice of treatment in com- parison with the language. Further statistical analysis see Table 5. Discussion b In case 1 it was noted that a proximal Chevron and a scarf osteotomy was associated with a higher rate of distal soft tissue procedure than the distal type of osteotomy. Fifty percent of the foot and ankle surgeons (F + A) per- formed a scarf while 52 % of the general surgeons did so. Joint replacement arthroplasty was not a very common choice for the management of case 3 (hallux rigidus) in 2 % of the F + A surgeons and 5 % of the general ortho- paedic surgeons, with most respondents preferring to perform a first MTPJ fusion. In case 2 with the severe bunion over 50 % of the F + A surgeons but only a third of the general orthopaedic surgeons would have chosen a scarf osteotomy in com- bination with distal soft tissue releases equal distribution of around 80 %. A Lapidus procedure was the preferred method to fix this condition in 32 % of all SGOT mem- bers while only 3 % of the F + A specialists did choose this here. The main difference was actually seen in the With all of the available treatments described in the literature for hallux valgus it was interesting to note in our study that there were only a few procedures which were preferred by the majority of the swiss orthopaedic surgeons. Five procedures (Scarf, distal chevron, prox- imal chevron, 1st MTPJ fusion and 1st MTPJ replace- ment) accounted for the major percentage of preferred treatments for each of the three case examples [6, 8, 12, 13, 15, 18, 24]. Discussion b Table 3 Results of the survey for procedure of choice for treatment in Case 2 (severe hallux valgus) Lapidus (TMT-I-fusion) 70/230 (31 %) MTP-I-fusion 48/230 (21 %) Scarf 38/230 (16 %) Other 30/230 (13 %) Ludloff 14/230 (6 %) prox Chevron 13/230 (6 %) distal Chevron 9/230 (4 %) Keller’s procedure 6/230 (2 %) none 2/230 (1 %) Table 3 Results of the survey for procedure of choice for treatment in Case 2 (severe hallux valgus) Table 3 Results of the survey for procedure of choice for treatment in Case 2 (severe hallux valgus) Lapidus (TMT-I-fusion) 70/230 (31 %) MTP-I-fusion 48/230 (21 %) Scarf 38/230 (16 %) Other 30/230 (13 %) Ludloff 14/230 (6 %) prox Chevron 13/230 (6 %) distal Chevron 9/230 (4 %) Keller’s procedure 6/230 (2 %) none 2/230 (1 %) The other interesting finding was the significant differ- ences in the practise of general orthopaedic surgeons and F + A specialists. The F + A specialists tend to per- form more modern surgeries such as the scarf osteotomy with additional distal soft tissue releases and usually don’t perform arthroplasties in the highly loaded first MTP joint. Discussion b A Lapidus procedure was the preferred method to fix this condition in 32 % of all SGOT mem- bers while only 3 % of the F + A specialists did choose this here. The main difference was actually seen in the Lap than A asso age or y oste olde was than likel age. repr of o but path Jo choi 2 % paed perf W liter our wer surg ima men trea 13, Th ence and form with don MT Con The age, Table 3 Results of the survey for procedure of choice for treatment in Case 2 (severe hallux valgus) Lapidus (TMT-I-fusion) 70/230 (31 %) MTP-I-fusion 48/230 (21 %) Scarf 38/230 (16 %) Other 30/230 (13 %) Ludloff 14/230 (6 %) prox Chevron 13/230 (6 %) distal Chevron 9/230 (4 %) Keller’s procedure 6/230 (2 %) none 2/230 (1 %) larger proportion of SFAS members choosing a Lapidus for case 2 or a MTP-I-joint fusion in the hallux rigidus over other type of operations [9]. Another demographic variable which we found to be associated with differing choices of treatments was the age of the surgeon. We grouped people into either older or younger than 50 years of age. In case 1, a Mitchell osteotomy was more likely to be performed by those older than fifty. In case 2, a Keller’s excision arthroplasty was more likely to be performed by those surgeons older than fifty whilst in case 3, a Cheilectomy alone was more likely to be performed by those greater than 50 years of age. These three findings were interesting as they may represent a change not only in the teaching and training of orthopaedic surgeons in Switzerland over the past years but possibly also a change in the concepts of aetiology and patho-physiology of hallux valgus as a condition. The scarf osteotomy has been described as a more involved and complex osteotomy than other types (ie distal chevron) and as such may be more commonly per- formed by the surgeon who more regularly performs, or has a more dedicated interest in foot and ankle surgery such as members of the SFAS. Many argue the merits of the scarf osteotomy are that it can provide significant degree of correction with less risk of metatarsal head avascular necrosis and better healing because of the biomechanical stability of the osteotomy [23]. Discussion b We were able to obtain a large number of respondents from our target population, which constituted a repre- sentative sample with an appropriate mix of fellows of the Swiss Orthopaedic Association. The large numbers of respondents and inclusion of orthopaedic surgeons who are not foot and ankle specialists provided results which gave us a good overview of the current treatment practices for forefoot deformity surgery in Switzerland. A weakness of the study is the low survey respond rate of only 35 % of all registered orthopaedic surgeons. On the other hand 71 % of the respondents (230 surgeons in a country with about 8 Mio inhabitants) were perform- ing foot and ankle surgery on a regular base. In a recent survey of academic American orthopaedic foot and ankle surgeons in mild bunion cases 87 % pre- ferring a distal metatarsal osteotomy, followed by a more proximal osteotomy and in 10 % augmented by an add- itional Akin osteotomy [22]. Compared to the in 2012 published survey of Australian orthopaedic surgeons we found less parallels in age and training as well as geographic/cultural differences as expected [9]. FAS = Swiss Foot & Ankle Society; 2 % did not report on membersh special interest in foot and ankle surgery and corre- sponding case loads. For case 1 (moderate hallux valgus) distal Chevron was the most commonly chosen procedure (41 %). 78 % would perform a distal soft tissue (McBride) procedure in addition. Scarf osteotomy was the next most com- monly chosen procedure (36 %). The correction was more likely to be accompanied by a McBride procedure It was interesting to note that the classical east–west differences in cultural and language in Switzerland is not correlated with the type of treatment chosen in forefoot surgery. Nevertheless being a member of the SFAS did appear to be a significant factor in the choice of the type of correction preferred in the more severe cases with a Page 5 of 7 Iselin et al. Discussion b BMC Musculoskeletal Disorders (2015) 16:292 Table 2 Results of the survey for procedure of choice for treatment in Case 1 (mild hallux valgus) distal Chevron 95/230 (41 %) Scarf 84/230 (36 %) other 23/230 (10 %) ReveL 9/230 (3.5 %) prox Chevron 8/230 (3.5 %) Lapidus (TMT-I fusion) 4/230 (2 %) Ludloff 4/230 (2 %) Keller’s procedure 2/230 (1 %) Bunionectomy 1/230 Table 4 Results of the survey for procedure of choice for treatment in Case 3 (hallux rigidus) Table 4 Results of the survey for procedure of choice for treatment in Case 3 (hallux rigidus) MTP-I-fusion 177/230 (78 %) Cheilectomy 20/230 (8 %) Other 19/230 (8 %) Joint replacement 5/230 (2 %) Keller’s procedure 2/230 (1 %) Interposition arthroplasty 2/230 (1 %) none 1/230 comparison of the french swiss to the German speeking swiss orthopaedic surgeons. It is clearly seen that the Lapidus procedure is more often used in the french part than in the German speeking. larger proportion of SFAS members choosing a Lapidus for case 2 or a MTP-I-joint fusion in the hallux rigidus over other type of operations [9]. The scarf osteotomy has been described as a more involved and complex osteotomy than other types (ie distal chevron) and as such may be more commonly per- formed by the surgeon who more regularly performs, or has a more dedicated interest in foot and ankle surgery such as members of the SFAS. Many argue the merits of the scarf osteotomy are that it can provide significant degree of correction with less risk of metatarsal head avascular necrosis and better healing because of the biomechanical stability of the osteotomy [23]. In case 1 it was noted that a proximal Chevron and a scarf osteotomy was associated with a higher rate of distal soft tissue procedure than the distal type of osteotomy. Fifty percent of the foot and ankle surgeons (F + A) per- formed a scarf while 52 % of the general surgeons did so. In case 2 with the severe bunion over 50 % of the F + A surgeons but only a third of the general orthopaedic surgeons would have chosen a scarf osteotomy in com- bination with distal soft tissue releases equal distribution of around 80 %. References 1. Coughlin MJ, Jones CP. Hallux Valgus: Demographics, Etiology, and Radiographic Assessment. Foot Ankle Int. 2007;28:759–77. g p 2. Hueter C. In Klinic der Gelenkkrankungen mit Einschluss der Orthopaedie. Vogel, Leipzig: Edited; 1870. 2. Hueter C. In Klinic der Gelenkkrankungen mit Einschluss der Orthopaedie. Vogel, Leipzig: Edited; 1870. 3. Mann RA, Coughlin MJ. Adult hallux valgus. In: Coughlin MJ, Mann RA, editors. Surgery of the Foot and Ankle. 7th ed. St. Louis: Mosby; 1999. p. 151–269. 3. Mann RA, Coughlin MJ. Adult hallux valgus. In: Coughlin MJ, Mann RA, editors. Surgery of the Foot and Ankle. 7th ed. St. Louis: Mosby; 1999. p. 151–269. 4. Silver D. The operative treatment of hallux valgus. J Bone Joint Surg. 1923;5:225. 4. Silver D. The operative treatment of hallux valgus. J Bone Joint Surg. 1923;5:225. 5. Nyska M. Principles of first metatarsal osteotomies. Foot Ankle Clin. 2001;6:399–408. 5. Nyska M. Principles of first metatarsal osteotomies. Foot Ankle Clin. 2001;6:399–408. 6. Plattner PF, Van Manen JW. Results of Akin type proximal phalangeal osteotomy for correction of hallux valgus deformity. Orthopedics. 1990;13:989–96. Additional file 1: Appendix: Questionnaire. (DOCX 32 kb) 8. Akin OF. The treatment of hallux valgus: a new operative procedure and its results. Med Sentinel. 1925;33:678–9. Competing interests The authors declare that they have no competing interests. 12. Johnson JE, Clanton TO, Baxter DE, Gottlieb MS. Comparison of chevron osteotomy and modified McBride bunionectomy for correction of mild to moderate hallux valgus deformity. Foot Ankle. 1991;12:61–8. Authors’ contributions LDI and GK equally contributed to the writing of the paper. LDI and GK carried out the background literature research. LDI drafted the manuscript PDG and PS revised the paper. LDI, GK, NE made substantive intellectual contributions to the planning of the study. DI was responsible for the statistical analysis. All authors read and approved the final manuscript. 13. Jones S, Al Hussainy HA, Ali F, Betts RP, Flowers MJ. Scarf osteotomy for hallux valgus. A prospective clinical and pedobarographic study. J Bone Joint Surg. 2004;86-B:830–6. 13. Jones S, Al Hussainy HA, Ali F, Betts RP, Flowers MJ. Scarf osteotomy for hallux valgus. A prospective clinical and pedobarographic study. J Bone Joint Surg. 2004;86-B:830–6. 14. Keller WL. The surgical treatment of bunions and hallux valgus. NY Med J. 1912;95:696. 14. Keller WL. The surgical treatment of bunions and hallux valgus. NY Med J. 1912;95:696. 15. Leventen EO. Etiology and treatment of hallux valgus. The chevron procedure. Orthopedics. 1990;13:973–6. Acknowledgements 16. Mann RA, Clanton TO. Hallux Rigidus: Treatment by Cheilectomy. J Bone Joint Surg. 1988;70-A(3):400–6. We would like to thank the Swiss Orthopaedic Society (SGOT aka swiss orthopaedics) for the support. 16. Mann RA, Clanton TO. Hallux Rigidus: Treatment by Cheilectomy. J Bone Joint Surg. 1988;70-A(3):400–6. 17. McBride ED. A conservative operation for bunions. J Bone Joint Surg. 1928;10-A:735–9. 17. McBride ED. A conservative operation for bunions. J Bone Joint Surg. 1928;10-A:735–9. Conclusion There are significant associations between the surgeons’ age, expertise and training and their preferred operative Iselin et al. BMC Musculoskeletal Disorders (2015) 16:292 Page 6 of 7 Table 5 Statistical analysis by Pearson's Chi squared test Comparison p-value Significant difference Language vs Choice Case 1 p-value = 0.47 No Language vs Choice Case 2 p-value = 0.28 No Language vs Choice Case 3 p-value = 0.11 No SFAS membership vs number of operations p-value <0.001 Yes (SFAS members do operate more F&A cases than not members SFAS membership vs Language background p-value = 0.86 No differences SFAS membership vs Choice Case 1 p-value = 0.89 No SFAS membership vs Choice in Case 2 p-value < 0.001 Yes SFAS membership vs Choice in Case 3 p-value = 0.014 Yes Age and operations p-value = 0.52 No differences in age and operations Konjunkturforschungsstelle, ETH Zürich, Zürich, Switzerland. 6Adelaide Orthosports Clinic, Adelaide, SA, Australia. intervention. Considerable differences were found in the practice of the general orthopaedic surgeons and the foot and ankle specialists. The cultural background and training is not demonstrating the expected classical Swiss east–west discrepancy. Despite the large number of surgical options available for hallux valgus, only a small number were preferred by the majority of surgeons. While we are all anecdotally aware that lesser deformity is treated with distal osteotomies and more severe deformity with a proximal osteotomy, we are aware of only limited data in the literature that verifies this. We could show dif- ferences in the swiss orthopaedic population in regard to the membership of the specialist’s society but the “Rösti- Graben” seems not to be as deep as mostly seen in the cultural and social correlation. Received: 1 March 2015 Accepted: 4 October 2015 Abbreviations 9. Barouk LS, Barouk P, Baudet B, Toullec E. The great toe proximal phalanx osteotomy: the final step of the bunionectomy. Foot Ankle Clin. 2005;10:141–55. SGOT: Schweizer Gesellschaft für Orthopädie und Traumatologie/Swiss Society of Orthopaedic Surgery and Tramatology; SFAS: Swiss foot and ankle society. 10. Bordelon RL. Technique and use of Akin osteotomy. Contemp Orthop. 1991;23:38–41. 10. Bordelon RL. Technique and use of Akin osteotomy. Contemp Orthop. 1991;23:38–41. 11. Hattrup SJ, Johnson KA. Chevron osteotomy: analysis of factors in patients’ dissatisfaction. Foot Ankle. 1985;5:327–32. 11. Hattrup SJ, Johnson KA. Chevron osteotomy: analysis of factors in patients’ dissatisfaction. Foot Ankle. 1985;5:327–32. Additional file 7. Regnauld B. Disorders of the Great Toe. In: Elson R, editor. The Foot: Pathology, Aetiology, Seminology, Clinical Investigation and Treatment. New York: Springer-Verlag; 1986. p. 269–81. 344 – 349. Pathology, Aetiology, Seminology, Clinical Investigation and Treatment. New York: Springer-Verlag; 1986. p. 269–81. 344 – 349. 8. Akin OF. The treatment of hallux valgus: a new operative procedure and its results. Med Sentinel. 1925;33:678–9. Additional file 1: Appendix: Questionnaire. (DOCX 32 kb) Author details 1 18. Mitchell CL, Fleming JL, Allen R, Glenney C, Sanford GA. Osteotomy- bunionectomy for hallux valgus. J Bone Joint Surg. 1958;40-A:41–58. 1Department of Orthopaedics and Traumatology, University Hospital Basel, Spitalstrasse 21, CH-3041 Basel, Switzerland. 2Department of Orthopaedics and Traumatology, Kantonsspital Luzern, Switzerland. 3FussInstitut Zürich, Zürich, Switzerland. 4First Orthopaedic Department, Aristotle, University of Thessaloniki, “G. Papanikolaou” Hospital, Thessaloniki, Greece. 5KOF 18. Mitchell CL, Fleming JL, Allen R, Glenney C, Sanford GA. Osteotomy- bunionectomy for hallux valgus. J Bone Joint Surg. 1958;40-A:41–58. 19. Raikin SM, Ahmad J, Pour AE, Abidi N. Comparison of Arthrodesis and Metallic Hemiarthroplasty of the Hallux Metatarsophalangeal Joint. J Bone Joint Surg. 2007;89:1979–85. 19. Raikin SM, Ahmad J, Pour AE, Abidi N. Comparison of Arthrodesis and Metallic Hemiarthroplasty of the Hallux Metatarsophalangeal Joint. J Bone Joint Surg. 2007;89:1979–85. Page 7 of 7 Page 7 of 7 Iselin et al. BMC Musculoskeletal Disorders (2015) 16:292 20. Thordarson DB, Ebramzadeh E, Moorthy M, Lee J, Rudicel S. Correlation of hallux valgus surgical outcome with AOFAS forefoot score and radiological parameters. Foot Ankle Int. 2005;26:122–7. 21. Iselin LD, Munt J, Symeonidis PD, Klammer G, Chehade M, Stavrou P. Operative management of common forefoot deformities: a representative survey of Australian orthopaedic surgeons. Foot Ankle Spec. 2012;5(3):188–94. 22. Pinney S, Song K, Chou L. Surgical Treatment of Mild Hallux Valgus Deformity: The State of Practice among Academic Foot and Ankle Surgeons. Foot Ankle Int. 2006;27:970–3. 23. Crevoisier X, Mouhsine E, Ortolano V, Udin B, Dutoit M. The scarf osteotomy for the treatment of hallux valgus deformity: a review of 84 cases. Foot Ankle Int. 2001;22:970–6. 24. Schneider W, Aigner N, Pinggera O, Knahr K. Chevron osteotomy in hallux valgus. Ten-year results of 112 cases. J Bone Joint Surg. 2004;86-B:1016–20. 20. Thordarson DB, Ebramzadeh E, Moorthy M, Lee J, Rudicel S. Correlation of hallux valgus surgical outcome with AOFAS forefoot score and radiological parameters. Foot Ankle Int. 2005;26:122–7. 21. Iselin LD, Munt J, Symeonidis PD, Klammer G, Chehade M, Stavrou P. Operative management of common forefoot deformities: a representative survey of Australian orthopaedic surgeons. Foot Ankle Spec. 2012;5(3):188–94. 22. Pinney S, Song K, Chou L. Surgical Treatment of Mild Hallux Valgus Deformity: The State of Practice among Academic Foot and Ankle Surgeons. Foot Ankle Int. 2006;27:970–3. 24. Schneider W, Aigner N, Pinggera O, Knahr K. Chevron osteotomy in hallux valgus. Ten-year results of 112 cases. J Bone Joint Surg. 2004;86-B:1016–20. Iselin et al. BMC Musculoskeletal Disorders (2015) 16:292 20. Thordarson DB, Ebramzadeh E, Moorthy M, Lee J, Rudicel S. Correlation of hallux valgus surgical outcome with AOFAS forefoot score and radiological parameters. Foot Ankle Int. 2005;26:122–7. 21. Iselin LD, Munt J, Symeonidis PD, Klammer G, Chehade M, Stavrou P. Operative management of common forefoot deformities: a representative survey of Australian orthopaedic surgeons. Foot Ankle Spec. 2012;5(3):188–94. 22. Pinney S, Song K, Chou L. Surgical Treatment of Mild Hallux Valgus Deformity: The State of Practice among Academic Foot and Ankle Surgeons. Foot Ankle Int. 2006;27:970–3. 23. Crevoisier X, Mouhsine E, Ortolano V, Udin B, Dutoit M. The scarf osteotomy for the treatment of hallux valgus deformity: a review of 84 cases. Foot Ankle Int. 2001;22:970–6. 24. Schneider W, Aigner N, Pinggera O, Knahr K. Chevron osteotomy in hallux valgus. Ten-year results of 112 cases. J Bone Joint Surg. 2004;86-B:1016–20. 23. Crevoisier X, Mouhsine E, Ortolano V, Udin B, Dutoit M. The scarf osteotomy for the treatment of hallux valgus deformity: a review of 84 cases. Foot Ankle Int. 2001;22:970–6. Author details 1 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
https://openalex.org/W4283726849	http://serd.artvin.edu.tr/tr/download/article-file/1887754	Turkish	null	Comparison of longitudinal studies in science education in terms of methodological characteristics	DergiPark (Istanbul University)	2,022	cc-by	6,415	* Bu çalışma 21-23 Haziran 2021’de Artvin İl Milli Eğitim Müdürlüğü ve Artvin Çoruh Üniversitesi iş birliği ile gerçekleştirilen Uluslararası Covid-19 Kongresi: Eğitimde Yeni Normlar II” kongresinde sözlü bildiri olarak sunulan ve özeti yayımlanmış olan bildirinin tam metnidir. 1 ORCID: 0000-0003-0299-9340, Aksaray Üniversitesi Fen Bilimleri Enstitüsü, suleymankahi@gmail.com 2 ORCID: 0000-0002-3247-0117, Aksaray Üniversitesi Fen Bilimleri Enstitüsü, recepcetn@gmail.com 3 ORCID: 0000-0003-3584-7132, Aksaray Üniversitesi Eğitim Fakültesi, rceken@gmail.com 4 ORCID: 0000-0002-4210-7733, Aksaray Üniversitesi Eğitim Fakültesi, hozcan@aksaray.edu.tr 5 ORCID: 0000-0002-3576-2633, Aksaray Üniversitesi Eğitim Fakültesi, tctunc@gmail.com * Bu çalışma 21-23 Haziran 2021’de Artvin İl Milli Eğitim Müdürlüğü ve Artvin Çoruh Üniversitesi iş birliği ile gerçekleştirilen Uluslararası Covid-19 Kongresi: Eğitimde Yeni Normlar II” kongresinde sözlü bildiri olarak sunulan ve özeti yayımlanmış olan bildirinin tam metnidir. Anahtar Kelimeler: Fen eğitimi, boylamsal araştırmalar, doküman analizi, uzaktan eğitim. Anahtar Kelimeler: Fen eğitimi, boylamsal araştırmalar, doküman analizi, uzaktan eğitim. Abstract Studies conducted to examining scientific research published in the field of science education are important in terms of contributing to the recognition of the methodological features that are frequently or limitedly preferred in related studies. From such factors, longitudinal studies are limited studies due to cost and long duration. In this study, longitudinal studies on science education were examined in terms of methodological features, year of publication, philosophical basis, data collection technique, number of participants, target group and data collection process. In the research, which was carried out in accordance with the qualitative study approach, the relevant researches carried out in Turkey from 2004 to 2021 were examined by document analysis. The results reveal that no longitudinal studies have been carried out for science education from the second half of 2020 to 2021. For the 21 longitudinal studies discussed, it can be stated that it is not carried out with secondary school students, who are among the target groups of science education. In addition, it has been determined that more place is given to pre-service teachers who continue their education in the relevant field and the teachers and teachers as the target participants in these studies. These results may offer some perspectives on how researchers working in the related field can identify the target group in their longitudinal research and how they can combine research processes and new opportunities provided by technology. Keywords: Science education, longitudinal studies, document analysis, distance education 1 1 Studies in Educational Research and Development, 2022, 6(1) Başvuru: 19.07.2021 Başvuru: 19.07.2021 Alıntılanma Önerisi: Çeken, R. (2022). Fen Eğitiminde Gerçekleştirilen Boylamsal Araştırmaların Yöntemsel Özellikler Bakımından Incelenmesi. Studies in Educational Research and Development, 6(1), 1- 19. Öz Fen bilimleri eğitimi alanında yayımlanmış olan bilimsel araştırmaların incelenmesine yönelik olarak gerçekleştirilen çalışmalar, ilgili araştırmalarda sıklıkla veya sınırlı düzeyde tercih edilen yöntemsel özelliklerin neler olduğunun fark edilmesine katkı sunması bakımından önem taşımaktadır. Bunlardan boylamsal araştırmalar, maliyetli olması ve uzun sürmesi gibi nedenler ile sınırlı düzeyde gerçekleştirilen çalışmalardır. Nitel araştırma desenine uygun olarak yürütülen bu çalışmada, 2004 ve 2021 yılları arasında yurt içinde gerçekleştirilmiş araştırmalar doküman analizi ile yayımlandığı yıl, dayandığı felsefi temel, veri toplama tekniği, katılımcı sayısı, hedef kitle ve veri toplama süreci bağlamında incelenmiştir. Sonuçlar, 2020 yılının ikinci yarısından 2021 yılına kadar fen eğitimine yönelik olarak herhangi bir boylamsal çalışmanın gerçekleştirilmediğini ortaya koymaktadır. Ele alınan 21 adet boylamsal çalışma için genel olarak fen eğitiminin hedef kitlelerinden olan ortaokul öğrencileri ile gerçekleştirilmediği ifade edilebilir. Bunun yanında, söz konusu çalışmalarda ilgili alanda öğrenimlerine devam etmekte olan öğretmen adaylarına ve öğretmenlere daha çok yer verildiği tespit edilmiştir. Bu sonuçlar, ilgili alana yönelik olarak çalışan araştırmacılarının boylamsal araştırmalarında hedef kitle Studies in Educational Research and Development, 2022, 6(1) 2 2 belirlemelerine ve araştırma süreçleri ile teknolojinin sağladığı yeni olanakları nasıl birleştirebileceklerine yönelik bazı bakış açıları sunabilir. belirlemelerine ve araştırma süreçleri ile teknolojinin sağladığı yeni olanakları nasıl birleştirebileceklerine yönelik bazı bakış açıları sunabilir. Giriş 2020 yılı ile birlikte tüm dünyada etkisi görülmeye başlanan koronavirüs pandemisi, yayılım hızındaki artış nedeniyle birçok alanda olumsuz sonuçlara yol açmıştır. İlk zamanlar anlık tedbirlerle süreç aşılmaya çalışılsa da söz konusu sürecin beklenilenden uzun sürmesi kalıcı çözümler için yeni arayışları gündeme getirmiştir. Eğitim alanı da gerek eğitim-öğretim faaliyeti gerekse eğitim araştırmaları bakımından süreçten en ciddi ölçüde etkilenen sektörler arasında yer almaktadır ( La Velle, Newman, Montgomery ve Hyatt, 2020). Bu nedenle bilim insanları eğitim ile ilgili çalışmalarını yürütebilmek için alternatif süreçler tasarlamaya başlamışlardır. Teknolojinin bu kadar yaygın olarak kullanıldığı günümüzde eğitim ve öğrenme süreçlerinin dijital ortamlara taşınması, bu sürecin aksamaması bakımından hızlı bir şekilde alınmış ve uygulamaya konulmuş olan kararlardan biri olmuştur. Ülkemiz de bu hızlı değişimden etkilenmiş, bu amaçla gerekli altyapı çalışmalarını hızlandırmış, eğitimi uzaktan bir şekilde Studies in Educational Research and Development, 2022, 6(1) 3 3 planlamaya ve uygulamaya başlamıştır. Dijital ortamda gerçekleştirilmekte olan uzaktan eğitim süreci öğrenci, öğretmen, ebeveyn, araştırmacılar ve diğer yetişkinler için çeşitli olumlu ve olumsuz durumlara da yol açmıştır (Williamson, Eynon ve Potter, 2020). Söz konusu dijital sürecin işleyişi ve katılımcılarına olası etkileri, eğitim araştırmalarına da konu olmuştur. planlamaya ve uygulamaya başlamıştır. Dijital ortamda gerçekleştirilmekte olan uzaktan eğitim süreci öğrenci, öğretmen, ebeveyn, araştırmacılar ve diğer yetişkinler için çeşitli olumlu ve olumsuz durumlara da yol açmıştır (Williamson, Eynon ve Potter, 2020). Söz konusu dijital sürecin işleyişi ve katılımcılarına olası etkileri, eğitim araştırmalarına da konu olmuştur. Eğitim sürecinin önemli özelliklerinden birisi de öğrencilerde etkili, kalıcı ve anlamlı öğrenmeyi sağlamaya yönelik olmasıdır. Öğrencilerin katılımcı oldukları söz konusu süreçlerde bu kalıcılığının ne derecede sağlandığının anlaşılabilmesi amacı ile farklı zaman dilimlerini içerecek şekilde tekrarlı testlerin uygulanması ve bunlardan derlenen verilerin karşılaştırılmasının yapılması gerekmektedir. Deneysel araştırmalar ve özellikle boylamsal çalışmalar sözü edilen tekrarlı ölçümler için araştırmacılara önemli fırsatlar sunmaktadır (Lei ve Zhao, 2007). Büyüköztürk vd. (2017) boylamsal araştırmaları, çalışma grubundan farklı zaman aralıklarıyla veri toplanarak, öğrenmelerin kalıcılığının ve bireylerin eğilimlerinin tespit edildiği araştırmalar olarak ifade etmektedirler. Tarama araştırmaları boylamsal, kesitsel, anlık ve geçmişe dönük çalışmalar olarak sınıflandırılmaktadır. Düşüncelerin, beklentilerin, algıların ve olayların zamanla nasıl gelişip değiştiği veya değişmediğini ortaya koyabilmek için boylamsal ya da kesitsel araştırmalara başvurulur. Boylamsal çalışmalarda incelenen hedef kitleye ait örneklem grubunun belli zaman dilimleri ile tekrarlı ölçme işlemlerine tabi tutulması ile gerçekleştirilir (Zeegers, 2001). Boylamsal tarama araştırmalarında eğilim belirlemeye, ortak özelliği olan bir grubu incelemeye ya da aynı kişilerin zamana bağlı değişimlerini, eğilimlerini araştırmaya odaklanılır (Büyüköztük vd., 2017). Boylamsal çalışmaların bazı olumsuz özellikleri de bulunmaktadır. Giriş Bu tür çalışmalar, bireyin eğitim ve öğrenme sürecindeki gelişimlerine yönelik olarak daha gerçekçi olabilen bulgular ortaya koysa da aynı veri toplama araçlarının belli bir zaman dilimi içinde tekrarlanarak aynı örneklem gruba uygulanması, ilgili grubun veri toplama aracına olan ilgisinin azalmasına yol açabilme olasılığı taşımaktadır. Buna ilave olarak, veri toplama süresinin uzun olması, veri kaybının yaşanması ihtimali, oldukça pahalı olabilmesi gibi zorlukların olması, boylamsal araştırmaların tercih edilmemesine yol açabilmektedir (Kıryak, Candaş, Çalık ve Zeybek, 2020). İlgili Araştırmalar Fen bilimleri eğitimi alanında yayımlanan sınırlı sayıda boylamsal çalışmalar incelendiğinde, söz konusu araştırmalarda şu konulara odaklandığı saptanmıştır: Emrahoğlu ve Öztürk (2009) yaptıkları çalışmada fen bilgisi öğretmen adaylarının temel astronomi konularındaki kavramlarını anlama seviyelerini ve kavram yanılgılarının, birinci sınıftan dördüncü sınıfa kadar değişimlerini incelemişlerdir. Öğretmen adaylarının birçok kavram yanılgılarıyla öğrenim hayatlarını tamamladıklarını ve öğretmen adayları ile ilköğretim çağındaki öğrencilerin benzer yanılgılara sahip olduklarını tespit etmişlerdir. Başka bir araştırmada, 7.sınıf öğrencilerinin sindirim sistemi konusunda sahip oldukları kavram yanılgılarının Studies in Educational Research and Development, 2022, 6(1) Studies in Educational Research and Development, 2022, 6(1) 4 kökenleri belirlenmeye çalışılmıştır (Güngör, 2009). Üç yıl süren çalışmada öğrencilerdeki kavram yanılgılarının kaynağı ile ilgili çeşitli tespitlerde bulunulmuştur. Kahraman (2020), fen bilimleri öğretmen adayları ile yaptığı çalışmada genetik mühendisliği, biyoteknoloji ve klonlama kavramlarına yönelik algılarındaki değişimi incelemiştir. Aldıkları derslerin de etkisiyle bu konulardaki bilişsel yapılarının geliştiği, kavram yanılgılarının azaldığı tespit edilmiştir. Ayrıca çalışmada veri toplama aracı olarak kullanılan kelime ilişkilendirme testlerinin eğitim materyali olarak kullanılması önerilmiştir. kökenleri belirlenmeye çalışılmıştır (Güngör, 2009). Üç yıl süren çalışmada öğrencilerdeki kavram yanılgılarının kaynağı ile ilgili çeşitli tespitlerde bulunulmuştur. Kahraman (2020), fen bilimleri öğretmen adayları ile yaptığı çalışmada genetik mühendisliği, biyoteknoloji ve klonlama kavramlarına yönelik algılarındaki değişimi incelemiştir. Aldıkları derslerin de etkisiyle bu konulardaki bilişsel yapılarının geliştiği, kavram yanılgılarının azaldığı tespit edilmiştir. Ayrıca çalışmada veri toplama aracı olarak kullanılan kelime ilişkilendirme testlerinin eğitim materyali olarak kullanılması önerilmiştir. Bursal (2013) çalışmasında, dördüncü sınıftan sekizinci sınıfa kadar öğrencilerin akademik başarılarını yılsonu notları üzerinden sınıf düzeyi ve cinsiyet değişkenlerini dikkate alarak incelemiştir. Kız öğrencilerin başarılarının erkek öğrencilerin başarılarından sınıf düzeyi yükseldikçe anlamlı olarak artış gösterdiğini tespit etmiş, cinsiyete dayalı farkın azaltılması için önerilerde bulunmuştur. Yapılan bir diğer araştırmada mesleğe yeni başlayan öğretmenlerle pedagojik ve epistemolojik inançlar üzerine çalışılmıştır. Üç yıl süren bu çalışmada araştırmacı, öğretmenlerin pedagojik değişimlerin mesleki tecrübe ile zamanla arttığını ancak epistemolojik inançlarının daha zor değiştiğini gözlemlemiştir (Doğan, 2014). Bilici ve Baran (2015) fen bilimleri öğretmenlerinin teknolojik pedagojik alan bilgisine yönelik öz yeterlik inançlarını belirlemek için yaptıkları çalışmada katılımcılara eğitim uygulanmış, uygulamanın hemen sonrasında öğretmenlerin teknolojik pedagojik öz yeterlik puanlarında artış olduğu, uygulamadan 6 hafta ve 1 yıl sonraki ölçümlerin ilk ölçümden anlamlı farklılığının olmadığı sonucuna varmışlardır. Verilen eğitimin uzun süreli bir etkisinin olduğu görülmüştür. Öğretmen adaylarının öz yeterlik ve başarı düzeylerinin zamanla nasıl değiştiğini anlamak için yapılan ve iki yıl süren bir araştırmanın sonucunda öğretmen adaylarının öz yeterliklerinin arttığı, başarı algılarının ise düştüğü gözlenmiştir (Gökdağ Baltaoğlu vd., 2015). İlgili Araştırmalar Pirpiroğlu ve Doğru (2015) yine öğretmen adayları üzerinde pedagojik alan bilgilerinin yıllara göre değişimini inceledikleri çalışmalarında katılımcıların alan bilgilerindeki değişimin kişiden kişiye farklılık gösterdiğini ve destekleyici eğitimler verilmesi ile öğretmen adaylarının pedagojik alan bilgilerinin gelişme gösterdiği sonucuna varmışlardır. Gencer vd. (2015) laboratuvar uygulamaları dersinde VEE diyagramının kullanılmasının fen bilimleri öğretmen adaylarının akademik başarılarına, öz yeterlik inançlarına ve tutumlarına olan etkisinin uzun sürede nasıl değiştiğini inceledikleri çalışmalarında, uyguladıkları ölçek sonucunda akademik başarının arttığını, ancak uygulanan anket sonucunda tutum puanlarında anlamlı bir artışın olmadığını gözlemlemişlerdir. Yapılan başka bir araştırmada ilköğretim öğrencilerinin Fen ve Matematik akademik başarılarının cinsiyet açısından yıllara göre değişimi yıl sonu puanları üzerinden araştırılmıştır (Bursal vd., 2015). Kız öğrencilerin iki dersten de akademik başarılarının erkek öğrencilerin puanlarından anlamlı olarak yüksek olduğunu tespit etmişlerdir. Sınıf düzeyinin artışıyla birlikte bu farkın daha da arttığını gözlemişlerdir. Bıkmaz (2017) öğretmen adayları ile yaptığı çalışmasında öğretme-öğrenme anlayışlarını ve bilimsel epistemolojik inançlarında zamanla meydana gelen değişimleri araştırmıştır. Sınıf öğretmenliği eğitiminde yer alan fizik, kimya ve fen öğretimi gibi Studies in Educational Research and Development, 2022, 6(1) 5 5 içeriklerde, öğretmen adaylarının bilimsel epistemolojik inançlarında öğretmenlik eğitiminin etkili olmadığı sonucuna varmıştır. Demirhan vd. (2018) fen bilimleri öğretmen adayları ile yaptıkları çalışmada bilimsel yaratıcılık ve akademik başarıların zamanla nasıl değiştiğini araştırmışlardır. Lisans eğitimi boyunca devam eden bu çalışmada adayların bilimsel yaratıcılıklarının ve akademik başarılarının arttığı gözlenmiştir. Taşdere (2018) fen bilimleri öğretmen adaylarının bilimin doğasına yönelik pedagojik alan bilgisinde yıllara göre meydana gelen değişimi incelemiş, var olan kavram yanılgılarının zamanla azaldığını gözlemiştir. Öğretmen adayları üzerinde yapılan başka bir çalışmada çevre sorunları üzerindeki algıların zamanla nasıl değiştiği incelenmiş, eğitim almalarına rağmen katılımcıların çevre sorunlarına ilişkin algılarının yeterli seviyeye ulaşmadığı belirlenmiştir (Yücel ve Özkan, 2018). Boylamsal araştırmalar gerçekleştirildiği süre olarak geniş bir zaman dilimini ifade edebilir. İlgili çalışmalar aylar hatta genel olarak yıllarca sürebilmektedir. Bu tür çalışmaların uzun zaman içinde gerçekleştirilebilmesi, verilerin incelenmekte olan gruptan toplanmasının gerekli olması ve yüksek maliyetli olması gibi zorlukları içerdiği bilinmektedir. Söz konusu sınırlı yanları nedeni ile bu tür araştırmaların gerçekleştirilmesi sürecinde genel olarak hedef kitle, zamanlama, yöntem seçimi gibi faktörler ile ilgili olarak sorunların yaşanması beklenmektedir. Eğitim ve öğrenme süreçlerinin dijital ortamlarda gerçekleştirilmeye başlanması da ilgili araştırmaların yöntemsel özelliklerinin yeniden uyarlanmasını gerektirecek bir değişim ve dönüşümü gerekli kılmaktadır. Bu araştırmada 2004’ten 2021’e fen eğitimine yönelik olarak gerçekleştirilen boylamsal çalışmalarda hedef kitle, yöntem, veri kaynağı ve veri toplama süresi gibi yöntemsel özellikler üzerinde durulmuştur. İlgili Araştırmalar Buna göre araştırmanın problem cümlesi şu şekilde belirlenmiştir: Problem cümlesi: 2004’ten günümüze fen eğitiminde gerçekleştirilen boylamsal araştırmaların yöntemsel özellikleri nasıl değişim göstermiştir? Alt Problemler: Araştırmanın Modeli Bu araştırmada, fen eğitimi alanında 2004-2021 yılları arasında yayımlanmış olan boylamsal çalışmalar incelenmiştir. İlgili çalışmaların yayımlanmış metinleri, nitel araştırma yöntemlerinden doküman incelemesi ile değerlendirilmiştir. Değerlendirilmesi yapılan çalışmalara ilişkin olarak genel bir tablonun sunulması amacı ile içerik analizi tekniğinden yararlanılmıştır. Fen bilimleri eğitimi alanında yayımlanmış olan bilimsel araştırmalara yönelik olarak yapılan içerik analizi çalışmaları (Çalık, Ünal, Coştu ve Karataş, 2008) ile alan yazındaki eğilimlerin neler olduğunun fark edilmesine katkı sunması ve tekrar eden konuların sürekli çalışılmasından kaçınılmasına destek olması nedeni ile alan yazına katkı sağlayabilmektedir. Örneğin; Göktaş vd., (2012) çalışmalarında, 19 adet dergide yayımlanmış olan 2115 adet eğitim araştırmasını farklı değişkenlere göre incelemişlerdir. Kıral ve Çilek (2020), Bektaş ve Zabun (2019), Kaya (2019), gerçekleştirmiş oldukları çalışmalarında resmi olarak yayımlanmış olan dokümanları, doküman analizi yöntemi ile incelemişlerdir. Bu araştırmada tez veri tabanları ile dergilerin internet adresleri ve tarama indekslerinden ulaşılabilen dokümanlar incelemeye tabi tutulmuştur. İncelemede üzerine odaklanılan kriterler, söz konusu makalelerde yer alması beklenen yöntemsel özelliklerden yayım yılı, felsefi temel, veri toplama teknikleri, katılımcı sayısı, hedef kitle ve veri toplama süresi üzerinde durulmuştur. Alt Problemler: Alt Problem 1. İlgili araştırmaların yayımlandığı yıllara göre dağılımı nasıldır? Alt Problem 2. İlgili araştırmalar hangi felsefi temeller üzerine kurgulanmıştır? Alt Problem 3. İlgili araştırmalarda hangi veri toplama teknikleri kullanılmıştır? Alt Problem 4. İlgili araştırmalarda yer alan katılımcı sayılarının dağılımı nasıldır? Alt Problem 5. İlgili araştırmalar hedef kitleye göre nasıl gruplandırılmaktadır? Alt Problem 6. İlgili araştırmalar veri toplama süresi nasıl dağılım göstermektedir? Studies in Educational Research and Development, 2022, 6(1) 6 Verilerin Toplanması Boylamsal çalışmalara veri tabanları (Google Akademik, Dergipark, TR Dizin, Ulusal Tez Merkezi, Türk Eğitim İndeksi) üzerinden ulaşılmıştır. Tarama ifadeleri, “boylamsal, doğrusal, lineer çalışmalar” olarak belirlenmiş ve bunlardan fen eğitimi ile ilgili olan araştırmalar, bu çalışmaya konu olmuştur. Çalışmada evren örneklem örtüşmesi söz konusudur. Çünkü fen eğitimi ile ilgili olarak sınırlı sayıda doğrusal çalışma gerçekleştirilmiş olması nedeni ile tarama sürecinde tespit edilmiş olan ilgili tüm araştırmalar, bu çalışmanın evrenini oluşturmuştur. Çalışma evrenine ulaşıldığına ilişkin farklı doğrulama mekanizmaları takip edilmiştir. Örneğin, ulaşılan her bir çalışmanın kaynakçasında yer alan ilgili araştırmalar saptanmaya çalışılmıştır. Bu şekilde tüm kaynakçalarda, diğer araştırmacılar tarafından atıf yapılan boylamsal çalışmalara ulaşılmıştır. Bu tarama süreci, bu çalışmanın yazarlarından ikisi tarafından birbirinden bağımsız olarak gerçekleştirilmiştir. Her iki araştırmacının tespitleri çalışmanın yazarlarından bir diğeri tarafından birleştirilmiştir. Birleştirme aşamasında, taramayı yapan araştırmacılardan birinin 21, diğerinin 19 adet çalışmaya ulaştığı görülmüştür. Studies in Educational Research and Development, 2022, 6(1) 7 7 Bu nedenle, çalışmada evren ve örneklemin örtüşmesi durumunun gerçekleştiği ifade edilebilir. Bu nedenle, çalışmada evren ve örneklemin örtüşmesi durumunun gerçekleştiği ifade edilebilir. Doküman analizi, yazılı olarak veya elektronik ortamdan edinilen belgelerin, belli bir sistematik çalışma düzenine göre incelenmesi, analiz edilmesi ve yorumlanması sürecidir. Araştırma kapsamında kullanılacak dokümanlar, araştırmacının müdahalesi olmayan yazılı ve görsel materyaller olabileceği gibi televizyon ve radyo programı kayıtları da veri kaynağı olarak kullanılabilmektedir (Kıral, 2020). Bu çalışmada verileri yurt içinde yayımlanmış olan makale ve lisansüstü tez çalışmaları oluşturmaktadır. İncelenen yöntemsel özellikler (YÖ) bağlamında, analiz birimi olarak “felsefi temel (FT), veri toplama teknikleri (VTT), katılımcı sayısı(KS), yayım yılı (YY), hedef kitle (HK) ve veri toplama süresi (VTS)” belirlenmiştir. İlgili analiz birimlerine ilişkin olarak saptanan veriler, karşılaştırmalı olarak tablolara aktarılmıştır. Tablolardan hareketle fen eğitimindeki boylamsal çalışmaların nicelik ve nitelikleri hakkında yorumlara gidilmiştir. Bilimsel çalışmaların taraması, 1 Mart 2021-1 Haziran 2021 tarihleri arasında gerçekleştirilmiştir. Araştırmada uyuşma düzeyi bakımından dokümanlarda yazılı olan içeriklere dayalı bir karşılaştırma süreci gerçekleştirilmiştir. Her bir bilimsel çalışmada FT, VTT, KS, YY, HK ve VTS’ye ilişkin veriler açık bir şekilde ifade edilmiştir. Araştırmacılar sadece bu tespitleri saptayıp tablolara aktarmışlardır. Bu nedenle her iki araştırmacı, FT, VTT, KS, YY, HK ve VTS’ ilişkin olarak Tablo 1’de yer alan saptamalar konusunda aynı fikirde olmuşlardır. Bulgular Bu bölümde çalışmanın bulgularına ve bu bulgulara ilişkin yorumlara yer verilmiştir. Tespit edilmiş çalışmalarda FT (nitel, nicel, karma), VTT (görgül, belgesel), KS (tek denekli, çok denekli), YY, HK (öğrenciler, öğretmen adayları ve öğretmenler) ve VTS’ye ilişkin olarak saptanan içerikler Tablo 1’de verilmiştir. Tablo 1’de saptanmış olan 21 adet boylamsal araştırma, FT bakımından nicel, nitel ve karma boyutları; VTT bakımından görgül ve belgesel olması; KS bakımından tek ve çok denekli olması; HK bakımından öğrencileri, öğretmen adayları ve öğretmenleri kapsaması; VTS bakımından kaç yıllık çalışma oldukları ve YY bakımından hangi yıllarda yayımlandıklarına ilişkin veriler yer almaktadır. 8 Tablo 1. verileri dikkate alınarak alt problem kapsamında ele alınan her bir değişkenin oluşturduğu kategoriye (FT, VTT, KS, YY, HK ve VTS) ilişkin bulgular, Tablo 2’de Tablo 1. İlgili Araştırmalarda YÖ’ye İlişkin Veriler (İlgili Araştırma, YY) FT VTT KS HK VTS (Yıl) Nicel Nitel Karma Görgül Belgesel Tek Çok Öğrenciler Öğretmen Adayları Öğretmenler (Erdaş, Aksüt ve Aydın, 2015) x x 13 (Canbazoğlu Bilici ve Baran, 2015) x x x x 1 (Savran Gencer, Sevim ve Kaska, 2015) x x x x 1 (Emrahoğlu ve Öztürk, 2009) x x x x 4 (Güngör, 2009) x x x x x 3 (Pirpiroğlu ve Doğru, 2015) x x x x 1 (Demirhan, Önder ve Beşoluk, 2018) x x x x 4 (Saraç, 2017) x x x x 1 Ay (Özata Yücel ve Özkan, 2018) x x x x 3 (Taşdere, Özsevgeç ve Türkmen, 2014) x x x x 3 Ay (Doğan, 2014) x x x x 3 (Parlaktaş, 2018) x x x x 3 Ay (Aktaş, 2016) x x x x 1 (Taşdere, 2018) x x x x 2 (Ayvar, 2019) x x x x 1 (Kahraman, 2020) x x x x 2 (Bursal, 2013) x x 5 (Çakır, 2004) x x x x 8 (Yücel ve Özkan, 2018) x x x x 4 (Yeşildağ-Hasançebi ve Günel, 2014) x x x x x 2 (Bursal, Buldur ve Dede, 2015) x x 5 TOPLAM 5 12 4 18 3 0 18 5 11 4 - Tablo 1. verileri dikkate alınarak alt problem kapsamında ele alınan her bir değişkenin oluşturduğu kategoriye (FT, VTT, KS, YY, HK ve VTS) ilişkin bulgular, Tablo 2’de özetlenmiştir. Kategorilere ilişkin frekans dağılımları Tablo 2’de yer almaktadır. Studies in Educational Research and Development, 2022, 6(1) 9 Tablo 2. Bulgular Çalışmalara ilişkin frekans dağılımları Kategoriler Alt Kategoriler Frekans Toplam Boylamsal Çalışmalara İlişkin YY’ler 2004 1 21 2009 2 2013 1 2014 3 2015 5 2016 1 2017 1 2018 5 2019 1 2020 1 Çalışmaların FT’ye ilişkin frekansları Nitel 12 21 Nicel 5 Karma 4 Çalışmaların VTT’ye göre sınıflandırılması Görgül 18 21 Belgesel 3 Çalışmaların KS’ye göre sınıflandırılması Çok denekli 18 18 Tek denekli 0 Çalışmaların HK’ye göre sınıflandırılması Ortaokul öğrencileri 5 20 Öğretmen adayları 11 Öğretmenler 4 Çalışmaların VTS’ye göre sınıflandırılması 1 yıldan az 3 21 1 Yıl 5 2 yıl 3 3 yıl 3 4 yıl 3 5 yıl ve üzeri 4 Tablo 2. Çalışmalara ilişkin frekans dağılımları Alt problemlere, Tablo 2. verileri dikkate alınarak yanıt verilmiştir. Her bir kategoride incelenen boylamsal çalışma sayısı ile alt kategorilere ilişkin frekans bilgilerine göre yorumlara yer verilmiştir. Alt problemlere ilişkin açıklamalar şunlardır: Studies in Educational Research and Development, 2022, 6(1) 10 Alt Problem 4. İlgili araştırmalarda yer alan katılımcı sayılarının (KS) dağılımı nasıldır? İlgili araştırmaların tek veya çok denekli araştırmalar olarak gerçekleştirilmesi durumuna göre incelenmesi sonucunda ulaşılan veriler, Tablo 2’de verilmiştir. İncelenen 21 adet çalışmanın 18’inin görgül olarak gerçekleştirilmiş olması, söz konusu çalışmalara uygulamalı olarak katılan deneklerin olduğu anlamına gelmektedir. İlgili katılımcıların tamamının çok denekli katılımcılar olduğu, 18 adet çalışmada tek denekli boylamsal araştırmanın yer almadığı görülmektedir. Alt Problem 3. İlgili araştırmalarda hangi veri toplama teknikleri (VTT) kullanılmıştır? Boylamsal araştırmaya göre kurgulanmış çalışmalarda VTT olarak hangi tekniğin kullanıldığına ilişkin veriler, Tablo 4’te yer almaktadır. VTT bakımından görgül ve belgesel odaklı olarak gerçekleştirilebilen bilimsel çalışmalar, bu araştırmada incelenen boylamsal çalışmalarda da kullanılmıştır. Buna göre incelenen 21 adet çalışmadan 18’i görgül, 3’ü belgesel odaklı olarak gerçekleştirilmiştir. Bu durumda araştırmacıların genel olarak uygulamalı çalışmalardan veriler toplama yolunu tercih etmiş oldukları anlaşılmaktadır. Alt Problem 1. İlgili araştırmaların yayımlandığı yıllara (YY) göre dağılımı nasıldır? Tablo 2’de, incelenmiş olan araştırmaların YY kriterine ilişkin veriler yer almaktadır. Tablo 2. verileri, ilgili araştırmaların 2015 ve 2018 yıllarında kümelendiği görülmektedir. 2020 yılının ilk yarısında 1 adet çalışmanın yayımlandığı, bundan sonra herhangi bir boylamsal çalışmanın yayımlanmamış olduğu görülmektedir. Alt Problem 2. İlgili araştırmalar hangi felsefi temeller (FT) üzerine kurgulanmıştır İncelenen 21 adet çalışanın FT bağlamındaki özellikleri Tablo 2’de verilmiştir. Tablo 2’de, 2004-2020 yılları arasında tespit edilmiş olan 21 adet çalışmanın dayandığı felsefi temele ilişkin veriler yer almaktadır. İlgili yıllar arasında yayımlanmış olan boylamsal fen eğitimi araştırmalarında 12’sinin nitel, 5’inin nicel ve 4’ünün de karma yöntemler ile gerçekleştirildiği görülmektedir. İlgili araştırmalarda daha çok nitel yöntemlerin benimsenmiş olduğu anlaşılmaktadır. Alt Problem 6. İlgili araştırmalar veri toplama süresi (VTS) nasıl dağılım göstermektedir? İlgili araştırmalarda verilerin ne kadarlık bir zaman dilimi süresince toplandığına ilişkin olan veriler Tablo 2’de verilmiştir. Tablo 2’de görüldüğü gibi, 2 adet çalışmada verilerin genel olarak 1 yıl ve üzeri zaman dilimi içinde toplandığı, 3 adet çalışmanın da 1 yıldan daha az süre boyunca derlendiği anlaşılmaktadır. Bu durum boylamsal çalışmaların verilerin toplanması süreci bakımından genel olarak yıllar alabilecek çalışmalar olduğu anlamına da gelebilir. Alt Problem 5. İlgili araştırmalar hedef kitleye (HK) göre nasıl gruplandırılmaktadır? İncelenen boylamsal araştırmaların hitap ettiği örneklem grubun özelliklerine ilişkin veriler, Tablo 2’de yer almaktadır. Tablo 2’de görüldüğü gibi, 2004-2020 yılları arasında fen eğitimine yönelik olarak gerçekleştirilmiş olan boylamsal araştırmaların 11’i öğretmen adayları yani, fen bilgisi öğretmenliği anabilim dalı öğrencileri ile, 5’i ortaokul öğrencileri ile ve 4’ü de öğretmenlerin katılımı ile gerçekleştirilmiştir. Güngör, (2009) ile Yeşildağ-Hasançebi ve Günel, (2014) çalışmalarında katılımcı grup olarak hem ortaokul öğrencilerini hem de öğretmenleri incelemişlerdir. Studies in Educational Research and Development, 2022, 6(1) 11 Tartışma, Sonuç ve Öneriler Bu araştırmada incelenen fen eğitimi ile ilgili olarak gerçekleştirilmiş 21 adet boylamsal çalışma, YY bakımından 2020 yılı öncesinde gerçekleştirilmiş araştırmalar olarak değerlendirilmektedir. Söz konusu araştırmaların yarısından fazlasının nitel çalışma yöntemleri ile gerçekleştirildiği anlaşılmaktadır. İncelene çalışmaların genel olarak görgül çalışmalar olduğu ve çok sayıda deneğin katılımı ile gerçekleştirildiği görülmektedir. Değerlendirmeler, incelenmiş olan 21 adet çalışmanın sadece 5’inin, fen eğitiminin odağındaki kitlelerden olan ortaokul öğrencileri ile gerçekleştirildiğini, HK bakımından genel olarak öğretmen adayları ile öğretmenlerin seçilmiş olduğunu ortaya koymaktadır. Fen eğitimi alanında yayımlanan boylamsal çalışmaların odağında yer alan hedef kitle olarak fen eğitiminin asıl hedef kitlelerinden olan ilk ve ortaokul öğrencileri üzerinde yapılan araştırmaların oldukça az olduğu değerlendirilmiştir. Araştırmacıların çalışmalarını öğrencilerin katılımı ile gerçekleştirilmesinden elde edilecek sonuçların, fen eğitimindeki arayışlara ışık tutacağı düşünülmektedir. Öğretmen adayları üzerinde yapılan çalışmaların ortaokul düzeyine göre fazla olması, akademik çalışma yapan araştırmacıların çalışmalarında kolay ulaşılabilir hedef kitle arayışından kaynaklanabilir. İlgili araştırmaların içeriğine bakıldığında hedef kitlenin genel olarak öğretmen adaylarının seçilmesinden dolayı en çok pedagojik alan bilgisi üzerine odaklanıldığı görülmektedir. Çalışmaların YY incelendiğinde, araştırmaların 2015 ve 2018 yıllarında yoğunlaştığı görülmektedir. Ulaşılan ilk çalışmanın 2004 yılında yapıldığı düşünülürse, sonraki yıllarda araştırmacıların boylamsal çalışmalar üzerindeki bilgi ve ilgisinin arttığı görülmektedir. Son yıllarda fen eğitiminde boylamsal çalışmanın oldukça sınırlı düzeyde kalması, söz konusu yönteme ilişkin olarak ifade edilen sınırlılıkların teknolojinin sunmuş olduğu olanaklar ile azaltılmasının yollarını aramayı gerektirmektedir. Alt problemlerden ikincisine ilişkin veriler değerlendirildiğinde, çalışmaların daha çok nitel çalışmalar olduğu görülmektedir. Nitel araştırma yönteminin kullanılması, bilginin derinlemesine incelenmesi açısından isabetli olabilir (Özcan ve Çalışkan, 2020). Göktaş vd. (2012) çalışmalarında, 125 adet eğitim araştırmasında yöntem bakımından betimsel ve nicel yöntemlerin daha çok kullanıldığı sonucuna ulaşılmıştır. Studies in Educational Research and Development, 2022, 6(1) 12 Üçüncü alt problemde ele alınan veriler değerlendirildiğinde, VTT olarak genellikle görgül çalışmaların tercih edildiği görülmektedir. Buna bağlı olarak dördüncü alt problemde katılımcı sayılarına yönelik veriler değerlendirildiğinde, uygulamalı çalışmaların tamamının çok sayıda katılımcı ile gerçekleştiği görülmektedir. Çalışmalardan 3 tanesi belgesel odaklı olarak gerçekleştirilmiştir. Beşinci alt probleme dayalı veriler değerlendirildiğinde, öğretmen adayları ve öğretmenlerle yapılan çalışmaların, ortaokul öğrencileri ile gerçekleştirilen çalışmalardan daha fazla olduğu görülmüştür. Göktaş vd. (2012) çalışmalarında, incelemiş oldukları eğitim araştırmalarında, örneklem bakımından lisans öğrencileri ve öğretmenlerin seçildiği çalışmaların daha fazla olduğu sonucuna ulaşılmıştır. Altıncı alt problemde ele alınan veriler değerlendirildiğinde en kısa süren boylamsal çalışmanın 13 hafta en fazla süren çalışmanın ise 8 yıl olduğu görülmektedir. Boylamsal çalışmalarda standart bir sürenin olmadığı da görülmektedir. Tartışma, Sonuç ve Öneriler Ancak uygulamaya dayalı olan ve görgül kaynaklardan verilerin toplandığı araştırmalar değerlendirildiğinde, uzun süreli çalışmaların geniş zaman dilimlerini kapsayan verileri içeren dokümanlar ile gerçekleştirilmiş araştırmalar olduğu görülmektedir. Fen bilimleri eğitiminde yapılan araştırmalarda nicel yöntemlerin nitel yöntemlere tercih edildiği görülmektedir (Filiz ve Kocakülah, 2020). Benzer şekilde ilgili araştırmalarda anlık ve kesitsel çalışmalar, boylamsal çalışmalara tercih edilmektedir. Bu araştırmada söz konusu yöntem ile ilgili olarak sınırlı sayıda çalışmanın tespit edilmiş olması, evren ve örneklem örtüşmesinin gerçekleşmesi, fen eğitimi ile ilgili bilimsel yayınlarda, boylamsal çalışmanın oldukça sınırlı sayıda olduğu anlamına gelmektedir. Özetle Türkiye’de fen eğitimi alanında yapılan boylamsal çalışmaların sınırlı olduğu görülmektedir. Boylamsal çalışmaların, öğrenmelerin kalıcılığının ve bireylerin eğiliminin tespiti açısından gerekli olduğu için fen eğitiminde kullanılmasının önemli olduğu düşünülmektedir. Bu araştırmanın sonuçları, fen eğitimi ile ilgili az sayıda boylamsal çalışmanın gerçekleştirildiğini ortaya koymaktadır. Bu nedenle araştırmacıların bu tür çalışmaları daha çok gerçekleştirmeleri gerekmektedir. İlgili araştırmaların, hedef kitle olarak öğretmen adaylarından, ilköğretim öğrencilerine doğru bir artış göstermesi gerekmektedir. Araştırmacıların fen eğitiminin odağında yer alan söz konusu hedef kitle ile ilgili olarak daha fazla sayıda bilimsel çalışma gerçekleştirmeleri önerilmektedir. İlgili araştırmaların görgül araştırmalara odaklandıkları değerlendirilmektedir. Ancak dokümanlara dayalı boylamsal çalışmalara daha çok yer verilmesi gerekmektedir. Uzaktan eğitim olanaklarının giderek yaygınlaştığı günümüzde öğrencilerin eğitim ve öğrenme süreçlerine daha kolay dâhil olabilmelerine rağmen daha az boylamsal çalışma gerçekleştirildiği anlaşılmaktadır. Araştırmacıların, bilimsel çalışmalarını hedef kitle, yöntem seçimi gibi metodolojik özellikler bakımından teknolojik gelişmelerin sunmuş olduğu olanaklara, özellikle uzaktan eğitim süreçlerine yeniden uyarlamaları beklenmektedir. Studies in Educational Research and Development, 2022, 6(1) 13 Kaynaklar Aktaş, S. (2016). Ortaokul 6, 7 ve 8. sınıf fen bilimleri dersi öğretim programlarının öğrencilerin bilimsel süreç becerileri, duygusal zekâları, bilişsel stilleri ve akademik başarılarına etkisi. Yüksek Lisans Tezi. Mustafa Kemal Üniversitesi, Fen Bilimleri Enstitüsü, Fen Bilgisi Eğitimi Ana Bilim Dalı, Hatay. Ayvar, İ. (2019). Etkili harmanlanmış öğrenme ortamının fen bilgisi öğretmen adaylarının bilimsel araştırma-sorgulama temalarını anlamaları üzerine etkisi. Yüksek Lisans Tezi. Uşak Üniversitesi, Fen Bilimleri Enstitüsü, İlköğretim Ana Bilim Dalı, Fen Bilgisi Eğitimi Bilim Dalı, Uşak. Bektaş, Ö. ve Zabun, E. (2019). Vatandaşlık eğitiminde değerler karşılaştırması: Türkiye ve Fransa. Değerler Eğitimi Dergisi, 17(37), 247-289. Bıkmaz, F. (2017). Öğretmen adaylarının öğretme-öğrenme anlayışları ve bilimsel epistemolojik inançlarının araştırılması: Boylamsal bir çalışma. Eğitim ve Bilim, 42(189). Bursal, M. (2013). İlköğretim öğrencilerinin 4-8. sınıf fen akademik başarılarının boylamsal incelenmesi: Sınıf düzeyi ve cinsiyet farklılıkları. Kuram ve Uygulamada Eğitim Bilimleri, 13(2), 1141-1156. Bursal, M., Buldur, S. ve Dede, Y. (2015). Alt sosyo-ekonomik düzeyli ilköğretim öğrencilerinin 4-8. sınıflar fen ve matematik ders başarıları: cinsiyet perspektifi. Eğitim ve Bilim, 40(179), 133-145. 01. 06. 2021 tarihinde http://egitimvebilim.ted.org.tr/index.php/EB/article/view/2913/1039 adresinden erişilmiştir. Büyüköztürk, Ş., Kılıç Çakmak, E., Ekgün, Ö.E., Karadeniz, Ş., ve Demirel, F. (2017). Bilimsel araştırma yöntemleri (23. Baskı). Ankara: Pegem Akademi Yayıncılık. Canbazoğlu Bilici, S. ve Baran, E. (2015). Fen bilimleri öğretmenlerinin teknolojik pedagojik alan bilgisine yönelik öz-yeterlik düzeylerinin incelenmesi: Boylamsal bir araştırma. Gazi Üniversitesi Gazi Eğitim Fakültesi Dergisi, 35(2), 285-306. 01.06.2021 tarihinde http://www.gefad.gazi.edu.tr/tr/pub/issue/6772/91170 adresinden erişilmiştir. Çakır, K. B. (2004). İlköğretimin farklı seviyelerinde bazı temel kimya kavramlarının anlaşılma düzeylerinin belirlenmesi: Boylamsal bir çalışma. Yüksek Lisans Tezi. Karadeniz Teknik Üniversitesi, Fen Bilimleri Enstitüsü, Ortaöğretim Fen ve Matematik Alanları Eğitimi Ana Bilim Dalı, Trabzon. Çalık, M., Ünal, S., Coştu, B. ve Karataş, F.Ö. (2008). Trends in Turkish science education. Essays in Education, 24, 23-45. Demirhan, E., Önder, İ. ve Beşoluk, Ş. (2018). Fen bilimleri öğretmen adaylarının bilimsel yaratıcılık ve akademik başarılarının yıllara göre değişimi. Kastamonu Education Journal, 26(3), 685-696. doi:10.24106/kefdergi.373323 Studies in Educational Research and Development, 2022, 6(1) 14 Doğan, Ö. K. (2014). Mesleğe yeni başlayan fen öğretmenlerinin pedagojik ve epistemolojik inançları ve sınıf içi uygulamaları: Boylamsal durum çalışması. Doktora Tezi. Marmara Üniversitesi Eğitim Bilimleri Enstitüsü Ortaöğretim Fen ve Matematik Alanları Anabilim Dalı Biyoloji Öğretmenliği Bilim Dalı, İstanbul. Emrahoğlu, Y. ve Öztürk, A. (2009). Fen bilgisi öğretmen adaylarının astronomi kavramlarını anlama seviyelerinin ve kavram yanılgılarının incelenmesi üzerine boylamsal bir araştırma. Çukurova Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 18(1), 165-180. 01.06.2021 tarihinde https://dergipark.org.tr/tr/download/article-file/70628 adresinden erişilmiştir. Erdaş, E., Aksüt, P. ve Aydın, F. Kaynaklar (2015). Fen ve teknoloji öğretim programlarının teknoloji okuryazarlığı boyutları açısından incelenmesi: boylamsal bir çalışma. Abant İzzet Baysal Üniversitesi Eğitim Fakültesi Dergisi, 15(2), 132-146. 01.06.2021 tarihinde https://dergipark.org.tr/tr/download/article-file/17434 adresinden erişilmiştir. Filiz, A. ve Kocakülah, M. S. (2020). Fen eğitiminde proje tabanlı öğrenme yaklaşımı ile ilgili yapılan araştırmaların içerik analizi. Ihlara Eğitim Araştırmaları Dergisi, 5(2), 175–194. Gencer, A. S., Sevim, S., ve Kaska, A. (2015). Genel biyoloji laboratuvarında Vee diyagramı uygulaması: Fen bilgisi öğretmen adaylarının akademik başarılarının, öz-yeterlik inançlarının ve tutumlarının boylamsal olarak değerlendirilmesi. Elektronik Sosyal Bilimler Dergisi, 14(52). 01.06.2021 tarihinde https://dergipark.org.tr/tr/download/article-file/70628 adresinden erişilmiştir. Gökdağ Baltaoğlu, M., Sucuoğlu, H. ve Yurdabakan, İ. (2015). Öğretmen adaylarının öz- yeterlik algıları ve başarı/başarısızlık yüklemeleri: Boylamsal bir araştırma. İlköğretim Online, 14(3), 803-814. DOI: 10.17051/io.2015.66489 Göktaş̧, Y., Küçük, S., Aydemir, M., Telli, E, Arpacık, Ö., Yıldırım, G. ve Reisoğlu, İ. (2012). Türkiye’de eğitim teknolojileri araştırmalarındaki eğilimler: 2000-2009 dönemi makalelerinin içerik analizi. Kuram ve Uygulamada Eğitim Bilimleri Dergisi, 12(1), 177-199. Güngör, B. (2009). İnsanda sindirim sistemi konusunda ilköğretim 7. sınıf öğrencilerinin kavram yanılgılarının kökenlerinin belirlenmesine yönelik boylamsal bir çalışma. Doktora Tezi, Balıkesir Üniversitesi Fen Bilimleri Enstitüsü, Balıkesir. Kahraman, S. (2020). Fen bilimleri öğretmen adaylarının biyoteknoloji, genetik mühendisliği ve klonlama kavramlarına ilişkin algılarının incelenmesi. Necatibey Eğitim Fakültesi Elektronik Fen ve Matematik Eğitimi Dergisi, 14(1), 57-83. 01.06.2021 tarihinde https://dergipark.org.tr/tr/pub/balikesirnef/issue/55528/576192 adresinden erişilmiştir. Kaya, H. P. (2019). Türkiye’de denetim alanında yazılmış olan doktora tezlerinin değerlendirilmesi (1995-2018). Karabük Üniversitesi Sosyal Bilimler Enstitüsü Dergisi, 9(2). 556-576. Studies in Educational Research and Development, 2022, 6(1) 15 Kıral, B. ve Çilek, A. (2020). 2023 Vizyon Belgesi’nin karakter eğitimi bakımından değerlendirilmesi. Milli Eğitim Dergisi, 49(225), 5-22. Kıryak, Z., Candaş, B., Çalık, M. ve Zeybek, Ö. (2020). Öğrencilerin fen bilimleri dersine yönelik zihinsel imajlarının belirlenmesi: Bir sınıflar arası karşılaştırma. Pamukkale Üniversitesi Eğitim Fakültesi Dergisi, 50, 468-490. Kula, F. ve Sadi, Ö. (2016). Türk fen bilimleri eğitiminde araştırma ve yönelimler: 2005 – 2014 yılları arası bir içerik analizi. İlköğretim Online, 15(2), 594-614. La Velle, L., Newman, S., Montgomery, C. ve Hyatt, D. (2020). Initial teacher education in England and the Covid-19 pandemic: Challenges and opportunities. Journal of Education for Teaching, 46(4), 596-608. Lei, J. ve Zhao, Y. (2007). Technology uses and student achievement: A longitudinal study. Computers & Education, 49(2), 284-296. Özata Yücel, E. ve Özkan, M. (2018). Fen bilimleri öğretmen adaylarının çevre sorunları algılarındaki değişimin incelenmesi: Kocaeli örneği. Pamukkale Üniversitesi Eğitim Fakültesi Dergisi, 44, 146-160. 01.06.2021 tarihinde https://dergipark.org.tr/tr/download/article-file/475578 adresinden erişilmiştir. Özcan, C. ve Çalışkan, İ. (2020). , P. (2001). Approaches to learning in science: A longitudinal study. British Journal of Educational Psychology, 71(1), 115-132. Introduction The coronavirus epidemic, which started to be seen all over the world with the year 2020, has led to negative results in many areas due to the increasing in the rate. Although the process was tried to be overcome with curent measures at the beginning, the fact that the epidemic lasted longer than the expected period brought new viewpoints for useful solutions to the problematic issues. The field of education is among the sectors that are seriously affected by the process in terms of both education and practices. For this reason, the researchers have started to create alternative environments in order to do their studies. One of the important features of the educational process is that it aims to provide effective, permanent and meaningful learning for students. In order to understand to what extent this permanence is achieved in the processes in which the students are participants, it is necessary to apply repeated exams in different time periods and to compare the data complated from them. Experimental studies and especially longitudinal ones offer important opportunities to researchers for repeated measurements. Longitudinal studies are defined as studies in which the persistence of learning and the tendencies of individuals are determined by collecting data at different time from the study group. Longitudinal research can refer to a wide period of time as the period in which it is carried out. Related studies can take months or even years in general. It is known that such studies involve difficulties such as being able to be carried out over a long period of time, collecting data from the group under investigation, and high cost. In recent years, it is expected that there have been some problems related to factors such as study groups, timing, method selection in general during the process of carrying out such research. Because the beginning of the education and learning processes in digital environments necessitates a change and transformation that will require the re-adaptation of the methodological features of the related researches. In this study, it was emphasized how longitudinal studies carried out with science education were affected by methodological features such as the year it was published from 2018 through 2021, philosophical basis, data collection technique, number of participants, study groups and data collection period. Extended Abstract Extended Abstract Kaynaklar Fen eğitimi alanındaki araştırmaların konu ve yöntem açısından incelenmesi. Ege Eğitim Dergisi, 2(1), 101-111. Parlaktaş, B. (2018). Fen bilgisi öğretmen adaylarının laboratuvar öğrenme ortamı, bilişsel esneklikleri ve sosyal sorun çözme becerilerine yönelik algılarının incelenmesi. Yüksek Lisans Tezi. Aydın Adnan Menderes Üniversitesi, Fen Bilimleri Enstitüsü, Matematik ve Fen Bilimleri Eğitimi Ana Bilim Dalı, Aydın. Pirpiroğlu, İ. ve Doğru, M. (2015). Fen bilimleri öğretmen adaylarının pedagojik alan bilgilerinin boylamsal olarak incelenmesi. Akdeniz İnsani Bilimler Dergisi, 2, 313-329. doi:10.13114/MJH.2015214575 Saraç, H. (2017). Temel eğitim düzeyindeki öğrencilerin dünya ve evren konularına ilişkin tutumlarının incelenmesi. Dokuz Eylül Üniversitesi Buca Eğitim Fakültesi Dergisi, 43, 25-40. 01.06.2021 tarihinde https://dergipark.org.tr/tr/download/article-file/432722 adresinden erişilmiştir. Savran Gencer, A, Sevim, S. ve Kaska, A. (2015). Genel biyoloji laboratuvarında Vee diyagramı uygulaması: Fen bilgisi öğretmen adaylarının akademik başarılarının, öz- yeterlik inançlarının ve tutumlarının boylamsal olarak değerlendirilmesi. Elektronik Sosyal Bilimler Dergisi, 14(52), 183-202. 01.06.2021 tarihinde https://dergipark.org.tr/tr/download/article-file/70628 adresinden erişilmiştir. Taşdere, A. (2018). Fen bilgisi öğretmen adaylarının bilimin doğasına yönelik pedagojik alan bilgisi gelişimlerinin incelenmesi. Doktora Tezi. Lisansüstü Eğitim Enstitüsü. Studies in Educational Research and Development, 2022, 6(1) 16 Karadeniz Teknik Üniversitesi, Eğitim Bilimleri Enstitüsü, İlköğretim Ana Bilim Dalı, Fen Bilgisi Eğitimi Bilim Dalı, Trabzon. Karadeniz Teknik Üniversitesi, Eğitim Bilimleri Enstitüsü, İlköğretim Ana Bilim Dalı, Fen Bilgisi Eğitimi Bilim Dalı, Trabzon. Taşdere, A., Özsevgeç, T. ve Türkmen, L. (2014). Bilimin doğasına yönelik tamamlayıcı bir ölçme aracı: Kelime ilişkilendirme testi. Fen Eğitimi ve Araştırmaları Derneği, Fen Bilimleri Öğretimi Dergisi, 2(2), 129-144. 01.06.2021 tarihinde https://dergi.fead.org.tr/wp-content/uploads/V-FenTekE_D1811-Ahmet-Tesdere.pdf adresinden erişilmiştir. Williamson, B., Eynon, R. ve Potter, J. (2020). Pandemic politics, pedagogies and practices: digital technologies and distance education during the coronavirus emergency. Learning, Media and Technology, 45(2), 107-114. https://doi.org/10.1080/17439884.2020.1761641 Yücel, E. Ö., ve Özkan, M. (2018). Fen bilimleri öğretmen adaylarının çevre sorunları algılarındaki değişimin incelenmesi: Kocaeli örneği. Pamukkale Üniversitesi Eğitim Fakültesi Dergisi, 44(44), 146-160. 01.06.2021 tarihinde https://dergipark.org.tr/tr/download/article-file/475578 adresinden erişilmiştir. Yeşildağ-Hasançebi, F. ve Günel, M. (2014). Farklı perspektiflerden argümantasyon tabanlı öğrenme yaklaşımının bilim öğrenme üzerine etkilerinin derinlemesine incelenmesi. Journal of Research in Education and Society, 1(1), 23-44. 01.06.2021 tarihinde https://dergipark.org.tr/tr/pub/etad/issue/26382/277988 adresinden erişilmiştir. Zeegers, P. (2001). Approaches to learning in science: A longitudinal study. British Journal of Educational Psychology, 71(1), 115-132. Studies in Educational Research and Development, 2022, 6(1) 17 Method In this study, longitudinal studies published between 2004-2020 in the field of science education were examined. Published articles of related studies were evaluated by document Studies in Educational Research and Development, 2022, 6(1) 18 analysis, one of the qualitative research methods. Content analysis technique was used in order to present a general table regarding the evaluated studies. analysis, one of the qualitative research methods. Content analysis technique was used in order to present a general table regarding the evaluated studies. Results It is seen that the similar researches were focused on in 2015 and 2018. It is seen that only one study was published in the first half of 2020, and no longitudinal studies have been published since then. The relevant period coincides with the period when the distance education practices were seen intensely in the society and education and teaching processes. In the longitudinal science education studies published between the relevant years, it is seen that 12 of them were carried out with qualitative methods, 5 with quantitative methods and 4 with mixed methods. Out of 21 studies, 18 were empirical and 3 were documentary oriented. It was determined that 18 of the 21 studies examined were empirical and a single-subject study was not carried out. Of the longitudinal studies conducted between 2004 and 2020 on science education, 11 were conducted with pre-service teachers, that is, with the students of the science teaching department, 5 with middle school students, and 4 with the participation of teachers. It is understood that in 2 studies, data were generally collected within a period of nearly 1 year, and 3 studies were compiled for less than 1 year. This may mean that longitudinal studies are studies that may take years in general in terms of the data collection process. The 21 longitudinal studies conducted on science education examined in this study are considered as studies had been complated before the year of 2020. It is understood that half of the researches in question were carried out with qualitative study methods. It is seen that the studies examined are generally empirical studies and were carried out with the participation of a large number of participants. Evaluations Show that only 5 of the 21 studies examined were carried out with middle school students, who are at the center of science education, and in these studies, teacher candidates and teachers were generally chosen as the target groups. Discussion It has been evaluated that there are very few studies on middle school students, who are among the main target groups of science education, as the target audience of longitudinal studies published in the field of science education. It is thought that the results of the researchers' study with the participation of students will shed light on the pursuits in science education. The fact that the number of studies on teacher candidates is higher than the middle school level is thought to be due to the search for an easily accessible target group in the studies of academic researchers. It is seen that researches in science education are generally pre and post test design or surveys. The fact that a limited number of studies were determined in the relevant field in this study and the fact that the universe and sample overlap means that there is a very limited number Studies in Educational Research and Development, 2022, 6(1) 19 of longitudinal studies in scientific publications related to science education. As a result, it is seen that longitudinal studies in the field of science education in Turkey are limited. Since longitudinal studies are necessary to determine the permanence of learning and the tendency of individuals, it is thought to be important to use in science education. Recommendations The results of this research reveal that few longitudinal studies on science education have been carried out. Therefore, researchers need to carry out more such studies. Relevant studies should show an increase in the study group from pre-service teachers to primary school students. It is recommended that researchers conduct more scientific studies regarding the target group, which is at the center of science education. It is considered that relevant studies focus on empirical research. However, longitudinal studies based on documents should be given more importance. It has been determined that fewer longitudinal studies have been carried out since 2018 due to involvement of target groups in distance education and education and learning processes. Researchers are expected to adapt their scientific studies to technology, that is, to distance education processes, in terms of methodological features such as target groups and methodological features.
https://openalex.org/W4286897152	https://zenodo.org/records/5920872/files/Diversification_MAP_PP-FI.pdf	English	null	MAP Position Paper (Finland) - Change in production and diversification of the rural economy	Zenodo (CERN European Organization for Nuclear Research)	2,021	cc-by	11,982	Authors Virpi Harilahti-Juola, Antonia Husberg, Mari Kattilakoski, Hanna-Mari Kuhmonen, Olli Lehtonen, Petri Rinne, Sami Tantarimäki, Taina Väre CHANGE IN PRODUCTION AND DIVERSIFICATION OF THE RURAL ECONOMY MAP Position Paper MAP Position Paper SHERPA has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No. 862448. The content of the document does not reflect the official opinion of the European Union. Responsibility for the information and views expressed therein lies entirely with the author(s). Editors Nordregio \| Michael Kull and Mats Stjernberg (Luke/Nordregio) Citation: Kull, M. & M., Stjernberg (eds.) (2021) MAP Position Paper (Finland) - Change in production and diversification of the rural economy. DOI: 10.5281/zenodo.5920872 Paper finalised in October 2021 Find out more about the Finnish Multi-Actor Platform! https://rural-interfaces.eu/maps/finland/ SHERPA Multi-Actor Platforms (MAPs) are invited to discuss three key questions: 1) What are the key needs for the development of the rural economy in your MAP, and how can they be addressed most effectively? 2) How can policy interventions support positive changes in the diversification of the rural economy, considering solutions that are needed at the local and national levels, and any implications for the wider policy framework (European Union level or others)? What can public administrations (at all levels) do to facilitate and encourage positive changes in the diversification of the rural economy? 1) What are the key needs for the development of the rural economy in your MAP, and how can they be addressed most effectively? 1) What are the key needs for the development of the rural economy in your MAP, and how can they be addressed most effectively? 2) How can policy interventions support positive changes in the diversification of the rural economy, considering solutions that are needed at the local and national levels, and any implications for the wider policy framework (European Union level or others)? What can public administrations (at all levels) do to facilitate and encourage positive changes in the diversification of the rural economy? The exercise is based on the following process: (i) preparation and reflection of the SHERPA Discussion Paper by each regional or national MAP, (ii) consultation with MAP participants, (iii) drafting of the MAP Position Paper, and (iv) synthesis of the regional and national MAP Position Papers for discussion at European Union level. 1. Contents SHERPA Multi-Actor Platforms (MAPs) are 1) What are the key needs can they be addressed most effectively? 2) How can policy interve economy, considering solutions that are n wider policy framework (European Union to facilitate and encourage positive chang 3) What are the research n The exercise is based on the following pro by each regional or national MAP, (ii) co Paper, and (iv) synthesis of the regional level. Background and motivation Green and digital transitions are at the core of the transformational changes needed in European rural areas. The diversification of rural economies has improved their resilience to the consequences of the COVID-19 pandemic, and will play a pivotal role in the recovery by offering development opportunities. The SHERPA process will support the gathering of evidence from across Europe, at multiple levels, regarding the directions of the diversification of rural economies, which are appropriate and feasible to address local needs. PA Multi-Actor Platforms (MAPs) are invited to discuss three key questions: 1 The strategic focal points of the programme are interdependence, environmental justice and a new knowledge-based economy. These are central throughout the different themes. The five themes in the Rural Policy Programme are: 1) Higher added value through sustainable utilisation of natural resources. 2) Rural actors as a part of the solution to a sustainable transition. 3) Strengthening competitiveness and viability. 4) Securing a fluent everyday life. 5) Strengthening participation and communality. 1. Introduction This Position Paper focuses on fundamental issues concerning the diversification of the rural economy, which are of particular interest for the Finnish SHERPA Multi-Actor Platform. These issues include entrepreneurship and the labour market, digitalisation, smart rural areas and smart adaptation, as well as economic diversification in rural areas. The National Rural Policy Programme for 2021–2027 is titled ”Countryside renewing with the times” (Ajassa uudistuva maaseutu) (Kattilakoski et al., 2021). This programme presents nationally shared views and guidelines for rural development in Finland for the coming years1. It sets out what kind of national rural policy is needed to meet various current challenges such as the global sustainability crisis and climate change, population ageing and shrinkage, digitalisation, location independence and increasing mobility of people. The programme highlights, among other things, the need for a transition from a fossil-based economy to a sustainable bioeconomy and circular economy. Managing the sustainability crisis requires a shift towards sustainable, low-carbon activities and economies. This change requires a broad debate on, for example, everyday life and entrepreneurship in rural areas, and the impact of decisions across the country and for different types of areas. 1 The strategic focal points of the programme are interdependence, environmental justice and a new knowledge-based economy. These are central throughout the different themes. The five themes in the Rural Policy Programme are: 1) Higher added value through sustainable utilisation of natural resources. 2) Rural actors as a part of the solution to a sustainable transition. 3) Strengthening competitiveness and viability. 4) Securing a fluent everyday life. 5) Strengthening participation and communality. 1. Contents Background and motivation ....................................................................................................... 3 1. Introduction ........................................................................................................................ 3 Dimensions of diversification ........................................................................................................... 5 2. Diversification of the rural economy: Entrepreneurship, employment & new business models ........................................................................................................................................ 5 2.1. Finnish Perspectives & good Practices ....................................................................................... 5 2.2. Opportunities, Challenges & Recommendations for and from Finland .......................................... 6 3. Smart rurality, smart communities and digitalisation ......................................................... 9 3.1. Finnish Perspectives & good Practices ....................................................................................... 9 3.2. Opportunities, Challenges & Recommendations for and from Finland ......................................... 11 4. Bio-economy and sustainable management of resources ................................................. 12 4.1. Finnish Perspectives, good practices & challenges .................................................................... 12 Recommendations and Conclusions ......................................................................................... 15 References ................................................................................................................................ 19 Appendix ................................................................................................................................... 20 Page \| 2 Background and motivation Green and digital transitions are at the core of the transformational changes needed in European rural areas. The diversification of rural economies has improved their resilience to the consequences of the COVID-19 pandemic, and will play a pivotal role in the recovery by offering development opportunities. The SHERPA process will support the gathering of evidence from across Europe, at multiple levels, regarding the directions of the diversification of rural economies, which are appropriate and feasible to address local needs. SHERPA Multi-Actor Platforms (MAPs) are invited to discuss three key questions: 1) What are the key needs for the development of the rural economy in your MAP, and how can they be addressed most effectively? 2) How can policy interventions support positive changes in the diversification of the rural economy, considering solutions that are needed at the local and national levels, and any implications for the wider policy framework (European Union level or others)? What can public administrations (at all levels) do to facilitate and encourage positive changes in the diversification of the rural economy? 3) What are the research needs and gaps? The exercise is based on the following process: (i) preparation and reflection of the SHERPA Discussion Paper by each regional or national MAP, (ii) consultation with MAP participants, (iii) drafting of the MAP Position Paper, and (iv) synthesis of the regional and national MAP Position Papers for discussion at European Union level. Background and motivat Green and digital transitions are at the co The diversification of rural economies ha pandemic, and will play a pivotal role in t The SHERPA process will support the gath the directions of the diversification of rur needs. 2 Diversification of the rural economy: Entrepreneurship, employment & new business models; Smart rurality, smart communities and digitalisation; Bioeconomy and sustainable management of resources. 3 See Appendix 1 for a compilation of reflections from participants. The Position Paper is the result of two key activities that the Finnish MAP was engaged in during this MAP cycle: The Position Paper is the result of two key activities that the Finnish MAP was engaged in during this MAP cycle: 1) The identification of topics and the writing of the paper as such. 2) The organisation of 4 workshops at the Finnish Rural Parliament 2021. Activity 1 evolved around several interlinked steps including:  an initial discussion of the SHERPA Discussion Paper;  identification of topics of relevance for the Finnish context;  formation of writing teams;  drafting of sub-chapters with feedback provided by the facilitator and monitor acting as editors of this paper and considering input from the Rural Parliament. Activity 2 concerned the organisation and implementation of 4 workshops at the Finnish Rural Parliament in September 2021 on the three dimensions addressed in this paper2 as well as one summary workshop held in Swedish3. The aim was: Activity 2 concerned the organisation and implementation of 4 workshops at the Finnish Rural Parliament in September 2021 on the three dimensions addressed in this paper2 as well as one summary workshop held in Swedish3. The aim was:  to share results from the draft Position Paper as well as from practice examples from Finland and the EU, including examples from MAP Denmark, MAP Poland and MAP Romania;  to exchange experiences and to identify inspiring examples and difficulties;  to identify and discuss needs and possibilities to further improve economic diversification;  to consider future potentials and develop recommendations. Overall, this paper has thus been the result of an initial brainstorming on the key dimensions of economic diversification in rural areas, debates among MAP members who have been centrally involved in co-writing the chapters as well as a fruitful exchange with workshop participants at the Finnish Rural Parliament. On the following pages, the Finnish MAP Position Paper presents some of the key issues related to the process of diversification of production and the rural economy in Finland. It provides a compilation of knowledge around three key dimensions for rural diversification. 2 Diversification of the rural economy: Entrepreneurship, employment & new business models; Smart rurality, smart communities and digitalisation; Bioeconomy and sustainable management of resources. 3 See Appendix 1 for a compilation of reflections from participants Page \| 3 Page \| 3 According to the vision of rural policy, a Finnish "diverse countryside is a national success factor. It provides a platform and solutions for a good life, innovation, entrepreneurship and a sustainable society. Finland will be developed as a whole, strengthening local opportunities". Page \| 4 Page \| 4 Dimensions of diversification There are three interrelated key dimensions of rural diversification that the Finnish MAP has focused on, i.e.:  Diversification of the rural economy: Entrepreneurship, employment & new business models  Diversification of the rural economy: Entrepreneurship, employment & new business models  Smart rurality, smart communities and digitalisation  Bio-economy and sustainable management of resources Each theme is addressed in a dedicated sub-chapter in the following. Chapter authors: Mari Kattilakoski, Taina Väre, Hanna-Mari Kuhmonen Keywords:  Knowledge economy and rural industries g y  Entrepreneurship, social entrepreneurship and social capital  Smart shrinking/adaptation, multi-locality, projects focusing on smart adaptation  Trends in other countries Smart adaptation as an alternative strategy for shrinking areas with declining and ageing population Rural areas have different prospects and therefore require different development strategies. A development focus on perpetual growth will not serve the whole country. The question is: can a region or place be viable without growth? The perspective of smart adaptation (“älykäs sopeutuminen”) or smart shrinking challenges us to think in an alternative way. It broadens the perspective on vitality compared to traditional growth-oriented ways of thinking. Smart adaptation is based on learning to manage change in areas where the population is decreasing. Adaptation and new innovative solutions and developments are needed to respond to these changes. Analyses of vitality should take into account perceived wellbeing and perceptions of “good life”. Change needs to be managed with consideration to all dimensions of sustainable development, including social, economic, ecological and cultural aspects. A resourceful way of thinking and acting is needed to support different strategies as different areas have different needs and conditions. There is a need for a knowledgebase and additional data that considers, for example, increasing mobility, location independence and digitalisation, as well as perceptions of a good life. Smart adaptation as a concept and practice offers approaches to facilitate the transition toward a sustainable society. Important areas of emphasis include  The need to improve statistics on rural types and the use of spatial information for municipal and regional development.  Establishing a regionally and locally applicable roadmap for smart adaptation. 2.1. Finnish Perspectives & good Practices This section addresses the themes of diversification of the rural economy, entrepreneurship, employment and new business models, and strongly relates to the previously mentioned National Rural Policy Programme for 2021–2027. The introduction and development of sustainable bioeconomy and circular economy models will provide new opportunities for rural economic activity. Sustainable solutions can be developed in areas such as in food systems, energy production and tourism. Decentralised models will be emphasised in the development work, which will also support security of supply. Finland's business and export industries rely heavily on the exploitation of natural resources in rural areas, and it is important to increase the local processing of raw materials. Local downstream processing creates entrepreneurship, employment and livelihoods in rural areas. In addition, the exploitation of natural resources (e.g. mining) must be based on minimising the potential negative effects of resource use and the benefits obtained must also be increasingly allocated to the local economy (environmental justice). Environmental justice principles shall be more strongly integrated into legislation and the allocation of budgetary resources. In Finland, public investment in the knowledge economy has been largely concentrated around university towns and cities, which has contributed to regional disparities. The transition to a just and sustainable society requires that national policies are more considerate of rural areas and actors in the knowledge economy. In rural areas, the knowledge economy is strongly linked to natural environments and the ability to combine high-tech skills and research knowledge with practical and local knowledge. The knowledge economy will be increasingly linked to a resource-efficient use of natural resources and technological innovation. For example, new innovations and new partnerships are needed to exploit the side streams of industrial manufacturing processes. Rural enterprises can play a key role in creating and implementing new models. The knowledge economy, based on tangible and intangible resources and knowledge located in rural areas, must be placed at the top of global value chains. The socio-economic structure of rural areas is influenced not only by demographic trends but also by educational levels and labour market structures. Demographic change, particularly in sparsely populated Page \| 5 rural areas, is challenging rural areas to look to the future from perspectives such as smart shrinkage and adaptation and the need for place-based approaches. Demographic change will have an impact on the supply of jobs, education and the provision of services. 2.1. Finnish Perspectives & good Practices As part of the transformation of work, flexible and diverse ways of anticipating skills needs are required. Skills shortages are a key challenge in rural areas. The labour needs of the diversifying rural economies need to be met through the provision of training and employment services. rural areas, is challenging rural areas to look to the future from perspectives such as smart shrinkage and adaptation and the need for place-based approaches. Demographic change will have an impact on the supply of jobs, education and the provision of services. As part of the transformation of work, flexible and diverse ways of anticipating skills needs are required. Skills shortages are a key challenge in rural areas. The labour needs of the diversifying rural economies need to be met through the provision of training and employment services. The transformation of work along with digitalisation are creating new opportunities for studying, working and for entrepreneurship in a location-independent way. Indeed, work is partly place-independent, offering more and more people the opportunity to live and work in rural areas. Increasing mobility is reflected in the daily lives and prospects of more and more people, as well as in working life and in business. Good practices / projects / tools /methods  YTYÄ: project focusing on societal entrepreneurship in rural areas (Yhteiskunnallinen yrittäjyys maaseudulla). Ruralia Institute, University of Helsinki.:https://www2.helsinki.fi/fi/ruralia- instituutti/yhteiskunnallinen-yrittajyys-maaseudulla-ytya  YTYÄ: project focusing on societal entrepreneurship in rural areas (Yhteiskunnallinen yrittäjyys maaseudulla). Ruralia Institute, University of Helsinki.:https://www2.helsinki.fi/fi/ruralia- instituutti/yhteiskunnallinen-yrittajyys-maaseudulla-ytya Ä /y y jyy y y  Äly: project focusing on smart specialisation in Finland. University of Eastern Finland. https://uefconnect.uef.fi/tutkimusryhma/mita-on-alykas-sopeutuminen-suomessa/  Remote working hubs as platforms for increasing vitality (Etätyöpisteet elinvoiman kasvualustoina). htt // it fi/h kk t/ ht i t t h j h kk t/ t t i t t li i k l t i 1  Remote working hubs as platforms for increasing vitality (Etätyöpisteet elinvoiman kasvualustoina). https://www.witas.fi/hankkeet/yhteistyotahojen_hankkeet/etatyopisteet_elinvoiman_kasvualustoina_1 119  Perspectives on smart specialisation from Scotland: https://uefconnect.uef.fi/wp- content/uploads/2021/08/Projektitiedote-20.8.2021.pdf  Perspectives on smart specialisation from Scotland: https://uefconnect.uef.fi/wp- content/uploads/2021/08/Projektitiedote-20.8.2021.pdf Recommendations:  Assessing the socio-economic and regional economic impact of available broadband connections at municipal level and businesses in rural areas.  Assessing the socio-economic and regional economic impact of available broadband connections at municipal level and businesses in rural areas.  Promote opportunities for location-independent work through (community-based) remote working facilities.  Promote opportunities for location-independent work through (community-based) remote working facilities.  Launch the preparation of a national strategy for remote work and accelerate the take-up of remote work and recruitment.  Launch the preparation of a national strategy for remote work and accelerate the take-up of remote work and recruitment.  Launch trials and development of remote work in cooperation between the public, private and third sectors.  Launch trials and development of remote work in cooperation between the public, private and third sectors. Multi-locality and place-independence as an opportunity Digitalisation and place-independence are linked to business, employment and labour market changes, as well as to new ways of organising, delivering and using services in rural areas. Digitalisation enables new place-independent ways of studying, working and entrepreneurship. The COVID-19 pandemic has accelerated the increase of location-independent work. The growing trend towards multi-locality is reflected in the daily lives and prospects of more and more people. Multi-locational living, linked to both leisure and work transition and seasonal work, brings new actors and activities to rural areas and also affects the vitality of municipalities and regions. It also increases the demand for public and private services. Recommendations:  There is a need for research on multi-locality as well as improving information systems for monitoring this phenomenon, both in connection to addressing current questions but also in the long-term.  Multi-locality needs to be taken into consideration in decision making. The demand for rural areas must be met by seeking new solutions.  Multi-locality needs to be taken into consideration in decision making. The demand for rural areas must be met by seeking new solutions.  There is a need for actions that support and enhance multi-local employment and entrepreneurship.  Identifying the bottlenecks of multi-locality: the use of geographic information and information systems, impact assessment, legislative bottlenecks, etc.  There is a need for actions that support and enhance multi-local employment and entrepreneurship.  Identifying the bottlenecks of multi-locality: the use of geographic information and information systems, impact assessment, legislative bottlenecks, etc. Page \| 6 Page \| 6 Technological developments and changes in lifestyles and attitudes are affecting the way that people work in different sectors. Work has become increasingly place-independent, allowing more and more people to live and work in rural areas. For example, professional work no longer determines where people live in the same way as it did in the past. This change will affect the opportunities for young people, highly qualified professionals and women in particular to live and work in rural areas. Fast and reliable communication technology and broadband internet will reduce the disadvantages connected with geographical distances for business, education and the provision of services. They enable people to work, be self-employed, study, combine work and leisure, and produce services. Fibre optic connections can contribute to positive demographic and business development in the area. High-speed, high-quality broadband is an investment in the vitality of rural areas. In Finland, fibre optic deployment appears to have been slower than expected and national and EU policy targets have not been met. Improving access to high-speed broadband in areas where it is not being built on market terms will continue to require public support and better national coordination. Co-construction of infrastructure also needs to be further promoted. Entrepreneurship in rural areas and new business models The rural economy comprises a multitude of different activities and industries in addition to the traditional ones such as agriculture, forestry and fishing. Some of these are tied to certain specific places in rural areas whereas others are location independent. It is also important to consider the various needs for renewal that businesses may face. It is therefore important to contribute to and support business restructuring as part of Recommendations:  Generating knowledge about the potentials that increasing local processing of natural resources may bring. g Promoting the development of business activities that rely on a sustainable use of nature. g  Promoting the development of business activities that rely on a sustainable use of nature.  Developing sustainable tourism as part of rural entrepreneurship (including multi-sectoral entrepreneurship). Tourism also provides benefits for other economic activities. R i b i t l l t  Developing sustainable tourism as part of rural entrepreneurship (including multi-sectoral entrepreneurship). Tourism also provides benefits for other economic activities. Removing barriers to local procurement.  Promoting dialogue between actors, e.g., in public procurement. Promoting dialogue between actors, e.g., in public procurement. New business models: sharing economy, platform economy, etc.  Supporting existing business opportunities: business services, training. pp g g pp , g  Mapping the scope and potential of social entrepreneurship in rural Finland. Information, activation and incentives (e.g. tax benefits) for social/community entrepreneurship are needed.  Mapping the scope and potential of social entrepreneurship in rural Finland. Information, activation and incentives (e.g. tax benefits) for social/community entrepreneurship are needed. ( g ) / y p p  In order to develop social entrepreneurship, competition law should take into account the overall regional economic impact of social entrepreneurship and the promotion of a sustainable transition in our society.  In order to develop social entrepreneurship, competition law should take into account the overall regional economic impact of social entrepreneurship and the promotion of a sustainable transition in our society.  Creating conditions for the development of cooperatives and social entrepreneurship in rural areas and exploring the possibilities of creating financial instruments to support service production in villages, between public and private funding.  Creating conditions for the development of cooperatives and social entrepreneurship in rural areas and exploring the possibilities of creating financial instruments to support service production in villages, between public and private funding.  Providing services not for profit, but for the benefit of the public good. So-called value-based enterprise: on the basis of associations or cooperatives.  Providing services not for profit, but for the benefit of the public good. So-called value-based enterprise: on the basis of associations or cooperatives.  The use of business models and forms of the platform economy should be increased as part of upgrading, product development and innovation in rural areas. 4 Services of General Economic Interest. On SGEI in Finland, see, for instance Ministry of Economic Affairs and Employment of Finland (2021). Page \| 7 Page \| 7 a sustainable transition, creating a pathway to a new livelihood and employment opportunities in rural areas. It is important to raise issues, including environmental justice and the local and economic effects of a sustainable use of natural resources, in the public debate. a sustainable transition, creating a pathway to a new livelihood and employment opportunities in rural areas. It is important to raise issues, including environmental justice and the local and economic effects of a sustainable use of natural resources, in the public debate. A culture that encourages economic activity and networks that support local and regional vitality are important. This can help attract new actors. Changes of ownership and generational patterns provide opportunities for young people and others interested in entrepreneurship in rural areas. Increasing productivity and supporting the development of new innovations requires, for instance, business advisory, consultancy, financial and internationalisation services that recognise the specificities of rural entrepreneurship. These services are needed to increase the level of upgrading and development of new innovations and they can be provided and developed through multiple service producers and new models and tools. For example, crowdfunding offers one option for rural start-ups and growth companies. Recommendations:  Exploring and promoting the introduction of the SGEI4 service obligation procedure for essential services that are not provided on market terms and “ensuring the availability of economic services that are important to citizens” (See Kettunen et al., 2015).  The multi-provider model (monituottajamalli) can, within the framework of the legislation, provide locally based solutions, make use of local entrepreneurship and community-based service provision. p  Develop multi-service centres to bring public, private and third sector services closer to rural residents, and to develop community-based entrepreneurship and teleworking. Developing the operation and funding models of multi-service centres. p  Develop multi-service centres to bring public, private and third sector services closer to rural residents, and to develop community-based entrepreneurship and teleworking. Developing the operation and funding models of multi-service centres.  Bringing a rural perspective and the perspectives of rural-based enterprises to the implementation of the roadmap based on the National Entrepreneurship Strategy.  Bringing a rural perspective and the perspectives of rural-based enterprises to the implementation of the roadmap based on the National Entrepreneurship Strategy. 3.1. Finnish Perspectives & good Practices The “Smart theme” and related processes and structures are still largely in their early stages from the rural perspective. However, much has already happened, is happening and is also under planning in Finland as well. This subchapter focuses mainly on describing the current situation. Smart growth needs also include smart adaptation / shrinkage perspective. Smart adaptation is one of the current key themes of the Rural Policy Programme 2021-2027, and one of the crosscutting operating principles of the National Island Programme 2020-2023, which will be promoted in the coming years. There are already projects and studies under way on this subject, concerning both the national, regional and municipal levels. In a broader sense, both social and governance/administrative Page \| 8 Page \| 8 Employment, access to workforce and transformation of labour The concentration/centralisation of education is contributing to widening regional disparities within and between regions. The negative impact is felt at company and sectoral level, reducing the willingness of companies to invest and develop their activities. Decisions in education policy have a great impact on training opportunities and the availability of skilled labour in rural areas. The shortage of skilled labour is a key challenge in rural areas. In order to attract skilled labour, ways must be found to secure secondary education for young people in rural areas. This can be done through the use of blended teaching (i.e. combining both contact and remote teaching), distance learning solutions to complement face-to-face teaching and fostered by increased cooperation between schools. There is a need to share good practice in training cooperation suitable for rural enterprises. Work-based learning should be supported by strengthening the resources, skills and approaches to guidance for both teachers and workplace representatives. Digital opportunities for continuous learning are evolving and changing the way we think about the spatial concentration of education and accessibility. Both digital opportunities and opportunities for cooperation between education and training organisations must be exploited to strengthen accessibility to higher education and vocational training, including in sparsely populated areas. Comprehensive training provision tailored and specific training, rural employment services and smoothly functioning labour migration are needed to meet labour needs. As part of the labour transition, there is a need to anticipate skills needs as well as for employment services and business advisory models adapted to rural contexts. 3. Smart rurality, smart communities and digitalisation Chapter authors: Sami Tantarimäki, Petri Rinne, Virpi Harilahti-Juola and Olli Lehtonen, Chapter authors: Sami Tantarimäki, Petri Rinne, Virpi Harilahti-Juola and Olli Lehtonen, Keywords:  Smart growth needs to also include smart adaptation  Local/village level development, LEADER, smart villages  Renewal of rural vitality  Using resources in a smart way  Digital technology  Social & governance aspects of smartness  Municipalities & digital societies / role of municipalities in digitalisation processes  Organisations promoting digital readiness  Opportunities of and needs for remote work  Regeneration of vitality  Structural changes in the labour market / re-structuring Smart Villages The social dimension and communality have also played a strong role in launching the concept of a smart village. The concept has been promoted, for example by the Smart Village theme group work (2019-2020) of the Finnish Rural Network, and by the smartest village in Finland competition (Rural Network of Finland). This also describes the development at the local level and lays the foundation for LEADER work in the forthcoming CAP period. In addition, a new Finnish-Swedish thematic group (of the Rural Network) “Smarta landsbygder i Svenskfinland och Norden” has started, and other activities have emerged, e.g. Smart Villages Facebook-group, cooperation between SV-villages and international cooperation SV-projects such as Smart Rural 21. In addition, Finnish rural areas/cases are included in (other than SHERPA) Horizon 2020 projects (e.g. AURORAL, DESIRA). Smart adaptation and resilience strive for and promote a good life. Therefore, a good life has become a key concept in rural speech and development, too. In addition to measuring the level of wellbeing, it is necessary to know where people’s good life stems from. This has been studied in several projects, such as HYVIS - Hyvän elämän jäljillä (looking for a good life, MDI), Kestävä maaseutu (Sustainable/resilient rural areas, Maaseudun Sivistyslitto MSL, UEF), SOMA - Sopeutuvat innovatiiviset maaseudut (Adapting innovative rural areas, UEF/SPATIA) and the HYMA network of the Rural Policy Council (UEF, MSL, Kajaani AMK). Page \| 9 Page \| 9 dimension of smartness and the role of municipalities in digitalisation processes have been at the centre of these projects. Table 1 displays projects, organisations involved and a link to people involved. Table 1. Finnish projects dealing with smart rurality and smart adaptation Acronym Finnish Title English Title Key partners and website ALKUVOIMA Esiselvitys: Älykäs sopeutuminen Pohjois- Karjalan maaseudulla Pre-study: Intelligent adaptation in rural areas in Northern Karelia University of Eastern Finland (UEF)/SPATIA https://uefconnect.uef.fi/en/group/smart- shrinking-in-the-north-karelian-rural- areas/ ÄLY Älykäs sopeutuminen Suomessa Smart adaptation in Finland UEF/SPATIA https://www.uef.fi/en/article/what-does- smart-shrinking-mean-in-finland PISARA Pienten kuntien strategiset ja luovat ratkaisut Strategic and creative solutions for small municipalities UEF/SPATIA Älykkäistä kylistä älykkäisiin suurkaupunkeihin From smart villages to smart cities University of Turku/Brahea Centre JÄRKEVÄ Järjestökenttä tietotalousosaamisen välittäjänä maaseudulla NGOs Imparting Knowledge Economy in Rural Areas UTU/Brahea Centre, TIEKE ry, Kainuun Nuotta DIGIKUNTA Varsinais-Suomen Kuntien digiyhteiskunta- ja post-covid –valmius South-West Finland municipalities' digital society and post- COVID preparedness UTU/Brahea Centre The reorganisation of mobility and transportation is also an important example in this context (e.g. MaasDigiboksi – the Digital agenda of rural transport and traffic). At the same time all these examples show how different organisations promote digital readiness. The reorganisation of mobility and transportation is also an important example in this context (e.g. MaasDigiboksi – the Digital agenda of rural transport and traffic). At the same time all these examples show how different organisations promote digital readiness. Remote work Opportunities of and needs for remote work have increased with COVID-19. It is a good example of the renewal or regeneration of the vitality of rural areas, supported strongly by the possibilities of smart solutions and digitalisation. Remote work has become an everyday practice also in rural areas. Attention has been paid to the importance of the matter and determined development has also been undertaken. The theme Concluding reflections In Finland, the strategic approach to the smart rurality, smart communities and digitalisation of rural areas is well advanced. Rural development and research networks are also strong. However, there is always room for improvement. Implementation is important for progress, success and future. In this context, among the key aspects that should be considered are financing, experiments, new companionships, participation and long-term development. Page \| 10 Page \| 10 Page \| 10 has been discussed in the Smart Village / Smart Rural dialogue, and projects have been launched such as Remote Working Hubs (Leader Aisapari) and Hubittaako (Municipality of Pihtipudas). Led by these, a national network of rural cooperation scenarios was established in 2020, which will spar, share experiences and information. A website has also been created (https://www.etatyotilat.fi/). Networking of remote work and co-working spaces will continue with new national network projects funded by the National Rural Policy Council and the Ministry of Economic Affairs and Employment of Finland. Structural changes in the labour market and the changing geography of work are now at an interesting stage. has been discussed in the Smart Village / Smart Rural dialogue, and projects have been launched such as Remote Working Hubs (Leader Aisapari) and Hubittaako (Municipality of Pihtipudas). Led by these, a national network of rural cooperation scenarios was established in 2020, which will spar, share experiences and information. A website has also been created (https://www.etatyotilat.fi/). Networking of remote work and co-working spaces will continue with new national network projects funded by the National Rural Policy Council and the Ministry of Economic Affairs and Employment of Finland. Structural changes in the labour market and the changing geography of work are now at an interesting stage. 3.2. Opportunities, Challenges & Recommendations for and from Finland .2. Opportunities, Challenges & Recommendations for and from Finlan In Finland the key challenge for smart rurality is related to the availability of broadband access. At the end of 2018 broadband was available in 27 191 population grids, which corresponds to 27.2% of inhabited population grids. Extending the geographical coverage of broadband availability and avoiding the deepening of the digital divide requires top-down coordination in broadband construction and more regionally tailored public funding. Research has shown that the telecommunication policy in Finland itself has been successful, and the aim to increase availability of broadband infrastructure in rural areas is encouraged in the future. This aim is supported by the fact that the mobile download speed varies a lot between regions. In addition, mobile download speed in rural areas has lagged behind those in urban and intermediate regions. Figure 1. Download speeds in Finland Source: Olli Lehtonen 2020 Source: Olli Lehtonen 2020 4.1. Finnish Perspectives, good practices & challenges 4.1. Finnish Perspectives, good practices & challenges Page \| 11 Page \| 11 Experiences are beginning to accumulate, and new projects will generate more knowledge, as well as research data. However, evaluation and research are both still needed. The forthcoming CAP 2021-2027 period will provide a key funding element to promote the theme, but other sources of funding are also possible and necessary. “Smart” is a multidisciplinary, wide-ranging and crosscutting theme and therefore the importance of coordination and cooperation is emphasised. Keywords:  Diversification of agriculture AND in forestry / fisheries sectors  Knowledge economy  Employment as enabler (cf. national strategic programmes)  Business models  Strategic links of rural policy to the bioeconomy / bioeconomy strategies  Social entrepreneurship & multi-locality  Regional perspectives on the bioeconomy  Diversification of agriculture AND in forestry / fisheries sectors  Knowledge economy  Employment as enabler (cf. national strategic programmes)  Business models  Strategic links of rural policy to the bioeconomy / bioeconomy strategies  Social entrepreneurship & multi-locality  Regional perspectives on the bioeconomy Social entrepreneurship and renewable energy A significant proportion of Europe's economy is organised via the Social Economy, and the socio-economic significance of social-economy enterprises is also widely recognised in European and national policies. One definition of the Social Economy is established by CIRIEC in the European Economic and Social Committee report “Social Economy in the European Union”: “The set of private, formally organised enterprises, with autonomy of decision and freedom of membership, created to meet their members' needs through the market by producing goods and providing services, insurance and finance, where decision-making and any distribution of profits or surpluses among the members are not directly linked to the capital or fees contributed by each member, each of whom has one vote, or at all events take place through democratic and participatory decision-making processes. The social economy also includes private, formally organised organisations with autonomy of decision and freedom of membership that produce non-market services for households and whose surpluses, if any, cannot be appropriated by the economic agents that create, control or finance them.” The social enterprises are an important tool for regional policy, where investments to the social enterprises in renewable energy offer development opportunities for distant and sparsely-populated rural areas. These areas often suffer from a one-sided economic structure, and they struggle with negative development trends, making these areas dependent on external subsidies and support. In these areas, the re-investments of revenues often offer an opportunity for organising local services, developing community businesses and investments in infrastructure and communication. For instance, the revenues can be used to secure basic services for instance in the fields of health and education in peripheral rural areas with declining and ageing populations, where service provision generally is a challenge. Positive attitudes toward renewable energy facilitate the potential use of social enterprises in improving local development, but government financing alone will not sustain community action on energy. Thus, the social enterprises also need local activity and full engagement, which could be linked with bottom-up place-based approaches to regional policy. Local people should be involved in the planning of re-investments, as they potentially have the best knowledge about the local strengths which constitute the basis of place-based thinking. Experiences from Scotland (Okkonen and Lehtonen, 2016) suggest that government support for social enterprises including, for instance, subsidies and grants to investments or infrastructure, can support local economic development. Introductory reflections village associations), can play a key role in introducing as well as creating such models, but this requires providing small operators/actors with opportunities. communities in rural areas. Small- and medium-sized enterprises (SMEs) and micro-enterprises, as well as local societies (e.g. village associations), can play a key role in introducing as well as creating such models, but this requires providing small operators/actors with opportunities. Finland's current bioeconomy strategy dates to 2014. The strategy was updated in 2021 to reflect changes in the operating environment and the government programme. The vision of the strategy is: “sustainably towards higher added value”. The strategy promotes social, ecological and economic sustainability, carbon neutrality and the circular economy. Introductory reflections According to the Rural Barometer 2020 study commissioned by the Rural Policy Council and produced by the Natural Resources Institute Finland, 68% of Finland’s citizens think that the role of rural areas will be emphasised in the future, as part of the green transition (Maaseutubarometri, 2020). According to the study, 77% of citizens see significant economic possibilities in the bioeconomy and the use of renewable resources. Furthermore, 68% of citizens believe that rural areas provide a good operating environment for innovative entrepreneurship. However, the study also shows, that 36% of citizens see substantial problems in the bioeconomy and in the use of renewable resources. The findings highlight the potential of the bioeconomy to strengthen rural areas (and society as a whole), but also point to the need for a broad dialogue between different actors. The Rural Policy Programme 2021-2027 (Kattilakoski et al., 2021) emphasises the need for place-based policy and development, and for retaining added value in rural areas. Bioeconomy is ultimately realised at the local level, but it is steered through multiple strategies, both at national and the EU level. Various strategies are integrated at the regional level, through cooperation within competence centres/clusters. Most of Finland’s natural resources are located in rural areas. As such, the Rural Policy Programme 2021- 2027 underlines the need for fair and just processes linked to the use of natural resources. The utilisation of renewable natural resources must be done through socially-sustainable processes and furthermore it should be linked to a broader perspective of fairness, according to which it is vital to safeguard the rights and opportunities of people and societies living in rural areas (including rights to basic services and infrastructure). In order to build and develop a bioeconomy and a society that is both carbon-neutral and sustainable we need scalable models that suit different places and environments. In order to introduce and develop such sustainable models we need dialogue and partnerships between different actors (public, private, people; education providers and enterprises) at different levels (local, regional, national and international), as well as research and inclusive development in order to translate new knowledge into practical applications. Bioeconomy and circular economy models have and will open new doors for both businesses and Page \| 12 communities in rural areas. Small- and medium-sized enterprises (SMEs) and micro-enterprises, as well as local societies (e.g. Local business models: The case of Eno Energy Cooperative The business concept of Eno Energy Cooperative5 focuses on the production of district heating energy by providing woodchips for the three heat production / distribution plants in the area. The energy cooperative is perhaps the most studied example of a local bioenergy system (for instance Lehtonen and Okkonen, 2019). Joint development work between the former municipality of Eno, the forest centre and forest owners made it possible to utilise forest energy locally and created a local production entity. The research findings from the Eno Energy Cooperative underlines that the use of forest energy can play a significant role in strengthening local development. Investments in the utilisation of local forest energy can have a significant impact on the number of jobs, the income level of the population and the availability of local services. In this way, the forest energy self-sufficiency system maintains economic activity and strengthens the prospects of local economies. It also enhances the sustainable use of local natural resources and reduces the local economy's dependence on externally produced energy. The experience from the Eno Energy Cooperative also proved that, in addition to traditional socioeconomic impacts, the energy cost savings, and their induced impacts, can create a remarkable additional increase to the local development benefits of bioenergy. As the example of Eno Energy Cooperative shows, local bioenergy production has the potential for significant benefits to the regional economy, but these are not self-evident (cf. common benefit / private benefit). Economic competition can lead to increasing costs and competition in the supply of raw materials, and if local energy is to be abandoned, regional economic benefits can also be lost. This can happen, for example, when local wood biomass is not cost-competitive for fuels, which are imported from outside the region. Therefore, local decision-making and policies also play a key role in defining the local energy production model and its success. Bioenergy entrepreneurship does not necessarily compete with other wood uses, as it can also support local forestry by creating a market for thinning small-diameter wood. This will improve tree growth and production in a thinned forest, generating increasing incomes for forest owners and raw material for the forest industry. Eno Energy Cooperative's research shows that regional economic impacts and socio-economic benefits can be maximised by: a) local operational business model and use of local raw materials (e.g. Key findings:  Previously known direct, indirect, and induced impacts of the investments and production are only a part of the socioeconomic impact of the production systems in the bioeconomy - the profitability can increase and even double the development benefits of bioeconomy investments.  Previously known direct, indirect, and induced impacts of the investments and production are only a part of the socioeconomic impact of the production systems in the bioeconomy - the profitability can increase and even double the development benefits of bioeconomy investments. p y  The regional economic impact of local bioenergy systems needs to be considered in the long term over economic cycles.  The regional economic impact of local bioenergy systems needs to be considered in the long term over economic cycles. 5 For more information please visit their website at Regional perspectives on bioeconomy strategies The purpose of this chapter is to demonstrate and discuss the socioeconomic impacts related to place-based bioeconomy development strategies, focusing on a large-scale biochar factory and its associated industries. The analysis focuses on the impacts of the development strategy on the incomes and employment in a small peripheral municipality in Pielinen Karelia, Finland. The analysed development strategy of the bioeconomy is Social entrepreneurship and renewable energy This highlights the importance of supporting the regional development functions of social enterprises, for instance, through the allocation of development funds or adjusting the financial support rates. The regional development functions of social enterprises could be coordinated by the community development agencies or local authorities to ensure common benefit. Key findings:  Social enterprises and a place-based development concept could be used in regions to roll-out its benefits and support the development of the disadvantaged distant areas which are not competitive enough themselves to generate endogenous growth.  Social enterprises and a place-based development concept could be used in regions to roll-out its benefits and support the development of the disadvantaged distant areas which are not competitive enough themselves to generate endogenous growth. g g g g  In peripheral regions, the local job growth in rural communities will be the primary means of improving their economic outcomes and keeping these communities inhabited.  In peripheral regions, the local job growth in rural communities will be the primary means of improving their economic outcomes and keeping these communities inhabited. Page \| 13 Local business models: The case of Eno Energy Cooperative Local business models: The case of Eno Energy Cooperative community entrepreneurship, cooperatives), b) sustainable use of local wood biomass without compromising future use of wood biomass (e.g. using forest residues), c) cost savings achieved by customers, and (d) reinvesting potential profits (e.g. other community activities) (Lehtonen, 2019). International examples of local socio- economic benefits can also be found, for example, in community energy projects in the UK and Canada (Simcock et al., 2016). Key findings: 5 For more information please visit their website at http://www.enonenergia.fi/Business concept of Eno For more information please visit their website at http://www.enonenergia.fi/Business_concept_of_Eno_Energy_Cooperative. 5 For more information please visit their website at http://www.enonenergia.fi/Business_concept_of_Eno_Energy_Cooperative. Page \| 14 Page \| 14 an example of a place-based policy, involving local actors and the knowledge of industrial heritage and processing, as well as global investments that support local natural resource utilisation with new technology. Our case study demonstrated (for an in-depth analysis see Lehtonen and Okkonen, 2016) that the positive development in resource peripheries can be supported by a local development bioeconomy strategy when these peripheries take advantage of the absolute advantages derived from natural resources and amenities by concentrating on local, highly competitive production factors. an example of a place-based policy, involving local actors and the knowledge of industrial heritage and processing, as well as global investments that support local natural resource utilisation with new technology. Our case study demonstrated (for an in-depth analysis see Lehtonen and Okkonen, 2016) that the positive development in resource peripheries can be supported by a local development bioeconomy strategy when these peripheries take advantage of the absolute advantages derived from natural resources and amenities by concentrating on local, highly competitive production factors. The successful development strategies highlight the potential for implementing “new initial” advantages in the green economy, emphasising a new way of utilising natural resources in rural areas. Therefore, the locational conditions of these types of economic activities should receive more attention in regional policy and planning, with a framework of place-based practices that ensures the participation of a committee of local actors and the inclusion of unique types of possibilities that each location offers. The positive development must be constructed locally and be customised for the specific conditions in different rural areas, as they are often weakly connected with growing regional urban centres and since they have various locational disadvantages. In addition, distant areas in sparsely-populated regions have notably limited opportunities to benefit from the agglomeration economies of centres and their competitive environments with spread effects. All this implies that their economic development should be constructed and supported locally. The dependency of a renewal of the local economy from the external actors in the resource periphery was evident from the fact that a high proportion of the development strategy's investments leaked outside the region. The results of the implementation of the bioeconomy strategy proved that the growth in resource peripheries not only depends on endogenous factors as regions face many development disadvantages and structural problems but also depend on imported technologies. Page \| 14 However, most of the socioeconomic impacts of production are directed at and induced by the local economy; therefore, the intensive use of local resources with new technologies can create development in the resource periphery and diversify the economic structure. Without bioeconomy strategies that aim toward new investments and upgrading industrial production, there is a high risk that stagnating areas would suffer from negative regional lock-in effects: increasing depopulation, declining employment and increasing unemployment. This lock-in does not inhibit new paths but has an impact on opportunities and limits possibilities for new directions, keeping the negative path in the region alive. Key findings:  A successful local bioeconomy requires local knowledge about the industrial traditions, resources, actors and cooperation partners, but it also requires external technology and capital to implement a development strategy.  A successful local bioeconomy requires local knowledge about the industrial traditions, resources, actors and cooperation partners, but it also requires external technology and capital to implement a development strategy. g  The new local bioeconomic-related strategies are noteworthy in supporting the development of industrial towns in transition in sparsely populated rural regions.  The new local bioeconomic-related strategies are noteworthy in supporting the development of industrial towns in transition in sparsely populated rural regions. Recommendations and Conclusions How could innovation be created in a rural-driven way?”  “COVID-19 permanently changed working life, so in addition to digitalisation and working from home, work/meeting spaces from the municipality/community are needed for meetings, networking, etc.”  “Best practice method: the LEADER method, the method of locally led development”  "All LEADER LAGs in Finland address community development or rural economies in their strategies, so I fully support that LAGs can do a lot in this area!"  "Local stakeholders know best their area and are working on daily concrete actions. Fighting with adaptation and future change management"  “Opportunities: bottom-up approach through locally led development. Public, private and third sector influence and act together, engage grassroots to be involved in the community and implement actions through EU and nationally funded projects Challenges: "trust" from authority, limited resources”  “Local experiences, dialogue between different actors and sectors of paramount important.”  “One man's rubbish can be another man's treasure, i.e. side streams and even waste can be an excellent raw material for someone else. And to keep transport distances short, the benefits come locally, low-threshold cooperation between entrepreneurs and getting to know each other is the key”  “Decentralised biogas production is an opportunity in rural Finland, because we can use a lot of biomass (e.g. grassland) in production. For example, selling biogas for transport can be an opportunity to diversify rural entrepreneurship.”  “Opportunities and challenges, the use of local social resources. Need to reach out beyond NGOs, to citizens at large. Associations are perhaps changing, people want to contribute to development here and there, perhaps not working 20 years in the same association. Challenge, how to involve citizens on a broad level, there are a lot of unused resources.”  “Decentralised biogas production is an opportunity in rural Finland, because we can use a lot of biomass (e.g. grassland) in production. For example, selling biogas for transport can be an opportunity to diversify rural entrepreneurship.” This paper has addressed the general topic of change in production and diversification of the rural economy with specific focus on Finland. This topic has been examined from the perspectives of three main thematic areas: entrepreneurship, employment and new business models; smart rurality, smart communities and digitalisation; bioeconomy and sustainable management of resources. Recommendations and Conclusions The purpose has been to highlight Finnish perspectives and good practices relating to these thematic areas as well as to identify key opportunities and challenges. Based on the examples discussed in the paper, the intention has also been to provide recommendations for rural areas in Finland and in Europe more broadly. Recommendations and Conclusions Finland and other rural areas in the EU and beyond are dealing with numerous interlinked challenges including climate change and sustainability of production, population ageing and shrinkage. At the same time, and according to the EU’s Long-Term Vision for Rural Areas, rural areas can become more prosperous by diversifying economic activities to new sectors with positive effects on employment (European Commission, 2021). In line with this image, most of the SHERPA MAPs shared the vision of a diversified rural economy thus “in 2040 the rural economy will be diversified, with non-agricultural activities adding to the sustainability of rural areas” (Chartier et al., 2021). Enablers to achieve this vision, and as highlighted in the new Finnish National Rural Policy Programme for 2021–2027, include digitalisation, location independence and increasing mobility of people, and finding ways and tools propelling a shift towards sustainable, low-carbon activities and economies. This change requires, Page \| 15 Page \| 15 as the authors of the Programme stressed, a broad debate on everyday life and entrepreneurship in rural areas and the impact of decisions across the country and for different types of areas. This paper can be seen as a contribution of our MAP members to this ongoing debate. Before we summarise their views, Figure 2 below gives the floor to participants of our four workshops at the Rural Parliament and to share their views (see Appendix for more). The debate will and has to continue. as the authors of the Programme stressed, a broad debate on everyday life and entrepreneurship in rural areas and the impact of decisions across the country and for different types of areas. This paper can be seen as a contribution of our MAP members to this ongoing debate. Before we summarise their views, Figure 2 below gives the floor to participants of our four workshops at the Rural Parliament and to share their views (see Appendix for more). The debate will and has to continue. 2. Selected messages from participants of 4 Sherpa workshops at the Finnish Rural Parliament, 28 ber 2021 “Diversification requires innovation which calls for networks and cooperation Changing the typical Figure 2. Selected messages from participants of 4 Sherpa workshops at the Finnish Rural Parliame September 2021 September 2021  “Diversification requires innovation, which calls for networks and cooperation. Changing the typical subordinate role of rural areas to cities in innovation development. Page \| 16 Overall, our paper highlights the importance of sustainable bioeconomy and circular economy models in providing new economic opportunities for rural areas and communities in a wide range of areas ranging from areas such as food systems to energy production and tourism. The importance of decentralised models is stressed, as they can also support security of supply. As argued in the first chapter focusing on entrepreneurship, employment and new business models, public investments in the knowledge economy are vital for diversifying the rural economy. However, in Finland, these investments have primarily been directed to university towns, which has contributed to increased socio-spatial disparities. Hence, transitioning towards a more just and sustainable society requires more consideration of the local conditions of Finland’s diverse rural areas. The socio-economic fortunes of rural areas are influenced by a multitude of factors. For instance, demographic change has an impact on the supply of jobs, education and service provision, and skills shortages are among the key challenges that Finnish rural areas are confronted with. This calls for the need for flexible ways of anticipating skills requirements, as well as diverse training and employment services. From the perspective of attracting and maintaining a skilled labour force, the chapter authors highlight the importance of securing secondary education for young people in rural areas. Multi-locality and new placed-independent ways of studying, working and entrepreneurship are aspects that present numerous opportunities for rural areas. In relation to this, our paper stresses the need for research on multi-locality and developing information systems for monitoring this phenomenon, both regarding the current situation and with consideration to long-term developments. In relation to entrepreneurship, something that is emphasised as important is a culture that encourages economic activity and networks that support local and regional vitality, as this can help attract new actors. So-called value-based enterprises where services are not provided for profit but for the benefit of the public good is another key aspect which could benefit local economies of rural communities. Additionally, changes of ownership and generational patterns may also provide opportunities for young people and others that can benefit entrepreneurship especially in rural areas dealing with challenges of demographic change. The second chapter dealing with smart rurality, smart communities and digitalisation highlights the importance of the smart adaptation concept, which is currently a core theme in Finnish rural policy. Page \| 16 II) Smart rurality, smart communities and digitalisation  Smart adaptation is one of the themes of the Rural Policy Programme 2021- 2027 and a cross-cutting principle of the National Archipelago Programme 2020-2023, which will be promoted in coming years.  Examples of relevant initiatives: the first wider national project “ÄLY- Älykäs sopeutuminen Suomessa” (smart adaptation in Finland, UEF/SPATIA); National thematic group on Smart Villages (2019-2020); the Smartest Village in Finland competition (2018-2020). III) Bioeconomy and sustainable  Success in bioeconomy requires the need for local knowledge about industrial traditions, resources, actors, and cooperation partners, but also external technology and capital for implementing strategies.  New local bioeconomy-related strategies support the development of sparsely populated rural regions. Selected key messages I) Diversification of the rural economy: Entrepreneurship, employment and new business models  Transformation of labour and digitalisation enable place-independent work, studying and entrepreneurship in new ways.  Digitalisation and place-independence are linked together with entrepreneurship, employment and transformation of labour as well as new ways of arranging, producing and using services in rural areas. II) Smart rurality, smart communities and digitalisation II) Smart rurality, smart communities and digitalisation II) Smart rurality, smart communities and digitalisation  Smart adaptation is one of the themes of the Rural Policy Programme 2021- 2027 and a cross-cutting principle of the National Archipelago Programme 2020-2023, which will be promoted in coming years.  Examples of relevant initiatives: the first wider national project “ÄLY- Älykäs sopeutuminen Suomessa” (smart adaptation in Finland, UEF/SPATIA); National thematic group on Smart Villages (2019-2020); the Smartest Village in Finland competition (2018-2020). III) Bioeconomy and sustainable Page \| 16 Smart adaptation is closely connected to the notion of promoting a good life for people in rural areas. As this concept has emerged and become increasingly established in the rural debate, the chapter authors emphasise the importance of understanding what constitutes the basis of well-being and quality of life as well as having ways of measuring these aspects. From the perspective of smart rurality, one of the core challenges is related to broadband access. Improving broadband internet access is important for overcoming a digital divide between urban and rural areas, and it requires top-down coordination in broadband construction accompanied by more regionally-tailored public funding. While telecommunication policy in Finland can generally be regarded as successful, one area of emphasis in future policy should be to improve the broadband infrastructure in rural areas. The chapter focusing on bioeconomy and a sustainable management of resources calls for socially sustainable processes in the utilisation of renewable natural resources. These should be linked to the broader perspective of fairness, which is anchored in the need to safeguard the rights and opportunities of people in rural communities, including the right to basic services and infrastructure. Furthermore, environmental justice perspectives and the local and economic effects of a sustainable use of natural resources are important aspects to bring into the public debate. Building and developing a bioeconomy and a society that is both carbon-neutral and sustainable requires scalable models that suit different places and environments. Introducing and developing such sustainable models demands dialogue and partnerships between different core actors and coordination between different territorial levels, as well as research and inclusive development as a means for translating new knowledge into practical applications. One of the key arguments presented in this chapter is that local job growth in rural communities in peripheral regions is vital for improving the economic prospects of these types of communities and keeping them inhabited. Page \| 17 Figure 3. Selected key messages relating to the three thematic areas. Selected key messages I) Diversification of the rural economy: Entrepreneurship, employment and new business models  Transformation of labour and digitalisation enable place-independent work, studying and entrepreneurship in new ways.  Digitalisation and place-independence are linked together with entrepreneurship, employment and transformation of labour as well as new ways of arranging, producing and using services in rural areas. III) Bioeconomy and sustainable  Success in bioeconomy requires the need for local knowledge about industrial traditions, resources, actors, and cooperation partners, but also external technology and capital for implementing strategies.  New local bioeconomy-related strategies support the development of sparsely populated rural regions. Page \| 18 References References Chartier, O., Salle, E., Irvine, K., Kull, M., Miller, D., Nieto, E., Vestergård, L.O., Potters, J. and Slätmo, E., Zomer, B., Iadecola, F. (2021). Long-Term Vision for Rural Areas: Contribution from SHERPA science-society- policy platforms. SHERPA Position Paper. DOI: 10.5281/zenodo.4557440. European Commission (2021). A long-term Vision for the EU's Rural Areas - Towards stronger, connected, resilient and prosperous rural areas by 2040. Available at https://ec.europa.eu/info/strategy/priorities-2019- 2024/new-push-european-democracy/long-term-vision-rural-areas_en#documents. Kattilakoski, M., Husberg, A., Kuhmonen, H., Rutanen, J., Vihinen, H., Töyli, P., Lukkari, T., Osmonen, E., Väre, T. and Åström, C. (2021) Ajassa uudistuva maaseutu – maaseutupoliittinen kokonaisohjelma 2021- 2027 (Countryside renewing with the times – Rural Policy Programme 2021-2027). Available at https://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/163350/MMM_2021_12.pdf?sequence=1 Kattilakoski, M., Husberg, A., Kuhmonen, H., Rutanen, J., Vihinen, H., Töyli, P., Lukkari, T., Osmonen, E., Väre, T. and Åström, C. (2021) Ajassa uudistuva maaseutu – maaseutupoliittinen kokonaisohjelma 2021- 2027 (Countryside renewing with the times – Rural Policy Programme 2021-2027). Available at https://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/163350/MMM_2021_12.pdf?sequence=1 Kettunen, M., Huikko, K., Pihlaja, R., Talvitie, H., Välikangas, A., Lukkari, T., Karppinen, V., Siirilä, H., Pellikka, M., Jutila, T. and Juntunen, A. (2015). Palveluvelvoite maaseudulla – SGEI palvelujen järjestämisen työkaluna (Service obligation in rural areas - SGEI services as a tool for organising services). Available at Lehtonen, O. and Okkonen, L. (2019). Energy cost reduction creates additional socioeconomic benefits - The case of Eno Energy Cooperative, Finland. Energy Policy 29: 352–359. Lehtonen, O. and Okkonen, L. (2016). Socio-economic impacts of a local bioenergy-based development strategy – The case of Pielinen Karelia, Finland. Renewable Energy 85: 610–619. Lehtonen, O. (2019) Population grid-based assessment of the impact of broadband expansion on population development in rural areas. Telecommunications Policy, 2020, vol. 44, issue 10 Maaseutubarometri (2020) Rural Barometer study. https://www.maaseutupolitiikka.fi/paatoksenteontueksi/tietoa/maaseutubaro (2020) Ministry of Economic Affairs and Employment of Finland (2021). SGEI-sääntelyn käyttö Suomessa - Työ- ja elinkeinoministeriön verkkopalvelu. Available online at https://tem.fi/sgei-saantelyn-kaytto-suomessa. Okkonen, L. and Lehtonen, O. (2016). Socio-economic impacts of community wind power projects in Northern Scotland. Renewable Energy 85: 826–833. Rural Network of Finland (2021). Smart villages in Finland. Available online at https://www.maaseutu.fi/en/the-rural-network/smart-villages. Simcock, N., Willis, R. and Capener, P. (2016). Cultures of Community Energy: International Case Studies. British Academy. Smart Rural 21 (2021). Smart Rural Areas of the 21 Century. Available online at https://www.smartrural21.eu/. Page \| 19 Page \| 19 Appendix Table 1. Compilation of noteworthy projects / initiatives / tools / methods implemented in Finland Name Contact & Internet address Aitojamakua site to promote entrepreneurship in the food sector https://aitojamakuja.fi/ruokasektorin-koordinaatiohanke/ Eno Energy Cooperative http://www.enonenergia.fi/Business_concept_of_Eno_Energy_C ooperative Etatyotilat.fi website showcasing workspaces for workers, entrepreneurs and students all around Finland. https://www.etatyotilat.fi/ Päijät-Häme Forestry Association harvesting products and cooperation with other natural product projects. www.mhy.fi/paijat-hame Regional resource flows and regions / City of Jyväskylä https://www.jyvaskyla.fi/ymparisto/resurssiviisaus/huominen- aina-tulevaisuutta/resurssiviisas-jyvaskyla-2040-ohjelma Remote working hubs as platforms for increasing vitality (Etätyöpisteet elinvoiman kasvualustoina). https://www.witas.fi/hankkeet/yhteistyotahojen_hankkeet/etaty opisteet_elinvoiman_kasvualustoina_1119. YTYÄ: project focusing on societal entrepreneurship in rural areas (Yhteiskunnallinen yrittäjyys maaseudulla) Ruralia Institute, University of Helsinki: https://www2.helsinki.fi/fi/ruralia-instituutti/yhteiskunnallinen- yrittajyys-maaseudulla-ytya Äly: project focusing on smart specialisation in Finland. University of Eastern Finland. https://uefconnect.uef.fi/tutkimusryhma/mita-on-alykas- sopeutuminen-suomessa/ Åland's sustainable food strategy https://landsbygd.ax/livsmedelsstrategin/ Table 2. Discussions from Rural Parliament workshops organised by MAP Finland Discussions from Rural Parliament workshops posted on the interactive Viima board Workshop on Diversification, 28.9.2021 Compilation of noteworthy projects / initiatives / tools / methods implemented in Finland Discussions from Rural Parliament workshops posted on the interactive Viima board Workshop on Diversification, 28.9.2021 Page \| 20 “Metsänhoitoyhdistys Päijät-Häme has done good work on the collection products and they have cooperated with other natural product projects. Joint workshops with processors, etc.” “Metsänhoitoyhdistys Päijät-Häme has done good work on the collection products and they have cooperate with other natural product projects. Joint workshops with processors, etc.” "We will put business-to-business cooperation and networks on our paper: production volumes and peer learning as objectives". "Good thinking is thinking in terms of regional resource flows and regions. For example, the City of Jyväskylä has been involved in developing this. https://www.jyvaskyla.fi/ymparisto/resurssiviisaus/huominen-aina- tulevaisuutta/resurssiviisas-jyvaskyla-2040-ohjelma" “Diversification requires innovation, which calls for networks and cooperation. Changing the typical subordinate role of rural areas to cities in innovation development: an opportunity - which innovations could be created in rural areas but not in cities? How could innovation be created in a rural-driven way?” “COVID19 permanently changed the working life, so in addition to digitalisation and working from home, work/meeting spaces from the municipality/community are needed for meetings, networking, etc. In addition, a low-threshold forum for local entrepreneurs, which does not require membership anywhere, but where you can get peer support, information about services you need, facilities, etc. Page \| 19 An idea workshop where local entrepreneurs can innovate local products and services, which could combine the skills of different entrepreneurs” “In Finland, a good example to promote entrepreneurship in the food sector; a national coordination project and provincial developers. Sharing information and taking things forward together, highlighting businesses on the Aitojamakuja.fi website.” https://aitojamakuja.fi/ruokasektorin-koordinaatiohanke/ “Regional development and the development of rural industries may no longer go hand in hand with the development of multi-location, etc... often lumped together.” “Focus on well-being and quality of life is important” Workshop on Smart rurality, 29.9.2021 “Best practice on method: the LEADER method, the method of locally led development” “Opportunities: bottom up approach through locally led development. Public, private and third sector influence and act together, engage grassroots to be involved in the community and implement actions through EU and nationally funded projects Challenges: "trust" from authority, limited resources” “Local experiences, dialogue between different actors and sectors of paramount important - better construct together than be sorry.” “Again, I'll add that local entrepreneurs' forum. One man's rubbish can be another man's treasure, i.e. side streams and even waste can be an excellent raw material for someone else. And to keep transport distances short, the benefits come locally, low-threshold cooperation between entrepreneurs and getting to know each other is the key” “The bioeconomy hype forgets biodiversity => Response Exactly. The countryside already has a bad reputation in the climate debate because of peat extraction, clear-cutting and the internationally very fast growing animal production. The use of various side streams from farming and companies and biogas are good things, but if the bioeconomy means less decaying wood in the forests, for example, then we are going in the wrong direction.” “Decentralised biogas production is an opportunity in rural Finland, because we can use a lot of biomass (e.g. grassland) in production. For example, selling biogas for transport can be an opportunity to diversify rural entrepreneurship.” “Good thinking is a way of thinking about regional resource flows and regions. For example, the City of Jyväskylä has been involved in developing this.” https://www.jyvaskyla.fi/ymparisto/resurssiviisaus/huominen- aina-tulevaisuutta/resurssiviisas-jyvaskyla-2040-ohjelma “Good thinking is a way of thinking about regional resource flows and regions. For example, the City of Jyväskylä has been involved in developing this.” https://www.jyvaskyla.fi/ymparisto/resurssiviisaus/huominen- aina-tulevaisuutta/resurssiviisas-jyvaskyla-2040-ohjelma Page \| 21 Page \| 21 "All LEADER LAGs in Finland address community development or rural economies in their strategies, so I fully support that LAGs can do a lot in this area!" Response => ""Yes, but not in all Europe..." "All LEADER LAGs in Finland address community development or rural economies in their strategies, so I fu support that LAGs can do a lot in this area!" Response => ""Yes, but not in all Europe..." "The role of the LAG´s is crucial when reaching grass-root and building on their strengths.” Response => "that is true there where LAG´s exist and which have Community development/local economies in their strategies" "Local stakeholders know best their area and are working on daily concrete actions. Fighting with adaptatio and future change management" Workshop on Bioeconomy, 28.9.2021 Page \| 22 "The best PR campaign for the mining industry, would be a good mining law." "The best PR campaign for the mining industry, would be a good mining law." "The best PR campaign for the mining industry, would be a good mining law." “Best practice on method: the LEADER method, the method of locally led development” Page \| 23 Page \| 23 Svenskspråkig (Swedish language) workshop, 29.9.2021 “Opportunities and challenges, the use of local social resources. Need to reach out beyond NGOs, to citizens at large. Associations are perhaps changing, people want to contribute to development here and there, perhaps not working 20 years in the same association. Challenge, how to involve citizens on a broad level, there are a lot of unused resources.” “Åland's sustainable food strategy” https://landsbygd.ax/livsmedelsstrategin/ “Best practice on method: LEADER method, the method for locally led development” “Opportunities – bottom-up approach through locally led development. Public, private and third sector influence and act together, engage grassroots to be involved in the community and implement actions through EU and nationally funded projects Challenges - "trust" from authority, limited resources” Svenskspråkig (Swedish language) workshop, 29.9.2021 Svenskspråkig (Swedish language) workshop, 29.9.2021 “Opportunities and challenges, the use of local social resources. Need to reach out beyond NGOs, to citizens at large. Associations are perhaps changing, people want to contribute to development here and there, perhaps not working 20 years in the same association. Challenge, how to involve citizens on a broad level, there are a lot of unused resources.” “Åland's sustainable food strategy” https://landsbygd.ax/livsmedelsstrategin/ “Åland's sustainable food strategy” https://landsbygd.ax/livsmedelsstrategin/ “Opportunities – bottom-up approach through locally led development. Public, private and third sector influence and act together, engage grassroots to be involved in the community and implement actions through EU and nationally funded projects Challenges - "trust" from authority, limited resources” “Opportunities – bottom-up approach through locally led development. Public, private and third sector influence and act together, engage grassroots to be involved in the community and implement actions through EU and nationally funded projects Challenges - "trust" from authority, limited resources” Page \| 24 Page \| 24 SHERPA has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No. 862448. The content of the document does not reflect the official opinion of the European Union. Responsibility for the information and views expressed therein lies entirely with the author(s). www.rural-interfaces.eu SHERPA has received funding from the European Union’s Horizon 2020 Research and Innovation Programme under Grant Agreement No. 862448. The content of the document does not reflect the official opinion of the European Union. Responsibility for the information and views expressed therein lies entirely with the author(s).
https://openalex.org/W2218580750	https://orbi.uliege.be/bitstream/2268/243353/1/Alexander_et_al-2018-British_Journal_of_Pharmacology.pdf	English	null	SLC6 neurotransmitter transporter family	null	2,013	cc-by	11,563	THE CONCISE GUIDE TO PHARMACOLOGY 2017/18: Overview D Harding4, ki6, nig11, SY, UK dom earch Triangle Park, NC 27709, USA alia Stephen PH Alexander1, Eamonn Kelly2, Neil V Marrion2, John A Peters3, Elena Faccenda4, Simon D Harding4, Adam J Pawson4, Joanna L Sharman4, Christopher Southan4, O Peter Buneman5, John A Cidlowski6, Arthur Christopoulos7, Anthony P Davenport8, Doriano Fabbro9, Michael Spedding10, Jörg Striessnig11, Jamie A Davies4 and CGTP Collaborators Stephen PH Alexander1, Eamonn Kelly2, Neil V Marrion2, John A Peters3, Elena Faccenda4, Simon D Harding4 Adam J Pawson4, Joanna L Sharman4, Christopher Southan4, O Peter Buneman5, John A Cidlowski6, Arthur Christopoulos7, Anthony P Davenport8, Doriano Fabbro9, Michael Spedding10, Jörg Striessnig11, Jamie A Davies4 and CGTP Collaborators 1School of Life Sciences, University of Nottingham Medical School, Nottingham, NG7 2UH, UK f f , y f g , g , , 2School of Physiology, Pharmacology and Neuroscience, University of Bristol, Bristol, BS8 1TD, UK 3N i Di i i M di l Ed i I i Ni ll H i l d M di l S h l U i 3Neuroscience Division, Medical Education Institute, Ninewells Hospital and Medical School, University of Du 4Centre for Integrative Physiology, University of Edinburgh, Edinburgh, EH8 9XD, UK 5Laboratory for Foundations of Computer Science, School of Informatics, University of Edinburgh, Edinburgh, EH8 9LE, United Kingdom 6National Institute of Environmental Health Sciences, National Institutes of Health, Department of Health and Human Services, Research Triangle Park, NC 27709, USA 7Monash Institute of Pharmaceutical Sciences and Department of Pharmacology, Monash University, Parkville, Victoria 3052, Australia 8Clinical Pharmacology Unit, University of Cambridge, Cambridge, CB2 0QQ, UK Q p , , 10Spedding Research Solutions SARL, Le Vésinet 78110, France Q p , , 10Spedding Research Solutions SARL, Le Vésinet 78110, France p g 11Pharmacology and Toxicology, Institute of Pharmacy, University of Innsbruck, A-6020 Innsbruck, Austria Abstract The Concise Guide to PHARMACOLOGY 2017/18 is the third in this series of biennial publications. This version provides concise overviews of the key properties of nearly 1800 human drug targets with an emphasis on selective pharmacology (where available), plus links to an open access knowledgebase of drug targets and their ligands (www.guidetopharmacology.org), which provides more detailed views of target and ligand properties. Although the Concise Guide represents approximately 400 pages, the material presented is substantially reduced compared to information and links presented on the website. It provides a permanent, citable, point-in-time record that will survive database updates. The full contents of this section can be found at http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full. In addition to this overview, in which are identiﬁed ‘Other protein targets’ which fall outside of the subsequent categorisation, there are eight areas of focus: G protein-coupled receptors, ligand-gated ion channels, voltage-gated ion channels, other ion channels, nuclear hormone receptors, catalytic receptors, enzymes and transporters. These are presented with nomenclature guidance and summary information on the best available pharmacological tools, alongside key references and suggestions for further reading. The landscape format of the Concise Guide is designed to facilitate comparison of related targets from material contemporary to mid-2017, and supersedes data presented in the 2015/16 and 2013/14 Concise Guides and previous Guides to Receptors and Channels. It is produced in close conjunction with the Nomenclature Committee of the Union of Basic and Clinical Pharmacology (NC-IUPHAR), therefore, providing ofﬁcial IUPHAR classiﬁcation and nomenclature for human drug targets, where appropriate. Table of contents Table of contents S1 Overview S6 Other Protein Targets S6 Adiponectin receptors S7 Blood coagulation components S8 Non-enzymatic BRD containing proteins S8 Carrier proteins S9 CD molecules S10 Methyllysine reader proteins S11 Fatty acid-binding proteins S13 Notch receptors S13 Regulators of G protein Signaling (RGS) proteins S14 Sigma receptors S15 Tubulins S17 G protein-coupled receptors S19 Orphan and other 7TM receptors S19 Class A Orphans S28 Class C Orphans S28 Taste 1 receptors S29 Taste 2 receptors S30 Other 7TM proteins S31 5-Hydroxytryptamine receptors S34 Acetylcholine receptors (musc S36 Adenosine receptors S37 Adhesion Class GPCRs S39 Adrenoceptors S43 Angiotensin receptors S44 Apelin receptor S45 Bile acid receptor S46 Bombesin receptors S47 Bradykinin receptors S48 Calcitonin receptors S50 Calcium-sensing receptor Table of contents S1 Overview S6 Other Protein Targets S6 Adiponectin receptors S7 Blood coagulation components S8 Non-enzymatic BRD containing proteins S8 Carrier proteins S9 CD molecules S10 Methyllysine reader proteins S11 Fatty acid-binding proteins S13 Notch receptors S13 Regulators of G protein Signaling (RGS) proteins S14 Sigma receptors S15 Tubulins S17 G protein-coupled receptors S19 Orphan and other 7TM receptors S19 Class A Orphans S28 Class C Orphans S28 Taste 1 receptors S29 Taste 2 receptors S30 Other 7TM proteins S31 5-Hydroxytryptamine receptors S34 Acetylcholine receptors (muscarinic) S36 Adenosine receptors S37 Adhesion Class GPCRs S39 Adrenoceptors S43 Angiotensin receptors S44 Apelin receptor S45 Bile acid receptor S46 Bombesin receptors S47 Bradykinin receptors S48 Calcitonin receptors S50 Calcium-sensing receptor Searchable database: http://www.guidetopharmacology.org/index.jsp Overview S Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S14 Sigma receptors S15 Tubulins S17 G protein-coupled receptors S19 Orphan and other 7TM receptors S19 Class A Orphans S28 Class C Orphans S28 Taste 1 receptors S29 Taste 2 receptors S30 Other 7TM proteins S31 5-Hydroxytryptamine receptors S34 Acetylcholine receptors (muscarinic) S36 Adenosine receptors S37 Adhesion Class GPCRs S39 Adrenoceptors S43 Angiotensin receptors S44 Apelin receptor S45 Bile acid receptor S46 Bombesin receptors S47 Bradykinin receptors S48 Calcitonin receptors S50 Calcium-sensing receptor e Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 e Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmaco S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Hepatocyte nuclear factor-4 receptors S215 2B. Retinoid X receptors S216 2C. Testicular receptors S216 2E. Tailless-like receptors S217 2F. COUP-TF-like receptors S218 3B. Estrogen-related receptors S218 4A. Nerve growth factor IB-like receptors S219 5A. Fushi tarazu F1-like receptors S220 6A. Germ cell nuclear factor receptors S220 0B. DAX-like receptors S221 Steroid hormone receptors S221 3A. Estrogen receptors S222 3C. 3-Ketosteroid receptors S225 Catalytic receptors S226 Cytokine receptor family S227 IL-2 receptor family S229 IL-3 receptor family S230 IL-6 receptor family S231 IL-12 receptor family S232 Prolactin receptor family S233 Interferon receptor family S234 IL-10 receptor family S235 Immunoglobulin-like family of IL-1 receptors S236 IL-17 receptor family S237 GDNF receptor family S237 Integrins S241 Natriuretic peptide receptor family S242 Pattern recognition receptors S243 Toll-like receptor family S244 NOD-like receptor family S246 Receptor tyrosine kinases (RTKs) S247 Type I RTKs: ErbB (epidermal growth factor) receptor family S248 Type II RTKs: Insulin receptor family S249 Type III RTKs: PDGFR, CSFR, Kit, FLT3 receptor family S250 Type IV RTKs: VEGF (vascular endothelial growth factor) receptor family S251 Type V RTKs: FGF (ﬁbroblast growth factor) receptor family S252 Type VI RTKs: PTK7/CCK4 S252 Type VII RTKs: Neurotrophin receptor/Trk family S253 Type VIII RTKs: ROR family S208 Nuclear hormone receptors S232 Prolactin receptor family S233 Interferon receptor family S234 IL-10 receptor family S235 Immunoglobulin-like family of IL-1 recept S236 IL-17 receptor family S237 GDNF receptor family S241 Natriuretic peptide receptor family S242 Pattern recognition receptors S243 Toll-like receptor family S244 NOD-like receptor family p y S246 Receptor tyrosine kinases (RTKs) yp p y S249 Type III RTKs: PDGFR, CSFR, Kit, FLT3 receptor fa S249 Type III RTKs: PDGFR, CSFR, Kit, FLT3 receptor family S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 S111 Tachykinin receptors S113 Thyrotropin-releasing hormone receptors S113 Trace amine receptor S114 Urotensin receptor S115 Vasopressin and oxytocin receptors S117 VIP and PACAP receptors S130 Ligand-gated ion channels S131 5-HT3 receptors S133 Acid-sensing (proton-gated) ion channels (ASICs) S135 Epithelial sodium channels (ENaC) S137 GABAA receptors S142 Glycine receptors S145 Ionotropic glutamate receptors S150 IP3 receptor S151 Nicotinic acetylcholine receptors S154 P2X receptors S156 ZAC S160 Voltage-gated ion channels S161 CatSper and Two-Pore channels S163 Cyclic nucleotide-regulated channels S164 Potassium channels S165 Calcium- and sodium-activated potassium channels S166 Inwardly rectifying potassium channels S169 Two P domain potassium channels S171 Voltage-gated potassium channels S175 Ryanodine receptor S176 Transient Receptor Potential channels S186 Voltage-gated calcium channels S189 Voltage-gated proton channel S190 Voltage-gated sodium channels S195 Other ion channels S196 Aquaporins S197 Chloride channels S197 ClC family S199 CFTR S200 Calcium activated chloride channel S201 Maxi chloride channel S202 Volume regulated chloride channels S204 Connexins and Pannexins S206 Sodium leak channel, non-selective S51 Cannabinoid receptors S52 Chemerin receptor S53 Chemokine receptors S57 Cholecystokinin receptors S58 Class Frizzled GPCRs S59 Complement peptide receptors S60 Corticotropin-releasing factor receptors S61 Dopamine receptors S63 Endothelin receptors S64 G protein-coupled estrogen receptor S65 Formylpeptide receptors S66 Free fatty acid receptors S67 GABAB receptors S69 Galanin receptors S70 Ghrelin receptor S71 Glucagon receptor family S72 Glycoprotein hormone receptors S73 Gonadotrophin-releasing hormone receptors S75 GPR18, GPR55 and GPR119 S76 Histamine receptors S77 Hydroxycarboxylic acid receptors S78 Kisspeptin receptor S79 Leukotriene receptors S81 Lysophospholipid (LPA) receptors S82 Lysophospholipid (S1P) receptors S83 Melanin-concentrating hormone receptors S84 Melanocortin receptors S85 Melatonin receptors S86 Metabotropic glutamate receptors S88 Motilin receptor S89 Neuromedin U receptors S90 Neuropeptide FF/neuropeptide AF receptors S91 Neuropeptide S receptor S92 Neuropeptide W/neuropeptide B receptors S93 Neuropeptide Y receptors S94 Neurotensin receptors S95 Opioid receptors S97 Orexin receptors S98 Oxoglutarate receptor S98 P2Y receptors S101 Parathyroid hormone receptors S101 Platelet-activating factor receptor S102 Prokineticin receptors S103 Prolactin-releasing peptide receptor S104 Prostanoid receptors S106 Proteinase-activated receptors S107 QRFP receptor S108 Relaxin family peptide receptors S110 Somatostatin receptors S111 Succinate receptor S208 Nuclear hormone receptors S209 1A. Thyroid hormone receptors S210 1B. Retinoic acid receptors S210 1C. Peroxisome proliferator-activated receptors S211 1D. Rev-Erb receptors S212 1F. Retinoic acid-related orphans S213 1H. Liver X receptor-like receptors S214 1I. Vitamin D receptor-like receptors S214 2A. S1 Overview p S13 Regulators of G protein Signaling (RGS) proteins Overview S1 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. p eptor family yp y g p S256 Type XIII RTKs: Ephrin receptor family yp p S257 Type XIV RTKs: RET yp S257 Type XV RTKs: RYK yp S258 Type XVI RTKs: DDR (collagen receptor) family yp ( g S258 Type XVII RTKs: ROS receptors yp p S259 Type XVIII RTKs: LMR family S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 British Journal of Pharm S254 Type IX RTKs: MuSK S254 Type X RTKs: HGF (hepatocyte growth factor) receptor family S255 Type XI RTKs: TAM (TYRO3-, AXL- and MER-TK) receptor family S255 Type XII RTKs: TIE family of angiopoietin receptors S256 Type XIII RTKs: Ephrin receptor family S257 Type XIV RTKs: RET S257 Type XV RTKs: RYK S258 Type XVI RTKs: DDR (collagen receptor) family S258 Type XVII RTKs: ROS receptors S259 Type XVIII RTKs: LMR family S259 Type XIX RTKs: Leukocyte tyrosine kinase (LTK) receptor family S260 Type XX RTKs: STYK1 S260 Receptor serine/threonine kinase (RSTK) family S261 Type I receptor serine/threonine kinases S262 Type II receptor serine/threonine kinases S262 Type III receptor serine/threonine kinases S262 RSTK functional heteromers S264 Receptor tyrosine phosphatase (RTP) family S266 Tumour necrosis factor (TNF) receptor family S272 Enzymes S275 Kinases (EC 2.7.x.x) S276 Rho kinase S276 Protein kinase C (PKC) S277 Alpha subfamily S277 Delta subfamily S278 Eta subfamily S278 FRAP subfamily S279 Cyclin-dependent kinase (CDK) family S279 CDK4 subfamily S279 GSK subfamily S280 Polo-like kinase (PLK) family S280 STE7 family S281 Abl family S281 Ack family S281 Janus kinase (JakA) family S282 Src family S283 Tec family S283 RAF family S284 Peptidases and proteinases S284 A1: Pepsin S284 A22: Presenilin S285 C14: Caspase S285 M1: Aminopeptidase N S285 M2: Angiotensin-converting (ACE and ACE2) S286 M10: Matrix metallopeptidase S286 M12: Astacin/Adamalysin S287 M28: Aminopeptidase Y S287 M19: Membrane dipeptidase S288 S1: Chymotrypsin S288 T1: Proteasome S289 S8: Subtilisin S289 S9: Prolyl oligopeptidase S290 Acetylcholine turnover S291 Adenosine turnover S292 Amino acid hydroxylases S293 L-Arginine turnover S294 2.1.1.- Protein arginine N-methyltransferases S294 Arginase S294 Arginine:glycine amidinotransferase S295 Dimethylarginine dimethylaminohydrolases S295 Nitric oxide synthases S296 Carboxylases and decarboxylases S297 Carboxylases S298 Decarboxylases S300 Catecholamine turnover S302 Ceramide turnover S303 Serine palmitoyltransferase S303 Ceramide synthase S304 Sphingolipid 4-desaturase S304 Sphingomyelin synthase S305 Sphingomyelin phosphodiesterase S305 Neutral sphingomyelinase coupling factors S306 Ceramide glucosyltransferase S306 Acid ceramidase S307 Neutral ceramidases S307 Alkaline ceramidases S308 Ceramide kinase S309 Chromatin modifying enzymes S309 2.1.1.- Protein arginine N-methyltransferases S310 3.5.1.- Histone deacetylases (HDACs) S310 Cyclic nucleotide turnover/signalling S310 Adenylyl cyclases (ACs) S312 Exchange protein activated by cyclic AMP (EPACs) S312 Nitric oxide (NO)-sensitive (soluble) guanylyl cyclase S313 Phosphodiesterases, 3’,5’-cyclic nucleotide (PDEs) S317 Cytochrome P450 S317 CYP1 family S318 CYP2 family S318 CYP3 family S319 CYP4 family S320 CYP5, CYP7 and CYP8 families S320 CYP11, CYP17, CYP19, CYP20 and CYP21 families S321 CYP24, CYP26 and CYP27 families S322 CYP39, CYP46 and CYP51 families S323 Endocannabinoid turnover S323 N-Acylethanolamine turnover Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S324 2-Acylglycerol ester turnover S325 Eicosanoid turnover S325 Cyclooxygenase S326 Prostaglandin synthases S327 Lipoxygenases S328 Leukotriene and lipoxin metabolism S329 GABA turnover S331 Glycerophospholipid turnover S331 Phosphoinositide-speciﬁc phospholipase C S332 Phospholipase A2 S334 Phosphatidylcholine-speciﬁc phospholipase D S335 Lipid phosphate phosphatases S335 Phosphatidylinositol kinases S336 1-phosphatidylinositol 4-kinase family S336 Phosphatidylinositol-4-phosphate 3-kinase family S337 Phosphatidylinositol 3-kinase family S337 Phosphatidylinositol-4,5-bisphosphate 3-kinase family S338 1-phosphatidylinositol-3-phosphate 5-kinase family S338 Type I PIP kinases (1-phosphatidylinositol-4-phosphate 5-kinase family) S339 Type II PIP kinases (1-phosphatidylinositol-5-phosphate 4-kinase family) S339 Haem oxygenase S340 Hydrogen sulphide synthesis S341 Hydrolases S342 Inositol phosphate turnover S342 Inositol 1,4,5-trisphosphate 3-kinases S343 Inositol polyphosphate phosphatases S343 Inositol monophosphatase S344 Lanosterol biosynthesis pathway S346 Nucleoside synthesis and metabolism S347 Sphingosine 1-phosphate turnover S348 Sphingosine kinase S348 Sphingosine 1-phosphate phosphatase S349 Sphingosine 1-phosphate lyase S349 Thyroid hormone turnover S350 1.14.11.29 2-oxoglutarate oxygenases S351 1.14.13.9 kynurenine 3-monooxygenase S352 2.4.2.30 poly(ADP-ribose)polymerases S352 2.5.1.58 Protein farnesyltransferase S353 3.5.1.- Histone deacetylases (HDACs) S354 3.5.3.15 Peptidyl arginine deiminases (PADI) S354 RAS subfamily S355 4.2.1.1 Carbonate dehydratases S355 5.99.1.2 DNA Topoisomerases S360 Transporters S362 ATP-binding cassette transporter family S362 ABCA subfamily S363 ABCB subfamily yp S254 Type X RTKs: HGF (hepatocyte growth factor) yp receptor family p y S251 Type V RTKs: FGF (ﬁbroblast growth factor) p y S251 Type V RTKs: FGF (ﬁbroblast growth factor) receptor family yp ( S252 Type VI RTKs: PTK7/CCK4 yp S252 Type VII RTKs: Neurotrophin receptor/Trk family S252 Type VII RTKs: Neurotrophin receptor/Trk family yp p S253 Type VIII RTKs: ROR family Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/ful Overview S2 Overview S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 y S260 Type XX RTKs: STYK1 p g p p p p S349 Sphingosine 1-phosphate lyase y S355 5.99.1.2 DNA Topoisomerases Overview S3 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/ful S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Introduction antagonists, inhibitors, activators, etc.) are included where they are both available (by donation or from commercial sources, now or in the near future) AND the most selective. The Concise Guide is divided into nine sections, which comprise pharmacological tar- gets of similar structure/function. These are G protein-coupled receptors, ligand-gated ion channels, voltage-gated ion channels, other ion channels, catalytic receptors, nuclear hormone recep- tors, enzymes, transporters and other protein targets. We hope that the Concise Guide will provide for researchers, teachers and students a state-of-the art source of accurate, curated information on the background to their work that they will use in the Introduc- tions to their Research Papers or Reviews, or in supporting their teaching and studies. We recommend that any citations to in- formation in the Concise Guide are presented in the following format: produce an authoritative consensus on nomenclature, which at- tempts to ﬁt in within the general guidelines from NC-IUPHAR. This current edition, the Concise Guide to PHARMACOLOGY 2017/18, is the latest snapshot of the database in print form, fol- lowing on from the Concise Guide to PHARMACOLOGY 2015/16. It contains data drawn from the online database as a rapid overview of the major pharmacological targets. Thus, there are many fewer targets presented in the Concise Guide compared to the online database. The priority for inclusion in the Concise Guide is the presence of quantitative pharmacological data. This means that often orphan family members are not presented in the Concise Guide, although structural information is available on the online database. The organisation of the data is tabular (where ap- propriate) with a standardised format, where possible on a single page, intended to aid understanding of, and comparison within, a particular target group. The Concise Guide is intended as an initial resource, with links to additional reviews and resources for greater depth and information. Pharmacological and structural data focus primarily on human gene products, wherever possible, with links to HGNC gene nomenclature and UniProt IDs. In a few cases, where data from human proteins are limited, data from other species are indicated. Pharmacological tools listed are prioritised on the basis of selectivity and availability. That is, agents (agonists, In order to allow clarity and consistency in pharmacology, there is a need for a comprehensive organisation and presentation of the targets of drugs. Introduction Thus, there are many fewer targets presented in the Concise Guide compared to the online database. The priority for inclusion in the Concise Guide is the presence of quantitative pharmacological data. This means that often orphan family members are not presented in the Concise Guide, although structural information is available on the online database. The organisation of the data is tabular (where ap- propriate) with a standardised format, where possible on a single page, intended to aid understanding of, and comparison within, a particular target group. The Concise Guide is intended as an initial resource, with links to additional reviews and resources for greater depth and information. Pharmacological and structural data focus primarily on human gene products, wherever possible, with links to HGNC gene nomenclature and UniProt IDs. In a few cases, where data from human proteins are limited, data from other species are indicated. Pharmacological tools listed are prioritised on the basis of selectivity and availability. That is, agents (agonists, Alexander SPH et al. (2017). The Concise Guide to PHARMACOL- OGY 2017/18: Overview. Br J Pharmacol 174: S1–S16. Alexander SPH et al. (2017). The Concise Guide to PHARMACOL- OGY 2017/18: Overview. Br J Pharmacol 174: S1–S16. In this overview are listed protein targets of pharmacological inter- est, which are not G protein-coupled receptors, ligand-gated ion channels, voltage-gated ion channels, ion channels, nuclear hor- mone receptors, catalytic receptors, transporters or enzymes. Acknowledgements g We are extremely grateful to the British Pharmacological Society and the International Union of Basic and Clinical Pharmacology, for ﬁnancial support of the website and for advice from the NC-IUPHAR subcommittees. We thank the University of Edinburgh, who host the www.guidetopharmacology.org website. Previously, the International Union of Basic and Clinical Pharmacology and the Wellcome Trust (099156/Z/12/Z]) also supported the initiation and expansion of the database. We are also tremendously grateful to the long list of collaborators from NC-IUPHAR subcommittees and beyond, who have assisted in the construction of the Concise Guide to PHARMACOLOGY 2017/18 and the online database www.GuideToPHARMACOLOGY.org. Further, we wish to thank Toni Wigglesworth for her assistance in the co-ordination of correspondence with these collaborators. Introduction This is the philosophy of the IUPHAR/BPS Guide to PHARMACOLOGY presented on the online free access database (http://www.guidetopharmacology.org/). This database is sup- ported by the British Pharmacological Society (BPS), the Interna- tional Union of Basic and Clinical Pharmacology (IUPHAR), the University of Edinburgh and previously the Wellcome Trust. Data included in the Guide to PHARMACOLOGY are derived in large part from interactions with the subcommittees of the Nomencla- ture Committee of the International Union of Basic and Clinical Pharmacology (NC-IUPHAR). A major inﬂuence on the develop- ment of the database was Tony Harmar (1951-2014), who worked with a passion to establish the curators as a team of highly in- formed and informative individuals, with a focus on high-quality data input, ensuring a suitably validated dataset. The Editors of the Concise Guide have compiled the individual records, in con- cert with the team of Curators, drawing on the expert knowl- edge of these latter subcommittees. The tables allow an indication of the status of the nomenclature for the group of targets listed, usually previously published in Pharmacological Reviews. In the absence of an established subcommittee, advice from several prominent, independent experts has generally been obtained to antagonists, inhibitors, activators, etc.) are included where they are both available (by donation or from commercial sources, now or in the near future) AND the most selective. The Concise Guide is divided into nine sections, which comprise pharmacological tar- gets of similar structure/function. These are G protein-coupled receptors, ligand-gated ion channels, voltage-gated ion channels, other ion channels, catalytic receptors, nuclear hormone recep- tors, enzymes, transporters and other protein targets. We hope that the Concise Guide will provide for researchers, teachers and students a state-of-the art source of accurate, curated information on the background to their work that they will use in the Introduc- tions to their Research Papers or Reviews, or in supporting their teaching and studies. We recommend that any citations to in- formation in the Concise Guide are presented in the following format: produce an authoritative consensus on nomenclature, which at- tempts to ﬁt in within the general guidelines from NC-IUPHAR. This current edition, the Concise Guide to PHARMACOLOGY 2017/18, is the latest snapshot of the database in print form, fol- lowing on from the Concise Guide to PHARMACOLOGY 2015/16. It contains data drawn from the online database as a rapid overview of the major pharmacological targets. S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 British Journal of Pharmacology (2017) 174, S1–S16 S365 ABCC subfamily S366 ABCD subfamily of peroxisomal ABC transporters S367 ABCG subfamily S368 F-type and V-type ATPases S368 F-type ATPase S368 V-type ATPase S369 P-type ATPases S369 Na+/K+-ATPases S369 Ca2+-ATPases S370 H+/K+-ATPases S370 Cu+-ATPases S370 Phospholipid-transporting ATPases S371 Major facilitator superfamily (MFS) of transporters S371 SLC superfamily of solute carriers S372 SLC1 family of amino acid transporters S372 Glutamate transporter subfamily S374 Alanine/serine/cysteine transporter subfamily S375 SLC2 family of hexose and sugar alcohol S375 Class I transporters S376 Class II transporters S377 Proton-coupled inositol transporter S377 SLC3 and SLC7 families of heteromeric amino acid transporters (HATs) S377 SLC3 family S378 SLC7 family S379 SLC4 family of bicarbonate transporters S380 Anion exchangers S380 Sodium-dependent HCO− 3 transporters S381 SLC5 family of sodium-dependent glucose transporters S381 Hexose transporter family S382 Choline transporter S383 Sodium iodide symporter, sodium-dependent multivitamin transporter and sodium-coupled monocarboxylate transporters S384 Sodium myo-inositol cotransporter transporters S385 SLC6 neurotransmitter transporter family S385 Monoamine transporter subfamily S386 GABA transporter subfamily S387 Glycine transporter subfamily S389 Neutral amino acid transporter subfamily S390 SLC8 family of sodium/calcium exchangers S390 SLC9 family of sodium/hydrogen exchangers S391 SLC10 family of sodium-bile acid co-transporters S392 SLC11 family of proton-coupled metal ion transporters S393 SLC12 family of cation-coupled chloride transporters S395 SLC13 family of sodium-dependent sulphate/carboxylate transporters S395 SLC14 family of facilitative urea transporters S396 SLC15 family of peptide transporters S398 SLC16 family of monocarboxylate transporters S399 SLC17 phosphate and organic anion transporter family S399 Type I sodium-phosphate co-transporters S400 Sialic acid transporter S400 Vesicular glutamate transporters (VGLUTs) S401 Vesicular nucleotide transporter S401 SLC18 family of vesicular amine transporters S403 SLC19 family of vitamin transporters S403 SLC20 family of sodium-dependent phosphate transporters S404 SLC22 family of organic cation and anion transporters S404 Organic cation transporters (OCT) S405 Organic zwitterions/cation transporters (OCTN) S406 Organic anion transporters (OATs) S407 Urate transporter S407 SLC23 family of ascorbic acid transporters S409 SLC24 family of sodium/potassium/calcium exchangers S409 SLC25 family of mitochondrial transporters S410 Mitochondrial di- and tri-carboxylic acid transporter subfamily S411 Mitochondrial amino acid transporter subfamily S412 Mitochondrial phosphate transporters S412 Mitochondrial nucleotide transporter subfamily S413 Mitochondrial uncoupling proteins S414 Miscellaneous SLC25 mitochondrial transporters S414 SLC26 family of anion exchangers S415 Selective sulphate transporters S415 Chloride/bicarbonate exchangers S416 Anion channels S416 Other SLC26 anion exchangers S417 SLC27 family of fatty acid transporters S418 SLC28 and SLC29 families of nucleoside transporters S418 SLC28 family S419 SLC29 family S420 SLC30 zinc transporter family S421 SLC31 family of copper transporters S422 SLC32 vesicular inhibitory amino acid transporter S422 SLC33 acetylCoA transporter S423 SLC34 family of sodium phosphate co-transporters S424 SLC35 family of nucleotide sugar transporters S425 SLC36 family of proton-coupled amino acid transporters S426 SLC37 family of phosphosugar/phosphate exchangers S427 SLC38 family of sodium-dependent neutral amino acid transporters S427 System A-like transporters S428 System N-like transporters S428 Orphan SLC38 transporters S429 SLC39 family of metal ion transporters S430 SLC40 iron transporter S430 SLC41 family of divalent cation transporters S431 SLC42 family of Rhesus glycoprotein ammonium transporters S432 SLC43 family of large neutral amino acid transporters S433 SLC44 choline transporter-like family S433 SLC45 family of putative sugar transporters S434 SLC46 family of folate transporters S435 SLC47 family of multidrug and toxin extrusion transporters S436 SLC48 heme transporter S436 SLC49 family of FLVCR-related heme transporters S437 SLC50 sugar transporter S438 SLC51 family of steroid-derived molecule transporters S438 SLC52 family of riboﬂavin transporters S439 SLCO family of organic anion transporting polypeptides S442 Patched family S389 Neutral amino acid transporter subfamily S390 SLC8 family of sodium/calcium exchangers S390 SLC9 family of sodium/hydrogen exchangers S391 SLC10 family of sodium-bile acid co-transporters S392 SLC11 family of proton-coupled metal ion transporters S393 SLC12 family of cation-coupled chloride transporters S395 SLC13 family of sodium-dependent sulphate/carboxyla transporters S395 SLC14 family of facilitative urea transporters S396 SLC15 family of peptide transporters S398 SLC16 family of monocarboxylate transporters S399 SLC17 phosphate and organic anion transporter family S399 Type I sodium-phosphate co-transporters S400 Sialic acid transporter S400 Vesicular glutamate transporters (VGLUTs) S401 Vesicular nucleotide transporter S401 SLC18 family of vesicular amine transporters S403 SLC19 family of vitamin transporters S403 SLC20 family of sodium-dependent phosphate transpor S404 SLC22 family of organic cation and anion transporters S404 Organic cation transporters (OCT) S405 Organic zwitterions/cation transporters (OCTN) S406 Organic anion transporters (OATs) S407 Urate transporter S407 SLC23 family of ascorbic acid transporters S409 SLC24 family of sodium/potassium/calcium exchanger S409 SLC25 family of mitochondrial transporters S410 Mitochondrial di- and tri-carboxylic acid transporter subfamily S411 Mitochondrial amino acid transporter subfamily S412 Mitochondrial phosphate transporters S412 Mitochondrial nucleotide transporter subfamily S413 Mitochondrial uncoupling proteins S414 Miscellaneous SLC25 mitochondrial transporters S414 SLC26 family of anion exchangers S415 Selective sulphate transporters S415 Chloride/bicarbonate exchangers S416 Anion channels S416 Other SLC26 anion exchangers y p p S385 SLC6 neurotransmitter transporter family p S385 Monoamine transporter subfamily S386 GABA transporter subfamily S416 Other SLC26 anion exchangers S387 Glycine transporter subfamily Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Overview S4 S.P.H. S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Conﬂict of interest The authors state that there are no conﬂicts of interest to disclose. c⃝2017 The Authors. British Journal of Pharmacology published by John Wiley & Sons Ltd on behalf of British Pharmacological Society. 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 origi Overview S5 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Overview S5 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharm Other Protein Targets Family structure – Heat shock proteins – Immunoglobulins – Inhibitors of apoptosis (IAP) protein family – Kelch-like proteins – Kinesins – Leucine-rich repeat proteins – Lymphocyte antigens – Mitochondrial-associated proteins – Myosin binding proteins – Non-catalytic pattern recognition receptors – Absent in melanoma (AIM)-like receptors (ALRs) – C-type lectin-like receptors (CLRs) – Other pattern recognition receptors S12 Notch receptors – Pentaxins – Serum pentaxins S13 Regulators of G protein Signaling (RGS) proteins S14 R4 family – Repulsive guidance molecules – Reticulons and associated proteins – Ribosomal factors S14 Sigma receptors S15 Tubulins – Tumour-associated proteins – WD repeat-containing proteins S6 Adiponectin receptors – B-cell lymphoma 2 (Bcl-2) protein family S7 Blood coagulation components – Bromodomain-containing proteins S7 Non-enzymatic BRD containing proteins S8 Carrier proteins S9 CD molecules – Chromatin-interacting transcriptional repressors S10 Methyllysine reader proteins – Circadian clock proteins – Claudins – EF-hand domain containing S11 Fatty acid-binding proteins – G-alpha family G(q) subfamily Adiponectin receptors p p Other protein targets →Adiponectin receptors appears to avoid G proteins; modelling based on the crystal struc- tures of the adiponectin receptors suggested ceramidase acivity, which would make these the ﬁrst in a new family of catalytic re- ceptors [93]. Overview: Adiponectin receptors (provisional nomen- clature, ENSFM00500000270960) respond to the 30 kDa complement-related protein hormone adiponectin (also known as ADIPOQ: adipocyte, C1q and collagen domain-containing protein; ACRP30, adipose most abundant gene transcript 1; apM-1; gelatin-binding protein: Q15848) originally cloned from adipocytes [49]. Although sequence data suggest 7TM domains, immunological evidence indicates that, contrary to typical 7TM topology, the carboxyl terminus is extracellular, while the amino terminus is intracellular [90]. Signalling through these receptors Nomenclature Adipo1 receptor Adipo2 receptor HGNC, UniProt ADIPOR1, Q96A54 ADIPOR2, Q86V24 Rank order of potency globular adiponectin (ADIPOQ, Q15848) > adiponectin (ADIPOQ, Q15848) globular adiponectin (ADIPOQ, Q15848) = adiponectin (ADIPOQ, Q15848) Comments: T-Cadherin (CDH13, P55290) has also been suggested to be a receptor for (hexameric) adiponectin [33]. Searchable database: http://www.guidetopharmacology.org/index.jsp Adiponectin receptors S6 Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Wang Y et al. (2017) Cardiovascular Adiponectin Resistance: The Critical Role of Adiponectin Re- ceptor Modiﬁcation. Trends Endocrinol Metab 28: 519-530 [PMID:28473178] Zhao L et al. (2014) Adiponectin and insulin cross talk: the microvascular connection. Trends Cardiovasc Med 24: 319-24 [PMID:25220977] g Other protein targets →Blood coagulation components Overview: Coagulation as a process is interpreted as a mechanism for reducing excessive blood loss through the generation of a gel-like clot local to the site of injury. The process involves the activation, adhesion (see Integrins), degranulation and aggregation of platelets, as well as proteins circulating in the plasma. The coagulation cascade involves multiple proteins being converted to more active forms from less active precursors, typically through proteolysis (see Proteases). Listed here are the components of the coagulation cascade targetted by agents in current clinical usage. Nomenclature coagulation factor V coagulation factor VIII serpin family C member 1 HGNC, UniProt F5, P12259 F8, P00451 SERPINC1, P01008 Selective activators – – heparin (pKd 7.8) [26], fondaparinux (pKd 7.5) [62], dalteparin [32], danaparoid [16, 56], enoxaparin [19], tinzaparin [20] Selective inhibitors drotrecogin alfa [36, 37] drotrecogin alfa [36, 37] – Further reading on Blood coagulation components Astermark J. (2015) FVIII inhibitors: pathogenesis and avoidance. Blood 125: 2045-51 [PMID:25712994] Girolami A et al. (2017) New clotting disorders that cast new light on blood coagulation and may play a role in clinical practice. J Thromb Thrombolysis 44: 71-75 [PMID:28251495] Rana K et al. (2016) Blood ﬂow and mass transfer regulation of coagulation. Blood Rev 30: 357-68 [PMID:27133256] serpin family C member 1 SERPINC1, P01008 Further reading on Adiponectin receptors Fisman EZ et al. (2014) Adiponectin: a manifold therapeutic target for metabolic syndrome, dia- betes, and coronary disease? Cardiovasc Diabetol 13: 103 [PMID:24957699] betes, and coronary disease? Cardiovasc Diabetol 13: 103 [PMID:24957699] Matsuda M et al. (2014) Roles of adiponectin and oxidative stress in obesity-associated metabolic and cardiovascular diseases. Rev Endocr Metab Disord 15: 1-10 [PMID:24026768] Ruan H et al. (2016) Adiponectin signaling and function in insulin target tissues. J Mol Cell Biol 8: 101-9 [PMID:26993044] betes, and coronary disease? Cardiovasc Diabetol 13: 103 [PMID:24957699] Matsuda M et al. (2014) Roles of adiponectin and oxidative stress in obesity-associated metabolic and cardiovascular diseases. Rev Endocr Metab Disord 15: 1-10 [PMID:24026768] R H t l (2016) Adi ti i li d f ti i i li t t ti J M l C ll Bi l 8 and cardiovascular diseases. Rev Endocr Metab Disord 15: 1-10 [PMID:24026768] Ruan H et al. (2016) Adiponectin signaling and function in insulin target tissues. J Mol Cell Biol 8: 101-9 [PMID:26993044] Nomenclature Adipo1 receptor Adipo2 receptor HGNC, UniProt ADIPOR1, Q96A54 ADIPOR2, Q86V24 Rank order of potency globular adiponectin (ADIPOQ, Q15848) > adiponectin (ADIPOQ, Q15848) globular adiponectin (ADIPOQ, Q15848) = adiponectin (ADIPOQ, Q15848) Comments: T-Cadherin (CDH13, P55290) has also been suggested to be a receptor for (hexameric) adiponectin [33]. Searchable database: http://www.guidetopharmacology.org/index.jsp Adiponectin receptors S6 Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Adipo2 receptor ADIPOR2, Q86V24 globular adiponectin (ADIPOQ, Q15848) = adiponectin (ADIPOQ, Q15848) Adipo1 receptor ADIPOR1, Q96A54 Adipo2 receptor ADIPOR2, Q86V24 mments: T-Cadherin (CDH13, P55290) has also been suggested to be a receptor for (hexameric) adiponectin [33] Adiponectin receptors S6 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Blood coagulation components S7 Further reading on Blood coagulation components Rana K et al. (2016) Blood ﬂow and mass transfer regulation of coagulation. Blood Rev 30: 357-68 [PMID:27133256] Astermark J. (2015) FVIII inhibitors: pathogenesis and avoidance. Blood 125: 2045-51 [PMID:25712994] [PMID:25712994] Girolami A et al. (2017) New clotting disorders that cast new light on blood coagulation and may play a role in clinical practice. J Thromb Thrombolysis 44: 71-75 [PMID:28251495] Blood coagulation components S7 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Blood coagulation components S7 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Brand M et al. (2015) Small molecule inhibitors of bromodomain-acetyl-lysine interactions. Chem. Biol. 10: 22-39 [PMID:25549280] Brand M et al. (2015) Small molecule inhibitors of bromodomain-acetyl-lysine interactions. ACS Chem. Biol. 10: 22-39 [PMID:25549280] Chem. Biol. 10: 22 39 [PMID:25549280] Fujisawa T et al. (2017) Functions of bromodomain-containing proteins and their roles in home- ostasis and cancer Nat Rev Mol Cell Biol 18: 246-262 [PMID:28053347] [ ] Nicholas DA et al. (2017) BET bromodomain proteins and epigenetic regulation of inﬂamma- tion: implications for type 2 diabetes and breast cancer. Cell Mol Life Sci 74: 231-243 [PMID:27491296] Non-enzymatic BRD containing proteins y g p Other protein targets →Bromodomain-containing proteins →Non-enzymatic BRD containing proteins Overview: Bromodomains bind proteins with acetylated lysine residues, such as histones, to regulate gene transcription. Listed herein are examples of bromodomain-containing proteins for which sufﬁcient pharmacology exists. Nomenclature bromodomain adjacent to zinc ﬁnger domain 2A bromodomain adjacent to zinc ﬁnger domain 2B CREB binding protein polybromo 1 SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily a, member 4 HGNC, UniProt BAZ2A, Q9UIF9 BAZ2B, Q9UIF8 CREBBP, Q92793 PBRM1, Q86U86 SMARCA4, P51532 Selective inhibitors GSK2801 (pKd 6.6) [73] GSK2801 (pKd 6.9) [73] I-CBP112 (pKd 6.8) [72] PFI-3 (pKd 7.3) [79] PFI-3 (pKd 7.1) [79] Further reading on Non-enzymatic BRD containing proteins Brand M et al. (2015) Small molecule inhibitors of bromodomain-acetyl-lysine interactions. ACS Chem. Biol. 10: 22-39 [PMID:25549280] Fujisawa T et al. (2017) Functions of bromodomain-containing proteins and their roles in home- ostasis and cancer Nat Rev Mol Cell Biol 18: 246-262 [PMID:28053347] Nicholas DA et al. (2017) BET bromodomain proteins and epigenetic regulation of inﬂamma- tion: implications for type 2 diabetes and breast cancer. Cell Mol Life Sci 74: 231-243 [PMID:27491296] Theodoulou NH et al. (2016) Clinical progress and pharmacology of small molecule bromodomain inhibitors. Curr Opin Chem Biol 33: 58-66 [PMID:27295577] Theodoulou NH et al. (2016) Progress in the Development of non-BET Bromodomain Chemical Probes. ChemMedChem 11: 477-87 [PMID:26749027] Overview: Bromodomains bind proteins with acetylated lysine residues, such as histones, to regulate gene sufﬁcient pharmacology exists. Further reading on Non-enzymatic BRD containing proteins Further reading on Non-enzymatic BRD containing proteins Theodoulou NH et al. (2016) Clinical progress and pharmacology of small molecule bromodomain inhibitors. Curr Opin Chem Biol 33: 58-66 [PMID:27295577] Theodoulou NH et al. (2016) Progress in the Development of non-BET Bromodomain Chemical Probes. ChemMedChem 11: 477-87 [PMID:26749027] Theodoulou NH et al. (2016) Clinical progress and pharmacology of small molecule bromodomain inhibitors. Curr Opin Chem Biol 33: 58-66 [PMID:27295577] Theodoulou NH et al. (2016) Progress in the Development of non-BET Bromodomain Chemical Probes. ChemMedChem 11: 477-87 [PMID:26749027] Theodoulou NH et al. (2016) Clinical progress and pharmacology of small molecule bromodomain inhibitors. Curr Opin Chem Biol 33: 58-66 [PMID:27295577] Theodoulou NH et al. (2016) Progress in the Development of non-BET Bromodomain Chemical Probes. ChemMedChem 11: 477-87 [PMID:26749027] Theodoulou NH et al. (2016) Clinical progress and pharmacology of small molecule bromodomain inhibitors. Curr Opin Chem Biol 33: 58-66 [PMID:27295577] Theodoulou NH et al. (2016) Progress in the Development of non-BET Bromodomain Chemical P b Ch M dCh 11 477 87 [PMID 26749027] Theodoulou NH et al. (2016) Clinical progress and pharmacology of small molecule bromodomain inhibitors. Curr Opin Chem Biol 33: 58-66 [PMID:27295577] Further reading on Carrier proteins Alshehri B et al. (2015) The diversity of mechanisms inﬂuenced by transthyretin in neuro- biology: development, disease and endocrine disruption. J Neuroendocrinol 27: 303-23 [PMID:25737004] Galant NJ et al. (2017) Transthyretin amyloidosis: an under-recognized neuropathy and cardiomy- opathy. Clin Sci (Lond) 131: 395-409 [PMID:28213611] [PMID:25737004] Delliere S et al. (2017) Is transthyretin a good marker of nutritional status? Clin Nutr 36: 364-370 [PMID:27381508] Galant NJ et al. (2017) Transthyretin amyloidosis: an under-recognized neuropathy and cardiomy- opathy. Clin Sci (Lond) 131: 395-409 [PMID:28213611] Carrier proteins prevent TTR dissociation is being pursued as a theapeutic strat- egy. To date one small molecule kinetic stabilising molecule (tafamidis) has been approved for FAP, and is being evaluated in clinical trials for other TTR amyloidoses. amyloidogenic mutants are linked to the development of patho- logical amyloidoses, including familial amyloid polyneuropathy (FAP) [4, 14], familial amyloid cardiomyopathy (FAC) [34], amy- loidotic vitreous opacities, carpal tunnel syndrome [54] and oth- ers. In old age, non-mutated TTR can also form pathological amyloid ﬁbrils [88]. Pharmacological intervention to reduce or Overview: Transthyretin (TTR) is a homo-tetrameric protein which transports thyroxine in the plasma and cerebrospinal ﬂuid and retinol (vitamin A) in the plasma. Many disease causing mutations in the protein have been reported, many of which cause complex dissociation and protein mis-assembly and depo- sition of toxic aggregates amyloid ﬁbril formation [63]. These Nomenclature transthyretin HGNC, UniProt TTR, P02766 Common abreviation TTR Carrier proteins S8 Carrier proteins S8 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S8 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharm S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Further reading on CD molecules Gabius HJ et al. (2015) The glycobiology of the CD system: a dictionary for translating marker desig- nations into glycan/lectin structure and function. Trends Biochem Sci 40: 360-76 [PMID:25981696] CD molecules Other protein targets →CD molecules These ligands surface peptides, normally involved in immune system regulation. Expression of PD-1 by cancer cells induces immune tolerance and evasion of immune system attack. Anti-PD-1 monoclonal dies are used to induce immune checkpoint blockade as a therapeutic intervention in cancer, effectively re-establishing immune vigilance. pembrolizumab was the ﬁrst anti-PD-1 antibody to be ed by the US FDA Methyllysine reader proteins y y p Other protein targets →Chromatin-interacting transcriptional repressors →Methyllysine reader proteins Overview: Methyllysine reader proteins bind to methylated proteins, such as histones, allowing regulation of gene expression. Nomenclature l(3)mbt-like 3 (Droso HGNC, UniProt L3MBTL3, Q96JM7 Selective agonists UNC1215 [35] l(3)mbt-like 3 (Drosophila) L3MBTL3, Q96JM7 UNC1215 [35] l(3)mbt-like 3 (Drosophila) L3MBTL3, Q96JM7 UNC1215 [35] UNC1215 [35] Teske KA et al. (2017) Methyllysine binding domains: Structural insight and small molecule probe development. Eur J Med Chem 136: 14-35 [PMID:28478342] Zahnow CA et al. (2016) Inhibitors of DNA Methylation, Histone Deacetylation, and Histone Demethylation: A Perfect Combination for Cancer Therapy. Adv Cancer Res 130: 55-111 [PMID:27037751] development. Eur J Med Chem 136: 14-35 [PMID:28478342] Zahnow CA et al. (2016) Inhibitors of DNA Methylation, Histone Deacetylation, and Histone Demethylation: A Perfect Combination for Cancer Therapy. Adv Cancer Res 130: 55-111 [PMID:27037751] CD molecules Other protein targets →CD molecules British Journal of Pharmacology (2017) 174, S1–S16 Nomenclature CD80 CD86 cytotoxic T-lymphocyte-associated protein 4 (CD152) programmed cell death 1 (CD279) CD300a HGNC, UniProt CD80, P33681 CD86, P42081 CTLA4, P16410 PDCD1, Q15116 CD300A, Q9UGN4 Common abreviation – – CTLA-4 PD-1 – Antibodies – – ipilimumab (pKd >9) [28], tremelimumab (pKd 8.9) [30] pembrolizumab (pKd ∼10) [11], nivolumab (pKd 9.1) [28, 38, 40] – Comment: The endogenous ligands for human PD-1 are programmed cell death 1 ligand 1 (PD-L1 aka CD274 (CD274, Q9NZQ7)) and programmed cell death 1 ligand 2 (PD-L2; PDCD1LG2). These ligands are cell surface peptides, normally involved in immune system regulation. Expression of PD-1 by cancer cells induces immune tolerance and evasion of immune system attack. Anti-PD-1 monoclonal antibodies are used to induce immune checkpoint blockade as a therapeutic intervention in cancer, effectively re-establishing immune vigilance. pembrolizumab was the ﬁrst anti-PD-1 antibody to be approved by the US FDA. Further reading on CD molecules Gabius HJ et al. (2015) The glycobiology of the CD system: a dictionary for translating marker desig- nations into glycan/lectin structure and function. Trends Biochem Sci 40: 360-76 [PMID:25981696] Methyllysine reader proteins Oth t i t t Ch ti i t ti t i ti l M th ll i d t i Nomenclature CD80 CD86 cytotoxic T-lymphocyte-associated protein 4 (CD152) programmed cell death 1 (CD279) CD300a HGNC, UniProt CD80, P33681 CD86, P42081 CTLA4, P16410 PDCD1, Q15116 CD300A, Q9UGN4 Common abreviation – – CTLA-4 PD-1 – Antibodies – – ipilimumab (pKd >9) [28], tremelimumab (pKd 8.9) [30] pembrolizumab (pKd ∼10) [11], nivolumab (pKd 9.1) [28, 38, 40] – programmed cell death 1 (CD279) CD300a PDCD1, Q15116 CD300A, Q9UGN4 PD-1 – pembrolizumab (pKd ∼10) [11], nivolumab (pKd 9.1) [28, 38, 40] – Comment: The endogenous ligands for human PD-1 are programmed cell death 1 ligand 1 (PD-L1 aka CD274 (CD274, Q9NZQ7)) and programmed cell death 1 ligand 2 (PD are cell surface peptides, normally involved in immune system regulation. Expression of PD-1 by cancer cells induces immune tolerance and evasion of immune system antibodies are used to induce immune checkpoint blockade as a therapeutic intervention in cancer, effectively re-establishing immune vigilance. pembrolizumab was th approved by the US FDA. ment: The endogenous ligands for human PD-1 are programmed cell death 1 ligand 1 (PD-L1 aka CD274 (CD274, Q9NZQ7)) and programmed cell death 1 ligand 2 (PD-L2; PDCD1LG2). CD molecules Other protein targets →CD molecules Other protein targets →CD molecules Overview: Cluster of differentiation refers to an attempt to catalogue systematically a series of over 300 cell-surface proteins associated with immunotyping. Many members of the group have identiﬁed functions as enzymes (for example, see CD73 ecto-5’-nucleotidase) or receptors (for example, see CD41 integrin, alpha 2b subunit). Many CDs are targetted for therapeutic gain using antibodies for the treatment of proliferative disorders. A full listing of all the Clusters of Differentiation is not possible in the Guide to PHARMACOLOGY; listed herein are selected members of the family targetted for ature CD2 CD3e CD20 (membrane-spanning 4-domains, subfamily A, member 1) CD33 CD52 niProt CD2, P06729 CD3E, P07766 MS4A1, P11836 CD33, P20138 CD52, P31358 n – – – SIGLEC-3 – nhibitors alefacept (Inhibition) [17, 53] – – – – s – catumaxomab (Binding) [43], muromonab-CD3 (Binding) [25], otelixizumab (Binding) [9] ofatumumab (Binding) (pKd 9.9) [47], rituximab (Binding) (pKd 8.5) [75], ibritumomab tiuxetan (Binding), obinutuzumab (Binding) [3, 66], tositumomab (Binding) lintuzumab (Binding) (pKd ∼10) [10], gemtuzumab ozogamicin (Binding) [7] alemtuzumab (Binding) [24, 79] Nomenclature CD2 CD3e CD20 (membrane-spanning 4-domains, subfamily A, member 1) CD33 CD52 HGNC, UniProt CD2, P06729 CD3E, P07766 MS4A1, P11836 CD33, P20138 CD52, P31358 Common abreviation – – – SIGLEC-3 – Selective inhibitors alefacept (Inhibition) [17, 53] – – – – Antibodies – catumaxomab (Binding) [43], muromonab-CD3 (Binding) [25], otelixizumab (Binding) [9] ofatumumab (Binding) (pKd 9.9) [47], rituximab (Binding) (pKd 8.5) [75], ibritumomab tiuxetan (Binding), obinutuzumab (Binding) [3, 66], tositumomab (Binding) lintuzumab (Binding) (pKd ∼10) [10], gemtuzumab ozogamicin (Binding) [7] alemtuzumab (Binding) [24, 79] Nomenclature CD2 CD3e CD20 (membrane-spanning 4-domains, subfamily A, member 1) CD33 CD52 HGNC, UniProt CD2, P06729 CD3E, P07766 MS4A1, P11836 CD33, P20138 CD52, P31358 Common abreviation – – – SIGLEC-3 – Selective inhibitors alefacept (Inhibition) [17, 53] – – – – Antibodies – catumaxomab (Binding) [43], muromonab-CD3 (Binding) [25], otelixizumab (Binding) [9] ofatumumab (Binding) (pKd 9.9) [47], rituximab (Binding) (pKd 8.5) [75], ibritumomab tiuxetan (Binding), obinutuzumab (Binding) [3, 66], tositumomab (Binding) lintuzumab (Binding) (pKd ∼10) [10], gemtuzumab ozogamicin (Binding) [7] alemtuzumab (Binding) [24, 79] Searchable database: http://www.guidetopharmacology.org/index.jsp CD molecules S9 Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full CD molecules S9 CD molecules S9 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. Further reading on Methyllysine reader proteins Liu K et al. (2015) Epigenetic targets and drug discovery Part 2: Histone demethylation and DNA methylation. Pharmacol. Ther. 151: 121-40 [PMID:25857453] Milosevich N et al. (2016) Chemical Inhibitors of Epigenetic Methyllysine Reader Proteins. Bio- chemistry 55: 1570-83 [PMID:26650180] Sadakierska-Chudy A et al. (2015) A comprehensive view of the epigenetic landscape part I: DNA methylation, passive and active DNA demethylation pathways and histone variants. Neurotox Res 27: 84-97 [PMID:25362550] Methyllysine reader proteins S10 Methyllysine reader proteins S10 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Fatty acid-binding proteins Other protein targets →Fatty acid-binding proteins y g p Other protein targets →Fatty acid-binding proteins acid receptors [70]) or for interaction with metabolic enzymes. Although sequence homology is limited, crystallographic stud- ies suggest conserved 3D structures across the group of binding proteins. the otherwise hydrophobic ligands to be mobile in aqueous me- dia. These binding proteins may perform functions extracellu- larly (e.g. in plasma) or transport these agents; to the nucleus to interact with nuclear receptors (principally PPARs and retinoic Overview: Fatty acid-binding proteins are low molecular weight (100-130 aa) chaperones for long chain fatty acids, fatty acyl CoA esters, eicosanoids, retinols, retinoic acids and related metabo- lites and are usually regarded as being responsible for allowing the otherwise hydrophobic ligands to be mobile in aqueous me- dia. These binding proteins may perform functions extracellu- larly (e.g. in plasma) or transport these agents; to the nucleus to interact with nuclear receptors (principally PPARs and retinoic Nomenclature fatty acid binding protein 1 fatty acid binding protein 2 fatty acid binding protein 3 fatty acid binding protein 4 HGNC, UniProt FABP1, P07148 FABP2, P12104 FABP3, P05413 FABP4, P15090 Rank order of potency stearic acid, oleic acid > palmitic acid, linoleic acid > arachidonic acid, α-linolenic acid [67] stearic acid > palmitic acid,oleic acid > linoleic acid > arachidonic acid, α-linolenic acid [67] stearic acid, oleic acid, palmitic acid > linoleic acid, α-linolenic acid, arachidonic acid [67] oleic acid, palmitic acid, stearic acid, linoleic acid > α-linolenic acid, arachidonic acid [67] Inhibitors fenoﬁbrate (pKi 7.6) [12] – Rat, fenoﬁbric acid (pKi 6.5) [12] – Rat, HTS01037 (pKi 5.1) [30] – Mouse – – – Selective inhibitors – – – HM50316 (pKi >9) [46] Comments A broader substrate speciﬁcity than other FABPs, binding two fatty acids per protein [82]. Crystal structure of the rat FABP2 [69]. Crystal structure of the human FABP3 [91]. – Nomenclature fatty acid binding protein 5 fatty acid binding protein 6 fatty acid binding protein 7 peripheral myelin protein 2 fatty acid binding protein 9 fatty acid binding protein 12 HGNC, UniProt FABP5, Q01469 FABP6, P51161 FABP7, O15540 PMP2, P02689 FABP9, Q0Z7S8 FABP12, A6NFH5 Comments Crystal structure of the human FABP5 [31]. Able to transport bile acids [95]. Crystal structure of the human FABP7 [5]. In silico modelling suggests that PMP2/FABP8 can bind both fatty acids and cholesterol [50]. Fatty acid-binding proteins Other protein targets →Fatty acid-binding proteins omenclature retinol binding protein 1 retinol binding protein 2 retinol binding protein 3 retinol binding protein 4 retinol binding protein 5 retinol binding protein 7 GNC, UniProt RBP1, P09455 RBP2, P50120 RBP3, P10745 RBP4, P02753 RBP5, P82980 RBP7, Q96R05 nk order of otency – stearic acid > palmitic acid, oleic acid, linoleic acid, α-linolenic acid, arachidonic acid [68] – – – – hibitors – – – A1120 (pIC50 7.8) [86] – – omenclature retinaldehyde binding protein 1 cellular retinoic acid binding protein 1 cellular retinoic acid binding protein 2 GNC, UniProt RLBP1, P12271 CRABP1, P29762 CRABP2, P29373 nk order of potency 11-cis-retinal, 11-cis-retinol > 9-cis-retinal, 13-cis-retinal, 13-cis-retinol, all-trans-retinal, retinol [15] tretinoin > alitretinoin stearic acid > palmitic acid, oleic acid, linoleic acid, α-linolenic acid, arachidonic acid [68] – ents: Although not tested at all FABPs, BMS309403 exhibits high afﬁnity for FABP4 (pIC50 8.8) compared to FABP3 or FABP5 (pIC50 <6.6) [21, 81]. HTS01037 is reported to interfere with FABP4 30]. Ibuprofen displays some selectivity for FABP4 (pIC50 5.5) relative to FABP3 (pIC50 3.5) and FABP5 (pIC50 3.8) [48]. Fenoﬁbric acid displays some selectivity for FABP5 (pIC50 5.5) relative to pIC50 4.5) and FABP4 (pIC50 4.6) [48]. Multiple pseudogenes for the FABPs have been identiﬁed in the human genome. retinol binding protein 4 cellular retinoic acid binding protein 2 CRABP2, P29373 Comments: Although not tested at all FABPs, BMS309403 exhibits high afﬁnity for FABP4 (pIC50 8.8) compared to FABP3 or FABP5 (pIC50 <6.6) [21, 81]. HTS01037 is action [30]. Ibuprofen displays some selectivity for FABP4 (pIC50 5.5) relative to FABP3 (pIC50 3.5) and FABP5 (pIC50 3.8) [48]. Fenoﬁbric acid displays some selectivity FABP3 (pIC50 4.5) and FABP4 (pIC50 4.6) [48]. Multiple pseudogenes for the FABPs have been identiﬁed in the human genome. lthough not tested at all FABPs, BMS309403 exhibits high afﬁnity for FABP4 (pIC50 8.8) compared to FABP3 or FABP5 (pIC50 <6.6) [21, 81]. HTS01037 is reported to interfere with FABP4 profen displays some selectivity for FABP4 (pIC50 5.5) relative to FABP3 (pIC50 3.5) and FABP5 (pIC50 3.8) [48]. Fenoﬁbric acid displays some selectivity for FABP5 (pIC50 5.5) relative to .5) and FABP4 (pIC50 4.6) [48]. Multiple pseudogenes for the FABPs have been identiﬁed in the human genome. Fatty acid-binding proteins Other protein targets →Fatty acid-binding proteins – – Searchable database: http://www.guidetopharmacology.org/index.jsp Fatty acid-binding proteins S11 Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Fatty acid-binding proteins S11 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Nomenclature retinol binding protein 1 retinol binding protein 2 retinol binding protein 3 retinol binding protein 4 retinol binding protein 5 retinol binding protein 7 HGNC, UniProt RBP1, P09455 RBP2, P50120 RBP3, P10745 RBP4, P02753 RBP5, P82980 RBP7, Q96R05 Rank order of potency – stearic acid > palmitic acid, oleic acid, linoleic acid, α-linolenic acid, arachidonic acid [68] – – – – Inhibitors – – – A1120 (pIC50 7.8) [86] – – Nomenclature retinaldehyde binding protein 1 cellular retinoic acid binding protein 1 cellular retinoic acid binding protein 2 HGNC, UniProt RLBP1, P12271 CRABP1, P29762 CRABP2, P29373 Rank order of potency 11-cis-retinal, 11-cis-retinol > 9-cis-retinal, 13-cis-retinal, 13-cis-retinol, all-trans-retinal, retinol [15] tretinoin > alitretinoin stearic acid > palmitic acid, oleic acid, linoleic acid, α-linolenic acid, arachidonic acid [68] – Comments: Although not tested at all FABPs, BMS309403 exhibits high afﬁnity for FABP4 (pIC50 8.8) compared to FABP3 or FABP5 (pIC50 <6.6) [21, 81]. HTS01037 is reported to interfere with FABP4 action [30]. Ibuprofen displays some selectivity for FABP4 (pIC50 5.5) relative to FABP3 (pIC50 3.5) and FABP5 (pIC50 3.8) [48]. Fenoﬁbric acid displays some selectivity for FABP5 (pIC50 5.5) relative to FABP3 (pIC50 4.5) and FABP4 (pIC50 4.6) [48]. Multiple pseudogenes for the FABPs have been identiﬁed in the human genome. Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Matsumata M et al. (2016) Fatty acid binding proteins and the nervous system: Their impact on mental conditions. Neurosci. Res. 102: 47-55 [PMID:25205626] Osumi T et al. (2016) Heart lipid droplets and lipid droplet-binding proteins: Biochemistry, physi- ology, and pathology. Exp Cell Res 340: 198-204 [PMID:26524506] Notch receptors p Other protein targets →Notch receptors p Other protein targets →Notch receptor been terminated following an unsuccessful Phase II single agent clinical trial in metastatic colorectal cancer [78]. been terminated following an unsuccessful Phase II single agent clinical trial in metastatic colorectal cancer [78]. receptor-ligand interactions to occur. Cleavage of the intracel- lular domain (ICD) of activated Notch receptors by γ-secretase is required for downstream signalling and Notch-induced tran- scriptional modulation [18, 57, 71, 89]. This is why γ-secretase inhibitors can be used to downregulate Notch signalling and explains their anti-cancer action. One such small molecule is RO4929097 [47], although development of this compound has Overview: The canonocal Notch signalling pathway has four type I transmembrane Notch receptors (Notch1-4) and ﬁve lig- ands (DLL1, 2 and 3, and Jagged 1-2). Each member of this highly conserved receptor family plays a unique role in cell-fate deter- mination during embryogenesis, differentiation, tissue pattern- ing, proliferation and cell death [2]. As the Notch ligands are also membrane bound, cells have to be in close proximity for receptor-ligand interactions to occur. Cleavage of the intracel- lular domain (ICD) of activated Notch receptors by γ-secretase is required for downstream signalling and Notch-induced tran- scriptional modulation [18, 57, 71, 89]. This is why γ-secretase inhibitors can be used to downregulate Notch signalling and explains their anti-cancer action. One such small molecule is RO4929097 [47], although development of this compound has Aberrant Notch signalling is implicated in a number of human cancers [41, 59, 74, 85]. Pharmaceutical inhibitors of Notch sig- nalling such as demcizumab and tarextumab are being actively investigated as novel anti-cancer agents [64]. Various types of activating and inactivating NOTCH1 mutations have been reported to be associated with human diseases, for example: aortic valve disease [23, 52], Adams-Oliver syndrome 5 [76], T-cell acute lymphoblastic leukemia (T-ALL) [87], chronic lymphocytic leukemia (CLL) [65] and head and neck squamous cell carcinoma [1, 77]. Further reading on Notch receptors Previs RA et al. (2015) Molecular pathways: translational and therapeutic implications of the Notch signaling pathway in cancer. Clin Cancer Res 21: 955-61 [PMID:25388163] Takebe N et al. (2015) Targeting Notch, Hedgehog, and Wnt pathways in cancer stem cells: clinical update. Nat Rev Clin Oncol 12: 445-464 [PMID:25850553] Borggrefe T et al. (2016) The Notch intracellular domain integrates signals from Wnt, Hedgehog, TGFbeta/BMP and hypoxia pathways. Biochim Biophys Acta 1863: 303-313 [PMID:26592459] Cheng YL et al. (2015) Emerging roles of the gamma-secretase-notch axis in inﬂammation. Phar- macol Ther 147: 80-90 [PMID:25448038] macol Ther 147: 80 90 [PMID:25448038] Palmer WH et al. (2015) Ligand-Independent Mechanisms of Notch Activity. Trends Cell Biol 25: 697-707 [PMID:26437585] Further reading on Fatty acid-binding proteins Matsumata M et al. (2016) Fatty acid binding proteins and the nervous system: Their impact on mental conditions. Neurosci. Res. 102: 47-55 [PMID:25205626] Osumi T et al. (2016) Heart lipid droplets and lipid droplet-binding proteins: Biochemistry, physi- ology, and pathology. Exp Cell Res 340: 198-204 [PMID:26524506] Gajda AM et al. (2015) Enterocyte fatty acid-binding proteins (FABPs): different functions of liver and intestinal FABPs in the intestine. Prostaglandins Leukot. Essent. Fatty Acids 93: 9-16 [PMID:25458898] [ ] Glatz JF. (2015) Lipids and lipid binding proteins: a perfect match. Prostaglandins Leukot. Essent. Fatty Acids 93: 45-9 [PMID:25154384] [ ] Glatz JF. (2015) Lipids and lipid binding proteins: a perfect match. Prostaglandins Leukot. Essent. Fatty Acids 93: 45-9 [PMID:25154384] y [ ] Hotamisligil GS et al. (2015) Metabolic functions of FABPs-mechanisms and therapeutic implica- tions. Nat Rev Endocrinol 11: 592-605 [PMID:26260145] Fatty acid-binding proteins S12 Fatty acid-binding proteins S12 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Sigma receptors g p Other protein targets →Sigma receptors g p Other protein targets →Sigma receptors Overview: Although termed ‘receptors’, the evidence for coupling through conventional signalling pathways is lacking. Initially described as a subtype of opioid re pharmacological overlap and no structural convergence with the G protein-coupled receptors; the crystal structure of the sigma1 receptor [94] suggests a trimeric structure domain traversing the endoplasmic reticulum membrane, with the bulk of the protein facing the cytosol. A wide range of compounds, ranging from psychoactive agen observed to bind to these sites. Comments: (-)-pentazocine also shows activity at opioid receptors. The sigma2 receptor has recently been repo protein, a 13TM cholesterol-binding protein. Regulators of G protein Signaling (RGS) proteins Other protein targets →Regulators of G protein Signaling (RGS) proteins Overview: Regulators of G protein signalling (RGS) proteins increase the deactivation rates of G protein signalling pathways through enhancing the GTPase activity of the G protein alpha subunit. Interactions through protein:protein interactions of many RGS proteins have been identiﬁed for targets other than heteromeric G proteins. The 20 RGS proteins are commonly divided into four families (R4, R7, R12 and RZ) based on sequence and domain homology. Described here is RGS4 for which a number of pharmacological inhibitors have been described. Regulators of G protein Signaling (RGS) proteins S13 Regulators of G protein Signaling (RGS) proteins S13 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Nomenclature regulator of G-protein signaling 4 HGNC, UniProt RGS4, P49798 Common abreviation RGS4 Selective inhibitors RGS4 inhibitor 11b (pIC50 7.8) [83], CCG-50014 (pIC50 7.5) [8, 83], RGS4 inhibitor 13 (pIC50 7.3) [83] regulator of G-protein signaling 4 RGS4, P49798 RGS4 RGS4 RGS4 inhibitor 11b (pIC50 7.8) [83], CCG-50014 (pIC50 7.5) [8, 83], RGS4 inhibitor 13 (pIC50 7.3) [83] Further reading on RGS proteins Sjogren B et al. (2010) Thinking outside of the "RGS box": new approaches to therapeutic targeting of regulators of G protein signaling. Mol Pharmacol 78: 550-7 [PMID:20664002] Turner EM et al. (2012) Small Molecule Inhibitors of Regulator of G Protein Signalling (RGS) Pro- teins. ACS Med Chem Lett 3: 146-150 [PMID:22368763] Sethakorn N et al. (2010) Non-canonical functions of RGS proteins. Cell Signal 22: 1274-81 [PMID:20363320] [ ] Sjogren B (2017) The evolution of regulators of G protein signalling proteins as drug targets - 20 years in the making: IUPHAR Review 21. Br J Pharmacol 174: 427-437 [PMID:28098342] Su TP et al. (2016) The Sigma-1 Receptor as a Pluripotent Modulator in Living Systems. Trends Pharmacol Sci 37: 262-78 [PMID:26869505] van Waarde A et al. (2015) Potential applications for sigma receptor ligands in cancer diagnosis and therapy. Biochim Biophys Acta 1848: 2703-14 [PMID:25173780] Further reading on Sigma receptors Chu UB et al. (2016) Biochemical Pharmacology of the Sigma-1 Receptor. Mol Pharmacol 89: 142-53 [PMID:26560551] Gris G et al. (2015) Sigma-1 receptor and inﬂammatory pain. Inﬂamm Res 64: 377-81 [PMID:25902777] Rousseaux CG et al. (2015) Sigma receptors [sigmaRs]: biology in normal and diseased states. J Recept Signal Transduct Res 1-62 [PMID:26056947] Chu UB et al. (2016) Biochemical Pharmacology of the Sigma-1 Receptor. Mol Pharmacol 89: 142-53 [PMID:26560551] Gris G et al. (2015) Sigma-1 receptor and inﬂammatory pain. Inﬂamm Res 64: 377-81 [PMID:25902777] Rousseaux CG et al. (2015) Sigma receptors [sigmaRs]: biology in normal and diseased states. J Recept Signal Transduct Res 1-62 [PMID:26056947] Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Sigma receptors S14 S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 Tubulins Other protein targets →Tubulins Overview: Tubulins are a family of intracellular proteins most commonly associated with microtubules, part of the cytoskeleton. They are exploited for therapeutic gain in cancer chemotherapy as targets for agents derived from a variety of natural products: taxanes, colchicine and vinca alkaloids. These are thought to act primarily through β-tubulin, thereby interfering with the normal processes of tubulin polymer formation and disassembly. Overview: Tubulins are a family of intracellular proteins most commonly associated with microtubules, part of the cytoskeleton. They are ex targets for agents derived from a variety of natural products: taxanes, colchicine and vinca alkaloids. These are thought to act primarily through β of tubulin polymer formation and disassembly. Nomenclature tubulin alpha 1a tubulin alpha 4a tubulin beta class I tubulin beta 3 class III tubulin beta 4B class IVb tubulin beta 8 class VIII HGNC, UniProt TUBA1A, Q71U36 TUBA4A, P68366 TUBB, P07437 TUBB3, Q13509 TUBB4B, P68371 TUBB8, Q3ZCM7 Inhibitors – – vinblastine (pIC50 9), vincristine, eribulin (pIC50 8.2) [58], paclitaxel (pEC50 8.1) [61], colchicine (pIC50 8) [13], cabazitaxel, docetaxel, ixabepilone combretastatin A4 (pIC50 8.2) [22] – – Further reading on Tubulins Gadadhar S et al. (2017) The tubulin code at a glance. J Cell Sci 130: 1347-1353 [PMID:28325758] Penna LS et al. (2017) Anti-mitotic agents: Are they emerging molecules for cancer treatment? Pharmacol Ther 173: 67-82 [PMID:28174095] ature tubulin alpha 1a tubulin alpha 4a tubulin beta class I tubulin beta 3 class III tubulin beta 4B class IVb tubulin beta 8 class VIII TUBA1A, Q71U36 TUBA4A, P68366 TUBB, P07437 TUBB3, Q13509 TUBB4B, P68371 TUBB8, Q3ZCM7 – – vinblastine (pIC50 9), vincristine, eribulin (pIC50 8.2) [58], paclitaxel (pEC50 8.1) [61], colchicine (pIC50 8) [13], cabazitaxel, docetaxel, ixabepilone combretastatin A4 (pIC50 8.2) [22] – – ing on Tubulins al. (2017) The tubulin code at a glance. J Cell Sci 130: 1347-1353 [PMID:28325758] Penna LS et al. (2017) Anti-mitotic agents: Are they emerging molecules for cancer treatment? Pharmacol Ther 173: 67-82 [PMID:28174095] Further reading on Tubulins Further reading on Tubulins code at a glance. J Cell Sci 130: 1347-1353 [PMID:28325758] Penna LS et al. (2017) Anti-mitotic agents: Are they emerging molecules for cancer treatment? Pharmacol Ther 173: 67-82 [PMID:28174095] Penna LS et al. (2017) Anti-mitotic agents: Are they emerging molecules for cancer treatment? Pharmacol Ther 173: 67-82 [PMID:28174095] Gadadhar S et al. (2017) The tubulin code at a glance. Tubulins Other protein targets →Tubulins Tubulins Other protein targets →Tubulins Tubulins J Cell Sci 130: 1347-1353 [PMID:28325758] Penna LS et al. (2017) Anti-mitotic agents: Are they emerging molecules for cancer treatmen Pharmacol Ther 173: 67-82 [PMID:28174095] Gadadhar S et al. (2017) The tubulin code at a glance. J Cell Sci 130: 1347-1353 [PMID:28325758] Searchable database: http://www.guidetopharmacology.org/index.jsp Tubulins S15 Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full Tubulins S15 Tubulins S15 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full S.P.H. Alexander et al. The Concise Guide to PHARMACOLOGY 2017/18: Overview. British Journal of Pharmacology (2017) 174, S1–S16 ( ) [ 14. Coelho T. (1996) [8894411] ( ) [ ] 64. Prezzavento O et al. (2007) [17328523] J ( ) [ ] 39. Korman AJ et al. (2006) Patent number: ( ) [ 14. Coelho T. (1996) [8894411] 84. Vicente Rabaneda EF et al. (2013) [23899231] ( ) [ ] 15. Crabb JW et al. (1998) [9541407] ( ) [ 33. Hug C et al. (2004) [15210937] ( ) [ ] 58. Narayan S et al. (2011) [21324687] ( ) [ 9. Bolt S et al. (1993) [8436176] 33. Hug C et al. (2004) [15210937] p bromodomains. Accessed on 10/11/2014. g ( ) [ ] 34. Jacobson DR et al. (1997) [9017939] y ( ) 59. Ntziachristos P et al. (2014) [24651013] ( ) 10. Caron PC et al. (1992) [1458463] ( ) [ 43. Linke R et al. (2010) [20190561] 89. Wilson JJ et al. (2006) [16530045] ( ) [ ] 70. Schroeter EH et al. (1998) [9620803] ( ) [ ] 44. Linsley PS et al. (1991) [1714933] ( ) [ ] 20. Friedel HA et al. (1994) [7528134] 90. Yamauchi T et al. (2003) [12802337] ( ) [ ] 71. I-CBP112 - a CREBBP/EP300-selective chem- ( ) [ ] 71. I-CBP112 - a CREBBP/EP300-selective chem- ical probe. Accessed on 03/03/2015. thesgc.org. y ( ) 45. Liu Q. (2013) Patent number: 21. Furuhashi M et al. (2007) [17554340] ( ) 91. Young AC et al. (1994) [7922029] J ( ) [ 57. Nam Y et al. (2006) [16530044] ( ) 8. Blazer LL et al. (2011) [21329361] 32. Holmer E et al. (1986) [3744129] g ( ) [ ] 79. PFI-3: Selective chemical probe for SMARCA g 79. PFI-3: Selective chemical probe for SMARCA bromodomains. Accessed on 10/11/2014. http://www.thesgc.org. p ( ) [ ] 38. Kline J et al. (2010) [21154117] p ( ) [ ] 38. Kline J et al. (2010) [21154117] g ( ) [ ] 13. Cifuentes M et al. (2006) [16504507] p ( ) [ ] 83. Turner EM et al. (2012) [22368763] gj 23. Garg V et al. (2005) [16025100] gj 23. Garg V et al. (2005) [16025100] g ( ) [ 92. Alon A et al. (2017) [28559337] g 92. Alon A et al. (2017) [28559337] 46. Liu X et al. (2011) [21481589] ( ) [ 65. Reslan L et al. (2013) [23537278] ( ) [ 65. Reslan L et al. (2013) [23537278] 85. Vilimas T et al. (2007) [17173050] ( ) [ ] 66. Richieri GV et al. (1994) [7929039] J ( ) [ ] 16. Cziraky MJ et al. (1993) [8137606] j ( ) [ 4. ANDRADE C. (1952) [12978172] ( ) [ ] 29. Hanson DC et al. (2004) Patent number: ( ) [ 53. Mitchell P. (2002) [12089534] ( ) [ ] 29. Hanson DC et al. (2004) Patent number: US6682736 B1. j 75. Stein R et al. (2004) [15102696] 28. Hall RD et al. (2013) [23302904] 28. Hall RD et al. (2013) [23302904] ( ) [ ] 52. McBride KL et al. (2008) [18593716] ( ) [ 3. Alduaij W et al. (2011) [21378274] 74. Sjölund J et al. (2008) [18079963] j ( ) [ ] 4. ANDRADE C. (1952) [12978172] http://www.thesgc.org. ( ) [ 60. Okuyama S et al. (1993) [7901723] J ( ) [ ] 35. James LI et al. (2013) [23292653] ( ) [ ] 11. Carven GJ et al. (2010) Patent num p g g 80. Su TP et al. (1991) [1658302] Q ( ) WO2013007052. 22. Gangjee A et al. (2013) [23895532] US6682736 B1. ( ) [ ] 5. Balendiran GK et al. (2000) [10854433] ( ) [ ] 54. Murakami K et al. (1999) [10403814] ( ) [ ] 76. Stittrich AB et al. (2014) [25132448] ( ) [ ] 55. Nagamatsu I et al. (2014) [24403446] 30. Hertzel AV et al. (2009) [19754198] ( ) [ 6. Berardi F et al. (1996) [8568804] 77. Stransky N et al. (2011) [21798893] 40. Latek R et al. (2009) [19300198] 86. Wang Y et al. (2014) [24835984] ( ) [ ] 41. Lefort K et al. (2007) [17344417] y ( ) [ ] 17. da Silva AJ et al. (2002) [11970990] ( ) [ ] 67. Richieri GV et al. (2000) [10852718] g ( ) [ ] 87. Weng AP et al. (2004) [15472075] J ( ) [ ] 18. De Strooper B et al. (1999) [10206645] ( ) [ ] 42. Lehmann BD et al. (2015) [25993190] ( ) [ ] 68. Sacchettini JC et al. (1989) [2671390] g ( ) [ ] 88. Westermark P et al. (1981) [7016817] p ( ) [ 19. Eriksson BI et al. (1995) [7667822] J ( ) [ ] 69. Schroeder F et al. (2008) [17882463] ( ) [ 43. Linke R et al. (2010) [20190561] Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full y ( ) [ ] 61. Ouyang X et al. (2006) [16377187] y ( ) [ ] 61. Ouyang X et al. (2006) [16377187] ( ) [ ] 36. Kanji S et al. (2001) [11714212] ( ) [ ] 81. Sulsky R et al. (2007) [17502136] 12. Chuang S et al. (2008) [18533710] j ( ) [ ] 37. Kapur S et al. (2001) [11463021] y g ( ) [ ] 62. Paolucci F et al. (2002) [12383040] y 82. Thompson J et al. (1997) [9054409] ( ) [ ] 63. Penchala SC et al. (2013) [23716704] g ( ) [ 56. Nakase J et al. (2009) [19398784] g ( ) [ 56. Nakase J et al. (2009) [19398784] ( ) [ ] 31. Hohoff C et al. (1999) [10493790] ( ) [ ] 7. Bernstein ID. (2000) [10720144] y ( ) [ ] 78. Strosberg JR et al. (2012) [22445247] J ( ) [ ] 57. Nam Y et al. (2006) [16530044] References 73. Shitara K et al. (2011) Patent number: US7923538 B2. 1. Agrawal N et al. (2011) [21798897] 50. Majava V et al. (2010) [20421974] 27. Halk EL et al. (2001) Patent number: WO2001014424. j ( ) [ ] 51. Matsumoto RR et al. (1995) [8566098] g ( ) [ ] 2. Al-Hussaini H et al. (2011) [20971825] 46. Liu X et al. (2011) [21481589] 93. Tanabe H et al. (2015) [25855295] 24. Ginaldi L et al. (1998) [9593475] 72. GSK2801: A Selective Chemical Probe for BAZ2B/A bromodomains. Accessed on 03/03/2015. thesgc.org. 47. Luistro L et al. (2009) [19773430] 94. Schmid HR et al. (2016) [27042935] ( ) [ 25. Goldstein G. (1987) [3105134] ( ) 48. Machbub B et al. (1988) [24248795] ( ) [ ] 26. Gotti R et al. (2013) [23598032] ( ) 95. Zwicker BL et al. (2013) [23603607] ( ) [ 49. Maeda K et al. (1996) [8619847] 95. Zwicker BL et al. (2013) [2 References S16 References S16 Searchable database: http://www.guidetopharmacology.org/index.jsp Full Contents of ConciseGuide: http://onlinelibrary.wiley.com/doi/10.1111/bph.13882/full
https://openalex.org/W2078030023	https://repository.ubn.ru.nl/bitstream/handle/2066/153528/1/153528.pdf	English	null	Novel pantothenate derivatives for anti-malarial chemotherapy	Malaria journal	2,015	cc-by	6,653	Novel pantothenate derivatives for anti-malarial chemotherapy Pett, H.E.; Jansen, P.A.M.; Hermkens, P.H.; Botman, P.N.M.; Beuckens-Schortinghuis, C.A.; Blaauw, R.H.; Graumans, W.; Vegte-Bolmer, M.G. van de; Koolen, K.M.; Rutjes, F.P.J.T.; Dechering, K.J.; Sauerwein, R.W.; Schalkwijk, J. 2015, Article / Letter to editor (Malaria Journal, 14, (2015), pp. 169, article 169) Doi link to publisher: https://doi.org/10.1186/s12936-015-0673-8 Version of the following full text: Publisher’s version Downloaded from: http://hdl.handle.net/2066/153528 Download date: 2024-10-24 Note: Note: To cite this publication please use the final published version (if applicable). To cite this publication please use the final published version (if applicable). Pett et al. Malaria Journal (2015) 14:169 DOI 10.1186/s12936-015-0673-8 * Correspondence: joost.schalkwijk@radboudumc.nl 2Department of Dermatology and Radboud Institute for Molecular Life Sciences, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands 6Pansynt B V, Nijmegen, The Netherlands Full list of author information is available at the end of the article © 2015 Pett et al.; licensee BioMed Central. 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. RESEARCH Open Access Novel pantothenate derivatives for anti-malarial chemotherapy Helmi E Pett1, Patrick AM Jansen2, Pedro HH Hermkens6, Peter NM Botman4, Christien A Beuckens-Schortinghuis4, Richard H Blaauw4, Wouter Graumans1, Marga van de Vegte-Bolmer1, Karin MJ Koolen5, Floris PJT Rutjes3,6, Koen J Dechering5, Robert W Sauerwein1,5 and Joost Schalkwijk2,6* Abstract Background: A number of synthetic pantothenate derivatives, such as pantothenamides, are known to inhibit the growth of the human malaria parasite Plasmodium falciparum, by interfering with the parasite Coenzyme A (CoA) biosynthetic pathway. The clinical use of pantothenamides is limited by their sensitivity to breakdown by ubiquitous human pantetheinases of the vanin family. Methods: A number of pantothenate derivatives (pantothenones) with potent and specific inhibitory activity against mammalian vanins were tested in a proliferation assay of asexual P. falciparum blood stages alone, and in combination with pantothenamides. Results: The vanin inhibitors were found to protect pantothenamides against breakdown by plasma vanins, thereby preserving the in vitro anti-malarial activity. Moreover, some of the vanin inhibitors showed in vitro anti-malarial activity in the low micromolar range. The most potent antimalarial in this series of compounds (RR8), was found to compete with pantothenate in a combination proliferation assay. No correlation, however, was found between anti-vanin and anti-malarial activity, nor was pantetheinase activity detected in P. falciparum extracts. Conclusions: Growth inhibition is most likely due to competition with pantothenate, rather than pantetheinase inhibition. As vanin inhibitors of the pantothenone class are stable in biological fluids and are non-toxic to mammalian cells, they may represent novel pantothenate-based anti-malarials, either on their own or in combination with pantothenamides. Keywords: Malaria, Anti-malarial, Plasmodium falciparum, Pantothenate, Pantothenic acid, Pantothenamide, Pantothenone, Coenzyme A Compounds Pantothenol was purchased from Sigma-Aldrich. De- tailed synthesis procedures of CJ-15,801 and CXP14.1- 060 are provided as Additional file 1. The methodologies for synthesis of CXP14.1-034, RR2, RR6, RR7, RR8 and SN12,601 have been previously published in Jansen et al. [13]. N5-Pan and N7-Pan were synthesized as described in the supplemental material of Jansen et al. [14]. N9- Pan was synthesized as N5-Pan and N7-Pan, but instead of using a pentylamine or heptylamine, a nonylamine was used for the synthesis of N9-Pan. The synthesis of SN14,621 and SN14,622 was performed as described in Winterbottom et al. [19]. The synthesis of phenethyl- Pan was performed as described by Spry et al. [9]. Background (pantothenic acid, vitamin B5), were tested for antiplas- modial activity as early as the 1940s [2]. These compounds included pantoyltaurine, substituted pantoyltaurylamides, sulphonamides, and pantothenones, according to the no- menclature used in a review on this subject by Spry et al. [3]. These and similar compounds were tested in different in vitro and in vivo malaria models from the 1960s and 1970s [4,5]. In 1976, Trager and Jensen published an art- icle describing the continuous culture of P. falciparum [6], allowing Divo et al. to discover that pantothenate is indeed the only water soluble vitamin that needs to be ex- ogenously available for P. falciparum survival [7]. Mean- while, Clifton et al. prepared a series of analogues with the general structure N1-(substituted) pantothenamide, and Each year over half a million people die of malaria, with Plasmodium falciparum being the primary cause of fatal malaria cases [1]. As the eradication of malaria is threat- ened by occurrence of clinical resistance to artemisinin derivatives, new drugs for malaria are sorely needed and so the search for new lead compounds continues [1]. Based on the observation, that addition of calcium pantothenate to Plasmodium lophurae cultures increased parasite viability, a selection of analogues of pantothenate Pett et al. Malaria Journal (2015) 14:169 Page 2 of 8 found them to have antibacterial activity due to being an- timetabolites of pantothenate [8]. Chemical structures for all of the compounds in this study are presented in Figure 1. Recent studies showed that some of the pantothena- mides were also active against P. falciparum in vitro, pro- vided that plasma pantetheinase activity was reduced [9]. This was discovered due to the observation that ‘aging’ P. falciparum growth media increased the anti-malarial ac- tivity of some pantothenamides [9]. Later, this same effect was achieved with heat inactivation of the parasite growth medium by de Villiers et al. [10]. The mechanism of break- down of pantothenamides by pantetheinases of the vanin family was elucidated in detail by Jansen et al. who discov- ered that combining pantothenamides with small molecule vanin inhibitors, protected pantothenamides against break- down, thereby dramatically increasing their antibacterial activity against both Staphylococcus aureus and Escherichia coli [11-14]. It has also been shown by de Villiers et al. that small modifications of the pantothenamide core structure could protect the molecule against pantetheinase-mediated degradation, albeit at a cost of a 100-fold decrease in anti- malarial potency [10]. Plasmodium falciparum culture The asexual stages of the NF54 strain P. falciparum were cultured as previously described [6], utilizing a shaker system with automated media change twice a day, parasites were kept in continuous culture within adapted Erlen- meyer flasks [20]. Erythrocytes were refreshed every two or three days to adjust haematocrit to 5% and parasitaemia to 0.5%. Human erythrocytes (blood type A) were obtained from healthy blood donors, with no history of malaria. Culture media consisted of RPMI 1640 with HEPES [5.94 g/l, hypoxanthine [0.05 g/l], 10% (v/v) pooled human serum (blood type A) obtained similarly to erythrocytes, and 0,2% (w/v) sodium bicarbonate. Temperature was set to 37°C and a low oxygen gas mixture was constantly flushed over the culture maintaining a stable atmosphere of 3% O2, 4% CO2, and 93% N2. Vanin activity assay h Vanin/pantetheinase activity assay with aminomethyl- coumarine (AMC) substrate and fluorescence readout was performed as described previously in Jansen et al. [13,21]. Human serum was used as a source of vanin enzymatic activity in assays to determine the anti-vanin activity of study compounds CJ-15,801, SN 12,601, SN 14,621, SN 14,622, CXP14.1-034, CXP14.1-060, RR2, RR6, RR7, and RR8. To determine whether P. falciparum para- sites harbour vanin activity, assays were performed on para- site extracts. To this end, 5 to 9 x 108 non-synchronous asexual NF54 strain P. falciparum parasites were pelleted by centrifugation at 4000 rpm for 10 minutes. Pellets were re-suspended in 5 ml of 0.06% saponin in phosphate buff- ered saline (PBS) and incubated on ice for 5 minutes to re- move erythrocytes. After this they were washed with PBS twice, with centrifugation in between as when pelleting cultures. Pellets from four different cultures were individually resuspended in a total volume of 300 μl of PBS and lysed by sonication (6 x 3 seconds). Vanin ac- tivity was determined by combining 29 μl of lysate with 1 μl of AMC substrate (final concentration of 333 μM) and incubation at room temperature. At 0, 1 and 19 hours, 3 μl of the reaction was diluted with 997 μl of PBS and fluorescence was measured in a 200 μl aliquot. Assay negative controls consisted of 29 μl of PBS and 1 μl of AMC substrate. Positive control consisted of 19 μl of PBS, 10 μl of human serum, and 1 μl of AMC substrate. Compounds, such as the pantothenamides in E. coli or the fungal product CJ-15,801 in S. aureus may hijack Coenzyme A (CoA) biosynthesis, being phosphorylated in the first step of the biosynthesis by pantothenate kin- ase (PanK) and eventually blocking CoA production or interfering with fatty acid synthesis downstream along the pathway [15-17]. Almost a decade ago, the fungal product CJ-15,801, was also discovered to have modest anti-malarial activity against asexual intra-erythrocytic stages of P. falciparum in vitro, and was demonstrated to inhibit parasite growth by a mechanism related to CoA biosynthesis or utilization [18]. In this study a selection of novel pantetheine analogues of the pantothenone class were investigated for potential use as anti-malarial chemotherapy. The investigated com- pounds are shown to be conceptually promising either as a monotherapy or in a combination of drugs. Plasmodium falciparum asexual blood stages assay with SYBR Green read-out A non-synchronous asexual NF54 strain P. falciparum culture was adjusted to a parasitaemia of 0.5-1% and a Pett et al. Malaria Journal (2015) 14:169 Page 3 of 8 Figure 1 Chemical structures of all the compounds tested in this study. Figure 1 Chemical structures of all the compounds tested in this study. haematocrit of 1-5%. The compound dilution curves were prepared in dimethylsulphoxide (DMSO) from 100 mM to 10 μM at half/log step dilutions. The established anti- malarial dihydroartemisinin (DHA) was used as a positive control and diluted in DMSO from 1 mM to 100 nM at half/log step dilutions. The DMSO dilutions were 500-fold diluted in growth medium to yield a final DMSO concen- tration of 0.2% (v/v), and final compound concentrations from 200 μM to 20 nM (2 μM to 200 pM for DHA). Higher concentrations for the experimental compounds would not have been possible to achieve due to limita- tions in final DMSO concentration (0.1% v/v). Fifty μl of each compound dilution was combined with 50 μl of parasite culture in a black, clear-bottomed or entirely black 96-well plate. The outermost wells were filled with sterile water to prevent evaporation. DHA at a concen- tration of 1 μM and 0.1% DMSO (negative control) were used to determine the assay window. The 96-well plates were incubated for 72 hours in a candle jar at 37°C. Read-out was done using the DNA-marker SYBR Green as described previously [22,23]. All assays were con- ducted at least in triplicate. haematocrit of 1-5%. The compound dilution curves were prepared in dimethylsulphoxide (DMSO) from 100 mM to 10 μM at half/log step dilutions. The established anti- malarial dihydroartemisinin (DHA) was used as a positive control and diluted in DMSO from 1 mM to 100 nM at half/log step dilutions. The DMSO dilutions were 500-fold diluted in growth medium to yield a final DMSO concen- tration of 0.2% (v/v), and final compound concentrations from 200 μM to 20 nM (2 μM to 200 pM for DHA). Higher concentrations for the experimental compounds would not have been possible to achieve due to limita- tions in final DMSO concentration (0.1% v/v). Fifty μl of each compound dilution was combined with 50 μl of parasite culture in a black, clear-bottomed or entirely black 96-well plate. The outermost wells were filled with sterile water to prevent evaporation. Plasmodium falciparum asexual blood stages assay with SYBR Green read-out DHA at a concen- tration of 1 μM and 0.1% DMSO (negative control) were used to determine the assay window. The 96-well plates were incubated for 72 hours in a candle jar at 37°C. Read-out was done using the DNA-marker SYBR Green as described previously [22,23]. All assays were con- ducted at least in triplicate. assays, however the final DMSO concentration was 0.2%, which was also adjusted for the positive and negative controls and was not found to cause differences in results obtained with single compounds in assay containing 0.1% DMSO. The pantothenamide phenethyl-Pan was combined with a 10−7 to 10−5 M concentration curve of CXP14.1-060; the other pantothenamides were combined only with a 5 μM concentration of the same vanin inhibi- tor. The pantothenone vanin inhibitor RR8 was addition- ally combined with 20 μM, 6.3 μM, 2.0 μM, and 0.6 μM of pantothenate to explore potential competition for the same molecular target in P. falciparum. All assays were conducted at least in triplicate. Anti-malarial activity of pantothenamides Using P. falciparum cultured in 10% human serum, the low level of anti-malarial activity of the pantothenamides N5-Pan, N7-Pan and phenethyl-Pan was confirmed, as shown in Figure 2A. None of these pantothenamides showed an IC50 below 50 μM. In addition, N9-Pan was tested. N9-Pan is a pantothenamide that has previously been shown to have modest antibacterial activity but had not been investigated for its anti-malarial activity [8]. N9-Pan was found to be clearly more potent than the other pantothenamides, with an IC50 of about 15 μM, in the presence of serum (Figure 2A). To investi- gate the effect of vanin inhibition on the potency of the pantothenamides N5-Pan, N7-Pan and N9-Pan, these Phenethyl-Pan was selected to further investigate the protective effect of the synthetic vanin inhibitor CXP14.1-060 on its anti-malarial activity. Phenethyl-Pan was previously shown to have an IC50 of 20 nM in ‘aged’ medium and is the most potent anti-malarial pantothena- mide known so far [9]. A dose range of this compound was tested in the presence of varying concentrations of CXP14.1-060 (Figure 3). The potency of phenethyl-Pan visibly improved upon addition of increasing concentra- tions of CXP14.1-060, reaching an IC50 of 23 nM (95% CI: 16.5-31.9 nM) upon addition of 10 μM of CXP14.1-060 (Figure 3). In the presence of 10 μM of vanin inhibitor CXP14.1-060, phenethyl-Pan showed a full inhibition of parasite growth, comparable to the efficacy of DHA, which was used as a reference compound and has an IC50 in the low nanomolar range. This experiment shows that small molecule inhibition of pantothenamide hydrolysis will allow the same level of protection found in heat- inactivated medium as shown by the comparable IC50 values [10]. Figure 2 Anti-malarial activity of pantothenamides. Four pantothenamides, phenethyl-Pan, but also N5-Pan, N7-Pan and N9-Pan in asexual blood-stages proliferation assay with SYBR Green read-out, without and in combination with 5 μM of vanin inhibitor CXP14.1-060. A) N5-Pan, N7-Pan, N9-Pan and phenethyl-Pan without vanin inhibitor CXP14.1-060. B) N5-Pan, N7-Pan, N9-Pan and phenethyl-Pan in combination with vanin inhibitor CXP14.1-060. C) DHA with and without 5 μM of vanin inhibitor CXP14.1.-060. Table 1 IC50 values for individual compounds Anti-malarial activity Anti-vanin activity* Compound IC50 (μM) 95% C.I. Statistical analyses Statistical analyses were performed using GraphPad Prism 5. This includes producing inhibition curves, nor- malizing them to percentage of inhibition, calculation of IC50 values, correlation coefficients (Spearman non- parametric correlation), and Schild analyses [24]. The IC50 is the concentration of the compound in question at which the compound reaches 50% inhibition relative to the DHA control curve, and is calculated using the interpolate function in GraphPad Prism 5. When combining two compounds in the same well the conditions were otherwise similar to the single compound Page 4 of 8 Pett et al. Malaria Journal (2015) 14:169 Page 4 of 8 Results were combined with 5 μM of the newly synthesized vanin inhibitor CXP14.1-060, that has nanomolar potency against human vanins, yet no anti-malarial activity up to the high- est concentration tested (100 μM) (Table 1). The potency of N5-Pan and N7-Pan increased upon vanin inhibition, but the effect was not as great as with phenethyl-Pan (Figure 2B). Remarkably, the potency of N9-Pan was not increased by the addition of the vanin inhibitor (Figure 2B). The potency of DHA did not change upon addition of 5 μM of the vanin inhibitor CXP14.1-060 (Figure 2C). Anti-malarial activity of pantothenamides (μM) IC50 (μM) Pantothenol >100 NA NT CJ-15,801 >100 NA 213.60 SN 12,601 6.1 4.5-8.7 17.61 SN 14,621 33.6 7.3-NA 33.70 SN 14,622 7.4 3.5-20.7 31.89 CXP14.1-034 24.1 11.3-NA 1.69 CXP14.1-060 >100 NA 0.039 RR2 2.6 2.1-3.3 2.85 RR6 14.5 5.3-NA 0.04 RR7 7.7 5.1-12.6 0.11 RR8 2.2 1.6-3.1 0.27 N5-Pan 45.1 34.8-60.2 NT N7-Pan 79.0 34.0-NA NT N9-Pan 14.6 11.0-20.1 NT Phenethyl-Pan 98.2 NA NT *Anti-vanin activity tested in human serum. NA: Not Applicable. NT: Not Tested. Table 1 IC50 values for individual compounds Anti-malarial activity of pantothenones inhibition is not the mode of action of the small mol- ecule inhibitors in P. falciparum asexual blood stages. In line with this finding, an assay using the AMC substrate to detect pantetheinase activity, did not detect hydrolytic activity in extracts of purified asexual stages of P. falcip- arum, even after a 19 h incubation period, suggesting that the parasite lacks such enzyme activity (Figure 6). This notion was corroborated by a BLAST search that did not reveal sequences in the P. falciparum genome database with homology to mammalian vanin genes. In order to investigate an alternate mode of action, compe- tition experiments with pantothenate were performed. To this end, a P. falciparum asexual blood stage growth assay was performed with eight concentrations of RR8 combined with eight concentrations of pantothenate. An increase in the IC50 of RR8 was found to occur at addition of increasing concentrations of pantothenate (Figure 7A). Upon performing a Schild analysis on the results of the concentrations, 20 μM, 6.3 μM, 2.0 μM, and 0.6 μM of pantothenate in combination with a dilution curve of RR8, the slope of the line in the Schild Plot was 1.087 ± 0.089, indicating that RR8 and pantothenate are competing for the same target in P. falciparum (Figure 7B). 15 pantothenate derivatives were synthesized and assayed for anti-malarial activity in the in the presence of 10% hu- man serum. The structures of these compounds are pre- sented in Figure 1, and their observed individual biological effects are presented in Table 1. Some of the newly synthe- sized vanin inhibitors were structurally similar to panto- thenate derivatives described in the 1940s, which were shown to have anti-malarial activity in avian malaria models [2]. Some of these reference compounds were re- synthesized, including the pantothenone SN12,601 and the sulphonamides SN 14,621 and SN 14,622. In this study, they were found to have moderate activity, and the effect of their dilution curves on P. falciparum growth are shown in Figure 4. In addition, pantothenol and CJ-15,801, two natural compounds with known weak activity against P. falciparum, were also tested [18,25]. CJ-15,801 was found to be a poor inhibitor of P. falciparum growth (Table 1 and Figure 4) and pantothenol showed no anti-malarial activity up to the high- est concentration tested (100 μM) (Table 1). Anti-malarial activity of pantothenones Out of the newly synthesized pantothenones, the vanin inhibitor RR8 was the most potent of these compounds, having an IC50 value of 2.2 μM (95% CI: 1.6-3.1 µM) (Table 1 and Figure 4). Table 1 IC50 values for individual compounds Figure 2 Anti-malarial activity of pantothenamides. Four pantothenamides, phenethyl-Pan, but also N5-Pan, N7-Pan and N9-Pan in asexual blood-stages proliferation assay with SYBR Green read-out, without and in combination with 5 μM of vanin inhibitor CXP14.1-060. A) N5-Pan, N7-Pan, N9-Pan and phenethyl-Pan without vanin inhibitor CXP14.1-060. B) N5-Pan, N7-Pan, N9-Pan and phenethyl-Pan in combination with vanin inhibitor CXP14.1-060. C) DHA with and without 5 μM of vanin inhibitor CXP14.1.-060. Pett et al. Malaria Journal (2015) 14:169 Page 5 of 8 Figure 3 Anti-malarial activity of phenetyl-Pan in the presence of varying concentrations of vanin inhibitor CXP14.1-060. Combination of anti-malarial pantothenamide phenethyl-Pan [9] with novel vanin inhibitor CXP14.1-060. Filled upright triangles represent phenethyl-Pan, with CXP14.1-060 added at a different concentration for every curve as indicated in figure. At a concentration of 10 μM of CXP14.1-060 phenetyl-Pan has an IC50 of 23 nM (95% CI: 16.5-31.9 nM). Figure 3 Anti-malarial activity of phenetyl-Pan in the presence of varying concentrations of vanin inhibitor CXP14.1-060. Combination of anti-malarial pantothenamide phenethyl-Pan [9] with novel vanin inhibitor CXP14.1-060. Filled upright triangles represent phenethyl-Pan, with CXP14.1-060 added at a different concentration for every curve as indicated in figure. At a concentration of 10 μM of CXP14.1-060 phenetyl-Pan has an IC50 of 23 nM (95% CI: 16.5-31.9 nM). Figure 3 Anti-malarial activity of phenetyl-Pan in the presence of varying concentrations of vanin inhibitor CXP14.1-060. Combination of anti-malarial pantothenamide phenethyl-Pan [9] with novel vanin inhibitor CXP14.1-060. Filled upright triangles represent phenethyl-Pan, with CXP14.1-060 added at a different concentration for every curve as indicated in figure. At a concentration of 10 μM of CXP14.1-060 phenetyl-Pan has an IC50 of 23 nM (95% CI: 16 5-31 9 nM) Figure 3 Anti-malarial activity of phenetyl-Pan in the presence of varying concentrations of vanin inhibitor CXP14.1-060. Combination of anti-malarial pantothenamide phenethyl-Pan [9] with novel vanin inhibitor CXP14.1-060. Filled upright triangles represent phenethyl-Pan, with CXP14.1-060 added at a different concentration for every curve as indicated in figure. At a concentration of 10 μM of CXP14.1-060 phenetyl-Pan has an IC50 of 23 nM (95% CI: 16.5-31.9 nM). Anti-malarial activity of pantothenones Mechanism of anti-malarial activity of pantothenones Mechanism of anti-malarial activity of pantothenones It was considered, that the pantetheinase inhibiting ac- tivity of the new compounds may contribute to anti- malarial activity. Therefore, the IC50 values of in vitro inhibition of asexual blood stages of P. falciparum were plotted against the IC50 values of anti-vanin activity in human serum (Figure 5). There was no correlation be- tween these activities, suggesting that pantetheinase This study underscores the potential of pantothenate de- rivatives for anti-malarial therapy, and demonstrates that the most potent serum-labile anti-malarial pantothena- mide (phenethyl-Pan) can be effectively protected against hydrolysis by serum pantetheinases using the novel vanin inhibitor CXP14.1-060. From a mechanistic point of view, this study indicates that the pantothenone RR8 Pett et al. Malaria Journal (2015) 14:169 Page 6 of 8 Figure 4 Anti-malarial activity of pantothenones and reference compounds. Inhibition curves for all compounds with IC50 values below 100 μM, and CJ-15,801, against asexual blood stages of P. falciparum in a proliferation assay with a SYBR Green read-out. Figure 4 Anti-malarial activity of pantothenones and reference compounds. Inhibition curves for all compounds with IC50 values below 100 μM, and CJ-15,801, against asexual blood stages of P. falciparum in a proliferation assay with a SYBR Green read-out. exerts its anti-malarial effect through competition with pantothenate. by de Villiers et al. showed that structural modifications of pantothenamides can be introduced to confer resistance to plasma-mediated breakdown [10]. These novel com- pounds, although less potent than the original pantothena- mides, are a starting point for further lead optimization studies [10]. Optimization of the potency would be import- ant to maximize the risk-benefit of a novel drug, as side ef- fects may be mediated by low-affinity, off-target effects. In addition, increasing the potency may lead to a lower effect- ive dose in humans, and hence impact the cost of treat- ment. Availability of affordable medicines is an important driver for success in malaria control, and the goal for de- velopment of novel drug therapies is to achieve effective treatment with a total cost of US$1 [26]. In that respect, the molecules described here provide attractive candidates as their chemistry is simple, which ensures a low cost of goods in a manufacturing process. In the 1940s, a number of chemical variations on panto- thenate were synthesized and tested for anti-malarial activ- ity [2]. These included pantothenones and sulphonamides, which were found to be active against avian malaria. Mechanism of anti-malarial activity of pantothenones This study shows that these compounds are also active against the human parasite P. falciparum. Although the new vanin inhibitor CXP14.1-060 effect- ively protected phenethyl-Pan, such a combination of drugs would be undesirable from a drug development per- spective. Clearly, the potency and/or stability of pantothen- ate derivatives needs to be improved before they can enter a drug development programme as therapeutic agents for human malaria infection. Nevertheless, the recent discov- ery of phenethyl-Pan with an IC50 of 20 nM is encouraging [9]. Although this compound is unstable in plasma, it illus- trates that it is realistic and feasible to aim for pantothenate derivatives active in the low nanomolar range. The study Many of the marketed anti-malarials and compounds in the clinical development portfolio originate from whole Pett et al. Malaria Journal (2015) 14:169 Page 7 of 8 Page 7 of 8 Figure 5 Lack of correlation between anti-malarial activity and anti-vanin activity. The IC50 for inhibition of vanin activity in human serum was plotted against the anti-malarial IC50 in asexual blood stages proliferation assay of P. falciparum with SYBR Green read-out. No signifi- cant correlation was observed. (Spearman rho=0.06079 (p=0.8651)). cell phenotypic screening efforts and exert their actions by inhibiting multiple targets or pathways of the parasite. Al- though such a polypharmacological profile may be im- portant to their efficacy, it is an undesirable feature in a rational medicinal chemistry approach. The exact target of the anti-malarial pantothenate derivatives has not been identified unequivocally but is it likely that they exert their effects by affecting targets dependent on pantothenate. In Figure 5 Lack of correlation between anti-malarial activity and anti-vanin activity. The IC50 for inhibition of vanin activity in human serum was plotted against the anti-malarial IC50 in asexual blood stages proliferation assay of P. falciparum with SYBR Green read-out. No signifi- cant correlation was observed. (Spearman rho=0.06079 (p=0.8651)). Figure 7 Pantothenate competes with RR8. Competition assay between RR8 and pantothenate (PA), results of Schild analysis. A) Shifting IC50: Combined data points from three experiments. RR8 with 20.0 μM PA, RR8 with 6.3 μM PA, RR8 with 2.0 μM PA, RR8 with 0.6 μM PA and RR8 alone. B) Schild plot: Combined data points from three experiments. A slope of 1 is indicative of a competitive antagonistic relationship between RR8 and PA. Figure 7 Pantothenate competes with RR8. Competition assay between RR8 and pantothenate (PA), results of Schild analysis. Mechanism of anti-malarial activity of pantothenones A) Shifting IC50: Combined data points from three experiments. RR8 with 20.0 μM PA, RR8 with 6.3 μM PA, RR8 with 2.0 μM PA, RR8 with 0.6 μM PA and RR8 alone. B) Schild plot: Combined data points from three experiments. A slope of 1 is indicative of a competitive antagonistic relationship between RR8 and PA. Figure 5 Lack of correlation between anti-malarial activity and anti-vanin activity. The IC50 for inhibition of vanin activity in human serum was plotted against the anti-malarial IC50 in asexual blood stages proliferation assay of P. falciparum with SYBR Green read-out. No signifi- cant correlation was observed. (Spearman rho=0.06079 (p=0.8651)). cell phenotypic screening efforts and exert their actions by inhibiting multiple targets or pathways of the parasite. Al- though such a polypharmacological profile may be im- portant to their efficacy, it is an undesirable feature in a rational medicinal chemistry approach. The exact target of the anti-malarial pantothenate derivatives has not been identified unequivocally but is it likely that they exert their effects by affecting targets dependent on pantothenate. In theory, the observed effects could still be mediated by ef- fects on red blood cell biology (e.g., red blood cell panto- thenate kinases (PANK)) rather than directly on the parasite. However, the recent discovery of a parasite- specific pantothenate transporter [27] leaves very little Figure 7 Pantothenate competes with RR8. Competition assay between RR8 and pantothenate (PA), results of Schild analysis. A) Shifting IC50: Combined data points from three experiments. RR8 with 20.0 μM PA, RR8 with 6.3 μM PA, RR8 with 2.0 μM PA, RR8 with 0.6 μM PA and RR8 alone. B) Schild plot: Combined data points from three experiments. A slope of 1 is indicative of a competitive antagonistic relationship between RR8 and PA. doubt that the parasite itself is the target of interfering with pantothenate dependent pathways. Future drug de- velopment efforts would benefit from information on the molecular targets of pantothenamides, which would in- clude both biosynthetic pathways (CoA synthesis, lipid synthesis, energy metabolism) and pantothenate transport systems. Structural information, which is available for mammalian and bacterial PANK, could guide medicinal chemistry strategies to achieve specific inhibition of the parasite enzyme and reduce side effects on the host. Figure 6 Lack of vanin activity in P. falciparum lysates. Vanin activity assays were performed using a fluorescent aminomethylcoumarine (AMC) substrate on lysates of P. Mechanism of anti-malarial activity of pantothenones falciparum cultures (on the right in white), and as a positive control human serum (on the left in black). The figure shows mean relative fluorescence units and standard deviations from measurements of four independent P. falciparum cultures. Values were corrected for background fluorescence signals measured in negative control (PBS) samples. Conclusions Pantothenamides with anti-malarial activity can be pro- tected from breakdown by ubiquitous pantetheinases of the vanin family with small molecule pantothenone vanin inhibitors. Some of these pantothenones exhibit anti- malarial activity in their own right. Compound series such as the one tested in this publication should be studied fur- ther for use as lead compounds for anti-malarial treatment. Figure 6 Lack of vanin activity in P. falciparum lysates. Vanin activity assays were performed using a fluorescent aminomethylcoumarine (AMC) substrate on lysates of P. falciparum cultures (on the right in white), and as a positive control human serum (on the left in black). The figure shows mean relative fluorescence units and standard deviations from measurements of four independent P. falciparum cultures. Values were corrected for background fluorescence signals measured in negative control (PBS) samples. References A survey of antimalarial drugs. Ann Arbor, Michigan: Edwards, J.W.; 1946. 27. Augagneur Y, Jaubert L, Schiavoni M, Pachikara N, Garg A, Usmani-Brown S, et al. Identification and functional analysis of the primary pantothenate transporter, PfPAT, of the human malaria parasite Plasmodium falciparum. 3. Spry C, Kirk K, Saliba KJ. Coenzyme a biosynthesis: an antimicrobial drug target. FEMS Microbiol Rev. 2008;32:56–106. J Biol Chem. 2013;288:20558–67. J Biol Chem. 2013;288:20558–67. 4. Trager W. Coenzyme a and the antimalarial action in vitro of antipantothenate against plasmodium lophurae, P. Coatneyi and P. Falciparum. Trans N Y Acad Sci. 1966;28:1094–108. 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 Abbreviations 10. de Villiers M. Structural modification of pantothenamides counteracts degradation by pantetheinase and improves antiplasmodial activity. ACS Med Chem Lett. 2013;4:784–9. CoA: Coenzyme A; DHA: Dihydroartemisinin; DMSO: Dimethylsulphoxide; PA: Pantothenic acid, pantothenate, vitamin B5; PanK: Pantothenate kinase (Plasmodium falciparum); PANK: Pantothenate kinase (Homo sapiens). 11. Jansen PA, Zeeuwen PL, Schalkwijk J, Rutjes FP, Ritzen B, Hermkens PH. Pantothenic acid derivatives and their use in the treatment of microbial infections. Patent Application number EP11725211, Publication number WO2011152720. 2011. Authors’ contributions HEP carried out the majority of P. falciparum proliferation assays and the majority of the writing of the manuscript. PAMJ carried out the vanin inhibition assays and wrote some of the manuscript. PHHH is responsible for design of active compounds and for writing of the supplemental material of the manuscript. PNMB and CABS synthesized active compounds. RHB was responsible for conception of chemical synthesis routes. WG and MV-B worked on early experimentation with compounds. KMJK performed the experiment that resulted in Figure 2C. FPJTR is responsible for design of active compounds. KD supervised experimental work of HEP and wrote a substantial part of the discussion section. RS approved work being done during experimental process. JS initiated the project, supervised experimental work of HEP and PAMJ, and wrote a substantial part of this publication. All authors have read and approved final version of the manuscript. 15. Strauss E, Begley TP. The antibiotic activity of N-pentylpantothenamide results from its conversion to ethyldethia-coenzyme a, a coenzyme a antimetabolite. J Biol Chem. 2002;277:48205–9. 16. Zhang YM, Frank MW, Virga KG, Lee RE, Rock CO, Jackowski S. Acyl carrier protein is a cellular target for the antibacterial action of the pantothenamide class of pantothenate antimetabolites. J Biol Chem. 2004;279:50969–75. 17. van der Westhuyzen R, Hammons JC, Meier JL, Dahesh S, Moolman WJ, Pelly SC, et al. The antibiotic CJ-15,801 is an antimetabolite that hijacks and then inhibits CoA biosynthesis. Chem Biol. 2012;19:559–71. 18. Saliba KJ, Kirk K. CJ-15,801, a fungal natural product, inhibits the intraerythrocytic stage of Plasmodium falciparum in vitro via an effect on pantothenic acid utilisation. Mol Biochem Parasitol. 2005;141:129–31. 19. Winterbottom R, Clapp JW, Miller WH, English JP, Roblin Jr RO. Studies in chemotherapy; amides of pantoyltaurine. J Am Chem Soc. 1947;69:1393–401. Acknowledgements Thi k 20. Ponnudurai T, Lensen AH, Meis JF, Meuwissen JH. Synchronization of Plasmodium falciparum gametocytes using an automated suspension culture system. Parasitology. 1986;93(Pt 2):263–74. This work was supported by the Radboud University Nijmegen Medical Center and by a grant of The Netherlands Genomics Initiative, Grant no. 93611013. Also, we extend our gratitude to Martijn Timmerman for expert technical advice in design and construction of drug assays in 96-well plates. 21. Ruan BH, Cole DC, Wu P, Quazi A, Page K, Wright JF, et al. A fluorescent assay suitable for inhibitor screening and vanin tissue quantification. Anal Biochem. 2010;399:284–92. Additional file Additional file 1: Supplemental Materials and Methods. The supplemental materials and methods explain how the compounds CJ-15,801 and CXP14.1-060 were synthesized. Page 8 of 8 Page 8 of 8 Pett et al. Malaria Journal (2015) 14:169 Received: 1 December 2014 Accepted: 4 April 2015 25. Saliba KJ, Ferru I, Kirk K. Provitamin B5 (pantothenol) inhibits growth of the intraerythrocytic malaria parasite. Antimicrob Agents Chemother. 2005;49:632–7. 26. Burrows JN, van Huijsduijnen RH, Mohrle JJ, Oeuvray C, Wells TN. Designing the next generation of medicines for malaria control and eradication. Malar J. 2013;12:187. Author details 1 22. Smilkstein M, Sriwilaijaroen N, Kelly JX, Wilairat P, Riscoe M. Simple and inexpensive fluorescence-based technique for high-throughput antimalarial drug screening. Antimicrob Agents Chemother. 2004;48:1803–6. 1Department of Medical Microbiology, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands. 2Department of Dermatology and Radboud Institute for Molecular Life Sciences, Radboud University Nijmegen Medical Center, Nijmegen, The Netherlands. 3Radboud University Nijmegen, Institute for Molecules and Materials, Nijmegen, The Netherlands. 4Chiralix B V, Nijmegen, The Netherlands. 5TropIQ Health Sciences, Nijmegen, The Netherlands. 6Pansynt B V, Nijmegen, The Netherlands. 23. Bennett TN, Paguio M, Gligorijevic B, Seudieu C, Kosar AD, Davidson E, et al. Novel, rapid, and inexpensive cell-based quantification of antimalarial drug efficacy. Antimicrob Agents Chemother. 2004;48:1807–10. 23. Bennett TN, Paguio M, Gligorijevic B, Seudieu C, Kosar AD, Davidson E, et al. Novel, rapid, and inexpensive cell-based quantification of antimalarial drug efficacy. Antimicrob Agents Chemother. 2004;48:1807–10. V, Nijmegen, The Netherlands. 5TropIQ Health Sciences, Nijmegen, The Netherlands. 6Pansynt B V, Nijmegen, The Netherlands. 24. Wyllie DJ, Chen PE. Taking the time to study competitive antagonism. Br J Pharmacol. 2007;150:541–51. 24. Wyllie DJ, Chen PE. Taking the time to study competitive antagonism. Br J Pharmacol. 2007;150:541–51. Received: 1 December 2014 Accepted: 4 April 2015 Competing interests KD and RS hold shares in TropIQ Health Sciences, a spin-off company of the RadboudUMC that aims to develop anti-malarial drugs. JS, FPJTR and PHHH hold shares in Pansynt, a spin-off company of the RadboudUMC that aims to develop pantothenate-based drugs for infectious diseases. Some of the compounds described in the manuscript are covered by a patent application (PCT/NL2011/050385) filed by the RadboudUMC (inventors: PAMJ, JS, RS, PHHH, FPJTR). KD and RS hold shares in TropIQ Health Sciences, a spin-off company of the RadboudUMC that aims to develop anti-malarial drugs. JS, FPJTR and PHHH hold shares in Pansynt, a spin-off company of the RadboudUMC that aims to develop pantothenate-based drugs for infectious diseases. Some of the 12. Jansen PA, Schalkwijk J, Rutjes FP, Sauerwein R, Hermkens PH. Derivatives of pantothenic acid and their use for the treatment of malaria. Patent application number EP11725211, publication number WO2011152721. 2011. 13. Jansen PA, van Diepen JA, Ritzen B, Zeeuwen PL, Cacciatore I, Cornacchia C, et al. Discovery of small molecule vanin inhibitors: new tools to study metabolism and disease. ACS Chem Biol. 2013;8:530–4. compounds described in the manuscript are covered by a patent application (PCT/NL2011/050385) filed by the RadboudUMC (inventors: PAMJ, JS, RS, PHHH, FPJTR). 14. Jansen PA, Hermkens PH, Zeeuwen PL, Botman PN, Blaauw RH, Burghout P, et al. Combination of pantothenamides with vanin inhibitors as a novel antibiotic strategy against Gram-positive bacteria. Antimicrob Agents Chemother. 2013;57:4794–800. References References 1. WHO. World malaria report 2013. Geneva: World Health Organization; 2013. 2. Wiselogle FY. A survey of antimalarial drugs. Ann Arbor, Michigan: Edwards, J.W.; 1946. 3. Spry C, Kirk K, Saliba KJ. Coenzyme a biosynthesis: an antimicrobial drug target. FEMS Microbiol Rev. 2008;32:56–106. 4. Trager W. Coenzyme a and the antimalarial action in vitro of antipantothenate against plasmodium lophurae, P. Coatneyi and P. Falciparum. Trans N Y Acad Sci. 1966;28:1094–108. 5. Trager W. Further studies on the effects of antipantothenates on malaria parasites (Plasmodium coatneyi and P. falciparum) in vitro. J Protozool. 1971;18:232–9. 6. Trager W, Jensen JB. Human malaria parasites in continuous culture. Science. 1976;193:673–5. 7. Divo AA, Geary TG, Davis NL, Jensen JB. Nutritional requirements of Plasmodium falciparum in culture. I. Exogenously supplied dialyzable components necessary for continuous growth. J Protozool. 1985;32:59–64. 8. Clifton G, Bryant SR, Skinner CG. N'-(substituted) pantothenamides, antimetabolites of pantothenic acid. Arch Biochem Biophys. 1970;137:523–8. 9. Spry C, Macuamule C, Lin Z, Virga KG, Lee RE, Strauss E, et al. Pantothenamides are potent, on-target inhibitors of Plasmodium falciparum growth when serum pantetheinase is inactivated. PLoS One. 2013;8, e54974. 1. WHO. World malaria report 2013. Geneva: World Health Organization; 2013. 1. WHO. World malaria report 2013. Geneva: World Health Organization; 2013. 2. Wiselogle FY. A survey of antimalarial drugs. Ann Arbor, Michigan: Edwards, J.W.; 1946. 3. Spry C, Kirk K, Saliba KJ. Coenzyme a biosynthesis: an antimicrobial drug target. FEMS Microbiol Rev. 2008;32:56–106. 4. Trager W. Coenzyme a and the antimalarial action in vitro of antipantothenate against plasmodium lophurae, P. Coatneyi and P. Falciparum. Trans N Y Acad Sci. 1966;28:1094–108. 5. Trager W. Further studies on the effects of antipantothenates on malaria parasites (Plasmodium coatneyi and P. falciparum) in vitro. J Protozool. 1971;18:232–9. 6. Trager W, Jensen JB. Human malaria parasites in continuous culture. Science. 1976;193:673–5. 7. Divo AA, Geary TG, Davis NL, Jensen JB. Nutritional requirements of Plasmodium falciparum in culture. I. Exogenously supplied dialyzable components necessary for continuous growth. J Protozool. 1985;32:59–64. 8. Clifton G, Bryant SR, Skinner CG. N'-(substituted) pantothenamides, antimetabolites of pantothenic acid. Arch Biochem Biophys. 1970;137:523–8. 9. Spry C, Macuamule C, Lin Z, Virga KG, Lee RE, Strauss E, et al. Pantothenamides are potent, on-target inhibitors of Plasmodium falciparum growth when serum pantetheinase is inactivated. PLoS One. 2013;8, e54974. 2. Wiselogle FY. A survey of antimalarial drugs. Ann Arbor, Michigan: Edwards J.W.; 1946. 2. Wiselogle FY. Submit your next manuscript to BioMed Central and take full advantage of: 5. Trager W. Further studies on the effects of antipantothenates on malaria parasites (Plasmodium coatneyi and P. falciparum) in vitro. J Protozool. 1971;18:232–9. g • 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 • Convenient online submission 6. Trager W, Jensen JB. Human malaria parasites in continuous culture. Science. 1976;193:673–5. • Thorough peer review 7. Divo AA, Geary TG, Davis NL, Jensen JB. Nutritional requirements of Plasmodium falciparum in culture. I. Exogenously supplied dialyzable components necessary for continuous growth. J Protozool. 1985;32:59–64. • No space constraints or color ﬁgure charges 8. Clifton G, Bryant SR, Skinner CG. N'-(substituted) pantothenamides, antimetabolites of pantothenic acid. Arch Biochem Biophys. 1970;137:523–8. 9. Spry C, Macuamule C, Lin Z, Virga KG, Lee RE, Strauss E, et al. Pantothenamides are potent, on-target inhibitors of Plasmodium falciparum growth when serum pantetheinase is inactivated. PLoS One. 2013;8, e54974. Submit your manuscript at www.biomedcentral.com/submit
https://openalex.org/W2285921512	https://www.redalyc.org/pdf/5798/579866787010.pdf	English	null	Present in absentia: Immigrant letters and requests for family reunification	História Unisinos	2,015	cc-by	8,475	Presença na ausência: cartas na imigração e cartas de chamada Maria Izilda Santos Matos1 mismatos@pucsp.br Oswaldo Mario Serra Truzzi2 truzzi@ufscar.br Abstract: Immigrant letters are a major source to learn not only about the circumstances and challenges involved in the migratory experience, but especially the mentality of historical actors involved in it. This article seeks first to emphasize the importance of the topic; it then discusses the expansion of studies on the migrants’ letters, to finish by illustrating the aspirations and hesitations involving family reunification, as well as the control and authority exerted at a distance through a particular type of letters – the so-called requests for family reunification – written by Portuguese immigrants in São Paulo in the early 20th century. Keywords: immigrants’ letters, Portuguese immigration, gender and family relations. Resumo: Cartas de imigrantes constituem uma fonte privilegiada para se apreender não apenas as circunstâncias e desafios envolvendo a experiência migratória, mas, sobretudo, a mentalidade dos agentes históricos nela envolvidos. O presente artigo procura em primeiro lugar ressaltar a relevância do tema; em seguida, discorre sobre a expansão dos estudos sobre os escritos dos e/imigrantes, para concluir ilustrando os anseios e hesitações envol- vendo a reunificação familiar, bem como o controle e a autoridade exercidos a distância por meio de um tipo particular de correspondência – as cartas de chamada – escrita por imigrantes portugueses em São Paulo no início do século XX. Palavras-chave: cartas de chamada, imigração portuguesa, cultura escrita popular, relações de gênero, relações familiares. 1 Pontiﬁ cal Catholic University of São Paulo. 2 Federal University of São Carlos. 1 Pontiﬁ cal Catholic University of São Paulo. Santos Matos, Maria Izilda; Serra Truzzi, Oswaldo Mario Present in absentia: Immigrant letters and requests for family reunifi cation História Unisinos, vol. 19, núm. 3, septiembre-diciembre, 2015, pp. 348-357 Universidade do Vale do Rio dos Sinos São Leopoldo, Brasil Santos Matos, Maria Izilda; Serra Truzzi, Oswaldo Mario Present in absentia: Immigrant letters and requests for family reunifi cation História Unisinos, vol. 19, núm. 3, septiembre-diciembre, 2015, pp. 348-357 Universidade do Vale do Rio dos Sinos São Leopoldo, Brasil Available in: https://www.redalyc.org/articulo.oa?id=579866787010 How to cite Complete issue More information about this article Journal's homepage in redalyc.org Scientific Information System Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Non-profit academic project, developed under the open access initiative How to cite Complete issue More information about this article Journal's homepage in redalyc.org Scientific Information System Network of Scientific Journals from Latin America, the Caribbean, Spain and Portugal Non-profit academic project, developed under the open access initiative História Unisinos 19(3):348-357, Setembro/Dezembro 2015 2015 Unisinos – doi: 10.4013/htu.2015.193.06.e This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits reproduction, adaptation, and distribution provided the original author and source are credited. Letters and migrations: a universe to be explored cities of Porto and Braga, in Portugal, and in the São Paulo Immigrant Museum (deposited in the Public Archive of the state of São Paulo). In the case of the district archives, the letters are in the files of proceedings to obtain passports, since it was mandatory to present them for this purpose, because, ac- cording to Portuguese law, married women and underage children could not emigrate without the permission of their husbands and fathers. Travel abroad was restricted, because the permanence of conjugal ties and keeping the family in Portugal acted as a support, increasing the possibility that the person would return and making it easier to continue sending money, which became essential to the family and to the country’s economy. On the basis of decree no. 7427 of April 30, 1921, the Portuguese gov- ernment induced a change in the usual practice of writing family letters with new formalities. The letters were re- placed by a standard and objective type of consular printed form, and no longer provided the previous references and information that are such a delightful historical source. The private writings that comprehend the so- called “ordinary writings” cover a variety of texts, including the writing of the popular classes (Chartier, 1991). Little attention has been given to them by historiography, prob- ably due to preconceived views that the popular classes (due to their low level of literacy) were not able to pro- duce significant records (Molinari, 1999). Initially these sources appeared in scientific investigations as alternative and/or complementary sources (Castillo Gómez, 2001); however, their use became disseminated when an opening was enabled by the emergence of the “other histories” (Matos, 2002), which expanded interest regarding the varied experiences of the past. These recent perspectives generated the need for new corpuses of documents and gave value to the “ordinary writings” that enabled fruitful discoveries regarding the histories of the popular classes, including their written culture. On the other hand, in the case of the small number of letters of reunification found in the files of the Immigrants’ Hostel (“Hospedaria dos Imigrantes”), these were the documental evidence that the immigrant had to present in the port of Santos to the Immigration Inspector, and they were attached to the landing lists filed at the Hostel. Letters and migrations: a universe to be explored From 1911 onwards, this document became mandatory for people over the age of 60 and who were not able to work, in an attempt by the Brazilian government to ensure family support for this more vulnerable group. The popular classes produced their own records, but often they have not been preserved in public archives, rather they were conserved over the times and kept secretly in attics and trunks, in a more affective sense, and aiming to preserve the family or group memory. These sources spell out multiple, exceptional experiences, personal ad- ventures, references to daily private life, and subjective issues and sensitivities. Since the middle of the Modern Era, in some countries in Europe, the popular classes exercised the ability to write because they had to deal with the bu- reaucratic demands of the modern state, which coincided with a certain dissemination of the literacy and schooling processes. Although the relation between the faster litera- cy/schooling processes and the increase of the number of popular texts is recognized, certainly the major factor for this expansion was the need for communication gener- ated by wars (especially from World War I onwards) and displacements (Blass, 2004). In the first case, these letters were generally ad- dressed by the husband (who had left before) to his wife or some other family member.3 Although access to the correspondence is only partial, since it covers only one di- rection (from the husband to the wife, and not vice-versa), even unilaterally it is often possible to infer the nature of the dialogue between them. The letters not only documented the worlds of origin and of destination; they were also an attempt to question silences, overcome distances, perpetuate affec- tions, reinforce ties and combat the feeling of homesick- ness and missing dear ones, reconfiguring relationships that had become vulnerable through the long distance and time of separation. The mobility required from the emigrants/immi- grants the exercise of reading/writing, and a broad textual spectrum became part of this universe. A set of handbooks and guidebooks, booklets, leaflets, periodicals, magazines, information regarding travel and the conditions in the countries to which they were going, directions about the bureaucratic procedures (passport and authorizations), and several other writings were circulated. Introduction Letters not only show processes of displacement and separation, they are also the product of both. These letters often became documents used in the immigration process to prove ties and make it possible to leave and/or to enter the countries of origin and of destination. In our case we worked with a corpus of unpublished documents, constituted by the so-called letters of reunification between Portuguese migrants and their families, found in district archives of the This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 3.0), which permits reproduction, adaptation, and distribution id d th i i l th d dit d Present in absentia: Immigrant letters and requests for family reuniﬁ cation 3 The same was observed by Seyferth (2005, p. 23) regarding letters from the Polish. Letters and migrations: a universe to be explored The present article first discusses the relevance of the topic, then the expansion of studies on the writings of the emigrants/immigrants, and finally illustrates the as- pirations and hesitations involved in family reunification, as well as the control and authority exerted at a distance by some letters of reunification analyzed here. 349 Likewise, an outstanding role is assigned to the displacements as elements stimulating the dissemination História Unisinos História Unisinos Maria Izilda Santos Matos, Oswaldo Mario Serra Truzzi and consolidation of reading/writing among the masses of emigrants/immigrants with a low level of literacy, who were challenged to produce documents invoking this process. The distance among family members was the main reason that led the popular classes to deal with the pen, moved by the desire to preserve domestic and family connections, with a need to maintain and/or construct chains of ink and links of paper, which became a practice, need, and moral obligation. problems, survival tactics, social ascension, difficulties in everyday life in the countryside and in the city, affective relations, subjectivities and sensitivities (suffering and anxieties, joys and frustrations). All these essential pieces of individual experiences shed light on the histories of emigration/immigration by explaining aspects that are difficult or almost impossible to perceive in other docu- mental corpuses (Caffarena, 2012, p. 19). While at the beginning of the 1800s letters be- tween Portugal and Brazil took approximately 60 days, this time was markedly reduced by the expansion of railways, steamships and mailbags. In the second half of the 19th century missives could arrive at their destination in up to 20 days (Rodrigues, 2013, p. 83). Through these missives, people tried to overcome the separation, control things at a distance, combat silences, perpetuate affections, reinforce family, kinship and friendship ties, be present in absentia, take up responsibilities and get around the feeling of homesickness and missing dear ones. The study of migrant writings enables us to under- stand the variety of uses and functions of what was written at this historical period, depending on the places where it was written and the reasons why, and, because of all this, on the material differences presented by the different documents. Looking at the protagonists of this phenomenon and using the sources that they themselves produced leads the historian to see emigration from a new, fundamental viewpoint. Letters and migrations: a universe to be explored In brief, to understand that these documents not only tell us about the experience of common men and women, but are also its product and direct consequence (Blass, 2004, p. 97). The letters not only portrayed the process of emigration/immigration and separation, they were also their product. The act of writing letters was considered privileged, free, secret, intimate, a true telling of individual experiences. However, when written at request, the letters were often read and reread in public, and also became col- lective experiences in the form of texts that were initially private and domestic and ended up being shared. “Al- though they were private and intimate, the letters took on public importance as a way of conveying messages, which reinforced the connection with the community of origin, thus enabling the transfer of other groups of immigrants” (Vendrame, 2010, p. 70). The letters were the means by which one could question silences, overcome distances, perpetuate affections and reinforce family, kinship and friendship ties. 4 Newspapers of the time published letters as advertisement for emigration or to denounce the conditions of the emigrants in the country. Meanings of the dialogues: family reunification, authority and control, said and not said In Italy, in the Archivio Ligure della Scritura Popolare, at the University of Genoa, there are outstanding studies led by Antonio Gibelli and Fabio Caffarena and, for the Italian emigrants to Brazil, studies by Federico Croci (Gibelli, 1989, 2002; Gibelli and Caffarena, 2001; Croci, 2008). Focusing on immigration to the south of the coun- try, Vendrame (2010) analyzed a few letters sent by Italian immigrants who settled in the colonial regions of Caxias do Sul and Santa Maria to their relatives in Italy, while Seyferth (2005) commented on the letters from the Polish compiled by Kula (1977) and Wachowicz (1981), as well as the letters from Germans compiled by Vendrame (2010). Vendrame (2010, p. 70) stresses the importance of the let- ters in attracting new immigrants, relatives to them or not, since these were the means par excellence of “relationship between those who stayed and those who left, [enabling] the understanding of choices, habits, beliefs and kinship ties of the families of the immigrants who established themselves in colonial settlements”. The same point is emphasized by Seyferth (2005, p. 47), who mentions that the letters and narratives “call attention to the maintenance of family ties with those who remained in the society of origin and the effort to bring the closest relatives to Brazil”. Although epistolary writing is old, it expanded with the increase in communications and the intensifi- cation of mobility, as already mentioned. Displacements, made easier by the development of transportation by trains and ships, became mass “phenomena”, and this historical migratory experience increased the distances between people, creating a need to communicate and efforts to approach. Letter-writing was disseminated, as already mentioned, including the popular classes, in a challenge to a less literate mass that, with great effort, tried to maintain ties. Thus new experiences of epistolary practice were disseminated and writing was “democratized”. Departures increased the feeling of separation, and even before the sight of the port of departure dissipated on the horizon, writing began in an effort to maintain ties; those who remained behind waited anxiously for the letters and complained of the lack of answers, showing the wish for dialogue. The moment of separation, the physical distance from home, the feeling of being far away when one has already arrived in the country of destination, trigger the decisive impulse to take up pen and paper and face the act of writing. Present in absentia: Immigrant letters and requests for family reuniﬁ cation families from whom they themselves descended. These authors discovered in the letters exchanged a possibility of capturing the experience of immigration through the very words of the people involved, which gave a special color to the work usually developed on the basis of other sources. In this way, collections of letters by immigrants were published by historians in the United States5, Europe, Canada and Australia throughout the 20th century due to their interest in studying social and political integration processes, changes in the family structure, involvements in revolutions and wars, and, in general, changes in the popular culture (Elliott et al., 2006). emigrated to Latin America, while Franco Ramella and Samuel Baily analyzed the correspondence exchanged between the members of the Sola family between Pied- monte and Buenos Aires (Franzina, 1979; Baily and Ramella, 1988). On the other hand, in Portugal the investigations by Henrique Rodrigues deserve to be mentioned. They concentrated on the letters of people who emigrated from Viana do Castelo (Rodrigues, 2013). The attempt to illustrate the life of emigrants and immigrants of French-language origin gave rise to the organization of the collection Envoyer et recevoir. Lettres et correspondances dans les diasporas francophones, published in 2006 (Frenette et al., 2006). Contemporaneously, there are outstanding studies developed at the University of Alcalá (Spain), at SIECE (Interdisciplinary Seminar on Studies of Written Culture), and it is worth mentioning the works by Veronica Blass and Laura Martínez Martín, who prioritize the letters of Spanish emigrants/immigrants to Latin America. The Center for Studies of the Galego Immigration at the University of Santiago de Compostela is also worth mentioning. There the research by Xosé M. Nuñez Seixas and Raúl Soutelo Vázquez (Nuñez Seixas and Soutelo Vázquez, 2005; Soutelo Vázquez, 2001, 2003) was developed. In comparison to such diversity, and despite their potential, the historiographic recognition of migrants’ letters in Brazil is still limited, especially compared to the research done in Italy, Great Britain, France, Spain and in receiving countries such as the USA and Argentina. 5 In the United States a very signiﬁ cant tradition was established regarding studies that try to focus on the experience of migration through personal letters. The Immigration History Research Center itself, located in Minnesota, has a great number of letter collections that have been used as an empirical base to develop the topic. Expansion and studies on the emigrants/immigrants’ writings From the end of the 19th century onwards, it is possible to trace studies that focused on the emigrants/ immigrants’ letters: in 1892, for instance, the Italian Commissioner for Emigration, Luigi Bodio, used mis- sives (700 letters) from Brazil to evaluate the conditions of the Italians who were coming to this country (Bodio, 1894).4 Likewise, in 1913 the physician and writer Filip- po Lussana highlighted aspects of the epistolography of emigration/immigration by analyzing this material (Lussana, 1913). Thanks to the circulation of the missives it is pos- sible to observe the displacement from the kaleidoscopic viewpoint of its protagonists, enabling an understanding of the social, cultural and identity changes. Besides the ties and the sociability that were created in the process (Chartier, 1991), the practices of writing, even derived from a popular and sometimes marginal culture, gained expression in their own literary form – epistolary literature, with specific styles, rhetoric and conventions. Considered a classic of the correspondence from emigrants and immigrants, William Isaac Thomas and Florian Znaniecki (1918-20) recompiled five volumes of letters in a work that, in its introduction, discusses a methodological approach to the use of such sources that is still constantly referenced today (Thomas and Znaniecki, 1958). In parallel with that, although with different concerns, a generation of North American historians trained in the 1920s and 1930s, all of them the children of immigrants, began to challenge the kind of narrative prevailing in North American history, intending to in- corporate the stories of common people and ordinary As a real “documentary treasure” of popular writing and memory (because of their volume and frequency), the letters of the emigrants/immigrants became of interest to scholars, enabling researchers to penetrate an invisible area that allows observing new projects, successes, financial 350 Vol. 19 Nº 3 - setembro/dezembro de 2015 Present in absentia: Immigrant letters and requests for family reuniﬁ cation Meanings of the dialogues: family reunification, authority and control, said and not said Letters exchanged between migrants and their families expressed different views about the household and everyday practices, processes of change, negotiation of identities, adaptation to the societies to which they came (there are letters announcing the intention of getting mar- ried or becoming a citizen), or the immigrants’ difficulties to integrate, how the “others” were constituted and what their attitude was towards them (Elliott et al., 2006, p. 9). In the first months, people were quick to answer, long texts were written, with plenty of details and expla- nations, narrating the first impressions; the topics were departure, the description of the voyage, arrival and first impressions, besides work, business and new relationships, and also requests for news of events in the family and in the village. In time, the regularity and length diminished as more routine prevailed in everyday life, but the letters were normally present on celebratory occasions or when major changes occurred. Even so, writing became more occasional, with a shorter text and less information. p It was relatively usual for the men to emigrate first, thus attenuating the impacts of change, and, in a preven- tive action against possible misfortunes, leave calling the remainder of the family to a more favorable time, when they were already settled and were in a better financial situation. Although most of the letters were written by men, there are constant mentions of the women, above all as addressees, in them. While the letters can indeed serve as a guide to recover the presence of women in the emigration/immigration process, on the other hand, between the lines, one can observe echoes of the silenced female voices, the resistance to departures, the demand for news or remittances. Although usually only one copy of each letter is available, one can observe that the rate of mailing was very variable. On the basis of the texts, in some cases one can see a regular, constant exchange, also delays in answering, complaints about the lack of news, months-long silences, or even an entire year, and requests to write more often. Systematizing the dates, one can observe that the months with the highest incidence were April, December and October, the two first because of Easter and Christmas celebrations. Meanings of the dialogues: family reunification, authority and control, said and not said The more intense exchange in October was due to the agricultural production cycle in Europe and denotes an interest in keeping up with agricultural work and, above all, the results of harvests in general and the vintage in particular (Rodrigues, 2013, p. 79). Men, through the missives, tried to make them- selves present in absentia, maintain control at a distance, request news, advice and guidance from family members, particularly from their wives (whether they should or not move to the city, change jobs, enlarge the business, etc.). Likewise it was common for letters to be sent for the purpose of obtaining a wife, since for several reasons gender imbalances or even ethnic and racial animosities at the place of arrival could lead men who had already emigrated to seek future wives in their home countries, often expressing the desire to preserve the original culture or honoring prior commitments. Phenomena such as weddings on the wharf, immediately after landing from the ship, were only possible by making prior arrangements through letters (Sinke, 2006). Time sometimes even caused estrangement and surprises of a physical nature, as in the correspondence of July 1914 from Antonio Teixeira Cardozo to Maria José, 47 years old, born in Porto: “The portraits you sent me I found them very good, and I see that the little girls are plump and grown, they hardly look the same” (Arquivo Distrital do Porto, Requisição de Passaportes [ADP-RP], process no. 1138). Letters become a representation of the absent authority that, despite the distance, reproduces family relations and hierarchies, interfering directly or indirectly in everyday life, in business, in problems with land and livestock, with a discourse on how to treat, what, for whom when and for how much to sell, and above all, since they were letters of reunification, how, when and why to come. Despite all efforts to remain close, gradually cultural differences appeared, due to the life-changing experiences such as crossing the ocean, the arrival in an unknown territory, facing challenges and deprivations. Meanings of the dialogues: family reunification, authority and control, said and not said Often the exchanges of letters with the relatives already begin on board the ships travel- 351 In 1979 Emilio Franzina had already published a collection of letters written by peasants from Veneto who História Unisinos Maria Izilda Santos Matos, Oswaldo Mario Serra Truzzi ling to the New World, or at the time of physical and mental departure of those who leave the family and social context (Caffarena, 2012, p. 21). permeated by power relations between the author and the addressee. It is possible to observe how gender relations were formed, reaffirmed or subverted by means of the process of writing and reading the letters. p g g Letters exchanged between migrants and their families expressed different views about the household and everyday practices, processes of change, negotiation of identities, adaptation to the societies to which they came (there are letters announcing the intention of getting mar- ried or becoming a citizen), or the immigrants’ difficulties to integrate, how the “others” were constituted and what their attitude was towards them (Elliott et al., 2006, p. 9). It was relatively usual for the men to emigrate first, thus attenuating the impacts of change, and, in a preven- tive action against possible misfortunes, leave calling the remainder of the family to a more favorable time, when they were already settled and were in a better financial situation. Although most of the letters were written by men, there are constant mentions of the women, above all as addressees, in them. While the letters can indeed serve as a guide to recover the presence of women in the emigration/immigration process, on the other hand, between the lines, one can observe echoes of the silenced female voices, the resistance to departures, the demand for news or remittances. As an ensemble, the missives record different experiences, revealing personal and family relations (dis- aggregation, separation and family reunion), involving solidarity in difficulties, support (rivalries, affections and friendships), showing interests, perspectives and possi- bilities, but their main theme is the effort to bring the family together again. Present in absentia: Immigrant letters and requests for family reuniﬁ cation I cannot stand how much I miss the children, and I have missed you too, since I had a cold that wouldn’t go away, but if you had been here, this would not happen to me [Antonio Teixeira Cardozo to Maria José, 47 years, born in Porto, on July 13, 1914 (ADP-RP, proc. no. 1138)]. bring you and our children with him” [Manuel dos Santos to Joana Rosa, 24 years old, Born in Guimarães, and three daughters, on June 20, 1914, (ADP-RP, proc. no. 1110)]. This applies both to crucial matters, such as the date of travel – “as soon as you receive this letter prepare yourself, because I want to see whether you will come to eat ‘rabana- das’ [a typical sweet] with me” [Antonio Ribeiro to Joaquina Maria da Conceição, on February 20, 1914 (ADP-RP, proc. no. 651)] – and for more run-of-mill topics: “Do not bring pots and pans or anything else, only bring what I told you to bring, nothing more” [Joaquim Teixeira to Rosa Moreira, on October 28, 1913 (ADP-RP, proc. no. 211)]. Anna, I need you to come and be with me, because I can’t be without you, in my life I need a woman, therefore come and join me and I’ll make a lot of money [Pinto Cardozo de Souza to his wife Anna da Silva, 29 years, born in Gaia, and daughter, in March 1914 (ADP-RP, proc. no. 167)]. Anna, I need you to come and be with me, because I can’t be without you, in my life I need a woman, therefore come and join me and I’ll make a lot of money [Pinto Cardozo de Souza to his wife Anna da Silva, 29 years, born in Gaia, and daughter, in March 1914 (ADP-RP, proc. no. 167)]. At the same time, anxious and maybe even inse- cure due to the time of absence and to the distance, it is relatively common for men to sign off seeking to reaffirm their male role in the couple’s gender relations: – “I’m your man”, wrote Manoel Gomes Pereira to Carolina, on November 14, 1913, or “ from this man of yours who soon wants to have you” ended the letter received by Matilde de Jesus, 55 years old, born in Lamego (ADP-RP, procs. no. 245 and 346). Certain husbands waited patiently, others pre- sented ultimatums. Meanings of the dialogues: family reunification, authority and control, said and not said 352 Sometimes the writing sounds like an order given by someone who only informs and decides, without accepting any challenges: “Joana, I inform you that I have written to your brother-in-law Silvestre to give him an order that when he returns to this country with the family he should The process of receiving the letters together with the tensions it involved should be underscored, ranging from doubts about the letters sent and not received to the topic of the different meanings grasped by the reader, Vol. 19 Nº 3 - setembro/dezembro de 2015 There are many other examples: There are many other examples: Albina, I’m very surprised that you are taking so long, after I told you to come so many times and sent money for travels and for expenses, I don’t know what you lack now. It is probably the will, but I won’t wait forever, so get a move on as soon as you get this [Albino Ribeiro to Albina de Jesus, 38 years, born in Figueira Verde, and six children, on January 4, 1914 (ADP- RP, proc. no. 852)]. […] I personally have faith in God that soon we will be in each other’s arms, and that only thus will my spirit be quietened, as soon as I have my love by my side, I have had so few joys in being far from the person I most esteem in this life, that I have been in a purgatory in this world because of your absence. You can’t imagine how my eyes are always full of tears because of my love from my heart … receive a thousand hugs and a thousand kisses from this your very humble husband, until God lets us embrace (in Sarmento, 1999, p. 291). Or else from a father to his son, who apparently was very hesitant about crossing the Atlantic: Now after you let me know that you had been set free at the inspection, I have written three letters to you telling you to come to me, and for this I have also already sent you money. Now once again I am writing you to come without wasting any time, I don’t know why you resist joining us and coming to lands of plenty, but if you do not come, it will be because you want nothing to do with your father, therefore very sorrowfully I tell you that I no longer want to have anything to do with you [Antonio de Freitas to his son Manoel de Freitas, 20 Now after you let me know that you had been set free at the inspection, I have written three letters to you telling you to come to me, and for this I have also already sent you money. Present in absentia: Immigrant letters and requests for family reuniﬁ cation They said that they would no longer write, that that would be the last attempt and intimidated their wives by threatening to abandon them if they did not come, as in the letter written by José Fernandes da Silva to his wife in 1918: […] you know very well that I have never liked to be gainsaid, and after I told you that you should not gainsay me, you pretended that I was not your man, well you had better know me, I told you to come, and your obligation was to come (ADP-RP, proc. no. 482). Or also, in an admonishing tone: “You know perfectly well that a woman’s place is with her husband”, wrote Alfredo Ferreira to Laura da Soledade, 27 years, born in Conselho de Arouca, on October 27, 1913 (ADP- RP, proc. no. 320). In the missives, some husbands wrote tenderly, showing love and affection, as Antonio de Almeida wrote his wife Teresa da Costa de Oliveira, in 1910: Or: […] you’ll have to accept the consequences to which I subject myself here, sometimes it is good, at others bad [Eduardo da Silva Pinto to his wife Angelina Rosa, seamstress, 29 years old, born in Concelho de Mesão Frio, and 9-year old son, on April 12, 1914 (ADP- RP, proc. no. 527)]. The husbands’ departure affected women’s everyday life, increasing their work and responsibility, because, besides household activities and taking care of the children, they took over the maintenance of the properties, the farming work, the trade and family businesses. Hence, in addition to their traditional roles they took over the family business, managed the couple’s assets and administered the use of the remittances, and also worked in the shops and in the field (they plowed, dug, harvested, prepared the products and took care of the livestock and their byproducts), besides other activities essential for survival, such as collecting wood for themselves and to sell (bakeries and potteries), and spin- ning and weaving (linen and wool), among other activities. The husbands themselves, although generally desirous of seeing their families again, also feared for the trip of their wife and children without the company of someone trustworthy to protect them. Take great care with the little girls, especially in Leix- ões and during the trip, so that nothing goes wrong 6 or 15 days before boarding ship [Antonio Teixeira Cardozo to Maria José, 47 years old, born in Porto, on July 13, 1914 (ADP-RP, proc. no. 1138)]. In the letters it can be noticed that they accused the husbands of remaining indifferent to problems and expressed jealousy because of gossip and slander. The family conflicts grew with distance and can be perceived via the complaints of wives who felt neglected, left help- less, alone with the children, often even going hungry. Sometimes a serious misfortune provided an argument for family reunification, as can be seen in a letter sent by José Gomes da Silva to his wife Maria da Conceição, trying to convince her to travel: […] you forget having no fear of coming, you come and also Preciosa, you listen to what she says, but the two of you have to come, so that when one of you goes out, the other stays with the children [Antonio Ribeiro to Joaquina Maria de Conceição, in 1913, and children aged 4 and 1 year (ADP-RP, proc. no. 651)]. Or else: [...] write me by return of mail and receive an embrace from this husband and friend, I miss you and am dying for you to come [Antonio Ribeiro to Joaquina Maria de Conceição, in 1913, and children aged 4 and 1 year, living in Porto (ADP-RP, proc. no. 651)]. Now once again I am writing you to come without wasting any time, I don’t know why you resist joining us and coming to lands of plenty, but if you do not come, it will be because you want nothing to do with your father, therefore very sorrowfully I tell you that I no longer want to have anything to do with you [Antonio de Freitas to his son Manoel de Freitas, 20 353 Others, less loving and more pragmatic, declared that they missed the women in everyday life, had many expenses with food and laundry, and really, really needed them to get on with life. História Unisinos Maria Izilda Santos Matos, Oswaldo Mario Serra Truzzi years old, resident in Vila do Conde, on December 10, 1913 (ADP-RP, proc. no. 50)]. […] here things are not good, but one will always find something to eat [José Gomes da Silva to his wife Ma- ria da Conceição, 25 years old, married, born in Castelo de Paiva, and daughter (ADP-RP, proc. no. 364)]. Manuel de Sousa Monteiro, on the other hand, in his letter in 1913, responding to the insistence of his wife, wrote that he would not call her “… because I have a very pretty mulatto girl with me, even you would be charmed if you saw her...” (Letter no. 181, of November 14, 1921. Fundo Hospedaria dos Imigrantes de São Paulo, APESP). Or: If perchance some family comes from there, you write to come with them, I think that your brother-in-law also wants to come to Brazil, if he comes, you come with him, if nobody is coming, I’ll go and fetch you [received by Maria Valmira Rodrigues Barbosa, 24 years, born in Viseca, and 1-year old son, on February 25, 1913 (ADP-RP, proc. no. 144)]. What made me feel the worst was the death of my daughter without even seeing her, it is my greatest sorrow, the same could happen to you or to me, therefore come to be with me, because we will always be happier living together than being separated, without knowing when I’ll be able to rejoin you (ADP-RP, proc. no. 364). […] always stay near the respectable women, both by day and by night, try to have a bed always near respect- able women, because a woman … must be respectable everywhere. Do not be afraid of boarding, that is child’s play, your Joaquim Teixeira (ADP-RP, proc. no. 211). Other wives, however, did not want to join their husbands – the husbands’ departure, despite the many things that the wives had to do, was a certain relief, because they felt more in charge of their own lives, they became used to dealing with money and with business, took on the role of manager of family affairs and were free from undesired pregnancies. Sometimes the wives made up excuses not to come (she was ill, a child was ill, her parents were ill), seeking for ways to put off the trip. Some spent the money that had been sent and did not leave; others, after many threats and complaints from their husbands, travelled out of fear that they would be abandoned. 354 There were women who expressed fear of the trip, of returning to domestic subservience and of the multiple uncertainties of an unknown country. These doubts were often fed, deliberately or not, by the letters they received from their husbands: Vol. 19 Nº 3 - setembro/dezembro de 2015 Present in absentia: Immigrant letters and requests for family reuniﬁ cation I know well that you find it difficult to leave your family, but remember that you are coming to your husband… Thus, I won’t bother you anymore... I hope that you will come soon, that it will not be necessary for me to write another letter. Or: Hear well what I say. You will have people there who tell you not to come, but you only have to listen to what I say, I cannot go there because right now I don’t have money to go and fetch you [Antonio Ribeiro to Joaqui- na Maria de Conceição, in 1913, and children aged 4 and 1 year (ADP-RP, proc. no. 651)]. Alone or with their children, they faced the travel across the Atlantic, with the prospect of an unknown country, in search of their dream to build up their family again. I don’t know whether you would be happy that I have you come? But be patient, because I want to be with you! I wrote that you should come, now let’s see what you tell me! If you do not come, do not send clothes, because I do not accept them!... I won’t even write anymore. I would not have you come if you never asked me. But now, whether you want it or not, you have a choice of two paths [Manoel Gomes Pereira to Carolina Augusta de Bastos, 19 years, born in Macieira de Co- imbra, and son, 2 years old (ADP-RP, proc. no. 245)]. Hear well what I say. You will have people there who tell you not to come, but you only have to listen to what I say, I cannot go there because right now I don’t have money to go and fetch you [Antonio Ribeiro to Joaqui- na Maria de Conceição, in 1913, and children aged 4 and 1 year (ADP-RP, proc. no. 651)]. Or: N.B.: If you do not come, this is the last letter that I’m writing you, and I will no longer care about you [Antonio de Castro to Maria da Conceição Pinto da Silva, born in Gaia, on December 23, 1913 (ADP-RP, proc. no. 157)]. On this day I write [...] and our dear children through my mother you must be surprised at my silence now I have to tell you that José Fernandes holds a promissory note with the amount needed for you and the children to travel to this country, because you must know that to send money there the exchange is very high therefore as I say come and our children so that we put an end to all this [Alfredo Ferreira to Laura da Soledade, 27 years, born in Conselho de Arouca, and 3-year old daughter on October 27, 1913 (ADP-RP, proc. no. 320)]. Or else: Since 15 months ago I told you to come and until now I have been [...] this is the last letter I wrote to you for this purpose, I also want you in my company and that you bring with you the three children, Adelaide, Lourianna and Frectoso, I even mentioned this [re- ceived by Matilde de Jesus, 55 years, born in Lamego (ADP-RP, proc. no. 346)]. Once they have decided to come, it is very com- mon for the husbands to advise their wives very precisely about what they should bring, as stipulated in the letters of October 28, 1913 and November 30, 1913 received by Rosa Maria and Maria Pinto, respectively: Others did not go as far as making threats, but complained: I am writing hoping that you will come, it is high time [Joaquim Rodrigues in a letter written from Monte Azul (SP) to his wife Rosa Tavares da Silva, 35 years old, born in Gaia, on April 22, 1914 (ADP-RP, proc. 511)]. Look, go and buy a box like that of Damião more or less ..., and then in it you put all your clothes and the bedclothes too, and afterwards only put in, for instance, forks and spoons, and 3 or 4 small plates, and [...] the small sewing kit, and all those little nonsenses that you want to bring, and hurry to have slippers made, and buy some stockings and put … delicate clothes in a bag to bring it in the steamer because they are very necessary (ADP-RP, proc. no. 211). Some wives, in turn, were anxious for the reunion, they insisted, pressured for the letter of reunification, threatened to leave for Brazil, and even without authori- zation they sought loopholes and alternatives. Some hus- bands responded with pacifying messages, others imposed conditions (not to bring her mother, not coming with her siblings, control her temper) and several ultimately gave in to the requests and told them to come. I received your letters to which I am answering, one of November 8 in which you say that you killed the sow, you did well, send me the loins, you ask whether you should bring it, there is a lot of meat here, and to bring it the customs taxes are very expensive and there are many eaters, bring a few apples and walnuts (ADP-RP, proc. no. 657). Since 15 months ago I told you to come and until now I have been [...] this is the last letter I wrote to you for this purpose, I also want you in my company and that you bring with you the three children, Adelaide, Lourianna and Frectoso, I even mentioned this [re- ceived by Matilde de Jesus, 55 years, born in Lamego (ADP-RP, proc. no. 346)]. g p http://dx.doi.org/10.1057/9780230601079 FRANZINA, E. 1979. Merica! Merica! Emigrazione e colonizzazione nelle lettere dei contandini Veneti in America Latina 1876-1902. FRANZINA, E. 1979. Merica! Merica! Emigrazione e colonizzazione nelle lettere dei contandini Veneti in America Latina 1876-1902. Milano, Feltrinelli, 272 p. Milano, Feltrinelli, 272 p. FRENETTE, Y. ; MARTEL, M. ; WILLIS, J. (dirs.). 2006. Envoyer et recevoir: Lettres et correspondances dans les diasporas francophones. Québec, Les Presses de l’Université Laval, 316 p. Québec, Les Presses de l’Université Laval, 316 p. GIBELLI, A. 2002. Emigrantes y soldados: la escritura como práctica de masas en los siglos XIX y XX. In: A. CASTILLO GÓMEZ (coord.), La conquista del alfabeto: escritura y clases populares. Gijón, Trea, p. 189-233. Finally, even acknowledging that these letters of reunification attempted to fulfill the requirements of government laws, they must also be seen in the context of aspirations of the parties involved, and they give us a possibility of observing the problem of silences, omis- sions, understatements and untruths, of leaving out the painful aspects, inconveniences, embarrassments. It is then clear that not everything that appears in the letters can be considered “the expression of the truth”, and also that not everything that might have been said appears in them. In this way they comprise a documentation that is strategic to analyze also what was not said and what was said between the lines, as we attempted to do here by exploring the aspirations for family reunification, the gaps that distance and time of separation imposed on relationships, and also the fears involving the trip and life in a new country. In these cases in which “narrative truths” are superposed on “factual truths”, the effort and commitment to maintain contact and relationship stand out, often at the expense of clarity and faithful reporting of experiences that were made. GIBELLI, A. 1989. “Fatemi un po sapere…”: scrittura e fotografia nella corrispondenza degli emigrante liguri. In: La via delle Americhe: l’emigrazione ligure tra evento e racconto. Catalogo della mostra. Genova, Sagep Editrice, p. 87-94. Genova, Sagep Editrice, p. 87-94. GIBELLI, A.; CAFFARENA, F. 2001. Le lettere degli emigrante. In: P. BEVILACQUA; A. DI CLEMENTI; E. FRANZINA (orgs.), Storia dell’emigrazione italiana. Roma, Donzelli, vol. 1, p. 563-574. KULA, M. 1977. Cartas dos emigrantes do Brasil. In: Anais da Co- munidade Brasileiro-Polonesa, Curitiba, Superintendência do Centenário da Imigração Polonesa ao Paraná, vol. VIII, p. 9-117. LUSSANA, F. 1913. g p http://dx.doi.org/10.1057/9780230601079 Lettere di illetterati: Note di psicologia sociale. Bo- logna, Zanichelli, 222 p. MATOS, M.Izilda S. 2014. Cotidiano e cultura: história, cidade e trabalho. 2. Ed.Bauru, EDUSC, 108 p. MOLINARI, A. 1999. L’emigrazione ligure: fonti autobiografiche/mem- orie dell identità. Cahiers de la Mediterranée: Mémoire et identité de la frontière: étude des migrations de proximité entre les Provinces Ligures et les Alpes-Maritimes. Nice, Centre de la Mediterranée Moderne et Contemporaine, 58, p. 7-17. NUÑEZ SEIXAS, X.; SOUTELO VÁZQUEZ,R. 2005 As cartas do destino. Vigo, Galaxia, 266 p. RODRIGUES, H.F. 2013. Epistolário popular e imagens da emigração oitocentista: uma abordagem ás cartas enviadas do Brasil para Viana do Castelo. In: H.F. RODRIGUES; E. PORTUGUÊS, Escritas privadas, da mobilidade e da guerra. Monção, Câmara Municipal de Monção, p. 59-123. Or else: […] here I received your letter and there I saw how clearly you let me know, since you are so anxious to join me, I have decided to do what you want [Antonio de Castro to his wife Maria da Conceição Pinto da Silva, 28 years old, born in Gaia, on December 23, 1913 (ADP-RP, proc. no. 157)]. […] here I received your letter and there I saw how clearly you let me know, since you are so anxious to join me, I have decided to do what you want [Antonio de Castro to his wife Maria da Conceição Pinto da Silva, 28 years old, born in Gaia, on December 23, 1913 (ADP-RP, proc. no. 157)]. 355 História Unisinos História Unisinos Maria Izilda Santos Matos, Oswaldo Mario Serra Truzzi York, Palgrave Macmillan, 320 p. Obviously in the missives there are also cases of husbands who did not express any wish for a reunion or referred in a vague manner to family reunification. They had come alone or with friends, enjoyed freedom (which was impossible in the society they came from) and no longer wished to return to the previous situation. Some men no longer sent news, never returned and founded new families in Brazil. The women waited endlessly and became “widows with living husbands”, with feelings of missing them. References BAILY, S.L.; RAMELLA, F. 1988. One Family, Two Worlds: An Italian Family’s Correspondence across the Atlantic, 1901-1922. New Brunswick, Rutgers University Press, 251 p. SARMENTO, C.M. 1999. “Minha querida marida”: subsídios para o estudo da família emigrante através das cartas de chamada 1890-1914. In: P. SÁ MACHADO; J.A. MAIA MARQUÉS (coords.), Maia. História Regional e Local. Maia, Câmara Mu- nicipal, II, p. 285-296. BLASS, V. 2004. ‘Puentes de papel’: apuntes sobre las escrituras de la emigración. Horizontes Antropológicos, 10(22):93-119. BODIO, L. 1894. Sulla emigrazione italiana e sul patronato degli emigranti. In: Atti Del primo Congresso geográfico italiano tenuto in Genova dal 18 al 25 settembre 1892, Genova, Tipografia Del Regio Istituto sordo-muti, vol. II, p. 109-148. SEYFERTH, G. 2005. Cartas e narrativas biográficas no estudo da migração. In: Z.B.F. DEMARTINI; O. TRUZZI, Estudos migratórios – perspectivas metodológicas. São Carlos, EdUFSCar, p. 13-51. CAFFARENA, F. 2012. Introducción. In: F. CAFFARENA; L.M. MARTÍN (org.), Escrituras migrantes: una mirada ítalo-española. Genova, Franco Angeli, p. 9-22. SINKE, S.M. 2006. Marriage through the Mail: North American Correspondence Marriage from Early Print to the Web. In: B.S. ELLIOTT; D.A. GERBER; S.M. SINKE, Letters Across Borders: The Epistolary Practices of International Migrants. New York, Palgrave Macmillan, p. 75-94. CASTILLO GÓMEZ, A. 2001. Escritura y clases subalternas: una mirada española. Oiartzun, Sendoa, 181 p. 356 SOUTELO VÁZQUEZ, R. 2001. De América para a casa: corresponden- cia familiar de emigrantes galegos no Brasil, Venezuela e Uruguay (1916-1969). Santiago de Compostela, Consello da Cultura Galega, 298 p. CHARTIER, R. 1991. Avant-propos. In: R. CHARTIER. La correspon- dance : Les usages de la lettre au XIXe siècle. Paris, Fayard, p. 5-15. CROCI, F. 2008. O chamado das cartas: migrações, cultura e identidade nas cartas de chamada dos italianos no Brasil, Locus, 14(2):13-39. ELLIOTT, B.; GERBER, D.; SINKE, S. (eds.). 2006. Letters across Borders – The Epistolary Practices of International Migrants. New SOUTELO VÁZQUEZ, R. 2003. La correspondencia familiar de los emigrantes gallegos durante el franquismo. In: A. CASTILLO Vol. 19 Nº 3 - setembro/dezembro de 2015 Present in absentia: Immigrant letters and requests for family reuniﬁ cation Present in absentia: Immigrant letters and requests for family reuniﬁ cation Primary sources GÓMEZ; F. MONTERO GARCÍA (dirs.), Franquismo y memoria popular: escrituras, voces y representaciones. Madrid, Siete Mares, p. 123-176. GÓMEZ; F. MONTERO GARCÍA (dirs.), Franquismo y memoria popular: escrituras, voces y representaciones. Madrid, Siete Mares, p. 123-176. Arquivo Distrital do Porto. Requisição de Passaportes (ADP-RP). Fundo Hospedaria dos Imigrantes de São Paulo (APESP). THOMAS, W.I.; ZNANIECKI, F. 1958. The Polish Peasant in Europe and America. New York, Dover, 2250 p. VENDRAME, M.I. 2010. “Nós partimos pelo mundo, mas para viver melhor”: redes sociais, família e estratégias migratórias. Metis, 9(17):69-82. VENDRAME, M.I. 2010. “Nós partimos pelo mundo, mas para viver melhor”: redes sociais, família e estratégias migratórias. Metis, 9(17):69-82. Submitted on August 16, 2015 Accepted on September 29, 2015 Submitted on August 16, 2015 Accepted on September 29, 2015 WACHOWICZ, R.C. 1981. O camponês polonês no Brasil. Curitiba, Fundação Cultural Casa Romário Martins, 149 p. História Unisinos Maria Izilda Santos Matos Pontifícia Universidade Católica de São Paulo Rua Monte Alegre, 984 05014-901, São Paulo, SP, Brasil Oswaldo Mario Serra Truzzi Universidade Federal de São Carlos Rod. Washington Luís, km 235 Caixa Postal 676 13565-905, São Carlos, SP, Brasil Maria Izilda Santos Matos Pontifícia Universidade Católica de São Paulo Rua Monte Alegre, 984 05014-901, São Paulo, SP, Brasil 357
https://openalex.org/W2907723698	https://hal-amu.archives-ouvertes.fr/hal-01986953/file/Prof_JPQE_L%20Escoubas.pdf	English	null	Design and realization of light absorbers using plasmonic nanoparticles	Progress in quantum electronics/Progress in Quantum Electronics	2,019	cc-by	20,618	To cite this version: Ludovic Escoubas, Miriam Carlberg, Judikaël Le Rouzo, Florent Pourcin, Jörg Ackermann, et al.. Design and realization of light absorbers using plasmonic nanoparticles. Progress in Quantum Elec- tronics, 2019, 10.1016/j.pquantelec.2018.12.001. hal-01986953 Design and realization of light absorbers using plasmonic nanoparticles Ludovic Escoubas, Miriam Carlberg, Judikaël Le Rouzo, Florent Pourcin, Jörg Ackermann, Olivier Margeat, Clement Reynaud, David Duche, Jean-Jacques Simon, Rose-Marie Sauvage, et al. Distributed under a Creative Commons Attribution 4.0 International License Design and realization of light absorbers using plasmonic nanoparticles HAL Id: hal-01986953 https://amu.hal.science/hal-01986953v1 Submitted on 19 Jan 2019 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 Progress in Quantum Electronics xxx (2019) xxx-xxx OF A B S T R A C T ECTED The applications of light absorbers concern photodetectors, optical filters, solar applications or flexible electronics. In this review, we will detail the application of such light absorbers and we will develop the main demonstrations of the use of metallic nanoparticles embedded within a host matrix to fabricate coatings aiming at harvesting light. We will explain how chemically syn- thetized silver nanoparticles of various shapes (spheres, cubes, …) and sizes allow controlling the optical properties of heterogeneous thin film layers. By coupling the optical characterizations with computer modeling, we will describe how the nanoparticles behave both individually and collectively. To control reflected and absorbed light by thin film layers containing nanoparticles several points have to be addressed: the relation between the shape of the nanoparticle and the absorptance of the layer, the interaction of light between nanoparticles and the collective behav- ior of aggregates. U ∗Corresponding author. Email address: ludovic.escoubas@im2np.fr (L. Escoubas) https://doi.org/10.1016/j.pquantelec.2018.12.001 Available online xxx 0079-6727/ © 2019. RO Design and realization of light absorbers using plasmonic nanoparticles PRO Miriam Carlberg⁠a, Judikael Le Rouzo⁠a, Florent Pourcin⁠b, Jorg Ackermann⁠b, ment Reynaud⁠a, David Duche⁠a, Jean-Jacques Simon⁠a, Rose-Marie Sauvage⁠c, PRO Ludovic Escoubas⁠a⁠, ⁠∗, Miriam Carlberg⁠a, Judikael Le Rouzo⁠a, Florent Pourcin⁠b, Jorg Ackermann⁠b, Olivier Margeat⁠b, Clement Reynaud⁠a, David Duche⁠a, Jean-Jacques Simon⁠a, Rose-Marie Sauvage⁠c, Gérard Berginc⁠d a Aix Marseille Univ, Université de Toulon, CNRS, IM2NP, Marseille, France b Aix Marseille Univ, CNRS, CINAM, Marseille, France c DGA/DS/MRIS, 75015, Paris, France d Thales Optronics, 78990, Elancourt, France a Aix Marseille Univ, Université de Toulon, CNRS, IM2NP, Marseille, France a Aix Marseille Univ, Université de Toulon, CNRS, IM2NP, Marseille, France b Aix Marseille Univ, CNRS, CINAM, Marseille, France c DGA/DS/MRIS, 75015, Paris, France d Thales Optronics, 78990, Elancourt, France a Aix Marseille Univ, Université de Toulon, CNRS, IM2NP, Marseille, France b Aix Marseille Univ, CNRS, CINAM, Marseille, France c DGA/DS/MRIS, 75015, Paris, France d Thales Optronics, 78990, Elancourt, France d Thales Optronics, 78990, Elancourt, France 1. Introduction UNCOR Many applications require having at disposal or designing materials, which are able to absorb the light, either specifically for some light incidences and spectral ranges chosen by users, or in broad-ranges of incidences and wavelengths. In most appli- cations the efficiency of absorption must be maximal and must be obtained with the thinnest materials, very often in thin lay- ers, and with the wish that these thin layers perfectly fit with the forms of the surfaces they cover. When one wishes to go be- yond the absorption ranges usually found in the materials in their natural state, one possibility is to design new synthetic mate- rials made of alloys of constituents, which are often of complex structures. The structuring of these materials, when the charac- teristic dimensions are close to the wavelengths to be absorbed, greatly influences the properties of absorption in terms of spec- tral range, of angular fields and of efficiency. To build such thin materials, which are ultra-absorbing and not available in nat- ural state, resonance phenomena of light found in Fabry-Perot cavities or plasmonic effects are very useful. These plasmonic ef- fects may appear in the gap between two thin metallic layers, and are called gap plasmons, or on the surface of structured materi- als (called surface plasmons), or even directly at the level of nanoparticles of metal or degenerate semiconductor (called localized plasmons). Thus, it is possible to obtain very strong light absorption with very small amount of material. Furthermore, associat https://doi.org/10.1016/j.pquantelec.2018.12.001 Available online xxx 0079-6727/ © 2019. L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx TED PROOF Fig. 1. a) Sketch of a switchable perfect absorber device. Arrays of square aluminum (Al) nanoantennas are stacked above a spacer layer of the phase change material GST-326 and an Al mirror. b) Amorphous to crystalline phase transition in GST. c) SEM images of two representative perfect absorber devices fabricated with antenna side lengths of d=400nm and d=450nm @Wiley (extracted with permission from ref 7 - https://doi.org/10.1002/adma.201502023). TED Fig. 1. a) Sketch of a switchable perfect absorber device. Arrays of square aluminum (Al) nanoantennas are stacked above a spacer layer of the phase change material GST-326 and an Al mirror. b) Amorphous to crystalline phase transition in GST. 1. Introduction c) SEM images of two representative perfect absorber devices fabricated with antenna side lengths of d=400nm and d=450nm @Wiley (extracted with permission from ref 7 - https://doi.org/10.1002/adma.201502023). RRECTE Fig. 2. (a) Schematic of the dual-band perfect absorber structure and the incident polarization configuration. (b) A SEM image of the designed structure. @OSA (ex- tracted with permission from ref 8 - https://doi.org/10.1364/OE.19.015221). RECT T ECTE RR Fig. 2. (a) Schematic of the dual-band perfect absorber structure and the incident polarization configuration. (b) A SEM image of the designed structure. @OSA (ex- tracted with permission from ref 8 - https://doi.org/10.1364/OE.19.015221). UNCORR Fig. 3. Photograph of resonant color filters. @ACS Publications (extracted with permission from ref 10 - https://doi.org/10.1021/ph500410u). UN Fig. 3. Photograph of resonant color filters. @ACS Publications (extracted with permission from ref 10 - https://doi.org/10.10 UN Fig. 3. Photograph of resonant color filters. @ACS Publications (extracted with permission from ref 10 - https://doi.org/10.1021/ph500410u). UN of resonant color filters. @ACS Publications (extracted with permission from ref 10 - https://doi.org/10.1021/ph500410u). U ing optical resonators, whose dimensions are close to the wavelength, allows electromagnetic interactions to be obtained, which opens new possibilities of controlling the light propagation in the materials. The aim of this article is firstly to describe the general working principles of light absorbers based on plasmonic effect, mainly obtained from nanoparticles. In a second step, we discuss the synthesis routes of plasmonic nanoparticles and we show U ing optical resonators, whose dimensions are close to the wavelength, allows electromagnetic interactions to be obtained, which opens new possibilities of controlling the light propagation in the materials.i U ing optical resonators, whose dimensions are close to the wavelength, allows electromagnetic interactions to be obtained, which opens new possibilities of controlling the light propagation in the materials.i The aim of this article is firstly to describe the general working principles of light absorbers based on plasmonic effect, mainly obtained from nanoparticles. In a second step, we discuss the synthesis routes of plasmonic nanoparticles and we show 2 2 L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx ROOF Fig. 4. (a) Schematic of a MIM plasmonic rectenna - (b) SEM image of nanocubes self-assembled into a periodic holed polymer matrix using dithiol molecule linkers localized between the base of the silver nanocubes and the gold surface. OO RO Fig. 4. 1. Introduction (a) Schematic of a MIM plasmonic rectenna - (b) SEM image of nanocubes self-assembled into a periodic holed polymer matrix using dithiol molecule linkers localized between the base of the silver nanocubes and the gold surface. ED TED PRO Fig. 5. Schematic of a double-stacked MIM (i.e. MIMIM) photodetectors operating in the near-infrared (NIR) spectrum up to 1200nm wavelength and presented in ref 17. The MIMIM comprised of metal nanoparticles (MNP), spacer insulator (Ispa), absorbing metal (Mabs), tunneling insulator (Itunnel) and bottom contact metal (Mcont), (b) SEM image of the nanoparticles after annealing 10nm Silver film for 20minat 500°C and (c) the size distribution of particles from SEM image of part (b) @Nature (extracted with permission from Ref. [17] - https://doi.org/10.1038/srep42349). ED PRO ED PRO TED Fig. 5. Schematic of a double-stacked MIM (i.e. MIMIM) photodetectors operating in the near-infrared (NIR) spectrum up to 1200nm wavelength and presented in ref 17. The MIMIM comprised of metal nanoparticles (MNP), spacer insulator (Ispa), absorbing metal (Mabs), tunneling insulator (Itunnel) and bottom contact metal (Mcont), (b) SEM image of the nanoparticles after annealing 10nm Silver film for 20minat 500°C and (c) the size distribution of particles from SEM image of part (b) @Nature (extracted with permission from Ref. [17] - https://doi.org/10.1038/srep42349). NCORRECT Fig. 6. a) Specular measured and Transfer Matrix Method (TMM) computed reflectances of Ag nanocubes:PVP composite layer deposited on Si* (b) TEM picture of silver nanocubes embedded in PVP c) Optical photograph of the Ag nanocubes:PVP composite on glass substrates. CORRECT CORRECT NC Fig. 6. a) Specular measured and Transfer Matrix Method (TMM) computed reflectances of Ag nanocubes:PVP composite layer deposited on Si* (b) TEM picture of silver nanocubes embedded in PVP c) Optical photograph of the Ag nanocubes:PVP composite on glass substrates. UN how it is possible to control their shapes, which can be spherical, prismatic or cubic for example, and thus affecting the inter- actions with the light. Then, we present analytical and numerical methods to model the interactions between these nanoparticles and the light waves: Mie theory, Discrete Dipole Approximation (DDA), Finite Difference Time Domain (FDTD). The modeling of the light propagation in these complex environments including nanoparticles of various forms requires the measurement of the optical complex indices in order to implement them in the computation codes. O 2. Design of plasmonic light absorbers NCO By definition, a light absorber is a device, whose optical properties are controlled. The specular and the diffuse reflectance, and also the amount of light transmitted through, are minimized. The device absorbs the incident light at a particular wavelength or a broad wavelength domain, ranging from the optical to the longest wavelength domain, and within the broadest solid angle for both linear s and p polarization states. Such absorbers are of interest for many applications, depending on their optical properties, as will be discussed in the following. Because of the large number of applications, extensive research has been done in order to define the optimal design of a light absorber. 1. Introduction ORR of two nanoparticles interacting in the Kerker's conditions or aggregates of particles (dimers, trimers, multimeres), whose 3D geome- try is perfectly controlled to create spectral absorption bands on the demand. ORR of two nanoparticles interacting in the Kerker's conditions or aggregates of particles (dimers, trimers, multimeres), whose 3D geome- try is perfectly controlled to create spectral absorption bands on the demand. ORR of two nanoparticles interacting in the Kerker's conditions or aggregates of particles (dimers, trimers, multimeres), whose 3D geome- try is perfectly controlled to create spectral absorption bands on the demand. 1. Introduction So, we will explain how spectroscopic ellipsom- etry measurement associated with Cauchy, Gaussian, or Lorentz models and the theory of effective medium allow extracting the values of the optical complex indices of these composite materials. Finally, we will devote the last part of the article to present achievements in which the light interactions between nanoparticles is exploited in very specific configurations: dimers consisting UN how it is possible to control their shapes, which can be spherical, prismatic or cubic for example, and thus affecting the inter- actions with the light. Then, we present analytical and numerical methods to model the interactions between these nanoparticles and the light waves: Mie theory, Discrete Dipole Approximation (DDA), Finite Difference Time Domain (FDTD). The modeling of the light propagation in these complex environments including nanoparticles of various forms requires the measurement of the optical complex indices in order to implement them in the computation codes. So, we will explain how spectroscopic ellipsom- etry measurement associated with Cauchy, Gaussian, or Lorentz models and the theory of effective medium allow extracting the values of the optical complex indices of these composite materials. Finally, we will devote the last part of the article to present achievements in which the light interactions between nanoparticles is exploited in very specific configurations: dimers consisting 3 ORRECTED PROOF L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx Fig. 7. Localized surface plasmon resonance (LSPR) frequency dependence on free carrier density and doping constraints @Nature (extracted with permission from ref 30 - https://doi.org/10.1038/nmat3004). Fig. 8. Seed and nanoparticle growth process. of two nanoparticles interacting in the Kerker's conditions or aggregates of particles (dimers, trimers, multimeres), whose 3D geome- try is perfectly controlled to create spectral absorption bands on the demand. 2. Design of plasmonic light absorbersl L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx PROOF D P Fig. 7. Localized surface plasmon resonance (LSPR) frequency dependence on free carrier density and doping constraints @Nature (extracted with permission from ref 30 - https://doi.org/10.1038/nmat3004). RRECTED 30 - https://doi.org/10.1038/nmat3004). Fig. 8. Seed and nanoparticle growth process. RRECTED Fig. 8. Seed and nanoparticle growth process. f t ti l i t ti i th K k ' diti t f ti l (di t i lti ) h 3D RR Fig. 8. Seed and nanoparticle growth process. RR Fig. 8. Seed and nanoparticle growth process. UN 2.1. Metal insulator metal (MIM) structure U A first highly-efficient absorber design concerns the metal-insulator-metal (MIM) structures [1–3]. The simplest MIM structure is composed of two metallic layers separated by a dielectric layer. The bottom layer is thick to act as a reflector and avoid opti- cal transmission. A simple MIM structure presents a single absorption band because there is a single resonant mode within the di- electric cavity. The incident light is trapped inside the cavity. The position of the absorption band depends on the chosen metal and dielectric materials. A typical choice for the metallic layer, in order for the device to absorb the light in the visible wave- length band, is silver or gold. Typical dielectric spacers are Si, SiO⁠2, AlO⁠3 or TiO⁠2. Depending on the aimed absorption band ED PROOF L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx Fig. 9. Schematic illustration of therapeutic nanoparticle platforms in preclinical development: (a) liposome, (b) polymer–drug conjugate, (c) polymeric nanoparticle, (d) dendrimer, and (e) iron oxide nanoparticle. The red dots represent hydrophilic drugs and the blue dots represent hydrophobic drugs @Nature Publishing Group (extracted with permission from ref 62 - https://doi.org/10.1038/sj.clpt.6100400). L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx ED Fig. 9. Schematic illustration of therapeutic nanoparticle platforms in preclinical development: (a) liposome, (b) polymer–drug conjugate, (c) polymeric nanoparticle, (d) dendrimer, and (e) iron oxide nanoparticle. The red dots represent hydrophilic drugs and the blue dots represent hydrophobic drugs @Nature Publishing Group (extracted with permission from ref 62 - https://doi.org/10.1038/sj.clpt.6100400). RRECTE Fig. 10. (a) Non uniform mesh and (b) TFSF disposition in the simulation window. RR Fig. 10. (a) Non uniform mesh and (b) TFSF disposition in the simulation window. ORR and applications, the suited materials and their thicknesses can be determined by computer simulations, such as finite difference time domain (FDTD) method, and numerical calculations, such as transfer matrix method (TMM) [1,4]. COR In order to absorb the light on a broader wavelength range, the number of layers can be increased to create multilayered MIM. Each dielectric cavity will be excited by a different wavelength of the incident light and therefore increase the absorption band of the device. Recent reports show up to 90% absorption from 400 to 1640nm for multilayered thin film stacks [4]. The advantages of multilayered MIM structures are undeniable, but the fabrication methods are an important factor in their industrial utilizations. UN 2.1. Metal insulator metal (MIM) structure NCO y p Deposition techniques such as sputtering [3] or thermal evaporation, sol-gel/spin coating [1] or Langmuir-Blodgett [5] methods, are of particular interest because of their simplicity and potential for low cost upscaling. Recent works have reported their ability to achieve efficient absorbers for large angles of incidence with a low polarization sensitivity.i NC g g p y Broadband multilayered MIM structure absorbers find applications in thermo and photovoltaic solar cells [6], radar technologies and thermal imaging [7] (see Fig. 1).l UN Experimental measurements show that the design of Fig. 1a exhibits low reflectance and hence very high resonant absorption (A>90%) in the MWIR spectral range. Geometrical tunability of the resonance wavelength is achieved by varying the side length d of the resonant square Al nanoantennas, with larger antennas corresponding to higher resonance wavelengths. When heating the sample above the germanium antimony telluride (GST) crystallization temperature, the resonance undergoes a pronounced phase-change-in- duced spectral redshift of up to 0.7μm while maintaining high absorbance. U Another absorption mechanism used in MIM structures is based on plasmon resonances [8,9]. The incident electromagnetic wave is coupled into the resonant dielectric gap in the form of a gap-plasmon and at the metal interface in the form of a surface-plasmon. Either the upper or the bottom metallic layer is structured to take advantage of the surface plasmon resonances. 5 L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx CTED PROOF Fig. 11. Example of the FDTD computed amplitude of the electric field enhancement on a 35nm edge size silver cube in a PVP polymer matrix at λ=350nm. TED PROOF Fig 11 Example of the FDTD computed amplitude of the electric field enhancement on a 35nm edge size silver cube in a PVP polymer matrix at λ=350nm TE Fig. 11. Example of the FDTD computed amplitude of the electric field enhancement on a 35nm edge size silver cube in a PV CTE Fig. 11. Example of the FDTD computed amplitude of the electric field enhancement on a 35nm edge size silver cube in a PVP polymer matrix at λ=350nm. NCORRECT Fig. 12. Example of coupled resonance calculated by FDTD: electric field enhancement at the dipolar resonance wavelength of silver nanoprisms displayed in a periodic squared pattern. The electromagnetic wave is travelling along x. NCORRECT NC Fig. 12. Table 1 T Table 1 Listing of the different optical models used to fit spectroscopic ellipsometry data adapted from Ref. [113]. UNCORRECT Listing of the different optical models used to fit spectroscopic ellipsometry data adapted from Ref. [113]. Sample Laws Meaning Ref Ag in PVA Lorentz Lorentz oscillator for main LSPR [101] Ag in PVP Lorentz + Cauchy Lorentz oscillator for main LSPR and Cauchy for the polymer host [112] Au nanospheres on gold substrate Lorentz Lorentz oscillator for main LSPR and Lorentz oscillator for background absorption [103] Ag nanospheres in Al⁠2O⁠3 Lorentz + Drude [109] Ag nanospheres and nanorods on Si substrates Lorentz + Tauc Lorentz Lorentz oscillator for main LSPR and Tauc-Lorentz for bulk silver [110] Au islands on glass substrate Gauss Gauss oscillator for main LSPR, Gauss oscillators for interband transitions of gold, Gauss oscillator for inhomogeneous broadening of LSPR [102] Ag islands on glass substrate Gauss + Tanguy Gauss oscillators for main LSPR, Gauss oscillator for bulk plasmon resonance and Tanguy oscillator for interband transitions [111] Fig. 15. (a) Complex refractive indices n and k of nanospheres and nanocubes blend in PVP and (b) the normalized extinction coefficient k compared to the one of nanospheres in PVP and nanocubes in PVP. @OSA (extracted with permission from Ref. [116]). RRECT Listing of the different optical models used to fit spectroscopic ellipsometry data adapted from Ref. [113]. Sample Laws Meaning Ref Ag in PVA Lorentz Lorentz oscillator for main LSPR [101] Ag in PVP Lorentz + Cauchy Lorentz oscillator for main LSPR and Cauchy for the polymer host [112] Au nanospheres on gold substrate Lorentz Lorentz oscillator for main LSPR and Lorentz oscillator for background absorption [103] Ag nanospheres in Al⁠2O⁠3 Lorentz + Drude [109] Ag nanospheres and nanorods on Si substrates Lorentz + Tauc Lorentz Lorentz oscillator for main LSPR and Tauc-Lorentz for bulk silver [110] Au islands on glass substrate Gauss Gauss oscillator for main LSPR, Gauss oscillators for interband transitions of gold, Gauss oscillator for inhomogeneous broadening of LSPR [102] Ag islands on glass substrate Gauss + Tanguy Gauss oscillators for main LSPR, Gauss oscillator for bulk plasmon resonance and Tanguy oscillator for interband transitions [111] UNCOR U Fig. 15. (a) Complex refractive indices n and k of nanospheres and nanocubes blend in PVP and (b) the normalized extinction coefficient k compared to the one of nanospheres in PVP and nanocubes in PVP. UN 2.1. Metal insulator metal (MIM) structure Example of coupled resonance calculated by FDTD: electric field enhancement at the dipolar resonance wavelength of silver nanoprisms displayed in a periodic squared pattern. The electromagnetic wave is travelling along x. UN As shown in Fig. 2, the structure is made from a film stack comprising: UN As shown in Fig. 2, the structure is made from a film stack comprising: U - a top metal layer composed of an elliptical gold nanodisk array, which geometrical parameters such as a, b and d can be tuned individually to control the interaction with light. U - a top metal layer composed of an elliptical gold nanodisk array, which geometrical parameters such as a, b and d can be tuned individually to control the interaction with light. U - a spacing dielectric SiO⁠2, MgF⁠2, or polymer layer with low permittivity to reach high absorption. The thickness of the dielectric layer is of high importance as it influences the dipole resonance. By increasing the thickness, the resonance effect diminishes and the overall absorption efficiency is reduced. 6 UNCORRECTED PROOF L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx Fig. 13. Simplified spectroscopic ellipsometry measurement setup. Fig. 14. Effective medium approach. Table 1 Listing of the different optical models used to fit spectroscopic ellipsometry data adapted from Ref. [113]. Sample Laws Meaning Ref Ag in PVA Lorentz Lorentz oscillator for main LSPR [101] Ag in PVP Lorentz + Cauchy Lorentz oscillator for main LSPR and Cauchy for the polymer host [112] Au nanospheres on gold substrate Lorentz Lorentz oscillator for main LSPR and Lorentz oscillator for background absorption [103] Ag nanospheres in Al⁠2O⁠3 Lorentz + Drude [109] Ag nanospheres and nanorods on Si substrates Lorentz + Tauc Lorentz Lorentz oscillator for main LSPR and Tauc-Lorentz for bulk silver [110] Au islands on glass substrate Gauss Gauss oscillator for main LSPR, Gauss oscillators for interband transitions of gold, Gauss oscillator for inhomogeneous broadening of LSPR [102] Ag islands on glass substrate Gauss + Tanguy Gauss oscillators for main LSPR, Gauss oscillator for bulk plasmon resonance and Tanguy oscillator for interband transitions [111] TED PROOF oubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx Fig. 13. Simplified spectroscopic ellipsometry measurement setup. Fig. 14. Effective medium approach. L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx R Fig 13 Simplified spectroscopic ellipsometry measurement setup TED PROO Fig. 13. UN 2.1. Metal insulator metal (MIM) structure Simplified spectroscopic ellipsometry measurement setup. Fig. 14. Effective medium approach. UNCORRECTED PRO Fig. 13. Simplified spectroscopic ellipsometry measurement setup. Fig. 14. Effective medium approach. Table 1 Listing of the different optical models used to fit spectroscopic ellipsometry data adapted from Ref. [113]. Sample Laws Meaning Ref Ag in PVA Lorentz Lorentz oscillator for main LSPR [101] Ag in PVP Lorentz + Cauchy Lorentz oscillator for main LSPR and Cauchy for the polymer host [112] Au nanospheres on gold substrate Lorentz Lorentz oscillator for main LSPR and Lorentz oscillator for background absorption [103] Ag nanospheres in Al⁠2O⁠3 Lorentz + Drude [109] Ag nanospheres and nanorods on Si substrates Lorentz + Tauc Lorentz Lorentz oscillator for main LSPR and Tauc-Lorentz for bulk silver [110] Au islands on glass substrate Gauss Gauss oscillator for main LSPR, Gauss oscillators for interband transitions of gold, Gauss oscillator for inhomogeneous broadening of LSPR [102] Ag islands on glass substrate Gauss + Tanguy Gauss oscillators for main LSPR, Gauss oscillator for bulk plasmon resonance and Tanguy oscillator for interband transitions [111] Fig. 15. (a) Complex refractive indices n and k of nanospheres and nanocubes blend in PVP and (b) the normalized extinction coefficient k compared to the one of nanospheres in PVP and nanocubes in PVP. @OSA (extracted with permission from Ref. [116]). Fig. 13. Simplified spectroscopic ellipsometry measurement setup. TED PRO Fig. 13. Simplified spectroscopic ellipsometry measurement setup. Fig. 14. Effective medium approach. TE Fig. 14. Effective medium approach. TE Table 1 Listing of the different optical models used to fit spectroscopic ellipsometry data adapted from Ref. [113]. TE Table 1 Listing of the different optical models used to fit spectroscopic ellipsometry data adapted from Ref. [113]. Table 1 @OSA (extracted with permission from Ref. [116]). 7 7 ORRECTED PROOF L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx Fig. 16. Scheme of the considered geometry. The system is composed of two silicon nanoparticles located at a distance d between them. The particle sizes are such that, while one nanoparticle has zero backscattering, the other one satisfied the MF scattering at the same incident wavelength (λ=700nm). The bottom figures show the 3-D spatial distribution of the light scattering of each isolated nanoparticle, with blue being the lowest intensity and red being the highest intensity. Depending on the incident direction, from right to the left (a) or vice versa (b), the light concentration on the gap drastically changes. @IEEE Photonics Society (extracted with permission from Ref. [123] - https://doi.org/10.1109/JPHOT.2016.2577714). L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx PRO TED P RECTE ORR Fig. 16. Scheme of the considered geometry. The system is composed of two silicon nanoparticles located at a distance d between them. The particle sizes are such that, while one nanoparticle has zero backscattering, the other one satisfied the MF scattering at the same incident wavelength (λ=700nm). The bottom figures show the 3-D spatial distribution of the light scattering of each isolated nanoparticle, with blue being the lowest intensity and red being the highest intensity. Depending on the incident direction, from right to the left (a) or vice versa (b), the light concentration on the gap drastically changes. @IEEE Photonics Society (extracted with permission from Ref. [123] - https://doi.org/10.1109/JPHOT.2016.2577714). OR - a bottom gold layer deposited on top of the substrate CO Fig. 2 (b) shows a SEM top-view image of the structure. Contrary to the component shown in Fig. 2, devices are generally highly polarization dependent and the absorption band is narrow, which makes them promising candidates for sensing applications, col- ored-optical filters [10] (see Fig. 3) and polarization detectors [11].ii CO Fig. 2 (b) shows a SEM top-view image of the structure. Contrary to the component shown in Fig. 2, devices are generally highly polarization dependent and the absorption band is narrow, which makes them promising candidates for sensing applications, col- ored-optical filters [10] (see Fig. 3) and polarization detectors [11].ii NCi As shown in Fig. 2.2. Plasmonic nanoparticles light absorbers ORRE Plasmonic nanoparticles display localized surface plasmon resonances [14,15] (LSPRs), which induce a selective light absorption and scattering depending on the material, size, shape and environment of the nanoparticle. Indeed, LSPRs excited in metallic nanopar- ticles are non-propagating plasmon excitations. Since the size of a metallic nanoparticle is on the same scale of the penetration depth of electromagnetic waves in metals (e.g., 20∼30nm for Ag and Au), the external field can penetrate the whole particle and shift the conduction electrons with respect to the rigid ion lattice. Thus, the charges are separated and this charge separation results in a restoring force and then an oscillation. The oscillation frequency is mainly related to effective electron mass, charge density, and geometry of the particle, as well as the properties of the surrounding medium. The amplitude of the induced electromagnetic field is much stronger than exciting fields (over 10 times). A comprehensive review on the applications of plasmonic effects to solar cells has been published by Atwater and Polman [16]. The nanoparticles can be: NCO - either deposited directly onto a thin dielectric spacer on a metallic substrate to fabricate a resonant absorbing MIM structure and the light is then absorbed through surface plasmon resonances [17] (see Fig. 5) in the cavity between the metal layer and the nanopar- ticles. The optical response of the deposited metasurfaces relies on the spacing and the nanocrystal size as well as nanoparticle density. This allows a large parameter space to fine tune the optical response [18,19]. It is essential here to carefully choose the underlying metal substrate as well as the thickness of the dielectric spacer to enhance optical couplings [20–22].i UN - or embedded within a host matrix. In this second configuration, the light is absorbed by the localized surface plasmon reso- nances of the nanoparticles inside the composite layer and the absorbed light can be locally converted in to heat [23]. The op- tical properties of the light absorber are mainly controlled by the size, density and shape of the nanoparticles and are indepen- dent of the choice of the substrate [24]. The role of the host materials, usually a transparent polymer, is to facilitate processing of homogeneous films over large areas and to control spacing and organization of the plasmonic absorber inside the nanocom- posite with dedicated optical properties. For example, Fig. Table 1 3, resonant optical filters can be fabricated from modified, asymmetric metal–insulator–metal (MIM) based Fabry–Perot cavities including plasmonic, lossy ultrathin (∼30nm) metallic films used as the top metallic layer. Different colors can be obtained by controlling the dielectric spacer thickness.i UN Because they can be confined over nanometric areas at dielectric-metal interfaces or into nanocavities, surface plasmons can create very high electric fields. This feature makes them particularly interesting for optical rectification purposes. One of the promises of optical rectification is to enable the fabrication of devices that convert light into electricity without relying on the photovoltaic effect. As a consequence, such a device – referred as optical rectennas (rectifying antenna) - would not be subject to the so-called Schockley Queisser limit that bounds the efficiencies of PV solar cells to 33% for single junctions. The concept of optical rectennas goes back to the 70's when Bailey [12] proposed that a nanoscale antenna coupled with a rectifier could har- vest electromagnetic waves in the visible and infrared region. Recent work have already demonstrated power production origi- nating from optical rectennas [13], but research in this field remains at stage of proof of concept. We have assembled very re- cently rectennas solar cells composed of plasmonic nanocubes (see Fig. 4a) associated with rectifying self-assembled molecular diodes that allow a plasmon cavity mode coupling between silver nanocubes and a gold plane. Thus, an electric field enhance 8 8 Progress in Quantum Electronics xxx (2019) xxx-xxx L. Escoubas et al. TED PROOF L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx Fig. 17. Distribution of the electric field in the incident and orthogonal planes (including the middle point of the gap) in the region between the nanoparticles consid- ering an incident beam of (λ=700nm). The geometry and view are shown in the center. (a) and (b) corresponding to the configuration of Fig. 16a with gap distances of d=375nm and d=200nm, respectively. (c) and (d) corresponding to the configuration of Fig. 16b with gap distances of d=120nm and d=445nm, respectively. @IEEE Photonics Society (extracted with permission from Ref. [123]- https://doi.org/10.1109/JPHOT.2016.2577714). TED Fig. 17. Distribution of the electric field in the incident and orthogonal planes (including the middle point of the gap) in the region between the nanoparticles consid- ering an incident beam of (λ=700nm). The geometry and view are shown in the center. (a) and (b) corresponding to the configuration of Fig. Table 1 16a with gap distances of d=375nm and d=200nm, respectively. (c) and (d) corresponding to the configuration of Fig. 16b with gap distances of d=120nm and d=445nm, respectively. @IEEE Photonics Society (extracted with permission from Ref. [123]- https://doi.org/10.1109/JPHOT.2016.2577714). ECT ment up to two orders of magnitude (intensity enhancement up to four orders of magnitude) is obtained that could enable the rectifi- cation process without any applied bias. The nanocubes shown in Fig. 4 are self-assembled thanks to dithiol molecules. By choosing a given molecule length, the gap thickness below the cube is controlled within a nanometer accuracy. A holed polymer matrix controls the periodicity of the nanocubes array (see Fig. 4b), which has a crucial role in the optical absorption by the device. 2.2. Plasmonic nanoparticles light absorbers 6a shows the specular reflectance spectra of a PVP layer with em- bedded Ag nanocubes. Deposited on Si, those nanocomposite layers produce a strong reflectance dip over the range of absorp 9 Progress in Quantum Electronics xxx (2019) xxx-xxx L. Escoubas et al. RRECTED PROOF 18. (A–C) Schematic and SEM images of nanocube assembly observed at embedding depths of 15nm, 42nm and 61nm of NCo. Scale bar=500nm. (D–F) Corre ding statistical analysis showing the population distributions of monomers, dimers, trimers, and multimers in each assembly. (G–I) Corresponding optical extinctio ra taken an incident illumination with a broadband white light source. The extinction spectrum of as-deposited NCo nanocubes prior to modular assembly is show ference (black lines). @Royal Society of Chemistry (extracted with permission from Ref. [124] - 10.1039/C5FD00134J). PROOF RECTED PR Fig. 18. (A–C) Schematic and SEM images of nanocube assembly observed at embedding depths of 15nm, 42nm and 61nm of NCo. Scale bar=500nm. (D–F) Corr sponding statistical analysis showing the population distributions of monomers, dimers, trimers, and multimers in each assembly. (G–I) Corresponding optical extinctio RECTE RRE Fig. 18. (A–C) Schematic and SEM images of nanocube assembly observed at embedding depths of 15nm, 42nm and 61nm of NCo. Scale bar=500nm. (D–F) Corre- sponding statistical analysis showing the population distributions of monomers, dimers, trimers, and multimers in each assembly. (G–I) Corresponding optical extinction spectra taken an incident illumination with a broadband white light source. The extinction spectrum of as-deposited NCo nanocubes prior to modular assembly is shown for reference (black lines). @Royal Society of Chemistry (extracted with permission from Ref. [124] - 10.1039/C5FD00134J). UNCOR Fig. 19. Schematic of a microbolometer membrane covered by a metamaterial consisting of a metal layer/dielectric (Si⁠3Nx)/periodic metal patch structure. tion where the plasmonic absorption of the Ag nanocubes occurs, while spin coating on glass leads to composite layers in which absorption is controlled by the amount of embedded cubes. The optical properties of such films depend not only on the natural, size and shape of the metal nanocrystals, but are also function of the spacing and arrangement of the nanocrystals U Fig. 19. Schematic of a microbolometer membrane covered by a metamaterial consisting of a metal layer/dielectric (Si⁠3Nx)/periodic metal patch structure. tion where the plasmonic absorption of the Ag nanocubes occurs, while spin coating on glass leads to composite layers in which absorption is controlled by the amount of embedded cubes. 2.2. Plasmonic nanoparticles light absorbers Thus degenerately do me nanostructure, opening up the p ing, nonlinear optics, and quantu c MIM structures or by solely takin ent optical phenomenon ROO Fig. 20. Schematic of a metamaterial perfect absorber achieving near-unity optical absorption using ultrathin plasmonic nanostructures with thicknesses smaller than the hot electron diffusion length. By integrating the metamaterial with a silicon substrate, a broadband and omnidirectional hot electron photodetector is obtained, showing a very high photoresponsivity. Dimensions of the metamaterial perfect absorber L=185nm–195nm; P=340nm–360nm, respectively and H=135nm. within the embedding film [25]. Near close-packed configuration of nanocrystals and their relative orientation in this packing cre- ate collective behaviors of unique optical signature [26,27]. Controlling aggregation and packing density of nanocrystals inside a hostmatrix can be expected to generate a new class of high performance plasmonic light absorber. ROO Fig. 20. Schematic of a metamaterial perfect absorber achieving near-unity optical absorption using ultrathin plasmonic nanostructures with thicknesses smaller than the hot electron diffusion length. By integrating the metamaterial with a silicon substrate, a broadband and omnidirectional hot electron photodetector is obtained, showing a very high photoresponsivity. Dimensions of the metamaterial perfect absorber L=185nm–195nm; P=340nm–360nm, respectively and H=135nm. within the embedding film [25]. Near close-packed configuration of nanocrystals and their relative orientation in this packing cre- ate collective behaviors of unique optical signature [26,27]. Controlling aggregation and packing density of nanocrystals inside a hostmatrix can be expected to generate a new class of high performance plasmonic light absorber. ROO Fig. 20. Schematic of a metamaterial perfect absorber achieving near-unity optical absorption using ultrathin plasmonic nanostructures with thicknesses smaller than the hot electron diffusion length. By integrating the metamaterial with a silicon substrate, a broadband and omnidirectional hot electron photodetector is obtained, showing a very high photoresponsivity. Dimensions of the metamaterial perfect absorber L=185nm–195nm; P=340nm–360nm, respectively and H=135nm. within the embedding film [25]. Near close-packed configuration of nanocrystals and their relative orientation in this packing cre- ate collective behaviors of unique optical signature [26,27]. Controlling aggregation and packing density of nanocrystals inside a hostmatrix can be expected to generate a new class of high performance plasmonic light absorber. ED PR The introduction of nanoparticles in the fluid of a thermal solar module increases the solar radiation absorption when compared to the fluid only. 2.2. Plasmonic nanoparticles light absorbers The material and size of the nanoparticles are carefully chosen in order to maximize the absorption and minimize the scattering of the nanoparticles in the spectral domain of interest. Indeed, the maximum solar irradiation intensity is located at 475nm. It has been reported, that the localized surface plasmon resonance absorption peak of core-shell Ag TiO⁠2 nanoparticles is centered around 474nm, making them an ideal candidate to increase the light absorption at the maximum solar irradiation intensity [28]. To increase the absorption on a larger wavelength band, from 250 to 1000nm, nanoparticles made of other materials can be used. It has been shown, that homogeneous Ti and core-shell Ti TiO⁠2, Ni NiO nanoparticles introduced in water at concentrations of 10⁠9 - 10⁠10cm⁠−3 are suitable to increase the absorption [29]. CTED LSPRs, can also be achieved in semiconductor quantum dots (QDs) with appreciable free carrier concentrations (see Fig. 7) allow- ing active on-chip control of LSPR responses. As shown in Fig. 7, the LSPR frequency can be tuned from near infrared (NIR) to far infrared (FIR) and even THz according the free carrier density and the nanosphere diameter. Thus degenerately doped semiconductor QDs allow realization of LSPRs and quantum-confined excitons within the same nanostructure, opening up the possibility of strong coupling of photonic and electronic modes, with implications for light harvesting, nonlinear optics, and quantum information pro- cessing [30]. CT Light absorbers are achieved by different means: MIM structures, plasmonic MIM structures or by solely taking advantage of the optical properties of nanoparticles. Each described absorber is based on a different optical phenomenon. EC The usage of plasmonic nanoparticles is of particular interest because of the large panel of optical properties they produce. In the recent years, the research on nanoparticle production made huge advances and nowadays a large variety of nanoparticle are produced chemically. This will be described in the following part. 2.2. Plasmonic nanoparticles light absorbers The optical properties of such films depend not only on the natural, size and shape of the metal nanocrystals, but are also function of the spacing and arrangement of the nanocrystals U Fig. 19. Schematic of a microbolometer membrane covered by a metamaterial consisting of a metal layer/dielectric (Si⁠3Nx)/periodic metal patch structure. tion where the plasmonic absorption of the Ag nanocubes occurs, while spin coating on glass leads to composite layers in which absorption is controlled by the amount of embedded cubes. The optical properties of such films depend not only on the natural, size and shape of the metal nanocrystals, but are also function of the spacing and arrangement of the nanocrystals U Fig. 19. Schematic of a microbolometer membrane covered by a metamaterial consisting of a metal layer/dielectric (Si⁠3Nx)/periodic metal patch structure. tion where the plasmonic absorption of the Ag nanocubes occurs, while spin coating on glass leads to composite layers in which absorption is controlled by the amount of embedded cubes. The optical properties of such films depend not only on the natural, size and shape of the metal nanocrystals, but are also function of the spacing and arrangement of the nanocrystals 10 L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx ROOF Fig. 20. Schematic of a metamaterial perfect absorber achieving near-unity optical absorption using ultrathin plasmonic nanostructures with thicknesses smaller than the hot electron diffusion length. By integrating the metamaterial with a silicon substrate, a broadband and omnidirectional hot electron photodetector is obtained, showing a very high photoresponsivity. Dimensions of the metamaterial perfect absorber L=185nm–195nm; P=340nm–360nm, respectively and H=135nm. within the embedding film [25]. Near close-packed configuration of nanocrystals and their relative orientation in this packing cre- ate collective behaviors of unique optical signature [26,27]. Controlling aggregation and packing density of nanocrystals inside a hostmatrix can be expected to generate a new class of high performance plasmonic light absorber. TED PROO g ultrathin plasmonic nanostructures with dband and omnidirectional hot electron ph m–195nm; P=340nm–360nm, respectivel tals and their relative orientation regation and packing density of na monic light absorber. reases the solar radiation absorpti en in order to maximize the absorp the maximum solar irradiation int orption peak of core-shell Ag Ti absorption at the maximum solar i 00nm, nanoparticles made of oth O nanoparticles introduced in wate reciable free carrier concentration quency can be tuned from near in ere diameter. RE 3. Chemical synthesis of nanoparticles of various shapes and sizes Organic and inorganic reducing agents, such as sodium citrate, ascorbate or sodium borohydride (NaBH⁠4), are used to reduce silver and gold [41–43]. Changing the reaction temperature, stirring speed and speed at which the solution is added induce a size and size dispersion change [59].i CTED Nanocubes and nanoprisms can be achieved by a two-step seed based synthesis [42,43,46]. The first step produces spherical seeds growing into the desired nanoparticle shape in the second step of the synthesis. Facet-specific capping agents or the solvent itself induce crystallographic defects on the seeds. These defects then induce a face selective growth in the second step of the nanoparticle growth. Indeed, the metal salt added in the second step of the synthesis will preferably be deposited on the other surfaces because the defects are the sites of highest energy [43,46]. Nanocubes are achieved by adding PVP while growing the seeds. In this case, the PVP protects the seeds from aggregating and, as described above, it preferentially binds to the {100} facets leading to a passivation of these facets [42]. RECT After synthesis, the nanoparticles are usually coated with additional surfactant to insure stability of size and shape and prevent aggregation of particles in solution. Common surfactants are sodium citrates [46], sodium borohydride [46] or PVP [42,60,61]. They are added in excess at the end of the synthesis to form a protective shell of a few nanometers around the nanoparticle. Aside the protecting properties, interesting optical properties arise when the polymer shell around the metallic core is modified. Indeed, as plas- monic responses of nanoparticles are environment sensitive, a 5nm thin layer of PVP around synthesized nanocubes can, for instance, shift the absorption peaks wavelength of several tens of nanometers. Other examples occur using inorganic core/shell nanoparticles and were applied for several applications in the biomedical domain [62] (see Fig. 9) such as imaging or sensing [54]. ORRE It must be noted that, besides the most studied gold and silver nanoparticles, other plasmonic metallic materials have been devel- oped. For instance, copper nanoparticles can be produced in large quantities by chemical synthesis for applications such as conductive inks [60]. Aluminum also can be synthesized as nanospheres by hydrolysis [64] or as nanowires and nanocubes by plasma arc dis- charge [65]. The major drawback of these materials, however, is their fast oxidation in presence of oxygen, even though the oxidation rate can be reduced [66]. RE 3. Chemical synthesis of nanoparticles of various shapes and sizes ORR The chemical and physical properties of nanoparticles depend on their material, size, shape and environment. Nanoparticles made from noble metals such as silver and gold are studied for a long time for their extraordinary optical responses attributed to localized plasmon resonances in the visible. As a classical example, the origin of the intriguing colors of the Lycurgus cup, made by ancient Romans, are embedded silver and gold nanoparticles, but the synthesis process was not reported, or got lost over the centuries, mak- ing us believe that it was rather an accidental than wanted process. Faraday conducted the first scientific study of synthesized noble metal nanoparticles in 1857. He reduced gold chloride by phosphorus and analyzed their optical properties. Following this, the ability to tune the optical properties of plasmonic nanoparticles leads the search on syntheses of nanoparticles of different sizes and shapes.i UNCO The synthesis methods can be separated into “bottom up” and “top down” approaches. The first approach starts with precursor material and through various reactions individual nanoparticles are produced. The produced nanoparticles present size dispersion, but the quantity is rather high and it is easy to implement. The second approach, “top down”, requires a large amount of energy, which will release nanoparticles from a target, e.g. laser ablation [31], arc discharge [32,33], chemical vapor condensation [33], ball milling [33], hydrogen plasma [33]. The main advantage of the “top down” methods is the high quality and size homogeneity of produced nanoparticles. However, the need for high energy makes the production expensive and mainly limited to applications where only a small quantity of nanoparticles is needed. As a large quantity of nanoparticles is required to efficiently modify the optical properties of thin film layers, this approach is not suitable in this case. On the contrary, chemical wet syntheses allow us to meet this criteria as the wet synthesis process is scalable and economically attractive. This process is therefore described more in details in the following. UN From the diversity of existing “bottom up” approaches, e.g. microemulsion [34], thermal decomposition [35], hydrothermal synthesis [36,37], sol-gel [38], sonochemical [39], radiolysis [40], the most commonly used method is chemical reduction of metallic ions by a chemical agent [41–43]. Since roman times, where church glasses were colored intentionally or not by metal 11 L. Escoubas et al. RE 3. Chemical synthesis of nanoparticles of various shapes and sizes Progress in Quantum Electronics xxx (2019) xxx-xxx OOF lic dust, tremendous progress has been made in the understanding and developing of wet-chemical syntheses of various materials, shapes and sizes [43]. The nanoparticle growth in chemical syntheses follows a model described in 1950 by LaMer [44] and schema- tized on Fig. 8. Metal salts are dissolved in a solvent and slowly added to the reducing agent solution. On the nanometer scale, ho- mogeneous nucleation occurs in a first step of the synthesis when the minimal concentration for nucleation is reached (C⁠min). The nucleus can then grow or vanish at any time depending on the critical radius in the system. Beyond this critical threshold, all formed nuclei are stable and will grow but below this critical threshold, nuclei are instable and vanish. In the nanoparticle growth step, the stable nuclei grow as long as further metal salts are added and as long as the concentration is above the minimal concentration for heterogeneous growth on seeds (C⁠min, seeds). The resulting solution changes then its color, as the nanoparticles grow. If the size dis- persion is high, a small growth or size reduction might take place over time to make the nanoparticle more uniform. The produced nanoparticles are said to be monodisperse in size when the standard deviation is equal or less than 5%. PROO p p q The most energetically favorable nanoparticle shape is a sphere. Other shapes are achieved by dividing the nanoparticle growth into two or more synthesis steps. In the specific case of silver nanoparticles, facet-specific capping agent can be used to control the shape evolution of silver nanoparticles in a seed-mediated synthesis. For instance, citrates and PVP were proven to selectively bind to the (111) and (100) facets of silver seeds, respectively, stabilizing these facets more than others and thus favoring the formation of silver nanoparticles with (111) or (100) facets exposed on the surface [45]. Following such strategies, silver nanoparticles could be obtained with various shapes. A non-exhaustive list of produced shapes are: spheres [46], disks [47], plates [42], prisms [48], dumbells [49,50], pyramids [51], cubes [42,43], cages [42], nanowires [42], nanoflowers [52–56], nanostars [57], multipods [58]. It seems that the only limit to nanoparticle synthesis is our imagination. Each of these nanoparticles has then different physical and chemical properties. D P For nanospheres, a single reduction step is required. RE 3. Chemical synthesis of nanoparticles of various shapes and sizes These materials also exhibit high losses at optical frequencies due to electron interband and intraband transitions and their electron densities are not easily tuned. This motivates the search for other plasmonic materials such as doped semiconductors with metal like behavior i.e. oxides, nitrides and chalcogenides [67–70]. COR The diversity of produced nanoparticles leads to a diversity of chemical and especially optical properties. The optical properties of the described nanoparticles can then either be experimentally characterized once produced or studied by computer modeling. In the following chapter, we firstly describe computer modeling as a convenient tool to foresee the optical properties of the nanoparticles and to validate the measured optical properties, obtained by spectroscopic ellipsometry for instance. In a second part, we describe the spectroscopic ellipsometry measurement technique. 4.2. Mie theory CTE The optical properties of plasmonic nanoparticles are numerically calculated using the Mie Theory [76]. Mie identified the need for a theory linking the particle size and shape to the optical properties of a colloidal metal solution for particles much smaller than the wavelength. He solved the Maxwell equations for spherical particles of sizes smaller than the wavelength by switching to spherical coordinates, giving an analytical solution for the scattering and absorption of metal nanospheres in any media. The Mie theory can then be applied to calculate the absorption and scattering cross sections. The dipolar absorption σabs and scattering σscat cross section of a spherical particle depends on the polarizability α of the particle: RR where the polarizability depends on the volume V of the particle and the dielectric constants ε of the sphere and εm of the medium. The derivation of the above equations is described in detail elsewhere [77].i RR where the polarizability depends on the volume V of the particle and the dielectric constants ε of the sphere and εm of th di Th d i ti f th b ti i d ib d i d t il l h [77] RR where the polarizability depends on the volume V of the particle and the dielectric RR the medium. The derivation of the above equations is described in detail elsewhere [7 COR The calculation of the cross sections gives us a first insight on the resonance wavelength of the nanoparticle and whether the nanoparticle will predominately absorb or scatter the incoming light at this wavelength. At the resonance wavelength, plasmonic nanoparticles strongly interact with the incoming light, i.e. the cross section exceeds the geometrical size of the nanoparticle. This simple calculation guides us in the choice of the right nanoparticle material and size with our application in mind. The knowledge of the size of the nanoparticle and the dielectric constant of the nanoparticle and its medium are necessary for the calculation. The cross sections of silver nanospheres are calculated by using the optical indices of bulk silver. NCO It is worth noting that the Mie theory describes the optical behavior of a single nanosphere in a homogeneous surrounding medium. The interaction and coupling between particles are not taken into account. The Mie theory is therefore useful for colloidal solutions, in which the nanoparticles typically do not interact, and for thin film layers with low nanoparticle density. NC 4. Computer modeling and optical characterization of embedded nanoparticles UN The optical properties of nanoparticles can be studied by different means. One common and convenient tool is computer modeling. Different simulation and numerical calculation solutions exist for this purpose and will be briefly described in the following: the Mie theory is limited to spherical particles, while the discrete dipole approximation and the finite difference time domain methods can be used for any shape.i U In the following, finite difference time domain (FDTD) simulations are described in detail as it is a convenient software to an- alyze the electric field enhancement on the nanoparticles and their optical properties. The geometry of the nanoparticle is freely chosen, as is the environment medium. In particular, the distance between two, or more, neighboring particles can be studied. 12 L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx F This feature is relevant to study aggregates of nanoparticles. Indeed, from a interparticle distance of a few nanometers, the enhanced electric fields around the nanoparticles couple and consequently alter the optical properties.i F This feature is relevant to study aggregates of nanoparticles. Indeed, from a interparticle distance of a few nanometers, the enhanced electric fields around the nanoparticles couple and consequently alter the optical properties.i OFi From an experimental point of view, spectroscopic ellipsometry of thin film layers is a powerful tool to obtain the complex optical indices of the nanoparticles in their medium. The measurement principle is described in the second part of this chapter. The measure- ment itself is straight forward, but an adequate diffusion model is required to derive the optical indices. 4.1. Computer modeling 4.1. Computer modeling ROO Computer simulations and numerical calculations are powerful tools to design optical absorbers. They can be used to study differ- ent design in order to optimize it and to confirm the measured optical properties. Calculations are cost effective and convenient tools, especially when the complexity of the design increases. Furthermore, they give us insight on the physical phenomena happening at the nanoscale. One example is the visualization of the electric field enhancement on a nanoparticle. RO Different calculation methods, having their advantages and drawbacks, co-exist and must be chosen with care depending on the application. An important feature in calculations and simulations is the material properties, which are either chosen from textbooks, such as from Johnson and Christy [71] or Palik [72], or experimentally measured by spectroscopic ellipsometry.i PR In simulation using finite element analysis, an important aspect for accurate results is the size and shape of the used mesh. De- creasing the mesh size improves the results, but considerably increases the calculation time. Therefore, the mesh has to be carefully chosen. D P The calculation method is chosen depending on the application and the structure of the device. The optical properties of MIM struc- tures can be obtained numerically by transfer matrix method (TMM) [73] and by electromagnetic computer simulations by full-wave solver based on the finite element method. The optical properties of plasmonic nanoparticles are either numerically calculated by Mie Theory, for spherical and ellipsoidal shapes, or simulated by time domain methods, such as FDTD [74], or frequency domain methods, such as FEM [75]. U where ξ0 and Χ are size dependent variables. 4.4. Finite difference time domain (FDTD) simulations 4.4. Finite difference time domain (FDTD) simulations CORRECT Finite difference time domain (FDTD) simulations [86] allows to simulate the light interaction of particles with any shapes. This is particularly interesting for non-spherical nanoparticles, such as nanocubes, nanoprisms or nanostars produced by facile chemical syn- theses. The optical properties of single nanoparticles, i.e. the absorption, scattering and extinction cross sections, are easily computed and the electric field enhancement on the nanoparticles are visualized. In order to compare the measured optical properties with the simulation, the optical properties of nanoparticles distributed in thin film layer are calculated. In this configuration, the electromag- netic coupling between the nanoparticles in the layer can be studied. FDTD simulations are a convenient tool to verify the optical properties of complex samples, such as multilayers or structured surfaces. The main drawback of FDTD simulations is calculation time. In order to increase the accuracy, especially for curved surfaces, very small mesh cells are chosen, leading to time consuming computation. To counter this, a solution is to work with non-uniform meshed regions with a fine meshing around the nanoparticle, as shown on Fig. 10a. A systematic two step analysis of nanoparticles can be efficiently used: first the nanoparticle alone is studied in different media, such as water and polymer, then the nanoparticles are arranged in a periodic pattern. In the case of non-interacting nanoparticles, in other words electromagnetically isolated particles from each other, the periodic pattern is sufficient. If the distance between the nanoparticles is a few nanometers, the enhanced electric fields will couple and the localized plasmon resonance is af- fected [87]. The study of the nanoparticle alone is mainly performed to obtain the absorption, scattering and extinction cross sections. For this, a total field scattered field source (TFSF) is used, as schematized on Fig. 10b. The electromagnetic radiation is along the blue arrow and its polarization along the purple arrows. The absorption monitor is placed around the particle inside the source region, region 1, and the scattering monitor is placed outside the source region, region 2. Outside the TFSF region, the incident fields are subtracted from the total fields, i.e. only the field scattered by the particle remains.i Cii The cross section is defined by the relation: UN where P is the scattered (respectively absorbed) power and I is the source intensity. 4.3. Discrete dipole approximation (DDA) 4.3. Discrete dipole approximation (DDA) TED PRO The absorption and scattering cross sections of arbitrary shaped nanoparticles can be numerically studied by using the discrete dipole approximation (DDA) method. This method was first described in 1964 by DeVoe [81] to calculate the optical properties of molecular aggregates and improved by Purcell and Pennymaker [82] to calculate the optical properties of interstellar dust. The DDA method approximates the studied object with a cubic array of dipoles, each dipole having a defined polarizability. In other words, the object is decomposed into a finite number of points for which the Maxwell equations are solved. The different dipole points interact, i.e. they are electromagnetically coupled. Therefore, the DDA method is sometimes referred to as the coupled dipole approximation (CDA). The accuracy of the calculation strongly depends on the number N of dipoles point chosen, especially for curved surfaces. Increasing N leads to accurate results and increases the computation time consequently. The DDA method is an accurate method to calculate the absorption and scattering cross sections [83]. Dunklin et al. [84] recently showed, that the optical properties of different densities of gold nanoparticles in polymer layers are successfully calculated by DDA. The method also allows a differentiation between dipolar and quadrupolar contribution, which is not straightforward with FDTD simulation. Zhou et al. [85] used DDA calculations to analyze the optical properties of silver nanocubes of edge sizes from 15 to 200nm. The physical origin of each resonance peak is determined, i.e. dipolar and/or quadrupolar resonances, and their size dependence is analyzed. DDA calculations are a convenient tool when used together with experimental measurements to identify the measured absorption and scattering peaks. 4.2. Mie theory An analytical formula has been derived for ellipsoidal particles and is typically called Modified Long Wavelength Approximation (MLWA) [78–80]. The formula takes into account the two axes of the particle, which lead to two distinct resonance wavelengths. N The polarizability for an ellipsoid of minor axis a and major axis b is: UN U where ξ0 and Χ are size dependent variables. 13 L. Escoubas et al. L. Escoubas et al. L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx Progress in Quantum Electronics xxx (2019) xxx-xxx OO The Mie theory offers simple equations to calculate the optical properties of nanospheres, but with the emergence of chemical and lithography methods to produce differently shaped nanoparticles such as nanocubes and nanoprisms, the use of the Mie theory is not sufficient anymore. For non-spherical nanoparticles, different methods are used. 4.4. Finite difference time domain (FDTD) simulations In the TFSF configuration, the σabs is simply the complementary to the total cross section measured in the region 1. As only the scattered light reaches region 2, σscat is measured in region 2. The extinction cross section is then the addition of the absorption and scattering cross sections: 14 14 L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx L. Escoubas et al. F Generally, the normalized scattered (respectively absorbed) efficiency is used to remove the size dependence. The efficiency sim- ply corresponds to the cross section normalized to the geometrical area A of the particle: OFi OF Once the absorption and scattering peaks for every shape are computed, the electric field enhancement, around and in the particle, are visualized to gain understanding on the origin and the nature of the peaks. ROO Perfect silver 35nm edge size nanocubes in PVP matrix exhibit an absorption peak at the wavelength 350nm. Several hypotheses are found in literature on the origin of this peak as shape dependent [88], quadrupole resonance [89], due to aggregates or to silver interband transition [90]. Fig. 11 below is an example of the FDTD computed amplitude of the electric field enhancement on a 35nm edge size silver cube in a PVP polymer matrix at λ=350nm. The enhancement is maximal at the corners of the cube. Therefore, we can conclude that the peak is related to the shape of the particle.i PRO Long et al. used the electric field enhancement visualization of silver nanospheres to investigate the lasing emission enhancement at the exciting wavelengths. The simulations confirmed the coupling between the plasmonic resonance of the silver nanosphere and the laser emission [91]. Sun et al. [92] analyzed the electric field enhancement of gold-silver core-shell nanorods deposited on dif- ferent thicknesses of a PMMA layer. It was showed that increasing the thickness of the polymer layer has an influence on the electric field enhancement and on the electric field distribution on the nanoparticle. The simulations allowed the authors to determine the optimal PMMA thickness, 56nm, to maximize the electric field enhancement, 27-fold enhancement [92]. ED P For systems where the nanoparticles do not interact, i.e. there is no electromagnetic coupling between the nanoparticles, the study of the nanoparticle alone is sufficient. 4.4. Finite difference time domain (FDTD) simulations For systems where the nanoparticles interact with each other, such as in thin film layers of randomly distributed nanoparticles, a second simulation step is necessary to study the effect of the coupling on the optical properties. Different simulation configurations are possible, as the nanoparticles can be distributed in a periodic lattice or pseudo-randomly. The squared lattice distribution is the simplest to implement, as it can be automatically generated, and the fastest in calculation time, as the calculation can be done on a single unit cell. Fig. 12 displays an example of a silver nanoprisms in a squared periodic pattern. The distance between the particles is small enough for the enhanced electric fields to interact.i CTEi For pseudo random distributions, different configurations have to be taken into account. Based on the periodic pattern of nanoprisms displayed on Fig. 12, a pseudo random configuration could be obtained by either modifying the orientation of one or more prisms. By turning a prism by 45° or 90°, the incident electromagnetic wave will not exalt the plasmonic resonance in the same manner. An alternative consists on modifying the distance between the nanoparticles to have coupled interactions and uncoupled interactions, i.e. nanoparticles behaving like single nanoparticles.i RECT As an example of specific optical properties obtained from a random distribution of interacting nanoparticles, in a recent theo- retical paper of B. X. Wang [93], a strong-backscattering phase function is studied. It is demonstrated that in particular conditions a disordered medium composed of randomly distributed silicon nanoparticles exhibits a strongly negative scattering asymmetry in the near infrared due to multiple light scattering. As the concentration of scattering particles rises, the backscattering is also enhanced. Predicting and controlling wave propagation in random particulate materials allows people to manipulate the scattering and absorp- tion of radiation. It paves the way to applications such as imaging through turbid media or radiative cooling of coatings by efficient reflection of incident solar power. RRl To conclude, numerical studies are a convenient way to obtain fast results on the optical properties of nanoparticles and to study their behavior either alone or when electromagnetic coupling occurs. 4.5. Characterization 4.5. Characterization UNCOR Following the chosen approach, computations either precede experimental characterization or confirm them. Synthesized nanopar- ticles can be optically characterized as prepared in solution and embedded in a polymer thin film layer. The optical measurements of solutions and of thin films are similar. In the following, only thin films will be considered. In a first step the optical properties, such as transmittance, total reflectance and diffuse reflectance, of the nanoparticles in different media are determined through spec- trophotometric measurements. These quantities are only intensities and strongly dependent on the characteristics of the samples: the thickness of the thin film layers, which changes with the deposition speed, the viscosity of the solution, the ambient temperature or the substrate. Spectroscopic ellipsometry measurements are used to obtain the complex optical indices of the thin film layers, which are independent on the thickness of the layer or the substrate. Spectroscopic ellipsometry is a powerful tool, which requires a general understanding, i.e. the wavelength and broadening of the different absorption or reflection peaks of the samples, as the measured data needs to be fitted with a dispersion model to obtain the optical indices. The general understanding is then either derived from computation or spectrophotometric measurements. UN 4.6. Spectroscopic ellipsometry 4.7. The data fit CORREC The optical model consists of mathematical laws, also called dispersion laws, describing the material of each layer. Once the right optical model is found, iterations verifying the Kramers-Krönig relations are necessary to validate the consistency of the determined indices [98]. Numerous dispersion laws, verifying the Kramers-Krönig relations, exist to account for multiple optical properties of probed materials. Ground knowledge of the optical properties, as the absorbance, is therefore required to choose the right laws and reduce the number of variables of the laws: the spectroscopic ellipsometry measurements are performed combined with spectropho- tometer measurements or numerical calculations. The free variables are then used as fitting parameters to increase the agreement of the calculated data with the experimental ones using the Levenberg-Marquardt method [99] as regression analysis. In the case of visible light absorbing silver nanoparticles dispersed in a non-absorbing polymer layer, the dispersion model is composed of a Cauchy law, accounting for the transparent non-absorbing polymer, and Gauss laws or Lorentz laws, accounting for the different ab- sorption peaks of the plasmonic nanoparticles [100–103]. The suited optical model is chosen by analyzing the model that minimizes the root-mean square error (RMSE) and maximize the coefficient of determination. Furthermore, the obtained indices will be used to compute the reflectance R through a transfer matrix method. The computed R is then compared to the measured R to validate the optical model. The difficulty in the data fit lies in the right choice of dispersion laws and in the number of parameters to vary for each law: a model with one Cauchy law and one Gauss law has six variables. The Cauchy, Gauss and Lorentz laws are described in detail in the following. UN 4.6. Spectroscopic ellipsometry an effective medium is sensed as sc PR The embedded nanoparticles in the thin film layers are small compared to the wavelength of the light therefore the measurement does not distinguish between the polymer and the nanoparticles, i.e. an effective medium is sensed as schematized in Fig. 14. PR een the polymer and the nanoparticles, i.e. an effective medium is sensed as schematized in Fig. 14. TED P The measured properties of the effective medium are an average of the properties of each material, i.e. the inclusions and the polymer matrix. It is supposed to recreate the experimental values and simplify the calculations [95]. Si substrates are typically used to maximize the optical index difference between the two layers. The substrate is assumed to be semi-infinite, i.e. there is no reflected light at the backside of the substrate, and its optical indices are known. Layer 2 on Fig. 14 represents a homogeneous layer of effective indices n⁠eff and k⁠eff, whose thickness is measured, for example, by a stylus profilometer. The described system has only two interfaces, but this can also be generalized to a multilayered stack. In the case of a multilayered stack, the thickness of each deposited layer has to be known. Furthermore, it is recommended, that the difference in optical indices of two neighboring layers should be large. If the indices are similar, the phase difference between two layers will not be significant enough to be sensed. The angle at which the measurement is suitably performed is chosen to be near the Brewster angle of the substrate to maximize the intensity of the reflected light [96]. Typically, variable angle spectroscopic ellipsometry (VASE) increases the precision [97].ii UN 4.6. Spectroscopic ellipsometry U Spectroscopic ellipsometry measurement is an indirect technique to determine either the complex optical indices or the thickness of transparent or semi-transparent thin film layers. These two variables are linked: the knowledge of one is necessary to determine the other. 15 L. Escoubas et al. L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx F The knowledge of the complex refractive index allows computing the reflectance and transmittance of any layer thickness on any substrate.i F The complex refractive index is the complex addition of the refractive index n and the extinction coefficient k: Fur- thermore, the complex refractive index is closely linked to the complex dielectric function: [94]. complex refractive index is the complex addition of the refractive index n and the extinction coefficien e, the complex refractive index is closely linked to the complex dielectric function: [94]. i e, the complex refractive index is closely linked to the complex dielectric function: [94]. OF A spectroscopic ellipsometry setup is schematized on Fig. 13. The light emitted by the light source is linearly polarized by the polarizer P. Upon reflecting onto the sample, the light becomes elliptically polarized.l OOl To compare the reflected light with the incident light, the light is once again linearly polarized by passing through the analyzer A before reaching the detector. In other words, the measurement consists of the analysis of the polarization change of a reflected light beam by a thin film layer. This is translated into a change of Fresnel reflection coefficients r⁠s and r⁠p, which is measured by the quantity ρ: PRO where ψ symbolizes the amplitude change and Δ the phase difference of the reflected light compared to the incident light. Each inter- face yields a different result.i PRO where ψ symbolizes the amplitude change and Δ the phase difference of the reflected light compared to the incident light. Each inter- face yields a different result.i PR rticles in the thin film layers are small compared to the wavelength of the light therefore the measuremen PR ded nanoparticles in the thin film layers are small compared to the wavelength of the light therefore the m guish between the polymer and the nanoparticles, i.e. an effective medium is sensed as schematized in Fig. PR The embedded nanoparticles in the thin film layers are small compared to the wavelength of the ligh does not distinguish between the polymer and the nanoparticles, i.e. 4.10. Lorentz law ECT Based on the simple model of a mass and spring system, the Lorentz model describes the classical theory of light-matter interaction and the frequency dependent polarization due to bound charges. The incident electromagnetic field induces vibrations of the electrons behaving as harmonic oscillators. EC g The Lorentz oscillator (LO) is characterized by an energy E⁠0, an oscillator strength f and a broadening Γ RREC OR The use of the Lorentz law therefore adds two variables to the analysis. OR A constant ε∞common to all the complex dielectric functions, e.g. LO and GO, is added to describe the oscillators outside of the measured spectral domain. This constant is another variable added to the optical model. 4.9. Gauss law 4.9. Gauss law RO The Gauss law is defined as an oscillator centered at an energy E⁠0, of amplitude Amp and broadening Br. The dielectric function is then defined as: RO ss law is defined as an oscillator centered at an energy E⁠0, of amplitude Amp and broadening Br. The dielei d as: RO as an oscillator centered at an energy E⁠0, of amplitude Amp and broadening Br. The dielectric function is D PR ED where is the full width at half maximum of the oscillator and D is the Dawson's integral [104] TED When describing a localized surface plasmon resonance absorption peak with a Gauss oscillator (GO), the energy E⁠O is related to the measured plasmon resonance energy of the nanoparticles. Amp is related to the intensity and Br to the width of the absorption peak. The use of the Gauss law therefore adds two variables to the analysis. 4.8. Cauchy law NC The Cauchy law typically describes transparent materials: U UNC UNC U where the parameter n∞is dimensionless and n(λ) tends to n∞at high energy, A and B characteriz U where the parameter n∞is dimensionless and n(λ) tends to n∞at high energy, A and B characterize the curvature and the 16 16 Progress in Quantum Electronics xxx (2019) xxx-xxx L. Escoubas et al. F amplitude in the visible and the UV respectively. The parameters C, D and E are similar to n∞, A and B for the extinction coefficient k.i F For non-absorbing materials, the extinction coefficient is simply set to zero over the whole spectral range. The parameters of a non-absorbing Cauchy law, as used in the following, are: OOF O The use of the Cauchy law therefore adds three variables to the analysis. OO The use of the Cauchy law therefore adds three variables to the analysis. 4 9 G l The use of the Cauchy law therefore adds three variables to the analysis. 4.11. Effective medium theory 4.11. Effective medium theory UNCO When using an effective medium approach, the samples composed of nanoparticles embedded in a host material are considered as a homogeneous material characterized by an effective medium εeff The Maxwell Garnett formula shown below, allows linking the effective medium dielectric function to the dielectric functions of each material constituting the effective medium [105,106]. This approach takes into account the first order approximation of the Rayleigh formulae. This simple theory, compared to other such as described in Ref. [107], does not take into account the multiple scattering of the particles in the layer and the polarization of the light. The multiple scattering induces macroscopic optical behavior and is therefore linked to a property of the effective medium. The random orientation of the nanoparticles in the layers should make the optical properties polarization independent. For the results presented in the fourth part of this article, the polymer host matrix is seen as a medium and the silver nanoparticles as inclusions within the medium, therefore: U 17 17 L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx F where δi is the volume fraction of the inclusions and εi the dielectric function of the inclusion, εm is the dielectric function of the medium and εeff the dielectric function of the effective medium. For the equation to be valid, the volume fraction should not exceed one third [81]. F where δi is the volume fraction of the inclusions and εi the dielectric function of the inclusion, εm is the dielectric function of the medium and εeff the dielectric function of the effective medium. For the equation to be valid, the volume fraction should not exceed one third [81]. OF Furthermore, the Maxwell-Garnett formula is only valid for spherical inclusions. For spheroids, a factor of depolarization has to be taken into account [99,105,108]. Analytical expressions for other shapes, such as nanoprisms and nanocubes, have not been described yet in the literature. ROO To conclude, there are a variety of modeling tools to study the optical properties of nanoparticles. FDTD calculations are used for nanoparticles of any shape in any medium, but it must be noted, that the calculation times are long. Then, the experimental character- ization method of spectroscopic ellipsometry is described. The measurement is an indirect technique to determine the effective optical indices of a layer. 5.1. Optical indices n and k D P The knowledge of the optical indices of a thin film layer, determined through spectroscopic ellipsometry, allows a complete un- derstanding of the interaction of light with the layer. In the scope of a highly efficient light absorber, the knowledge of the optical indices of a single layer allows the calculation of multilayers to optimize the absorption. As described above, the determination of the optical indices depends on the right choice of the dispersion model. Different samples and their adapted dispersion models are discussed below. 4.11. Effective medium theory In order to obtain the optical indices, it is necessary to derive an optical model composed on several dispersion laws. In the following chapter, silver nanospheres and nanocubes randomly embedded in a PVP layer are studied with a suited optical model. 5. Optical properties of plasmonic nanoparticles absorbers 5.1. Optical indices n and k EC 5.3. Non-electromagnetically coupled nanospheres and nanocubes in PVP EC 5.3. Non-electromagnetically coupled nanospheres and nanocubes in PVP NCORRE One interesting example concerns the spectroscopic ellipsometry characterization of a blend of silver nanospheres and silver nanocubes embedded in a PVP thin film. The silver nanospheres in PVP and nanocubes in PVP are firstly characterized independently. Then both shapes, nanospheres and nanocubes, are randomly distributed within the same thin film. In the scope of spectroscopic ellip- sometry measurements, the knowledge of the optical properties of each shape is necessary in order to choose an appropriated optical model. Indeed, the model is composed of a Cauchy law, accounting for the optical properties of the non-absorbing host matrix (PVP in this example), and several Lorentz laws centered at the localized plasmon resonance peaks of the nanospheres and nanocubes. Fig. 15a describes the optical indices of such a layer of nanospheres and nanocubes. The peak centered at 420nm is associated with the dipolar resonance of the nanospheres, while dipolar resonance of the nanocubes is situated at 450nm. The peak centered at 350nm is linked to the cubic shape of the nanocubes. The intensity of each peak is related to the concentration of each type of nanoparticles. For example, by adding more nanocubes to the layer, the intensity of the peaks at 350nm and 450nm is increased. As the nanoparticles in the layer are not electromagnetically interacting, the optical model used to fit the spectroscopic ellipsometry data is a simple ad- dition of the model used for each geometry, nanosphere or nanocube. This is visualized on Fig. 15b, where the normalized extinction coefficient of the layer containing nanospheres and nanocubes is compared to the individual extinction coefficients of nanospheres and nanocubes in separated layers. Electromagnetic coupling between particles in a layer would occur if the distance between the particles is in the order of a few nanometers. This can be achieved by increasing the concentration of nanoparticles in the layer or by forming aggregates [114,115]. The electromagnetic coupling then induces a shifting of the plasmon resonances. UN The values of the optical indices n and k (Fig. 15a) are directly linked to the concentration of nanoparticles within the PVP thin film. By increasing the concentration, the absorption at the plasmonic resonance wavelength is increased. In this example, as shown in Fig. 15b, using a blend with two differently shaped nanoparticles, nanospheres and nanocubes, allows a broadening of the absorption from 400 to 600nm. 5.2. The dispersion models CTE The spectroscopic ellipsometry measurements of noble metal nanoparticles on substrates and embedded within various dielectric matrices are studied by several authors [100–103,109–111]. Concerning the non-absorbing host matrix, the use of a Cauchy law is well established. This is not the case for the optical properties of noble metal nanoparticles as shown on Table 1, which lists the dif- ferent laws used and their physical meaning. CT overview of the different laws used and a guideline for determining the optical model. Depending on th mple, the right model has to be chosen. CT Table 1 gives a broad overview of the different laws used and a guideline for determining the opti structure of the probed sample, the right model has to be chosen. EC In the following, the model composed of a Cauchy law and several Lorentz laws is described in more EC In the following, the model composed of a Cauchy law and several Lorentz laws is described in more detail. 5.4. Interaction of light between nanoparticles D PROOF Controlling the scattered field of nanoparticles in interaction is of high interest for new applications needing dynamic devices such as in optical communications and in the foreseen optical computing. Components such as the all-optical nano-switch based on the accurate control of the interaction between neighboring scatterers, semiconductor or dielectric nanoparticles, have been predicted and experimentally demonstrated [117–119]. The spatial distribution of the scattered fields of the nanoparticle and the distance be- tween them allows controlling the electromagnetic interaction of the nanoparticles. Indeed, at the beginning of the 1980s, Kerker et al. published the basis of the modeling of interacting scatterers in specific conditions [120]. Considering spherical particles exhibiting both electric and magnetic responses, in the Rayleigh limit, they studied the relation between the scattering coefficients of the Mie theory, which is a multipolar decomposition to calculate the scattering and absorption cross sections. This decomposition involves coefficient associated to the electric behavior and other coefficients corresponding to the magnetic one. For dipole-like particles, only the two first Mie coefficients, one electric and one magnetic, are not negligible. Kerker's conditions correspond to interferences of the dipole scattered fields producing a zero scattering in either the forward or the backward direction. A directional control over the global scattered field can be achieved using the coherent interaction between electric and magnetic resonances. The shape [121] and the size [122] of the nanoparticles and their distance are the main parameters governing Kerker's conditions. In a recent paper by R. Vergaz [123], two nanoparticles satisfying Kerker's conditions in an optimal configuration are presented (see Figs. 16 and 17). Indeed with this dimer of nanoparticles, two interferential effects are possible, one between the scattered field of each nanoparticle and the background-incoming field and one between the scattered fields of each component of the dimer. Thus constructive and destructive interferences appear depending on the distance between the nanoparticles. Then one can control the spatial distribution of light, and more precisely obtain a maximum variation of light intensity in the gap region, by manipulating these interferences. This dimer con- figuration can be used as the base for the design of the all-optical nano-switch. CTED g g p Fig. 16 describes the system composed of two spherical silicon nanoparticles separated by a distance d, one particle having a radius R⁠1 of 82nm and a second one with a radius R⁠2 of 97nm. 5.5. Collective behavior of aggregates NCOR One example of creating collective response of silver nanoparticle in a controlled way was presented by G. K. Laxminarayana et al. [124]. A novel, modular approach to Ag nanoparticle self-assembly utilizes polymer templating to control meta-atom size and geometry. Colloidal nano-crystals (NCs) are deposited onto the polymer support. They serve as the initial binding platform and are called NCo. Using solvent or thermal annealing of the polymer allows then immobilizing and embedding the NCo into the polymer. Their exposed surfaces are chemically modified with a covalent molecular linker such as a dithiol. Finally, by introducing a second particle (NC1), which reacts with the molecular linker, NC meta-atoms are formed. Repeating this protocol allows producing hierar- chical or dendritic NC motifs (see Fig. 18A–C). Horizontal and vertical nanocube dimers were successfully fabricated with remarkably high yield. As it can be seen in Fig. 18 (G-I), the assembly of the Ag nanocubes into aggregates of controlled number of cubes and organization generates specific optical responses related to their collective properties. EC 5.3. Non-electromagnetically coupled nanospheres and nanocubes in PVP U The knowledge of such optical indices allows one to design the structure depending on the applications. 18 Progress in Quantum Electronics xxx (2019) xxx-xxx L. Escoubas et al. L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx 5.4. Interaction of light between nanoparticles The radius has been chosen such that the first particle (R⁠1=82nm) satisfies the zero-backward (ZB) scattering condition (see Fig. 156 on the left), while the second one (R⁠2=97nm) fulfills the minimum for- ward scattering condition (MF) at the incident wavelength 700nm (see Fig. 16a on the right), but there is still an appreciable electric field (red color of the sphere on the illuminated side). Thus the scattered radiation by the nanoparticles can be directed very selec- tively, which allows controlling the overlapping of the fields in a dimer: as a function of the direction of illumination of the dimer, the scattered light can be directed outwards or towards the gap between the nanoparticles. Thus, depending on the illumination side of the dimer, a maximum or a minimum of the scattered field could be observed in gap. For high values of the gap “d”, nanoparticles are considered as isolated and there is no interaction between them but smaller distances allow interferential phenomena to be obtained.i REC In the configuration of Fig. 16a, with light impinging the particle with the largest radius (97nm) from the right, the electromag- netic fields scattered by the particles in the gap are low and as seen in Fig. 17a and b, there is now interference phenomena in the gap. On the other hand, Fig. 17c, which is corresponding to Fig. 16b with light impinging the particle with the smallest radius (82nm) from the left, we observe strong electromagnetic fields scattered by the particles in the gap, but there is a destructive interference phenomena in the gap. By changing the “d” value from 120nm to 445nm, constructive interference can be obtained in the gap as shown in Fig. 17d. Progress in Quantum Electronics xxx (2019) xxx-xxx 6.2. Photodetection D PR T. Maier and H. Brueckl [127] associate a microbolometer, made from a Si⁠3Nx membrane, with a metamaterial absorbing incident wavelengths in a resonance domain. The metamaterial is directly built above the microbolometer membrane (Fig. 19). It consists of a metal layer/dielectric (Si⁠3Nx)/periodic metal patch structure. The resonance peaks of this structure allow the microbolometer to become wavelength selective by only optimizing the geometry of the metamaterial elements. This selectivity can be tuned between 2.9μm and 7.7μm by adjusting the dimensions of the metal patchs and their periodicity. The absorption peak can reach up to 88%. The structure is very weakly sensitive to the angle of incidence of radiation because the resonance mode used is of the gap-plasmon type [128] (see part 1). In addition, the use of a metallic layer covering the entire surface changes the conduction heat in the structure by increasing its heat capacity, which leads to an improvement of the response time of the microbolometer. ECTED After excitation, surface plasmons can lose their energy in the form of photon re-emission or non-radiatively in the form of en- ergetic electrons or 'hot electrons’. Recently, hot carriers (hot electrons) have sparked a strong interest because they can be useful for many applications such as photodetection, photovoltaic devices, photocatalysis or surface imaging. Hot carriers are used typically to the photodetection with a structure of Schottky type barrier consisting of a thin metallic layer in contact with a semiconductor material. In the article by W. Li and J. Valentine [129], the authors study the addition of a perfectly absorbing metamaterial at the top of the Schottky barrier (Fig. 20). Thus, they are able to strongly increase the photoresponsivity of the device and obtain a sili- con photodetector sensitive in the infrared domain well below the silicon bandgap energy. In addition, the photodetector response is broadband with a photoresponsivity larger than 1.8mA/W for wavelengths ranging from 1200nm to 1500nm and insensitive to the polarization state through the use of square resonators. 6.1. Spectral filtering 6.1. Spectral filtering ROOF It is possible to create artificial materials able to transmit the visible light and totally absorb the infrared (IR) one. Applications concerns IR blocking plasmonic glass windows. In the articles presented by Y. Qin et al. [125] and L.V. Besteiro et al. [126], the authors use a hexagonal matrix of polystyrene latex spherical nanoparticles (PSL) assembled on a flat silicon substrate. They first perform a reactive ion etching (RIE) process to tune the PSL shape, followed by a magnetron sputtering process of a thin copper or aluminum layer, covering the PSL. By then dissolving the PSL in toluene, they obtain a solution containing copper or aluminum ‘nanocups’. These nanocups, which metal thickness can be controlled by the sputtering time and curvature by the dimensions of the polystyrene nanospheres, exhibit localized plasmons. By adjusting plasmonic absorption peaks, through the nanocup geometric para- meters, authors can thus control the spectral transmission of the windows on which the nanocups are deposited and thus maintain transparency properties in the visible domain while blocking IR. NC 6. Few applications of plasmonic light absorbers UN Among the very many articles dealing with localized plasmons and surface plasmons, we highlight here a few applications of par- ticular interest and dealing with spectral filtering and photodetection. 19 Progress in Quantum Electronics xxx (2019) xxx-xxx L. Escoubas et al. 7. Conclusion NCORRE Beyond simply absorbing light and dissipating this energy into thin composite layers, new applications, such as the all-optical nano-switch previously described, are emerging and are using the coupling of light with nanoparticles. These nanoparticles may ex- hibit various shapes, be alone or apart from each other and therefore without electromagnetic interaction, or even very close in inter- action at a controlled distance and so in resonant electromagnetic regime. They can be aggregated and their morphology controlled by chemical routes, in dimer and multimer forms. They can be in interaction with the substrates and thus constituting Fabry-Perot resonators. These nanoparticles can also be metallic and thus allowing localized plasmons to be excited or be used to generate gap plasmons by the electromagnetic interaction with planar substrates. We also find dielectric nanoparticles presenting electric or mag- netic responses under light excitation and leading to remarkable effects under certain conditions, for example in the Kerker's condi- tions. Thus, by combining the expertise of chemists, able to find original synthesis routes to create, to shape, and to connect these nanoparticles together or with substrates of different natures, and the analysis and models of physicists to understand the interaction of light with these nano-objects, it is already possible to create original components, for example the rectenna or ‘hot electrons’ pho- todetector which directly transforms the light energy into an electrical current. But in the future, it will be possible to create much more complex objects, that we probably can not imagine yet, and whose operating principles will certainly rely on controlled assem- blies of nanoparticles with not only specific optical and magnetic properties but also with quantum properties. Uncited references [63]. 20 L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx Acknowledgments OF This work was carried out thanks to the support of the A*MIDEX project (No. ANR-11-IDEX-0001-02) funded by the “Investisse- ments d’Avenir” French Government program and managed by the French National Research Agency (ANR). We also thank the Di- rection Générale de l'Armement (DGA) for a financial support to Miriam Carlberg and Florent Pourcin. References Stefanakos, Design and fabrication of metal-insulator-metal diode for high frequency applications, In: Infrared Technol- ogy and Applications XLIII 10177, International Society for Optics and Photonics, 2017, p. 101772J. PR gy y [6] Chihhui Wu, Neuner Burton III, Jeremy John, Andrew Milder, Byron Zollars, Steve Savoy, Gennady Shvets, Metamaterial-based Integrated Plasmonic Ab- sorber/emitter for Solar Thermo-Photovoltaic systems,Published, IOP Publishing Ltd, 12 January 2012, Journal of Optics, Volume 14, Number 2. PR [6] Chihhui Wu, Neuner Burton III, Jeremy John, Andrew Milder, Byron Zollars, Steve Savoy, Gennady Shvets, Metamaterial-based Integrated Plasmonic Ab- sorber/emitter for Solar Thermo-Photovoltaic systems,Published, IOP Publishing Ltd, 12 January 2012, Journal of Optics, Volume 14, Number 2. PR [6] Chihhui Wu, Neuner Burton III, Jeremy John, Andrew Milder, Byron Zollars, Steve Savoy, Gennady Shvets, Metamaterial-based Integrated Plasmonic Ab- sorber/emitter for Solar Thermo-Photovoltaic systems,Published, IOP Publishing Ltd, 12 January 2012, Journal of Optics, Volume 14, Number 2. PR hihhui Wu, Neuner Burton III, Jeremy John, Andrew Milder, Byron Zollars, Steve Savoy, Gennady Shvets, Metamaterial-based Integra rber/emitter for Solar Thermo-Photovoltaic systems,Published, IOP Publishing Ltd, 12 January 2012, Journal of Optics, Volume 14, N PR [6] Chihhui Wu, Neuner Burton III, Jeremy John, Andrew Milder, Byron Zollars, Steve Savoy, Gennady Shvets, Metamaterial-base b / i f S l Th Ph l i P bli h d IOP P bli hi L d 12 J 2012 J l f O i V l P [6] Chihhui Wu, Neuner Burton III, Jeremy John, Andrew Milder, Byron Zollars, Steve Savoy, Gennady Shvets, M sorber/emitter for Solar Thermo-Photovoltaic systems,Published, IOP Publishing Ltd, 12 January 2012, Jour D P [7] Andreas Tittl, Ann-Katrin U. Michel, Schäferling Martin, Xinghui Yin, Behrad Gholipour, long Cui, Matthias Wuttig, Thomas Taubner, Frank Neubrech, Harald Giessen, A switchable mid-infrared plasmonic perfect absorber with multispectral thermal imaging capability, Adv. Mater. 27 (31) (August 19, 2015) 4597–4603, https://doi.org/10.1002/adma.201502023. D [8] Bingxin Zhang, Yanhui Zhao, Hao Qingzhen, Brian Kiraly, Iam-Choon Khoo, Shufen Chen, Tony Jun Huang, Polarization-i fect absorber based on a metal-dielectric-metal elliptical nanodisk array, Optic Express 19 (16) (2011) 15221–15228 https ED [8] Bingxin Zhang, Yanhui Zhao, Hao Qingzhen, Brian Kiraly, Iam-Choon Khoo, Shufen Chen, Tony Jun Huang, Polarization-independent dual-band infrared per- fect absorber based on a metal-dielectric-metal elliptical nanodisk array, Optic Express 19 (16) (2011) 15221–15228 https://doi.org/10.1364/OE.19.015221. [9] Abul K. Azad, J. Wilton, M. Kort-Kamp, Milan Sykora, Nina R. Weisse-Bernstein, Ting S. Luk, Antoinette J. Taylor, Diego A.R. References CT [16] H. Atwater, A. Polman, Plasmonic for improved photovoltaic devices, Nat. Mater. 9 (3) (2010) 205–213, https://doi.org/10.1038/nmat2629. EC [17] Sina Abedini Dereshgi, Zulkarneyn Sisman, Topalli Kagan, Ali Kemal Okyay, Plasmonically enhanced metal–insulator multista t b ti d ll ti j ti f i f d li ti S i R 7 (2017) 42349 htt //d i /10 1038/ EC rate absorption and collection junctions for near-infrared applications, Sci. Rep. 7 (2017) 42349, https://doi.org/10.1038/srep42349. [18] J. Kim, H. Yang, P.F. Green, Langmuir 28 (2012) 9735–9741. EC p j pp p p g p [18] J. Kim, H. Yang, P.F. Green, Langmuir 28 (2012) 9735–9741. 18] J. Kim, H. Yang, P.F. Green, Langmuir 28 (2012) 9735–9741. EC [19] F. Huang, S. Drakeley, M.G. Millyard, A. Murphy, R. White, E. Spigone, J. Kivioja, J. Baumberg, J. Adv. Opt. Mater. 4 (20 RE [20] M.K. Hedayati, M. Javaherirahim, B. Mozooni, R. Abdelaziz, A. Tavassolizadeh, V.S.K. Chakravadhanula, V. Zaporojtchenko, T. Strunkus, F. Faupel, M. El- bahri, Adv. Mater. 23 (2011) 5410–5414. R [21] D. Prezgot, A. Ianoul, J. Phys. Chem. C 119 (2015), 3293–3. R [22] A. Moreau, C. Ciracì, J.J. Mock, R.T. Hill, Q. Wang, B.J. Wiley, A. Chilkoti, D.R. Smith, Nature 492 (2012) 86–89. RR [23] L. De Sio, et al., Next-generation thermo-plasmonic technologies and plasmonic nanoparticles in optoelectronics, Prog. RR [23] L. De Sio, et al., Next-generation thermo-plasmonic technologies and plasmonic nanoparticles in optoelectronics, Prog. Quant. Electron. 41 (2015) 23–70. RR [23] L. De Sio, et al., Next-generation thermo-plasmonic technologies and plasmonic nanoparticles in optoelectronics, Prog. Quant. Electron. 41 (2015) 23–70. [24] G M Ak l d J H T B H P T B L S D R S ith M H Mikk l Ad M t (2015) 1 7 RR [25] J. Wei, N. Schaeffer, P.A. Albouy, M.P. Pileni, Chem. Mater. 27 (2015) 5614–5621. R [26] B. Gao, G. Arya, A.R. Tao, Nat. Nanotechnol. 7 (2012) 433–437. OR [26] B. Gao, G. Arya, A.R. Tao, Nat. Nanotechnol. 7 (2012) 433–437. OR [27] A. Bottomley, D. Prezgot, J.P. Coyle, A. Ianoul, Nanoscale 8 (2016) 11168–11176. OR [28] H. Li, Y. He, Z. Liu, B. Jiang, Y. Huang, Rapid synthesis of broadband Ag@ TiO 2 core–shell nanoparticles for solar energy conversion, Sol. Energy Mater. Sol. Cells 166 (2017) 52–60.l CO [29] V.K. Pustovalov, L.G. Astafyeva, W. References Fritzsche, Light-absorption selection of nanoparticles and nanofluids containing nanoparticles for their effective heating by solar radiation, Nanotechnol. Environ. Eng. 2 (2017) 7. CO , g [30] Joseph M. Luther, Prashant K. Jain, Trevor Ewers, A. Paul Alivisatos, Localized Nat. Mater. 10 (2011) 361–366, https://doi.org/10.1038/nmat3004. CO g [30] Joseph M. Luther, Prashant K. Jain, Trevor Ewers, A. Paul Alivisatos, Localized surface plasmon resonances arising from fr CO Nat. Mater. 10 (2011) 361–366, https://doi.org/10.1038/nmat3004. CO Nat. Mater. 10 (2011) 361–366, https://doi.org/10.1038/nmat3004. C [31] S. Eliezer, et al., Synthesis of nanoparticles with femtosecond laser pulses, Phys. Rev. B 69 (2004) 144119. C [31] S. Eliezer, et al., Synthesis of nanoparticles with femtosecond laser pulses, Phys. Rev. B 69 (2004) 144119. NC Novel technique for preparing a nano-silver water suspension by th NC 32] D.C. Tien, et al., Novel technique for preparing a nano-silver NC q p p g p y g [33] A. Tavakoli, M. Sohrabi, A. Kargari, A review of methods for synthesis of nanostructured metals with emphasis on iron compounds, Chem. Pap. 61 (2007) 151–170. NC [33] A. Tavakoli, M. Sohrabi, A. Kargari, A review of methods for synthesis of nanostructured metals with emphasis on iro 151–170. N . Ji, H. Xu, P. Simon, Z. Wu, Regularly shaped, single-crystalline ZnO nanorods with wurtzite structure, J. Am. Chem. Soc. 124 (2002 65. N [34] L. Guo, Y.L. Ji, H. Xu, P. Simon, Z. Wu, Regularly shaped, single-crystalline ZnO nanorods with wurtzite structure, J. Am. 14864–14865. UN 14864 14865. [35] Y. Wang, Y. Xia, Bottom-up and top-down approaches to the synthesis of monodispersed spherical colloids of low melting-point metals, Nano Lett. 4 (2004) 2047–2050. UN [35] Y. Wang, Y. Xia, Bottom-up and top-down approaches to the synthesis of monodispersed spherical colloids of low melting- 2047–2050. U [36] W. Lu, F. Liao, Y. Luo, G. Chang, X. Sun, Hydrothermal synthesis of well-stable silver nanoparticles and their application for enzymeless hydrogen peroxide detection, Electrochim. Acta 56 (2011) 2295–2298. U [37] B. Liu, H.C. Zeng, Hydrothermal synthesis of ZnO nanorods in the diameter regime of 50 nm, J. Am. Chem. Soc. 125 (200 U B. Liu, H.C. Zeng, Hydrothermal synthesis of ZnO nanorods in the U [38] I.A. Rahman, V. Padavettan, Synthesis of silica nanoparticles by sol-gel: size-dependent properties, surface modification nanocomposites—a review, J. Nanomater. 8 (2012), 2012. [39] H. Xu, B.W. Zeiger, K.S. Suslick, Sonochemical synthesis of nanomaterials, Chem. Soc. Rev. 42 (2013) 2555–2567. References Dalvit, Hou-Tong Chen, Meta- surface broadband solar absorber, Sci. Rep. 6 (2016) https://doi.org/10.1038/srep20347, Article number: 20347.ii ED [8] Bingxin Zhang, Yanhui Zhao, Hao Qingzhen, Brian Kiraly, Iam-Choon Khoo, Shufen Chen, Tony Jun Huang, Polarization-independent dual-band infrared per- fect absorber based on a metal-dielectric-metal elliptical nanodisk array, Optic Express 19 (16) (2011) 15221–15228 https://doi.org/10.1364/OE.19.015221. [9] Abul K. Azad, J. Wilton, M. Kort-Kamp, Milan Sykora, Nina R. Weisse-Bernstein, Ting S. Luk, Antoinette J. Taylor, Diego A.R. Dalvit, Hou-Tong Chen, Meta- surface broadband solar absorber, Sci. Rep. 6 (2016) https://doi.org/10.1038/srep20347, Article number: 20347.ii ED [9] Abul K. Azad, J. Wilton, M. Kort-Kamp, Milan Sykora, Nina R. Weisse-Bernstein, Ting S. Luk, Antoinette J. Taylor, Diego A.R. Dalvit, Hou-Tong Chen, Meta- surface broadband solar absorber, Sci. Rep. 6 (2016) https://doi.org/10.1038/srep20347, Article number: 20347.ii D [9] Abul K. Azad, J. Wilton, M. Kort-Kamp, Milan Sykora, Nina R. Weisse-Bernstein, Ting S. Luk, Antoinette J. Taylor, Diego A surface broadband solar absorber Sci Rep 6 (2016) https://doi org/10 1038/srep20347 Article number: 20347 surface broadband solar absorber, Sci. Rep. 6 (2016) https:// ED , p ( ) p // g/ / p , [10] Zhongyang Li, Serkan Butun, Koray Aydin, Large-area, lithography-free super absorbers and color filters at visible frequencies using ultrathin metallic films, ACS Photonics 2 (2) (2015) 183–188, https://doi.org/10.1021/ph500410u. ED , p ( ) p // g/ / p , [10] Zhongyang Li, Serkan Butun, Koray Aydin, Large-area, lithography-free super absorbers and color filters at visible frequencies using ultrathin metallic films, ACS Photonics 2 (2) (2015) 183–188, https://doi.org/10.1021/ph500410u. ED [10] Zhongyang Li, Serkan Butun, Koray Aydin, Large-area, lithography-free super absorbers and color filters at visible frequen ACS Photonics 2 (2) (2015) 183–188, https://doi.org/10.1021/ph500410u. TE [11] D. Wu, et al., Ultra-narrow band perfect absorber and its application as plasmonic sensor in the visible region, Nanoscale R [12] R.L. Bailey, J. Eng. Power 94 (2) (1972) 73. ] D. Wu, et al., Ultra-narrow band perfect absorber and its applic [12] R.L. Bailey, J. Eng. Power 94 (2) (1972) 73. TE y g [13] X. Chen, et al., Nature Nanotechnology, July, 20171–7. [13] X. Chen, et al., Nature Nanotechnology, July, 20171–7. CTE [14] E. Hutter, J. H. Fendler, Exploitation of Localized Surface Plasmon Resonance, Adv. Mater. 16 (19) (October, 2004) 1685–1706 https://doi.org/10.1002/ adma.200400271. CT [15] Katherine A. Willets and Richard P. Van Duyne, Localized surface plasmon resonance spectroscopy and sensing, Annu. Rev (Volume publication date 5 May 2007) https://doi.org/10.1146/annurev.physchem.58.032806.104607. References OO Kenanakis, et al., Perfect absorbers based on metal–insulator–metal structures in the visible region: a simple approach for practical a 123 (2017) 77. OO [2] G. Kajtár, M. Kafesaki, E.N. Economou, C.M. Soukoulis, Theoretical model of homogeneous metal–insulator–metal perfect multi-band absorbers for the visible spectrum, J. Phys. Appl. Phys. 49 (2016), 055104. OO [2] G. Kajtár, M. Kafesaki, E.N. Economou, C.M. Soukoulis, Theoretical model of homogeneous metal–insulator–metal perfect multi-band absorbers for the visible spectrum, J. Phys. Appl. Phys. 49 (2016), 055104. OO [2] G. Kajtár, M. Kafesaki, E.N. Economou, C.M. Soukoulis, Theoretical model of homogeneous metal–insulator–metal perfect multi-band absorbers for the visible spectrum, J. Phys. Appl. Phys. 49 (2016), 055104. RO [3] Y.-J. Jen, W.-C. Liu, T.-K. Chen, S. Lin, Y.-C. Jhang, Design and deposition of a metal-like and admittance-matching metamaterial as an ultra-thin perfect ab- sorber, Sci. Rep. 7 (2017). RO [3] Y.-J. Jen, W.-C. Liu, T.-K. Chen, S. Lin, Y.-C. Jhang, Design and deposition of a metal-like and admittance-matching metamaterial as an ultra-thin perfect ab- sorber, Sci. Rep. 7 (2017). RO Y.-J. Jen, W.-C. Liu, T.-K. Chen, S. Lin, Y.-C. Jhang, Design and deposition of a metal-like and admittance-matching metamaterial a sorber, Sci. Rep. 7 (2017). RO S.A. Dereshgi, A. Ghobadi, H. Hajian, B. Butun, E. Ozbay, Ultra-broadband, lithography-free, and large-scale compatible perfect abs choice of metal layers in metal insulator multilayer stacks Sci Rep 7 (2017) 14872 RO [4] S.A. Dereshgi, A. Ghobadi, H. Hajian, B. Butun, E. Ozbay, Ultra-broadband, lithography-free, and large-scale compatible perfec choice of metal layers in metal-insulator multilayer stacks, Sci. Rep. 7 (2017) 14872. PR [5] I. Azad, M.K. Ram, D.Y. Goswami, E. Stefanakos, Design and fabrication of metal-insulator-metal diode for high frequency applications, In: Infrared Technol- ogy and Applications XLIII 10177 International Society for Optics and Photonics 2017 p 101772J PR [5] I. Azad, M.K. Ram, D.Y. Goswami, E. Stefanakos, Design and fabrication of metal-insulator-metal diode for high frequency applications, In: Infrared Technol- ogy and Applications XLIII 10177, International Society for Optics and Photonics, 2017, p. 101772J. PR [5] I. Azad, M.K. Ram, D.Y. Goswami, E. Stefanakos, Design and fabrication of metal-insulator-metal diode for high frequ ogy and Applications XLIII 10177, International Society for Optics and Photonics, 2017, p. 101772J. PR [5] I. Azad, M.K. Ram, D.Y. Goswami, E. L. Escoubas et al. References Raj, Seedless, surfactantless room temperature synthesis of single crystalline fluorescent gold nanoflow catalytic activity, Chem. Mater. 20 (2008) 3546–3548.l RO . Bohidar, Synthesis of branched Ag nanoflowers based on a bioinspired technique: their surface enhanced Raman scattering and an- ys. Chem. C 113 (2009) 14753–14758. RO [53] B.K. Jena, B.K. Mishra, S. Bohidar, Synthesis of branched Ag nanoflowers based on a bioinspired technique: their surface enhan tibacterial activity, J. Phys. Chem. C 113 (2009) 14753–14758. RO [54] Z. Wang, J. Zhang, J.M. Ekman, P.J. Kenis, Y. Lu, DNA-mediated control of metal nanoparticle shape: one-pot synthesis and cellular uptake of highly stable and functional gold nanoflowers, Nano Lett. 10 (2010) 1886–1891.l R gl , ( ) [55] J. Xie, Q. Zhang, J.Y. Lee, D.I. Wang, The synthesis of SERS-active gold nanoflower tags for in vivo applications, ACS Nano 2 (2008) 2473–2480.l Rl . Lee, D.I. Wang, The synthesis of SERS-active gold nanoflower tags for in vivo applications, ACS Nano 2 (2008) 2473–2480.l R [55] J. Xie, Q. Zhang, J.Y. Lee, D.I. Wang, The synthesis of SERS-active gold nanoflower tags for in vivo applications, ACS Nan l [56] J. Xu, et al., Synthesis and catalytic properties of Au–Pd nanoflowers, ACS Nano 5 (2011) 6119–6127. PR [57] A. Guerrero-Martínez, S. Barbosa, I. Pastoriza-Santos, L.M. Liz-Marzán, Nanostars shine bright for you: colloidal synthesis, properties and applications of branched metallic nanoparticles, Curr. Opin. Colloid Interface Sci. 16 (2011) 118–127. P urphy, Room temperature, high-yield synthesis of multiple shapes of gold nanoparticles in aqueous solution, J. Am. Chem. Soc. 126 P [58] T.K. Sau, C.J. Murphy, Room temperature, high-yield synthesis of multiple shapes of gold nanoparticles in aqueous solution, J. 8648–8649. D P glein, M. Giersig, Formation of colloidal silver nanoparticles: capping action of citrate, J. Phys. Chem. B 103 (1999) 9533–9539. D P [59] A. Henglein, M. Giersig, Formation of colloidal silver nanoparticles: capping action of citrate, J. Phys. Chem. B 103 (1999) 9533–9539. D [60] Y. Lee, J. Choi, K.J. Lee, N.E. Stott, D. Kim, Large-scale synthesis of copper nanoparticles by chemically controlled reduction fo electronics, Nanotechnology 19 (2008) 415604. D [61] F. Fievet, J.P. Lagier, M. Figlarz, Preparing monodisperse metal powders in micrometer and submicrometer sizes by the polyol process, MRS Bull. 14 (1989) 29–34. ED [62] L. Zhang, F.X. Gu, J.M. Chan, A.Z. Wang, R.S. Langer, O.C. Farokhzad, Nanoparticles in medicine: therapeutic applications and development, Clin. Pharmacol. Ther. 83 (5) (2008) 761–769. References Xia, Quantitative analysis of dipole and quadrupole excitation in the surface plasmon resonance of metal nanoparticles, J. Phys. Chem. C 112 (2008) 20233–20240. N [86] Lumerical solutions, inc, Accessed: 15th March 2017), http://www.lumerical.com/tcad-products/fdtd/. UN [87] C.A. Downing, E. Mariani, G. Weick, Retardation effects on the dispersion and propagation of plasmons in metallic nanoparticle chains, J. Phys. Condens. Mat- ter 30 (2) (2017), 025301. UN [87] C.A. Downing, E. Mariani, G. Weick, Retardation effects on the dispersion and propagation of plasmons in metallic nanoparticle chains, J. Phys. Condens. Mat- ter 30 (2) (2017), 025301. UN [88] I.O. Sosa, C. Noguez, R.G. Barrera, Optical properties of metal nanoparticles with arbitrary shapes, J. Phys. Chem. B 107 ( U [89] A.A. Ashkarran, S. Daemi, Tuning the plasmon of metallic nanostructures: from silver nanocubes toward gold Nanoboxes, Plasmonics 11 (2016) 1011–1017. U [89] A.A. Ashkarran, S. Daemi, Tuning the plasmon of metallic nanostructures: from silver nanocubes toward gold Nanoboxes, Plasmonics 11 (2016) 10 U [89] A.A. Ashkarran, S. Daemi, Tuning the plasmon of metallic nanostructures: from silver nanocubes toward gold Nanoboxes, Plasmonics 11 (2016) 1011–1017 [90] N E Christensen The band structure of silver and optical interband transitions Phys Status Solidi B 54 (1972) 551–563 U [90] N.E. Christensen, The band structure of silver and optical interband transitions, Phys. Status Solidi B 54 (1972) 551–563. U [91] L. Long, et al., Localized surface plasmon resonance improved lasing performance of Ag nanoparticles/organic dye random laser, J. Alloy. Comp. 693 (2017) 876–881.l U [91] L. Long, et al., Localized surface plasmon resonance improved lasing performance of Ag nanoparticles/organic dye random laser, J. Alloy. Comp. 693 (2017) 876–881.l [92] J. Sun, et al., Uniform and reproducible plasmon-enhanced fluorescence substrate based on PMMA-coated, large-area Au@ 11 (2018) 953–965. 11 (2018) 953–965. [93] B.X. Wang, C.Y. Zhao, Achieving a strongly negative scattering asymmetry factor in random media composed of dual-dipolar particles, Phys. Rev. 97 (2018)023836htt //d i /10 1103/Ph R A 97 023836 ( ) [93] B.X. Wang, C.Y. Zhao, Achieving a strongly negative scattering asymmetry factor in random media composed of dual-dipolar particles, Phys. Rev. 97 (2018)023836https://doi.org/10.1103/PhysRevA.97.023836. [93] B.X. Wang, C.Y. Zhao, Achieving a strongly negative scattering asymmetry factor in random media composed of dual-dipolar particles, Phys. Rev. 97 (2018)023836https://doi.org/10.1103/PhysRevA.97.023836. [93] B.X. Wang, C.Y. Zhao, Achieving a strongly negative scattering asymmetry factor in random media composed of dual-dipo 97 (2018)023836https://doi.org/10.1103/PhysRevA.97.023836. [94] F. References Jin, The Finite Element Method in Electromagnetics, John Wile RR g y [76] G. Mie, Beiträge zur Optik trüber Medien, speziell kolloidaler Metallösungen, Ann. Phys. 330 (1908) 377–445. RR [77] L. De Sio, T. Placido, R. Comparelli, M.L. Curri, M. Striccoli, N. Tabiryan, T.J. Bunning, Next-generation thermo-plasmoni nanoparticles in optoelectronics, Prog. Quant. Electron. 41 (2015) 23–70.l OR [78] K.L. Kelly, E. Coronado, L.L. Zhao, G.C. Schatz, others, The optical properties of metal nanoparticles: the influence of size, J. Phys. Chem. B-Condens. Phase 107 (2003) 668–677. OR E. Coronado, L.L. Zhao, G.C. Schatz, others, The optical properties of metal nanoparticles: the influence of size, shape, and dielectric e em. B-Condens. Phase 107 (2003) 668–677. OR [79] W.-H. Yang, G.C. Schatz, R.P. Van Duyne, Discrete dipole approximation for calculating extinction and Raman intensities for small particles with arbitrary shapes, J. Chem. Phys. 103 (1995) 869–875. OR [79] W.-H. Yang, G.C. Schatz, R.P. Van Duyne, Discrete dipole approximation for calculating extinction and Raman intensities for small particles with arbitrary shapes, J. Chem. Phys. 103 (1995) 869–875. CO [80] L.A. Golovan, V.Y. Timoshenko, P.K. Kashkarov, Optical properties of porous-system-based nanocomposites, Phys.-Uspekh CO an, V.Y. Timoshenko, P.K. Kashkarov, Optical properties of porous-system-based nanocomposites, Phys.-Uspekhi 50 (2007) 595. CO [80] L.A. Golovan, V.Y. Timoshenko, P.K. Kashkarov, Optical properties of porous-system-based nanocomposites, Phys.-Uspekhi 50 (2007) 595. CO [81] H. DeVoe, Optical properties of molecular aggregates. I. Classical model of electronic absorption and refraction, J. Chem. P CO [82] E.M. Purcell, C.R. Pennypacker, Scattering and absorption of light by nonspherical dielectric grains, Astrophys. J. 186 (1 CO [83] B.T. Draine, P.J. Flatau, Discrete-dipole approximation for scattering calculations, JOSA A 11 (1994) 1491–1499.ii CO [83] B.T. Draine, P.J. Flatau, Discrete-dipole approximation for scattering calculations, JOSA A 11 (1994) 1491–1499.ii C [83] B.T. Draine, P.J. Flatau, Discrete-dipole approximation for scattering calculations, JOSA A 11 (1994) 1491–1499.ii C Flatau, Discrete-dipole approximation for scattering calculations, J C 83] B.T. Draine, P.J. Flatau, Discrete-dipole approximation for sca NC [84] J.R. Dunklin, C. Bodinger, G.T. Forcherio, D.K. Roper, Plasmonic extinction in gold nanoparticle-polymer films as film thic decrease below resonant wavelength, J. Nanophotonics 11 (2017), 016002. NC [85] F. Zhou, Z.-Y. Li, Y. Liu, Y. Xia, Quantitative analysis of dipole and quadrupole excitation in the surface plasmon resonanc Chem. C 112 (2008) 20233–20240. NC [85] F. Zhou, Z.-Y. Li, Y. Liu, Y. References E [63] G. Mie, Beiträge zur Optik trüber Medien, speziell kolloidaler Metallösungen, Ann. Phys. 330 (1908) 377–445. TE [64] G.V. Naik, V.M. Shalaev, A. Boltasseva, Alternative plasmonic materials: beyond gold and silver, Adv. Mater. 25 (2013) TE [65] P.R. West, et al., Searching for better plasmonic materials, Laser Photon. Rev. 4 (2010) 795–808. TE [65] P.R. West, et al., Searching for better plasmonic materials, Laser Photon. Rev. 4 (2010) 795–808. CTE , , g p , ( ) [66] clément Barrière, Kilian Piettre, Virginie Latour, olivier Margeat, cédric-olivier Turrin, Bruno chaudret, Pierre Fau, Ligand effects on the air stability of cop nanoparticles obtained from organometallic synthesis, J. Mater. Chem. 22 (2012) 2279–2285. CTE [66] clément Barrière, Kilian Piettre, Virginie Latour, olivier Margeat, cédric-olivier Turrin, Bruno chaudret, Pierre Fau, Ligand nanoparticles obtained from organometallic synthesis, J. Mater. Chem. 22 (2012) 2279–2285. CT G.V. Naik, J. Kim, A. Boltasseva, Oxides and nitrides as alternative plasmonic materials in the optical range [Invited], Opt. Mater. E 1090–1099. CT , R.H. Dinegar, Theory, production and mechanism of formation of monodispersed hydrosols, J. Am. Chem. Soc. 72 (1950) 4847–485 CT [68] V.K. LaMer, R.H. Dinegar, Theory, production and mechanism of formation of monodispersed hydrosols, J. Am. Chem. Soc. 72 (1950) 4847–4854. CT [68] V.K. LaMer, R.H. Dinegar, Theory, production and mechanism of formation of monodispersed hydrosols, J. Am. Chem. Soc EC [69] C.B. Murray, C.R. Kagan, M.G. Bawendi, Synthesis and characterization of monodisperse nanocrystals and close-packed nanocr Mater. Sci. 30 (2000) 545–610. EC [70] N.D. Burrows, et al., Anisotropic nanoparticles and anisotropic surface chemistry, J. Phys. Chem. Lett. 7 (2016) 632–641. EC [70] N.D. Burrows, et al., Anisotropic nanoparticles and anisotropic surface chemistry, J. Phys. Chem. Lett. 7 (2016) 632–641. E [71] P.B. Johnson, R.-W. Christy, Optical constants of the noble metals, Phys. Rev. B 6 (1972) 4370. RE [72] E.D. Palik, Handbook of Optical Constants of Solids, vol. 429, Academic, San Diego, 1998, Google Sch. RE [73] C.C. Katsidis, D.I. Siapkas, General transfer-matrix method for optical multilayer systems with coherent, partially coheren Opt. 41 (19) (2002) 3978–3987. R [74] D.M. Sullivan, Electromagnetic Simulation Using the FDTD Method, John Wiley & Sons, 2013. RR g g y [75] J.M. Jin, The Finite Element Method in Electromagnetics, John Wiley & Sons, 2015. RR e Finite Element Method in Electromagnetics, John Wiley & Sons, 2 RR .M. References [40] S.S. Joshi, S.F. Patil, V. Iyer, S. Mahumuni, Radiation induced synthesis and characterization of copper nanoparticles, Nanostruct. Mater. 10 (1998) 1135–1144. [41] E. Abbasi, et al., Silver nanoparticles: synthesis methods, bio-applications and properties, Crit. Rev. Microbiol. 42 (2016) 173– [42] B. Wiley, Y. Sun, B. Mayers, Y. Xia, Shape-controlled synthesis of metal nanostructures: the case of silver, Chem. Eur J. 11 (2005) 454–463 gach, Nonspherical noble metal nanoparticles: colloid-chemical syn 43] T.K. Sau, A.L. Rogach, Nonspherical noble metal nanoparticle [44] Victor K. LaMer, Robert H. Dinegar, Theory, production and mechanism of formation of monodispersed hydrosols, J. Am. Chem. Soc. 72 (11) (1950) 4847–4854, https://doi.org/10.1021/ja01167a001. 4847–4854, https://doi.org/10.1021/ja01167a001. Progress in Quantum Electronics xxx (2019) xxx-xxx L. Escoubas et al. [45] Xiaohu Xia, Jie Zeng, Qiang Zhang, Christine H. Moran, Younan Xia, Recent developments in shape-controlled synthesis of silver nanocrystals, J. Phys. Chem. C 116 (41) (2012) 21647–21656.i [46] L. Mulfinger, et al., Synthesis and study of silver nanoparticles, J Chem. Educ. 84 (2007) 322. L. Mulfinger, et al., Synthesis and study of silver nanoparticles, J C F g , , y y p , ( ) [47] E. Hao, K.L. Kelly, J.T. Hupp, G.C. Schatz, Synthesis of silver nanodisks using polystyrene mesospheres as templates, J. Am. Chem. Soc. 124 (2002) 15182 15183 i [47] E. Hao, K.L. Kelly, J.T. Hupp, G.C. Schatz, Synthesis of silver nanodisks using polystyrene mesospheres as templa 15182–15183. Fi [47] E. Hao, K.L. Kelly, J.T. Hupp, G.C. Schatz, Synthesis of silver nanodisks using polystyrene mesospheres as templates, J. Am. Chem. Soc. 124 (2002) 15182–15183. OF [48] D. Aherne, D.M. Ledwith, M. Gara, J.M. Kelly, Optical properties and growth aspects of silver nanoprisms produced by a highly reproducible and rapid synthe- sis at room temperature, Adv. Funct. Mater. 18 (2008) 2005–2016. [49] M. Li, Z.S. Zhang, X. Zhang, K.Y. Li, X.F. Yu, Optical properties of Au/Ag core/shell nanoshuttles, Optic Express 16 (2008) O [50] C.-J. Huang, P.-H. Chiu, Y.-H. Wang, C.-F. Yang, Synthesis of the gold nanodumbbells by electrochemical method, J. Colloid Interface Sci. 303 (2006) 430–436. O [51] J. Burgin, M. Liu, P. Guyot-Sionnest, Dielectric sensing with deposited gold bipyramids, J. Phys. Chem. C 112 (2008) 19279–1 OO [52] B.K. Jena, C.R. Raj, Seedless, surfactantless room temperature synthesis of single crystalline fluorescent gold nanoflowers with pronounced SERS and electro- catalytic activity, Chem. Mater. 20 (2008) 3546–3548.l [52] B.K. Jena, C.R. References Wooten, Optical Properties of Solids, Academic press, 2013. [94] F. Wooten, Optical Properties of Solids, Academic press, 2013. [95] T.C. Choy, Effective Medium Theory: Principles and Applications, vol. 165, Oxford University Press, 2015. Spectroscopic Ellipsometry: Principles and Applications, John Wil j y y [97] J.A. Woollam, et al., Overview of variable-angle spectroscopic ellipsometry (VASE): I. Basic theory and typical applications, Optical Metrology (1999) 3–28. [98] H. Fujiwara, Spectroscopic Ellipsometry: Principles and Applications, John Wiley & Sons, 2007.i [97] J.A. Woollam, et al., Overview of variable-angle spectroscopic ellipsometry (VASE): I. Basic theory and typ [97] J.A. Woollam, et al., Overview of variable-angle spectroscopic ellipsometry (VASE): I. Basic theory and typical applications, Optical Metrology (1999) 3–2 [98] H. Fujiwara, Spectroscopic Ellipsometry: Principles and Applications, John Wiley & Sons, 2007. [99]i L. Escoubas et al. Progress in Quantum Electronics xxx (2019) xxx-xxx [101] M. Voué, N. Dahmouchene, J. De Coninck, Annealing of polymer films with embedded silver nanoparticles: effect on optical properties, Thin Solid Films 519 (2011) 2963–2967.i [101] M. Voué, N. Dahmouchene, J. De Coninck, Annealing of polymer films with embedded silver nanoparticles: effect on optical properties, Thin Solid Films 519 (2011) 2963–2967.i F [102] M. Lončarić, J. Sancho-Parramon, H. Zorc, Optical properties of gold island films—a spectroscopic ellipsometry study, Thin Solid Films 519 (2011) 2946–2950.i OF D. Evans, J.R. Henderson, Spectroscopic ellipsometric evaluation of gold nanoparticle thin films fabricated using layer-by-layer self-as (2003) 531–534. 103] H.-L. Zhang, S.D. Evans, J.R. Henderson, Spectroscopic ellipsometric evaluation of gold nanoparticle thin films fabricated using Adv. Mater. 15 (2003) 531–534. OF [104] M. Abramowitz, I.A. Stegun, Handbook of Mathematical Functions, Dover Publications, N. Y, 1972361. [104] M. Abramowitz, I.A. Stegun, Handbook of Mathematical Functions, Dover Publications, N. Y, 1972361. 04] M. Abramowitz, I.A. Stegun, Handbook of Mathematical Func O Levy, D. Stroud, Maxwell Garnett theory for mixtures of anisotropic inclusions: application to conducting polymers, Phys. Rev. B 5 [105] O. Levy, D. Stroud, Maxwell Garnett theory for mixtures of anisotropic inclusions: application to conducting 106] S.G. Moiseev, Active Maxwell–Garnett composite with the unit refractive index, Phys. B Condens. Matter 405 (2010) 3042–304 OO [107] G. Berginc, Coherent light scattering of heterogeneous randomly rough films and effective medium in the theory of electromagnetic wave multiple scattering, Quant. Electron. 43 (2013) 1055. OO [108] I.L. Skryabin, A.V. Radchik, P. Moses, G.B. Smith, The consistent application of Maxwell–Garnett effective medium theory to anisotropic composites, Appl. Phys. Lett. References 70 (1997) 2221–2223. [109] V.P. Drachev, et al., The Ag dielectric function in plasmonic metamaterials, Optic Express 16 (2008) 1186–1195. RO [110] V.G. Kravets, S. Neubeck, A.N. Grigorenko, A.F. Kravets, Plasmonic blackbody: strong absorption of light by metal nanoparticles embedded in a dielectric ma- trix, Phys. Rev. B 81 (2010) 165401. RO [110] V.G. Kravets, S. Neubeck, A.N. Grigorenko, A.F. Kravets, Plasmonic blackbody: strong absorption of light by metal nanop trix, Phys. Rev. B 81 (2010) 165401. PRO [111] T.W. Oates, M. Ranjan, S. Facsko, H. Arwin, Highly anisotropic effective dielectric functions of silver nanoparticle arrays, Optic Express 19 (2011) 2014–2028. [112] M. Perera, D. Schmidt, W.E.K. Gibbs, S. Juodkazis, P.R. Stoddart, Effective optical constants of anisotropic silver nanoparticle films with plasmonic properties, Opt. Lett. 41 (2016) 5495–5498. PR [112] M. Perera, D. Schmidt, W.E.K. Gibbs, S. Juodkazis, P.R. Stoddart, Effective optical constants of anisotropic silver nanoparticle films with plasmonic properties Opt. Lett. 41 (2016) 5495–5498. PR [113] R.R. Bhat, J. Genzer, Using spectroscopic ellipsometry for quick prediction of number density of nanoparticles bound to non-transparent solid surfaces, Surf. Sci. 596 (2005) 187–196. P yana, M. Rozin, S. Smith, A.R. Tao, Modular, polymer-directed nanoparticle assembly for fabricating metamaterials, Faraday Discuss –502. P 114] G.K. Laxminarayana, M. Rozin, S. Smith, A.R. Tao, Modular, polymer-directed nanoparticle assembly for fabricating metamate 186 (2016) 489–502. D P [115] T.A.F. König, P.A. Ledin, M. Russell, J.A. Geldmeier, M.A. Mahmoud, M.A. El-Sayed, V.V. Tsukruk, Silver nanocube aggregation gradient materials in search for total internal reflection with high phase sensitivity, Nanoscale 7 (2015) 5230–5239.i Dl [116] M. Carlberg, et al., Spectroscopic ellipsometry study of silver nanospheres and nanocubes in thin film layers, Opt. Mater. Express 7 (12) (2017) 4241–4248, https://doi.org/10.1364/OME.7.004241. https://doi.org/10.1364/OME.7.004241. ED g arov, et al., Tuning of magnetic optical response in a dielectric nanoparticle by ultrafast photoexcitation of dense electron-hole plasm p. 2015) 6187–6192. ED [118] B. García-Cámara, et al., All-optical nanometric switch based on the directional scattering of semiconductor nanoparticles, J. Phys. Chem. C 119 (2015) 19558–19564. TE [119] J.M. Geffrin, et al., Magnetic and electric coherence in forward- and back-scattered electromagnetic waves by a single dielectric subwavelength sphere, Nat. Commun. 3 (2013), Art. no. 1171. TE [119] J.M. Geffrin, et al., Magnetic and electric coherence in forward- and back-scattered electromagnetic waves by a single diel Commun. 3 (2013), Art. no. 1171. TE 120] M. Kerker, D.S. Wang, C.L. UNCOR References Gilles, Electromagnetic scattering by magnetic spheres, J. Opt. Soc. Am. 73 (6) (1983) 765–767. CT [121] I. Staude, et al., Tailoring directional scattering through magnetic and electric resonances in subwavelength silicon nanodisks, ACS Nano 7 (9) (2013) 7824–7832. CT [122] B. García-Cámara, J.F. Algorri, A. Cuadrado, V. Urruchi, J.M. Sánchez-Pena, R. Vergaz, Size dependence of the directional ductor nanoparticles, IEEE Photon. Technol. Lett. 27 (19) (Oct. 2015) 2059–2062. EC [123] R. Vergaz, J.F. Algorri, A. Cuadrado, J.M. Sánchez-Pena, B. García-Cámara, Control of the light interaction in a semiconductor nanoparticle dimer through scattering directionality, IEEE Photonics Journal 8 (3) (2016) 4501410, https://doi.org/10.1109/JPHOT.2016.2577714. C [123] R. Vergaz, J.F. Algorri, A. Cuadrado, J.M. Sánchez-Pena, B. García-Cámara, Control of the light interaction in a semicondu tt i di ti lit IEEE Ph t i J l 8 (3) (2016) 4501410 htt //d i /10 1109/JPHOT 2016 2577714 EC [123] R. Vergaz, J.F. Algorri, A. Cuadrado, J.M. Sánchez-Pena, B. García-Cámara, Control of the light interaction in a se scattering directionality, IEEE Photonics Journal 8 (3) (2016) 4501410, https://doi.org/10.1109/JPHOT.2016.25 EC g y, ( ) ( ) , p g [124] Gurunatha K. Laxminarayana, Matthew Rozin, Stephanie Smith, Andrea R. Tao, Modular, polymer-directed nanoparticle assembly for fabricating metamateri- als, Faraday Discuss 186 (2016) 489. EC [124] Gurunatha K. Laxminarayana, Matthew Rozin, Stephanie Smith, Andrea R. Tao, Modular, polymer-directed nanoparticle a als, Faraday Discuss 186 (2016) 489. RE [125] Y. Qin, et al., Near-infrared Plasmonic Copper Nanocups Fabricated by Template-Assisted Magnetron Sputtering, ACS Pho 1021/acsphotonics.7b00866.i RE ear-infrared Plasmonic Copper Nanocups Fabricated by Template-Assisted Magnetron Sputtering, ACS Photonics, 2017https://doi.org nics.7b00866.i RE [126] L.V. Besteiro, et al., Plasmonic glasses and films based on alternative inexpensive materials for blocking infrared radiation, Nano Lett. 18 (5) (2018) 3147–3156. RR [127] T. Maier, H. Brueckl, Multispectral microbolometers for the midinfrared, Opt. Lett. 35 (22) (2010) 3766–3768. RR [128] C.A. Reynaud, D. Duché, J. Le Rouzo, A. Nasser, L. Nony, F. Pourcin, O. Margeat, J. Ackermann, G. Berginc, C.A. Nijhuis, L. Escoubas, J.-J. Simon, Enhancing reproducibility and Nonlocal effects in film-coupled nanoantennas, Adv. Optical Mater. (2018) 1801177, https://doi.org/10.1002/adom.201801177. [129] W. Li, J. Valentine, Metamaterial perfect absorber based hot electron photodetection, Nano Lett. 14 (2014) 3510–3514. RR A. Reynaud, D. Duché, J. Le Rouzo, A. Nasser, L. Nony, F. Pourcin, O. Margeat, J. Ackermann, G. Berginc, C.A. Nijhuis, L. Escoubas, producibility and Nonlocal effects in film-coupled nanoantennas, Adv. Optical Mater. References (2018) 1801177, https://doi.org/10.1002/ado Li J Valentine Metamaterial perfect absorber based hot electron photodetection Nano Lett 14 (2014) 3510–3514 RR [128] C.A. Reynaud, D. Duché, J. Le Rouzo, A. Nasser, L. Nony, F. Pourcin, O. Margeat, J. Ackermann, G. Berginc, C.A. Nijhuis, reproducibility and Nonlocal effects in film-coupled nanoantennas, Adv. Optical Mater. (2018) 1801177, https://doi.org/1 RR reproducibility and Nonlocal effects in film-coupled nanoantennas, Adv. Optical Mater. (2018) 1801177, https://doi.org/10.1002/adom.201801177. [129] W. Li, J. Valentine, Metamaterial perfect absorber based hot electron photodetection, Nano Lett. 14 (2014) 3510–3514. RR p yi p , p ( ) , p g [129] W. Li, J. Valentine, Metamaterial perfect absorber based hot electron photodetection, Nano Lett. 14 (2014) 3510–3514. Ri [129] W. Li, J. Valentine, Metamaterial perfect absorber based hot electron photodetection, Nano Lett. 14 (2014) 3510–3514.
https://openalex.org/W4321011486	https://zenodo.org/records/7522951/files/611-622.pdf	Russian	null	АБРАЗИВ ЕЙИЛИШ	Zenodo (CERN European Organization for Nuclear Research)	2,023	cc-by	4,612	Educational Research in Universal Sciences ISSN: 2181-3515 Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 АННОТАЦИЯ Ажралувчи штамплар йўналтирувчи элементлари прецизион жуфтлари ишининг бузилиши ва эскиришининг асосий сабаби абразив ейилиши аниқланди. Бу маълумотлар асосида штамп қолипларига ўзгартириш киритиш муҳимлиги алоҳида аҳамият касб этади. Штамп пн лари ейилишнинг турли кўринишда: толиқиш тирқишлари юзага келиши ва металнинг майда зарралари ажралиши билан боғлиқ емирилиши аниқланади. Бу ишқаланувчи деталлар юза қатламлари парчаланишининг асосий сабаблари ўрганиш учун хизмат қилади. т сўзлар: Абразив, пн, штамп, пресс, ейилиш, эмперик момент, Калит сўзлар: Абразив, пн, штамп, пресс, ейилиш, эмперик моме Қисқартмалар: МЧЖ- масуляти чекланган жамият, АЖ- акциядорлик жамияти, Абразив сақловчи аралашмаларни пресслаш учун штамп пн лари асосан абразив ейилиш шароитида ишлайди. Прессланаётган аралашманинг абразивли хоссаларини аниқловчи қуйидаги факторлар мавжуд: - аралашма таркибидаги абразив зарраларни қаттиқлиги ва механик мустаҳкамлиги; - заррачалар шакли; рр р - аралашма таркибидаги доналари катта, пластик бўлмаган материаллар концентрацияси; - намлиги - аралашма таркибидаги доналари катта, пластик бўлмаган материаллар концентрацияси; - намлиги. Штамп пн ларини эскириш тезлигига пресслашда зичланадиган массанинг ён ва олд пн ларига тўқнашув босими катта таъсир кўрсатади. Штамп пн ларини эскириш тезлигига пресслашда зичланадиган массанинг ён ва олд пн ларига тўқнашув босими катта таъсир кўрсатади. Саноатда ишлаб чиқариш шароитида пн ларни алмаштиришда технологик критерияга амал қилинади, яъни зичланган хом-ашё подкладкаларининг ён юза тирналиши ва ёриқлар ҳосил бўлиши билан ифодаланади. Рухсат этилган ейилиш катталиги 0,8...1,0 мм ни ташкил этади. 611 Multidisciplinary Scientific Journal Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 1-расм. Штампнинг пресс дастгоҳига ўрнатилган ҳолати 1-расм. Штампнинг пресс дастгоҳига ўрнатилган ҳолати Пн ларни ейилишига кўп сонли факторлар таъсир этади, уларни кўпчилиги тасодифий факторлардир. Шу сабабли ейилган юза ҳолатини объектив тасвирини қуриш учун тадқиқот натижаларига статистик ишлов бериш усулидан фойдаланилган. Штамп пн лари юзасида ейилиш катталиги характеристикаси ва тақсимланишини текшириш, пн ларни ейилиш қонуниятига асосан пресслаш меъёри ва шароитига боғлиқ эканлигини аниқлашга имкон беради ҳамда буюмларни пресслаш йўли билан тайёрлаш учун пн ларни мустаҳкамлашни самарали технологиясини ишлаб чиқиш зарур. “Ўзбекистон темир йўллари” АЖ тасарруфидаги “Фарғона механика заводи” МЧЖ да сифатли конструкцион пўлат 20 материалидан фойдаланилди. Дастлаб нитроцементитлаш, сўнг чиниқтириш жараёнидан ўтказилиб, борлаш йўли билан мустаҳкамланган пўлат пн лар тайёрлаб олинди. Пн комплектлари керамик тўшама (подкладка)лар ишлаб чиқарувчи пресс қолипга ўрнатилган. Ейилиш катталигини20 донадан иборат бўлиб, олд ва ён пн лар учун 10 донадан иборат ўлчаш ишлари амалга оширилди (1-расм). Штампларда тайёрланган керамик тўшама (подкладка) лар миқдори 1400-1500 донани ташкил этган. АННОТАЦИЯ Пн лар ейилишини ўлчаш учун фойдаланилган ўлчов асбоб ва воситалари: катта назорат плитаси, кичик назорат плитаси, индикаторли устун ва микрометрик индикатор шкала бўлими 0,001 мм. Ёрдамичи асбоблар сифатида чизғич, штангенциркул ва чегарали узунлик ўлчагичи ишлатилган [1]. Пн лар ейилишини текшириш учун махсус стенд ишлаб чиқилган, унинг схемаси илова Е да келтирилган (2-расм). Кичик назорат плитаси, катта назорат плитаси устига қўйилган, унга индикаторли устун монтаж қилинган. Назорат қилинаётган плита кичик назорат плитасига жойлаштирилган. Индикаторни ростлаш қуйидаги равишда бажарилган: Ан индикатор узунлик ўлчовли https://t.me/Erus_uz Multidisciplinary Scientific Journal December, 2022 61 612 Multidisciplinary Scientific Journal December, 2022 612 December, 2022 Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 тўплами ёрдамида маълум ўлчамда А назорат қилинаётган пн қалинлигидан бирмунча кичкина оралиқда ўрнатилади [3]. Шундай қилиб деталнинг хақиқий қалинлиги А емирилган участкада ва ҳар бир нуқтада ўлчанган катталик пн юзасида кўриб чиқилаётган нуқтада абсолют ейилиш катталиги ҳисобланади. Пн нинг ички юзасида 8 та кўндаланг қирқим белгиланган ва 6 та узунасига, бир текисда барча пн лар бўйича жойлаштирилган. Ейилиш катталигини пн ишчи юзасининг 48 нуқтасида ўлчанади (белгилаш схемаси 2-расмда ён томони учун, 3-расмда олд томони учун). Пн ларда нуқталарни чизғич ва штангенциркуль ёрдамида белгиланган. Ўлчаш жараёнида кўндаланг йўналишда пн ни ўзи силжиган (ўлчам объекти), узунасига эса катта назорат плитаси бўйича устун индикатори билан биргаликда силжиган. Бу пн барча юзаси бўйича ўлчаш имконини яратган. Пн тайёрлашда қўйилган хатоликларга боғлиқ ҳолда унинг қалинлиги доимий эмас, шу сабабли ростлаш катталиги АХ ни ҳақиқий катталик А га боғлиқ ҳолда коррекциялашга тўғри келган. Ушбу ишда Ах катталик қиймати қилиб қуйидагилар қабул қилинган: 10±0,05 мм. Ейилиш катталиги қуйидаги равишда аниқланган. А ва Ах орасидаги фарқ ҳисобланган А −АХ = ∆А, сўнг ∆Аф назорат қилинаётган ейилиш майдонининг i-нуқтасида ўлчанган ҳақиқий ўлчам Аф орасида кўрилган ўлчам АХ орасидаги фарқ ҳисобланган, яъни А −АХ = ∆А. Бунда ейилиш катталиги U ишчи юзанинг ҳар қандай нуқтасида қуйидаги кўринишда ифодаланиши мумкин. U = ∆A −∆Аф. (1) (1) Ростлаш катталиги АХ ўзгарувчан, бу вақтда ҳақиқий ўлчамларни Аф ни кўрилган ҳар бир ўлчам қиймати АХ учун ҳисоб ҳажмини камайтириш ўртачасини олиш зарур. Масалан координаталари III (4) бўлган назорат қилинаётган пн нуқтаси учун (расм 4.8) Ан=10 мм бўлганда ҳосил қиламиз. ∆Аф ўрт. АННОТАЦИЯ = ∑ ∆Aфi n=20 i=1 20 , (2) (2) бу ерда n-20 рақами ўлчамлари 10 мм бўлган пн лар сони (20та назорат қилинаётган пн лар орасидан); бу ерда n-20 рақами ўлчамлари 10 мм бўлган пн лар сони (20та назорат қилинаётган пн лар орасидан); ∆Аф ўрт.- ейилишнинг ўртача қиймати, мкм. Multidisciplinary Scientific Journal December, 2022 613 613 Multidisciplinary Scientific Journal Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 ∆Аф ўр = (150 + 675 + 876 + 338 + 300 + 311 + 173 + 969 + 177 + 738 + +985 + 187 + 1611 + 417 + 541 + 587 + 642 + 867 + 598 + 681 + 388 + +518 + 1059 + 365 + 73 + 113 + 901)/20 = 503мкм. ∆Аф ўр = (150 + 675 + 876 + 338 + 300 + 311 + 173 + 969 + 177 + 738 + +985 + 187 + 1611 + 417 + 541 + 587 + 642 + 867 + 598 + 681 + 388 + +518 + 1059 + 365 + 73 + 113 + 901)/20 = 503мкм. Шундай қилиб,ейилиш назорат қилинаётган майдон барча нуқталарда ўлчанган натижаларни ва Ан нинг барча қийматлари учун ўртача қиймати аниқлаш олиб борилган. Ҳар бир қўйилган ўлчам Ан учун (10,01; 10,03; 10,05; 10,07 мм) ∆А ўлчам ҳисоблаб чиқилган. Бунинг учун емирилмаган четки участкадан пн дастлабки ҳолатини аниқлаш учун текислик ўтказилган (пн тайёрлашда олинган). Масалан Ах=10,05мм ли пн учун ∆Афқиймати V(l) қирқимда қиймати 611, 576, 625, 618, 576 ва 575 мкм га тенг. И (чекка) қирқимда эса ∆Аф қиймати 532, 583, 643, 665, 653, 597 мкм га тенг бўлганда ∆А=640мкм ҳосил бўлади. Шундай қилиб ∆А қиймати пн лар бошқа гуруҳлари учун ҳар бир қурилган ўлчам учун ҳисоблаб чиқилган. Сўнг ейилиш катталиги Uҳар бир кўрилган ўлчам учун Ах алоҳида, масалан Ан =10,05 мм пн лар учун Сўнг ейилиш катталиги Uҳар бир кўрилган ўлчам учун Ах алоҳида, масалан Ан =10,05 мм пн лар учун = 615 мкм; ∆Аф = 470 мкм – I (5) координата нуқталари учун; Бунда бешинчи нуқта учун I қирқимда (4.8-расм) ∆Аф = 470 мкм – I (5) координата нуқталари учун; Бунда бешинчи нуқта учун I қирқимда (4.8-расм) ∆𝐴5 = 5 ∙ ∆𝐴11−∆𝐴1 11 + ∆𝐴1. (3) ∆𝐴5 = 5 ∙640 −615 11 + 615 = 627,5 мкм. ∆𝐴5 = 5 ∙ ∆𝐴11−∆𝐴1 11 + ∆𝐴1. АННОТАЦИЯ (3) ∆𝐴5 = 5 ∙ ∆𝐴11−∆𝐴1 11 + ∆𝐴1. ∆𝐴5 = 5 ∙ ∆𝐴11−∆𝐴1 11 + ∆𝐴1. (3) ∆𝐴5 = 5 ∙640 −615 11 + 615 = 627,5 мкм. Ейилиш катталиги U(4.5) билан мос равишда ташкил этади. Ейилиш катталиги U(4.5) билан мос равишда ташкил этади. U=627,5 - 470=157,5 мкм. Ейилиш катталиги U(4.5) билан мос равишда ташкил этади. Ейилиш катталиги U(4.5) билан мос равишда ташкил этади. U=627,5 - 470=157,5 мкм. Ейилишни барча олинган катталиклари Ан катталигига боғлиқ жадвал (Е илова) га киритилган. Ейилишни барча олинган катталиклари Ан катталигига боғлиқ жадвал (Е илова) га киритилган. Йигирмата пн учун ейилишни ўртача катталиги аниқлашнинг охирги операцияси Ах турли нуқталари учун олинган ейилиш қийматларини ўртачасини аниқлашдир. 𝑈𝛴= ∑ 𝑈𝑖 𝑛 𝑖=1 𝑛 , 𝑈𝛴= ∑ 𝑈𝑖 𝑛 𝑖=1 𝑛 , (4) 𝑈𝛴= ∑ 𝑈𝑖 𝑛 𝑖=1 𝑛 , (4) (4) (4) бунда 𝑈𝑖−i-гуруҳ ишчи юзасидаги ҳар бир назорат қилинадиган нуқтасидаги ейилиш;n – пн гуруҳлари сони. бунда 𝑈𝑖−i-гуруҳ ишчи юзасидаги ҳар бир назорат қилинадиган нуқтасидаги ейилиш;n – пн гуруҳлари сони. Масалан, 4 (II) координатали нуқта учун 450 + 400 + 530 + 672 + 274 Масалан, 4 (II) координатали нуқта учун 614 https://t.me/Erus_uz https://t.me/Erus_uz Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 Ҳисоблаш натижалари 4.5-жадвалга киритилган. Шунга ўҳшаш ҳисоб штамп пн лари олд томони учун бажарилган, уларда ейилиш майдонида нуқталар сони 36 тани ташкил этган. Ҳисоблаш натижалари 4.5-жадвалга киритилган. Шунга ўҳшаш ҳисоб штамп пн лари олд томони учун бажарилган, уларда ейилиш майдонида нуқталар сони 36 тани ташкил этган. 1-жадвал Штамп ён томон пн лари юзасининг ейилиш катталиги, мкм; Ён томон пн ларининг ўртача ейилиши, мкм № 1 2 3 4 5 6 7 8 I 36 127 142 163 108 65 44 58 II 17 133 155 186 124 89 68 66 III 26 244 145 174 145 97 88 82 IV 11 178 174 131 164 67 114 64 V 24 184 189 201 177 68 74 54 VI 16 104 104 152 87 89 66 24 2-жадвал Штамп олд томон пн лари юзасининг ейилиш катталиги, мкм; 2-жадвал Штамп олд томон пн лари юзасининг ейилиш катталиги, мкм; Штамп олд томон пн лари юзасининг ейилиш катталиги, мкм; Штамп олд томон пн лари юзасининг ейилиш катталиги, мкм; Олд томон пн ларининг ўртача ейилиши, мкм № 1 2 3 4 5 6 I 69 74 40 61 52 33 II 41 61 119 44 51 49 III 39 47 124 68 74 81 IV 19 55 107 78 68 36 V 18 78 116 84 64 18 VI 65 87 81 63 41 19 Пн лар ейилиш катталигини ўлчаш натижалари статистик ишлов бериш натижасида пн қирқимлари бўйича ейилиш катталигини тақсимланиш назарий қонунини аниқлаш учун ейилишни тарқалиш эмперик эгри чизиғини қуриш зарур ва назарий тарқалишини эмперикка яқинлик даражасини аниқлаш лозим. Бу мақсадда координаталари 3(II); 3(V); 5(II); 5(V)нуқталарда ейилишни тўрт қатор қиймати танлаб олинган. 615 December, 2022 Multidisciplinary Scientific Journal https://t.me/Erus_uz Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 4.10-расм. MathCalc дастурининг бошқариш алгоритми блок-схемаси Ҳисоб жадвал кўринишида келтирилган Е1, Е4, Е7 ва Е10 иловаларда келтирилган. У ерда барча тўрт қатор учун ейилишни тақсимланиш эгри чизиғи частоталар полигони орқали ифодаланган (Е3, Е4, Е5, Е6-расмлар). σ- дан ташқари тақсимланиш характерловчи шундай кўрсатгич: 𝑎𝑠- ў ў б ё 4.10-расм. MathCalc дастурининг бошқариш алгоритми блок-схемаси 4.10-расм. MathCalc дастурининг бошқариш алгоритми блок-схемаси Ҳисоб жадвал кўринишида келтирилган Е1, Е4, Е7 ва Е10 иловаларда келтирилган. У ерда барча тўрт қатор учун ейилишни тақсимланиш эгри чизиғи частоталар полигони орқали ифодаланган (Е3, Е4, Е5, Е6-расмлар). σ- дан ташқари тақсимланиш характерловчи шундай кўрсатгич: 𝑎𝑠- ассиметрия ўлчами ва 𝑒𝑘- эксцесс кўрсатгичиишда баён этилган усулда ҳисобланган [8]. Уларни тўлиқ кўриб чиқамиз. имметрия ўлчами 𝑎𝑠 - қиймат қуйидаги формула бўйича ҳисобланади: 𝑎𝑠= 𝑚3 𝜎𝑇 3, 𝑎𝑠= 𝑚3 𝜎𝑇 3, (5) (5) бу ерда 𝑚3 – учинчи тартибли марказий эмпирик момент [52]. 𝑀′ 3 𝑀′ 𝑀′ 2 (𝑀′)3 (6) ∙𝑀2 ′ ∙𝑀1 ′ + 2 ∙(𝑀1 ′)3; (6) 𝑀1 ′ = ∑𝑛𝑖𝑢𝑖 𝑛 ; 𝑀2 ′ = ∑𝑛𝑖𝑢𝑖 2 𝑛 ; 𝑀3 ′ = ∑𝑛𝑖𝑢𝑖 3 𝑛 ; 𝑛𝑖 – барча йиғиндилар сони, 𝑛𝑖 – барча йиғиндилар сони, 𝑢𝑖 –шартли варианта, 𝑢𝑖= 𝑛𝑖−𝐶 ℎ ; 𝑢𝑖 –шартли варианта, 𝑢𝑖= 𝑛𝑖−𝐶 ℎ ; С – энг катта частотали варианта, (С=0) h – иккита қўшни варианта айирмаси h – иккита қўшни варианта айирмаси Эксцесс кўрсатгичи 𝑒𝑘- ушбу кўрсатгич кўриб чиқилаётган қаторнинг https://t.me/Erus_uz Multidisciplinary Scientific Journal December, 2022 616 https://t.me/Erus_uz 616 Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 тақсимлаш маркази олдида алоҳида қийматлар концентрацияси бўйича меъёрий кўрсатгичдан фарқ қилишини акс эттиради. Ушбу кўрсатгич қуйидаги формула билан аниқланади. Штамп олд томон пн лари юзасининг ейилиш катталиги, мкм; 𝑒𝑘= 𝑚4 𝜎𝑇 4 −3, (7) бунда 𝑚4 –тўртинчи тартибли марказий эмпирик момент [9]. 𝑚4 = 𝑀4 ′ −4 ∙𝑀3 ′ ∙𝑀1 ′ + 6 ∙𝑀2 ′ ∙(𝑀1 ′)2 −3 ∙(𝑀1 ′)4; (8) 𝑀1 ′ = ∑𝑛𝑖𝑢𝑖 𝑛 ; 𝑀2 ′ = ∑𝑛𝑖𝑢𝑖 2 𝑛 ; 𝑀3 ′ = ∑𝑛𝑖𝑢𝑖 3 𝑛 ; 𝑀4 ′ = ∑𝑛𝑖𝑢𝑖 4 𝑛 ; (7) да 𝑚4 –тўртинчи тартибли марказий эмпирик момент [9]. 𝑀4 ′ −4 ∙𝑀3 ′ ∙𝑀1 ′ + 6 ∙𝑀2 ′ ∙(𝑀1 ′)2 −3 ∙(𝑀1 ′)4; (8) 𝑀1 ′ = ∑𝑛𝑖𝑢𝑖 𝑛 ; 𝑀2 ′ = ∑𝑛𝑖𝑢𝑖 2 𝑛 ; 𝑀3 ′ = ∑𝑛𝑖𝑢𝑖 3 𝑛 ; 𝑀4 ′ = ∑𝑛𝑖𝑢𝑖 4 𝑛 ; 𝑛𝑖 – барча йиғиндилар сони, 𝑢𝑖 – шартли варианта, 𝑢𝑖= 𝑥𝑖−𝐶 ℎ; 𝑢𝑖 – шартли варианта, 𝑢𝑖= 𝑥𝑖−𝐶 ℎ; С – энг катта частотали варианта, (С=0), С – энг катта частотали варианта, (С=0), С – энг катта частотали варианта, (С=0), С – энг катта частотали варианта, (С=0), h – иккита қўшни варианта айирмаси. h – иккита қўшни варианта айирмаси. Агар 𝑒𝑘>3 бўлса эксцесс ижобий (эгри чизиқ чўққиси меъёрдан баланд) ва аксинча 𝑒𝑘<3 бўлса бунда тақсимланиш эгри чизиғи чўққиси меъёрдан пастда жойлашган бўлади ва эксцесс салбий бўлади. Эксцесс йўқ бўлса 𝑒𝑘=3 қийматга эга бўлади. Координаталари 5(V) нуқта учун 𝑒𝑘= 0,587 0,0115 −3 = 7 Координаталари 5(V) нуқта учун 𝑒𝑘= 0,587 0,0115 −3 = 7 Координаталари 5(V) нуқта учун 𝑒𝑘= 0,587 0,0115 −3 = 7 Қолган 𝑎𝑠 ва 𝒆𝒌 қийматлари Е1, Е4, Е7, Е10 жадвалларда келтирилган. 𝑒𝑘>0 бўлган эксцесснинг барча қийматлари, яъни тақсимлаш эгри чизиқлари частоталар полигонида ижобий эксцесс кузатилади(Е3 – Е6-расмлар) [5]. Ассимметрия ўлчамига 𝑎𝑠 – келсак, бунда координаталари 3(II) ва 3(V) нуқталар учун иккита салбий қиймат ва иккита 5(II) и 5(V) координата нуқталари учун ижобий қиймат олинган. Бу салбий ёки ижобий ассиметриялиги ҳақида маълумот беради. Ижобий ассимметрияда ўлчамлар ўрта арифметик қиймати модадан ўнгроқда жойлашган, салбийда эса чапроқда жойлашади [9]. Ейилиш турли қаторлардаги натижаларни горизонтал ва вертикал бўйича таққослаш масаласи жуда катта аҳамиятга эга, чунки у қаторлар орасида, натижаларда қандайдир тасодифий ходисалар борлигини аниқлашга имкон беради. Масалан, пресслаш жараёнида зичлашни турли босқичларида турли катталикдаги босим ёки зичлашда пуансоннинг турли тезлик билан ҳаракатланишига боғлиқ ҳолда амалга оширилади [10]. Бунинг учун ейилиш қаторлари бўйича ўртача қийматларни таққослаш усулидан фойдаланилади. Multidisciplinary Scientific Journal December, 2022 617 617 December, 2022 Multidisciplinary Scientific Journal Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 4.11-расм. Ён пн лар ейилган юзатопографияси 4.12-расм. Олд пн лар ейилган юза топографияси Қаторлар бўйича ейилишни ўртача қийматини таққослаш таҳлили кўрсатишича кки горизонтал қаторлар орасида (координаталари 3(II) ва 5(II); 3(V) ва 5(V)) андайдир аҳамиятга молик бўлмаган тасодифий хатоликлар таъсири келтириб иқарган бўлиши мумкин. Вертикал қатордаги фарқ координаталари 3(II) ва 3(V) ўлган нуқтада пресслаш жараёнини характерлайдиган аниқ қонуният асосида елиб чиққан. Шундай қилиб ейилиш характери пн лар горизонтал қатори ўйича ўхшаш, вертикал қаторлар бўйича турли ташқи факторлар таъсиридан елиб чиққан сезиларли равишдаги фарқ мавжуд [45 104] 4.11-расм. Ён пн лар ейилган юзатопографияси 4.11-расм. Ён пн лар ейилган юзатопографияси 4.11-расм. Ён пн лар ейилган юзатопографияси 4.12-расм. Олд пн лар ейилган юза топографияси 4.12-расм. Олд пн лар ейилган юза топографияси Қаторлар бўйича ейилишни ўртача қийматини таққослаш таҳлили кўрсатишича икки горизонтал қаторлар орасида (координаталари 3(II) ва 5(II); 3(V) ва 5(V)) қандайдир аҳамиятга молик бўлмаган тасодифий хатоликлар таъсири келтириб чиқарган бўлиши мумкин. Вертикал қатордаги фарқ координаталари 3(II) ва 3(V) бўлган нуқтада пресслаш жараёнини характерлайдиган аниқ қонуният асосида келиб чиққан. Шундай қилиб ейилиш характери пн лар горизонтал қатори бўйича ўхшаш, вертикал қаторлар бўйича турли ташқи факторлар таъсиридан келиб чиққан сезиларли равишдаги фарқ мавжуд [45, 104]. Қурилган топографиялар таҳлили кўрсатишича штамп ён пн лар ейилиши олд пн ларга нисбатан деярли икки баробар узунроқ (4.10-расм). Координаталари 5(V) нуқта учун 𝑒𝑘= 0,587 0,0115 −3 = 7 Корхона технологияси бўйича нитроцементитлангандан сўнг чиниқтириш (бўшатиш) йўли билан ишлаб чиқарилган штамп пн лари ейилиш юзасини юқоридаги усулда топографиясини таҳлил қилинганда https://t.me/Erus_uz Multidisciplinary Scientific Journal December, 2022 618 618 December, 2022 Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 ейилишга чидамлилиги амалда 3 мартага пасайди. Бир сменада ишлаб чиқарилган керамик подкладка миқдори пн лар алмаштирилмаганда 20 минг донани ташкил этди. Бундан сўнг ишлаб чиқарилган маҳсулотда прессланаётган керамик тўшама (подкладка) ларда дефектлар ҳосил бўлиш сабаби, пн ларни алмаштиришга зарурат пайдо бўлганлигига зарурат сезилганлигидир [5]. ейилишга чидамлилиги амалда 3 мартага пасайди. Бир сменада ишлаб чиқарилган керамик подкладка миқдори пн лар алмаштирилмаганда 20 минг донани ташкил этди. Бундан сўнг ишлаб чиқарилган маҳсулотда прессланаётган керамик тўшама (подкладка) ларда дефектлар ҳосил бўлиш сабаби, пн ларни алмаштиришга зарурат пайдо бўлганлигига зарурат сезилганлигидир [5]. Шундай қилиб ўлчанган натижаларни статик ишлов беришни Калмогоров критерияси бўйича баҳолаш кўрсатишича штамп пўлат пн ларининг ейилиш катталигини тақсимланиши меъёрий тақсимланиш қонунига бўйсунади [6]. Ейилиш юзасининг топографияси таҳлили асосида шундай хулосага келиш мумкин: пн лар ейилиши бир текисда бўлмаслиги ва рухсат этилган ейилиш чегарасини аниқлаш, маҳсулотга қўйилган ўлчам талабларига боғлиқ ҳолда асосий ўлчамдан ташқи тарафга 0,3 дан 0,5 мм гачани ташкил этади. ХУЛОСАЛАР 1. Абразив заррачалари мавзуд аралашмаларни зичлашда ҳосил бўладиган зўриқишни аниқлашга имкон берадиган назарий тадқиқотлар бажарилди. 2. Абразив заррачалари мавжуд аралашмаларни пресслаш йўли билан зичлашда 1. Абразив заррачалари мавзуд аралашмаларни зичлашда ҳосил бўладиган зўриқишни аниқлашга имкон берадиган назарий тадқиқотлар бажарилди. 2. Абразив заррачалари мавжуд аралашмаларни пресслаш йўли билан зичлашда вертикал, горизонтал ва тўқнашув зўриқишлар аналитик боғлиқликлари олинди. 3. Пуансон ҳаракатида абразив заррачаларнинг штамп пластиналари ён ва олд томонларига босимни тарқалиши ўрганилди. 1. Абразив заррачалари мавзуд аралашмаларни зичлашда ҳосил бўладиган зўриқишни аниқлашга имкон берадиган назарий тадқиқотлар бажарилди. 1. Абразив заррачалари мавзуд аралашмаларни зичлашда ҳосил бўладиган зўриқишни аниқлашга имкон берадиган назарий тадқиқотлар бажарилди. 2. Абразив заррачалари мавжуд аралашмаларни пресслаш йўли билан зичлашда вертикал, горизонтал ва тўқнашув зўриқишлар аналитик боғлиқликлари олинди. 3. Пуансон ҳаракатида абразив заррачаларнинг штамп пластиналари ён ва олд томонларига босимни тарқалиши ўрганилди. 2. Абразив заррачалари мавжуд аралашмаларни пресслаш йўли билан зичлашда вертикал, горизонтал ва тўқнашув зўриқишлар аналитик боғлиқликлари олинди. 3. Пуансон ҳаракатида абразив заррачаларнинг штамп пластиналари ён ва олд томонларига босимни тарқалиши ўрганилди. 4. Штамп пластиналаридаги босим катталигига, қолипга аралашма солиниш баландлигини ростлашдаги ўзгариши аниқланди. 4. Штамп пластиналаридаги босим катталигига, қолипга аралашма солиниш баландлигини ростлашдаги ўзгариши аниқланди. 5. Штамп пластиналари ейилиш бардошлигини ошириш борасида диффузияли борлаш технологияси жараёни назарий тадқиқоти ва технологик факторларга боғлиқ бўлган математик модели ишлаб чиқилди. Бу бор концентрацияси ва диффузия вақтига боғлиқ ҳолда мустаҳкамланган қатлам қалинлигини аниқлаш имконини берувчи ва ресурстежамкорлигини таъминловчи технологияларни ишлаб чиқиш имконини беради. 6. Диффузияли борлаш технологияси учунюқори ейилишбардошликка эга бўлган FeB ва Fe2B темир боридларидан ташкил топган таркиб ишлаб чиқилди. Бу эса кукунсимон бор сақловчи аралашманинг янги таркибидан фойдаланиш, пўлат пластиналари юзасида барқарор икки фазали қатламни ҳосил қилишга имкон яратади. 7. Штамп пластиналари учун пўлат 20 материалини қўллаш асосида унинг юзасига диффузияланадиган қатлами 20-25 мкмга таъминланиши аниқланди. Буштамп пластиналарини ишлаб чиқариш шароитига тадбиқ этиш имконини беради. 8. Керамик маҳсулотларни пресслашда технологик факторларга боғлиқ бўлган https://t.me/Erus_uz Multidisciplinary Scientific Journal December, 2022 619 https://t.me/Erus_uz Multidisciplinary Scientific Journal December, 2022 619 619 Educational Research in Universal Sciences ISSN: 2181-3515 Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 ва емирилиш катталиги технологик факторлар орасидаги ўзаро ФОЙДАЛАНИЛГАН АДАБИЁТЛАР: (REFERENCES) 1. Todjiboyev R.K., Ulmasov A.A., & Muxtorov Sh. (2021). 3M structural bonding tape 9270. Science and Education, 2 (4), 146-149. 2. Toshkoziyeva, Z., & Muxtorov, S. (2022). DESIGN ANALYSIS FOR THE PRODUCTION OF PLATE HANDLES FOR CAR WINDSHIELDS. Journal of Integrated Education and Research, 1(1), 164–172. Retrieved from https://ojs.rmasav.com/index.php/ojs/article/view/34 3. ХУЛОСАЛАР АРРАЛИ ДЖИН КОЛОСНИКЛАРИ АЛМАШУВЧИ ЭЛЕМЕНТЛАРИНИ КОНСТРУКТИВ ЎЛЧАМЛАРИНИ АСОСЛАШ. Nazariy va amaliy tadqiqotlar xalqaro jurnali, 2 3, 88-97. doi: 10.5281/zenodo.6503659odo.6503605 17. Бахадиров, Г. А., Цой, Г. Н., Набиев, А. М., & Эргашев, И. О. (2022). Экспериментальный Отжим Капиллярно-Пористого Материала На Металлокерамической Опорной Плите. Central Asian Journal of Theoretical and Applied Science, 3(5), 100-109. Retrieved from https://cajotas.centralasianstudies.org/index.php/CAJOTAS/article/view/499 18. Fayzimatov Shukhrat Nomonovich, Ergashev Ilhomjon Olimjonovich, & Valikhonov Dostonbek Azim o’g’li (2022) Effects Of Crushing on Cutting and НИТЕЙ. Educational Research in Universal Sciences, 1(6), 524–533. Retrieved from http://erus.uz/index.php/er/article/view/820 НИТЕЙ. Educational Research in Universal Sciences, 1(6), 524–533. Retrieved from http://erus.uz/index.php/er/article/view/820 11. Xusanboyev, A., & Muxtorov, S. (2022). NOSOZLIKLAR SONINI TAQSIMLASH VA KANALIZATSIYA TARMOQLARI ELEMENTLARINI TIKLASH MUDDATI. Educational Research in Universal Sciences, 1(6), 617–625. Retrieved from http://erus.uz/index.php/er/article/view/831 12. Abdullayeva, D., & Muxtorov, S. (2022). SEYSMIK HUDUDLARDA KANALIZATSIYA TARMOQLARINI ISHONCHLILIGINI BAHOLASH. Educational Research in Universal Sciences, 1(6), 514–523. Retrieved from http://erus.uz/index.php/er/article/view/818 13. Toshqo‘ziyeva, Z., & Muxtorov, S. (2022). KANALIZATSIYA TARMOQLARI ELEMENTLARINING ISHONCHLILIGI KO‘RSATKICHLARINING SON QIYMATLARINI ANIQLASH. Educational Research in Universal Sciences, 1(6), 609–616. Retrieved from http://erus.uz/index.php/er/article/view/830 14. Khusanboyev, A., & Mukhtorov, S. (2022). IMPROVING THE STRENGTH OF DETAILS BY CHROMING THE SURFACES. Educational Research in Universal Sciences, 1(6), 626–634. Retrieved from http://erus.uz/index.php/er/article/view/832 15. Бахадиров, Гайрат Атаханович , Эргашев, Илхомжон Олимжонович, Цой, Герасим Николаевич, & Набиев, Айдер Мустафаевич (2022). УСТРОЙСТВО ДЛЯ ОПРЕДЕЛЕНИЯ СИЛЫ ВТЯГИВАНИЯ ПЛОСКОГО МАТЕРИАЛА МЕЖДУ РАБОЧИМИ ВАЛКОВЫМИ ПАРАМИ. Nazariy va amaliy tadqiqotlar xalqaro jurnali, 2 (3), 66-73. doi: 10.5281/zen 16. Эргашев, Илхомжон Олимжонович (2022). АРРАЛИ ДЖИН КОЛОСНИКЛАРИ АЛМАШУВЧИ ЭЛЕМЕНТЛАРИНИ КОНСТРУКТИВ 17. Бахадиров, Г. А., Цой, Г. Н., Набиев, А. М., & Эргашев, И. О. (2022). Экспериментальный Отжим Капиллярно-Пористого Материала На Металлокерамической Опорной Плите. Central Asian Journal of Theoretical and Applied Science, 3(5), 100-109. Retrieved from 17. Бахадиров, Г. А., Цой, Г. Н., Набиев, А. М., & Эргашев, И. О. (2022). Экспериментальный Отжим Капиллярно-Пористого Материала На Металлокерамической Опорной Плите. Central Asian Journal of Theoretical and Applied Science, 3(5), 100-109. Retrieved from p j g p p 18. Fayzimatov Shukhrat Nomonovich, Ergashev Ilhomjon Olimjonovich, & Valikhonov Dostonbek Azim o’g’li. (2022). Effects Of Crushing on Cutting and Cleaning of Surface Facilities in Cutting and Processing of Polymer Materials. Eurasian Research Bulletin, 4, 17–21. Retrieved from https://www.geniusjournals.org/index.php/erb/article/view/353 18. Fayzimatov Shukhrat Nomonovich, Ergashev Ilhomjon Olimjonovich, & Valikhonov Dostonbek Azim o’g’li. (2022). Effects Of Crushing on Cutting and Cleaning of Surface Facilities in Cutting and Processing of Polymer Materials. Eurasian Research Bulletin, 4, 17–21. Retrieved from https://www.geniusjournals.org/index.php/erb/article/view/353 19. ХУЛОСАЛАР Toshkoziyeva, Z., & Muxtorov, S. (2022). ANALYSIS OF THE REQUIREMENTS FOR MODERN HEAT EXCHANGERS AND METHODS OF PROCESS INTENSIFICATION. Journal of Integrated Education and Research, 1(1), 140–149. Retrieved from https://ojs.rmasav.com/index.php/ojs/article/view/30 4. Sherzod Sobirjon, O. G. ’Li Muxtorov, & Islombek Ikromjon O’G’Li Qoxxorov (2022). Issiqlik almashuvchi qurulmalar va ularda jarayonni intensivlash usullari tahlili. Science and Education, 3(5), 370-378. 5. Toshqo’ziyeva, Z., & Muxtorov, S. (2022). AVTOMABILLARNI 3M STRUKTURALI ULASH LENTASI BILAN MAXKAMLANUVCHI PLASTINA TUTQICHI KONSTRUKSIYALARINI TAXLILI. Journal of Integrated Education and Research, 1(1), 114–125. Retrieved from https://ojs.rmasav.com/index.php/ojs/article/view/27 6. Махмудов, А., & Мухторов , Ш. (2022). ВЛИЯНИЕ ИНДИВИДУАЛЬНОГО УВЛАЖНИТЕЛЯ НА ОБРЫВНОСТЬ НИТЕЙ ОСНОВЫ В ПРОЦЕССЕ ТКАЧЕСТВА. Eurasian Journal of Academic Research, 2(13), 884–890. извлечено от https://www.in-academy.uz/index.php/ejar/article/view/7639 7. Махмудов, А., & Мухторов , Ш. (2022). ИССЛЕДОВАНИЕ ОСНОВНОГО ПЛАНЕТАРНОГО РЕГУЛЯТОРА. Eurasian Journal of Academic Research, 2(13), 879–883. извлечено от https://in-academy.uz/index.php/ejar/article/view/7638 8. Mukhtorov, S. S. ugli, & Rustamova, M. M. (2022). AN ANALYSIS OF THE IMPACT OF CONFIDENCE ON THE RELIABILITY OF EARTHQUAKE DETECTION UNDERGROUND. Educational Research in Universal Sciences, 1(6), 480–487. Retrieved from http://erus.uz/index.php/er/article/view/813 9. Mukhtorov, S. S. ugli, & Rustamova, M. M. (2022). IMPROVING THE STRENGTH OF DETAILS BY CHROMING THE SURFACES. Educational Research in Universal Sciences, 1(6), 488–496. Retrieved from http://erus.uz/index.php/er/article/view/814 10. Нурматова C. С., & Myхторов Ш. С. (2022). В ПРОЦЕССЕ ПЛЕТЕНИЯ ВЛИЯНИЕ ТОЧНОГО СМАЧИВАНИЯ НА ОБРЫВ СОЕДИНИТЕЛЬНЫХ VOLUME 1 \| ISSUE 7 \| 2022 ва емирилиш катталиги технологик факторлар орасидаги ўзаро 620 Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 НИТЕЙ. Educational Research in Universal Sciences, 1(6), 524–533. Retrieved from http://erus.uz/index.php/er/article/view/820 11. Xusanboyev, A., & Muxtorov, S. (2022). NOSOZLIKLAR SONINI TAQSIMLASH VA KANALIZATSIYA TARMOQLARI ELEMENTLARINI TIKLASH MUDDATI. Educational Research in Universal Sciences, 1(6), 617–625. Retrieved from http://erus.uz/index.php/er/article/view/831 12. Abdullayeva, D., & Muxtorov, S. (2022). SEYSMIK HUDUDLARDA KANALIZATSIYA TARMOQLARINI ISHONCHLILIGINI BAHOLASH. Educational Research in Universal Sciences, 1(6), 514–523. Retrieved from http://erus.uz/index.php/er/article/view/818 13. Toshqo‘ziyeva, Z., & Muxtorov, S. (2022). KANALIZATSIYA TARMOQLARI ELEMENTLARINING ISHONCHLILIGI KO‘RSATKICHLARINING SON QIYMATLARINI ANIQLASH. Educational Research in Universal Sciences, 1(6), 609–616. Retrieved from http://erus.uz/index.php/er/article/view/830 14. Khusanboyev, A., & Mukhtorov, S. (2022). IMPROVING THE STRENGTH OF DETAILS BY CHROMING THE SURFACES. Educational Research in Universal Sciences, 1(6), 626–634. Retrieved from http://erus.uz/index.php/er/article/view/832 15. Бахадиров, Гайрат Атаханович , Эргашев, Илхомжон Олимжонович, Цой, Герасим Николаевич, & Набиев, Айдер Мустафаевич (2022). УСТРОЙСТВО ДЛЯ ОПРЕДЕЛЕНИЯ СИЛЫ ВТЯГИВАНИЯ ПЛОСКОГО МАТЕРИАЛА МЕЖДУ РАБОЧИМИ ВАЛКОВЫМИ ПАРАМИ. Nazariy va amaliy tadqiqotlar xalqaro jurnali, 2 (3), 66-73. doi: 10.5281/zen 16. Эргашев, Илхомжон Олимжонович (2022). ХУЛОСАЛАР Ilhom Olimjonovich Ergashev, Rustam Jaxongir O’G’Li Karimov, Ravshan Xikmatullayevich Karimov, & Salimaxon Sobirovna Nurmatova (2021). KOLOSNIK 19. Ilhom Olimjonovich Ergashev, Rustam Jaxongir O’G’Li Karimov, Ravshan Xikmatullayevich Karimov, & Salimaxon Sobirovna Nurmatova (2021). KOLOSNIK 621 December, 2022 https://t.me/Erus_uz Educational Research in Universal Sciences ISSN: 2181-3515 VOLUME 1 \| ISSUE 7 \| 2022 ALMASHINUVCHI MASHINASI ELEMENTI EGILISHINING NAZARIY TADQIQOTLARI. Scientific progress, 2 (7), 83-87. ALMASHINUVCHI MASHINASI ELEMENTI EGILISHINING NAZARIY TADQIQOTLARI. Scientific progress, 2 (7), 83-87. 20. Ergashev Ilhomjon Olimjonoviсh, & Mahmudov Nasimbek Odilbekoviсh. (2022). Сalсulation of Сarrier and Interсhangeable Element Сombination. Eurasian Journal of Engineering and Technology, 5, 68–73. Retrieved from https://www.geniusjournals.org/index.php/ejet/article/view/1162 21. Мухаммадиев, Д. М., Ахмедов, Х. А., & Эргашев, И. О. (2020). Расчет перемещений вставки относительно колосник. In Инновационные исследования: теоретические основы и практическое применение (pp. 103-105). 22. Мухаммадиев, Д. М., Ахмедов, Х. А., Эргашев, И. О., Жамолова, Л. Ю., & Мухаммадиев, Т. Д. (2020). Силовой расчет соединений колосника пильного джина со вставкой. Известия высших учебных заведений. Технология текстильной промышленности, (1), 137-143. 23. Sherzod Sobirjon O’G’Li Muxtorov, & Islombek Ikromjon O’G’Li Qoxxorov (2022). Issiqlik almashuvchi qurulmalar va ularda jarayonni intensivlash usullari tahlili. Science and Education, 3 (5), 370-378 24. Toxir Yusupovich Radjabov, Akbar Turg’Unboyevich Ergashev, Ilhomjon Yusufjonovich Mirzaolimov, & Abdulaziz Ikhtior Ugli Karshiboev (2022). EXAMPLE OF CALCULATION OF REINFORCED CONCRETE BEAM SPANS FOR TEMPORARY (A-14 and NK-100) AND PERMANENT LOADS. Academic research in educational sciences 3 (TSTU Conference 1) 908 913 21. Мухаммадиев, Д. М., Ахмедов, Х. А., & Эргашев, И. О. (2020). Расчет перемещений вставки относительно колосник. In Инновационные исследования: теоретические основы и практическое применение (pp. 103-105). 22. Мухаммадиев, Д. М., Ахмедов, Х. А., Эргашев, И. О., Жамолова, Л. Ю., & Мухаммадиев, Т. Д. (2020). Силовой расчет соединений колосника пильного джина со вставкой. Известия высших учебных заведений. Технология текстильной промышленности, (1), 137-143. 22. Мухаммадиев, Д. М., Ахмедов, Х. А., Эргашев, И. О., Жамолова, Л. Ю., & Мухаммадиев, Т. Д. (2020). Силовой расчет соединений колосника пильного джина со вставкой. Известия высших учебных заведений. Технология текстильной промышленности, (1), 137-143. 23. Sherzod Sobirjon O’G’Li Muxtorov, & Islombek Ikromjon O’G’Li Qoxxorov (2022). Issiqlik almashuvchi qurulmalar va ularda jarayonni intensivlash usullari tahlili. Science and Education, 3 (5), 370-378 23. Sherzod Sobirjon O’G’Li Muxtorov, & Islombek Ikromjon O’G’Li Qoxxorov (2022). Issiqlik almashuvchi qurulmalar va ularda jarayonni intensivlash usullari tahlili. Science and Education, 3 (5), 370-378 24. Toxir Yusupovich Radjabov, Akbar Turg’Unboyevich Ergashev, Ilhomjon Yusufjonovich Mirzaolimov, & Abdulaziz Ikhtior Ugli Karshiboev (2022). ХУЛОСАЛАР EXAMPLE OF CALCULATION OF REINFORCED CONCRETE BEAM SPANS FOR TEMPORARY (A-14 and NK-100) AND PERMANENT LOADS. Academic research in educational sciences, 3 (TSTU Conference 1), 908-913. 25. Д. М. Мухаммадиев, Ф. Х. Ибрагимов, О. Х. Абзоиров, & Л. Ю. Жамолова (2022). Расчет устойчивости междупильной прокладки при сжатии. Современные инновации, системы и технологии, 2 (4), 0301-0311. doi: 10.47813/2782-2818-2022-2-4-0301-0311 26. Mukhammadiev, D. M., Akhmedov, K. A., Ergashev, I. O., Zhamolova, L. Y., & Abdugaffarov, K. J. (2021, April). Calculation of the upper beam bending of a saw gin. In Journal of Physics: Conference Series (Vol. 1889, No. 4, p. 042042). IOP Publishing. 27. Мухаммадиев, Д. М., Ахмедов, Х. А., Примов, Б. Х., Эргашев, И. О., Мухаммадиев, Т. Д., & Жамолова, Л. Ю. (2019). Влияние радиуса кривизны лобового бруса и фартука рабочей камеры на показатели пильного джина с набрасывающим барабаном. Известия высших учебных заведений. Технология текстильной промышленности, (5), 105-110. 27. Мухаммадиев, Д. М., Ахмедов, Х. А., Примов, Б. Х., Эргашев, И. О., Мухаммадиев, Т. Д., & Жамолова, Л. Ю. (2019). Влияние радиуса кривизны лобового бруса и фартука рабочей камеры на показатели пильного джина с набрасывающим барабаном. Известия высших учебных заведений. Технология текстильной промышленности, (5), 105-110. 28. Nodirjon Ibragimovich Otaboyev, Abbosjon Sharofidin Ogli Qosimov, & Xudoyberdi Xasanboy Ogli Xoldorov (2022). AVTOPOEZD TORMOZLANISH JARAYONINI O`RGANISH UCHUN AVTOPOEZD TURINI TANLASH. Scientific progress, 3 (5), 87-92. 28. Nodirjon Ibragimovich Otaboyev, Abbosjon Sharofidin Ogli Qosimov, & Xudoyberdi Xasanboy Ogli Xoldorov (2022). AVTOPOEZD TORMOZLANISH JARAYONINI O`RGANISH UCHUN AVTOPOEZD TURINI TANLASH. Scientific progress, 3 (5), 87-92. 28. Nodirjon Ibragimovich Otaboyev, Abbosjon Sharofidin Ogli Qosimov, & Xudoyberdi Xasanboy Ogli Xoldorov (2022). AVTOPOEZD TORMOZLANISH JARAYONINI O`RGANISH UCHUN AVTOPOEZD TURINI TANLASH. Scientific progress, 3 (5), 87-92. Multidisciplinary Scientific Journal December, 2022 622 622 December, 2022 Multidisciplinary Scientific Journal https://t.me/Erus_uz https://t.me/Erus_uz
https://openalex.org/W3211537607	https://journals.ontu.edu.ua/index.php/atbp/article/download/2144/2326	Chinese	null	Замкнуті САР з прогнозуванням: аналіз альтернативних варіантів сруктур	Avtomatizaciâ tehnologičeskih i biznes-processov	2,021	cc-by	4,138	Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ [4].3DWHQW1Rȱ3C7 '.RDNVLDOމQ\\SURW\SRWRNRY\\URWDWVL\Q\\S\ORYLG-GLOމQ\N>Coaxial counter-current rotational saw-cut@ / S. I. Movchan, L. M. Datsenko, V. P. Skiba, N. V. Tarusova, A. O. Angelovska. u201910374; app. 10/15/2019; publ. 06/25/2020, bul. No. 12. >in Ukraine@. [4].3DWHQW1Rȱ3C7 '.RDNVLDOމQ\\SURW\SRWRNRY\\URWDWVL\Q\\S\ORYLG-GLOމQ\N>Coaxial counter-current rotational saw-cut@ / S. I. Movchan, L. M. Datsenko, V. P. Skiba, N. V. Tarusova, A. O. Angelovska. u201910374; app. 10/15/2019; publ. 06/25/2020, bul. No. 12. >in Ukraine@. [5].3DWHQW ʋ ,3& %' .RDNVLDOމQ\\ SURW\SRWRNRY\\ URWDWVL\Q\\ S\OR-YLGGLOމQ\N >Coaxial counterflow rotary dust separator@ / SI Movchan, VP Skiba, NM Voznyuk, OO Dereza, A. Yu. Yakunicheva; Owner: Tavriya State Agrotechnological University. Dmitry Motorny. u202006663. application 10/16/2020; publ. EXOʋ>in Ukraine@. [6].Sandler, A. K., Opryshko, M. O. (2020). Volokonno-optycheskyy datchyk kontrolya sostoyanyya tekhnycheskykh zhyd-kostey y hazov. >Fiber-optic sensor for monitoring the state of technical liquids and gases@ // X International Scientific and Methodological Conference "Ship Electrical Engineering, Electronics and Automation", 24.11.2020 - November 25, 2020: conference materials. - Odessa: NU "OMA". - P. 63-68. >in Ukraine@. [6].Sandler, A. K., Opryshko, M. O. (2020). Volokonno-optycheskyy datchyk kontrolya sostoyanyya tekhnycheskykh zhyd-kostey y hazov. >Fiber-optic sensor for monitoring the state of technical liquids and gases@ // X International Scientific and Methodological Conference "Ship Electrical Engineering, Electronics and Automation", 24.11.2020 - November 25, 2020: conference materials. - Odessa: NU "OMA". - P. 63-68. >in Ukraine@. [7].Sandler, A. K. (2021). Metod pidvyshchennya efektyvnosti diahnostuvannya tekhnichnoho stanu sudnovykh hazoturbin-nykh ustanovok na osnovi volokonno-optychnykh tekhnolohiy. >Method of increasing the efficiency of diagnosing the technical condition of ship gas turbine plants on the basis of fiber-optic technologies@: dis. ... cand. tech. Science: 05.22.20. – Ʉyiv: - 159 p. >in Ukraine@. [7].Sandler, A. K. (2021). Metod pidvyshchennya efektyvnosti diahnostuvannya tekhnichnoho stanu sudnovykh hazoturbin-nykh ustanovok na osnovi volokonno-optychnykh tekhnolohiy. >Method of increasing the efficiency of diagnosing the technical condition of ship gas turbine plants on the basis of fiber-optic technologies@: dis. ... cand. tech. Science: 05.22.20. – Ʉyiv: - 159 p. >in Ukraine@. Ɉɬɪɢɦɚɧɚ ɜ ɪɟɞɚɤɰɿʀ 10.08.2021. ɉɪɢɣɧɹɬɚ ɞɨ ɞɪɭɤɭ 25.08.2021. Received 10 August 2021. Approved 25 August 2021. Available in Internet 31 September 2021. Ɉɬɪɢɦɚɧɚ ɜ ɪɟɞɚɤɰɿʀ 10.08.2021. ɉɪɢɣɧɹɬɚ ɞɨ ɞɪɭɤɭ 25.08.2021. Received 10 August 2021. Approved 25 August 2021. Available in Internet 31 September 2021. ɍȾɄ [621.867.3:622.612]:658.5 ɁȺɆɄɇɍɌȱɋȺɊɁɉɊɈȽɇɈɁɍȼȺɇɇəɆȺɇȺɅȱɁ ȺɅɖɌȿɊɇȺɌɂȼɇɂɏȼȺɊȱȺɇɌȱȼɋɊɍɄɌɍɊ ɋɬɟɩɚɧɨɜ Ɇ.Ɍ. ɈɇȺɏɌ (ɍɤɪɚʀɧɚ) ORCID: https://orcid.org/0000-0003-1297-5537 E-mail: stepanov197818@gmail.com Copyright © 2021 by author and the journal “Automation of technological and business – processes”. This work is licensed under the Creative Commons Attribution International License (CC BY). http://creativecommons.org/licanses/by/4.0 DOI: Ⱥɧɨɬɚɰɿɹ. ɍɫɬɚɬɬɿɪɨɡɝɥɹɞɚɽɬɶɫɹɫɢɫɬɟɦɢɚɜɬɨɦɚɬɢɱɧɨɝɨɪɟɝɭɥɸɜɚɧɧɹɹɤɿɪɟɚɥɿɡɭɸɬɶɚɥɝɨɪɢɬɦɢɤɟɪɭɜɚɧɧɹɡ ɩɪɨɝɧɨɡɭɜɚɧɧɹɦɫɤɥɚɞɨɜɢɯɜɿɥɶɧɨɝɨɬɚɜɢɦɭɲɟɧɨɝɨɪɭɯɭɧɚɱɚɫɡɚɩɿɡɧɟɧɧɹɜɩɟɪɟɞɜɡɚɦɤɧɭɬɨɦɭɤɨɧɬɭɪɿ ɉɪɨɜɨɞɢɬɶɫɹ ɩɨɪɿɜɧɹɥɶɧɢɣ ɚɧɚɥɿɡ ʀɯ ɪɨɛɨɬɢ ɭ ɩɟɪɟɯɿɞɧɢɯ ɬɚ ɫɬɚɥɢɯ ɪɟɠɢɦɚɯ ɪɨɛɨɬɢ ɚ ɬɚɤɨɠ ɡɚɩɚɫɿɜ ɫɬɿɣɤɨɫɬɿ ɹɤɿ ɜɨɧɢ ɡɚɛɟɡɩɟɱɭɸɬɶ Ɉɛ ɽɤɬɢ ɬɟɯɧɨɥɨɝɿɱɧɨɝɨ ɬɢɩɭ ɞɨɫɢɬɶ ɱɚɫɬɨ ɦɚɸɬɶ ɜɟɥɢɤɭ ɿɧɟɪɰɿɣɧɿɫɬɶ ɜ ɤɚɧɚɥɚɯ ɪɟɝɭɥɸɜɚɧɧɹ ɹɤɚ ɩɨɜ ɹɡɚɧɨʀɧɟɬɿɥɶɤɢɡɱɢɫɬɢɦɡɚɩɿɡɧɟɧɧɹɦɚɥɟɛɿɥɶɲɨɸɦɿɪɨɸɡɚɤɭɦɭɥɹɰɿɽɸɪɟɱɨɜɢɧɢɿɟɧɟɪɝɿʀɬɚɤɡɜɚɧɢɦɽɦɧɿɫɧɢɦ ɡɚɩɿɡɧɟɧɧɹɦ ɉɨɜɧɚɚɛɨɱɚɫɬɤɨɜɚɤɨɦɩɟɧɫɚɰɿɹɰɿɽʀɿɧɟɪɰɿɣɧɨɫɬɿɦɨɠɟɜɡɧɚɱɧɿɣɦɿɪɿɩɨɥɿɩɲɢɬɢɹɤɿɫɬɶɪɟɝɭɥɸɜɚɧɧɹ ɞɥɹɬɚɤɢɯɨɛ ɽɤɬɿɜɇɚɩɪɚɤɬɢɰɿɞɥɹɤɨɦɩɟɧɫɚɰɿʀɜɩɥɢɜɭɡɚɩɿɡɧɟɧɧɹɧɚɞɢɧɚɦɿɤɭɜɥɚɫɧɨɝɨɪɭɯɭɱɚɫɬɨɜɢɤɨɪɢɫɬɨɜɭɸɬɶ ɫɢɫɬɟɦɢɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚɹɤɿɡɧɚɱɧɨɪɨɡɲɢɪɸɸɬɶɡɚɩɚɫɫɬɿɣɤɨɫɬɿɫɢɫɬɟɦɿɡɚɛɟɡɩɟɱɭɸɬɶʀɯɩɪɚɰɟɡɞɚɬɧɿɫɬɶ ɜ ɭɦɨɜɚɯ ɧɟɫɬɚɰɿɨɧɚɪɧɢɯ ɜɥɚɫɬɢɜɨɫɬɟɣ ɨɛ¶ɽɤɬɚ ɤɟɪɭɜɚɧɧɹ Ɍɚɤɨɠ ɩɪɨɝɧɨɡɭɜɚɧɧɹ ɜɢɤɨɪɢɫɬɨɜɭɽɬɶɫɹ ɭ ɫɢɫɬɟɦɚɯ ɤɟɪɭɜɚɧɧɹɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɜɹɤɢɯɤɟɪɭɸɱɢɣɜɩɥɢɜɧɚɤɨɠɧɨɦɭɤɪɨɰɿɪɨɡɪɚɯɨɜɭɽɬɶɫɹɡɚɪɚɯɭɧɨɤɜɢɪɿɲɟɧɧɹ ɨɩɬɢɦɿɡɚɰɿɣɧɨʀ ɡɚɞɚɱɿ ɧɚ ɨɫɧɨɜɿ ɦɚɬɟɦɚɬɢɱɧɨʀ ɦɨɞɟɥɿ ɨɛ¶ɽɤɬɚ ɤɟɪɭɜɚɧɧɹ ɐɿ ɫɢɫɬɟɦɢ ɬɚɤɨɠ ɡɚɫɬɨɫɨɜɭɸɬɶ ɞɥɹ ɤɟɪɭɜɚɧɧɹɨɛ’ɽɤɬɚɦɢɬɟɯɧɨɥɨɝɿɱɧɨɝɨɬɢɩɭɡɨɤɪɟɦɚɪɟɤɨɦɟɧɞɭɸɬɶɞɨɡɚɫɬɨɫɭɜɚɧɧɹɩɪɢɤɟɪɭɜɚɧɧɿɛɚɝɚɬɨɤɚɧɚɥɶɧɢɦɢ ɨɛ¶ɽɤɬɚɦɢ ɤɚɧɚɥɢ ɹɤɢɯ ɩɨɜ¶ɹɡɚɧɿ ɦɿɠ ɫɨɛɨɸ ɱɟɪɟɡ ɞɿɸ ɩɟɪɟɯɪɟɫɧɢɯ ɡɜ¶ɹɡɤɿɜ ȼ ɹɤɨɫɬɿ ɚɥɶɬɟɪɧɚɬɢɜɢ ɜɤɚɡɚɧɢɦ ɫɢɫɬɟɦɚɦɡɚɩɪɨɩɨɧɨɜɚɧɚɫɢɫɬɟɦɚɪɟɝɭɥɸɜɚɧɧɹɡɩɪɨɝɧɨɡɭɜɚɧɧɹɦɜɢɦɭɲɟɧɨɝɨɪɭɯɭɜɡɚɦɤɧɭɬɢɣɤɨɧɬɭɪɹɤɨʀɜɜɟɞɟɧɨ ɚɥɝɨɪɢɬɦ ɩɪɨɝɧɨɡɭɜɚɧɧɹ ɜ ɪɟɚɥɶɧɨɦɭ ɱɚɫɿ ɧɚ ɨɫɧɨɜɿ ɤɭɛɿɱɧɨɝɨ ɫɩɥɚɣɧɭ ɉɪɨɜɟɞɟɧɨ ɫɬɪɭɤɬɭɪɧɢɣ ɬɚ ɨɩɬɢɦɚɥɶɧɢɣ ɩɚɪɚɦɟɬɪɢɱɧɢɣ ɫɢɧɬɟɡ ɚɥɶɬɟɪɧɚɬɢɜɧɢɯ ɜɚɪɿɚɧɬɿɜ ɫɢɫɬɟɦ ɚɜɬɨɦɚɬɢɱɧɨɝɨ ɪɟɝɭɥɸɜɚɧɧɹ ȼ ɹɤɨɫɬɿ ɛɚɡɨɜɨɝɨ ɪɟɝɭɥɹɬɨɪɚɛɭɥɨɨɛɪɚɧɨɬɢɩɨɜɢɣɉȱȾ-ɪɟɝɭɥɹɬɨɪɉɨɪɿɜɧɹɥɶɧɢɣɚɧɚɥɿɡɨɩɬɢɦɚɥɶɧɢɯɫɢɫɬɟɦɩɪɨɜɟɞɟɧɢɣɜɱɚɫɨɜɿɣɿ ɱɚɫɬɨɬɧɢɯɨɛɥɚɫɬɹɯɩɨɤɚɡɚɜɩɟɪɟɜɚɝɭɫɢɫɬɟɦɢɪɟɝɭɥɸɜɚɧɧɹɳɨɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɤɟɪɭɜɚɧɧɹɡɚɩɪɨɝɧɨɡɨɦɧɚɨɫɧɨɜɿ ɤɭɛɿɱɧɨɝɨ ɫɩɥɚɣɧɭ ɉɪɢ ɚɧɚɥɿɡɿ ɪɨɛɨɬɢ ɫɢɫɬɟɦ ɡɚ ɤɚɧɚɥɨɦ ɞɿʀ ɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɫɢɫɬɟɦɚ ɪɟɝɭɥɸɜɚɧɧɹ ɡ ɩɪɨɝɧɨɡɭɜɚɧɧɹɦ ɩɨ ɤɭɛɿɱɧɨɦɭ ɫɩɥɚɣɧɭ ɡɚɛɟɡɩɟɱɭɽ ɡɧɢɠɟɧɧɹ ɿɧɬɟɝɪɚɥɶɧɨɝɨ ɿ ɩɪɹɦɢɯ ɩɨɤɚɡɧɢɤɿɜ ɹɤɨɫɬɿ ɩɟɪɟɯɿɞɧɢɯ ɩɪɨɰɟɫɿɜɞɨɉɟɪɟɜɿɪɤɚɧɚɝɪɭɛɿɫɬɶɫɢɫɬɟɦɚɜɬɨɦɚɬɢɱɧɨɝɨ ɪɟɝɭɥɸɜɚɧɧɹɩɨɤɚɡɚɥɚɳɨɫɢɫɬɟɦɚɚɜɬɨɦɚɬɢɱɧɨɝɨ ɪɟɝɭɥɸɜɚɧɧɹɡɩɪɨɝɧɨɡɭɜɚɧɧɹɦɪɟɝɭɥɶɨɜɚɧɨʀɡɦɿɧɧɨʀɡɚɤɭɛɿɱɧɢɦɫɩɥɚɣɧɨɦɦɚɽɩɪɢɛɥɢɡɧɨɨɞɧɚɤɨɜɢɣɡɚɩɚɫɫɬɿɣɤɨɫɬɿɡɚ ɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹɬɚɬɪɨɲɤɢɧɢɠɱɢɣɡɚɩɚɫɫɬɿɣɤɨɫɬɿɡɚɤɨɟɮɿɰɿɽɧɬɨɦɩɟɪɟɞɚɱɿɨɛ¶ɽɤɬɚɤɟɪɭɜɚɧɧɹ ɧɿɠɫɢɫɬɟɦɚɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚ 38 Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ Ʉɥɸɱɨɜɿ ɫɥɨɜɚ: ɩɪɨɝɧɨɡɭɜɚɧɧɹ ɤɭɛɿɱɧɢɣ ɫɩɥɚɣɧ ɩɪɨɝɧɨɡɭɸɱɚ ɦɨɞɟɥɶ ɫɢɫɬɟɦɚ ɚɜɬɨɦɚɬ ɡɚɩɚɫɫɬɿɣɤɨɫɬɿ Keywords: prediction, cubic spline, prediction model, automatic control system, robust control. Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ Abstract. The article considers automatic control systems that implement control algorithms with prediction of the components of free and forced motion for the time of delay in a closed loop. A comparative analysis of their work in transient and stable modes of operation, as well as the reserves of stability that they provide. Objects of a technological type quite often have a large inertia in the control channels associated not only with a pure delay, but, to a greater extent, with the accumulation of matter and energy, the so-called capacitive delay. Full or partial compensation of this inertia can significantly improve the quality of regulation for such objects. In practice, Smith's anticipatory systems are often used to compensate for the effect of the delay on the dynamics of self-motion, which significantly expand the stability of the systems and ensure their operability in conditions of non-stationary properties of the control object. Forecasting is also used in control systems with a predictive model, in which the control effect at each step is calculated by solving an optimization problem based on a mathematical model of the control object. These systems are also used for the control of technological objects, in particular, recommended for use in the control of multi-channel objects whose channels are interconnected through the action of cross-links. As an alternative to these systems, a control system with predicted forced motion in a closed loop is proposed, which introduces a real-time prediction algorithm based on a cubic spline. The structural and optimal parametric synthesis of alternative variants of automatic control systems is carried out. A typical PID controller was chosen as the basic controller. Comparative analysis of optimal systems, conducted in the time and frequency domains, showed the advantage of the control system, which implements the principle of control by forecast based on the cubic spline. When analyzing the operation of systems along the channel of action of uncontrolled perturbations, the control system with prediction of the cubic spline provides a reduction of integrated and direct quality indicators of transients up to 40%. A check on the roughness of the automatic control systems showed that the automatic control system with prediction of the variable on the cubic spline has approximately the same margin of safety over dead time and a slightly lower margin of stability for the transfer factor of the control object than the Smith biased system. ȼɫɬɭɩ ɭ Ɉɫɧɨɜɧɢɦɢɯɚɪɚɤɬɟɪɢɫɬɢɤɚɦɢ ɡɚɦɤɧɭɬɢɯɋȺɊɽɹɤɿɫɬɶɪɟɝɭɥɸɜɚɧɧɹɿɫɬɿɣɤɿɫɬɶəɤɿɫɬɶɪɟɝɭɥɸɜɚɧɧɹɜɢɡɧɚɱɚɽɬɶɫɹ ɞɢɧɚɦɿɱɧɨʀɬɚɫɬɚɬɢɱɧɨʀɬɨɱɧɿɫɬɸɋȺɊɆɚɤɫɢɦɚɥɶɧɨɞɨɫɹɠɧɚɞɢɧɚɦɿɱɧɚɬɨɱɧɿɫɬɶɋȺɊɛɚɝɚɬɨɜɱɨɦɭɨɛɦɟɠɭɽɬɶɫɹ ɡɚɩɿɡɧɟɧɧɹɦ ɿ ɿɧɟɪɰɿɣɧɿɫɬɸ ɤɚɧɚɥɿɜ ɨɛ ɽɤɬɚ ɤɟɪɭɜɚɧɧɹ Ɉɛ ɽɤɬɢ ɬɟɯɧɨɥɨɝɿɱɧɨɝɨ ɬɢɩɭ ɞɨɫɢɬɶ ɱɚɫɬɨ ɦɚɸɬɶ ɜɟɥɢɤɭ ɿɧɟɪɰɿɣɧɿɫɬɸɜɤɚɧɚɥɚɯɪɟɝɭɥɸɜɚɧɧɹɩɨɜ ɹɡɚɧɨʀɧɟɬɿɥɶɤɢɡɱɢɫɬɢɦɡɚɩɿɡɧɟɧɧɹɦɚɥɟɛɿɥɶɲɨɸɦɿɪɨɸɡɚɤɭɦɭɥɹɰɿɽɸ ɪɟɱɨɜɢɧɢɿɟɧɟɪɝɿʀɬɚɤɡɜɚɧɢɦɽɦɧɿɫɧɢɦɡɚɩɿɡɧɟɧɧɹɦɉɨɜɧɚɚɛɨɱɚɫɬɤɨɜɚɤɨɦɩɟɧɫɚɰɿɹɰɿɽʀɿɧɟɪɰɿɣɧɨɫɬɿɦɨɠɟɜɡɧɚɱɧɿɣ ɦɿɪɿɩɨɥɿɩɲɢɬɢɹɤɿɫɬɶɪɟɝɭɥɸɜɚɧɧɹɞɥɹɬɚɤɢɯɨɛ ɽɤɬɿɜɇɚɩɪɚɤɬɢɰɿɞɥɹɤɨɦɩɟɧɫɚɰɿʀɜɩɥɢɜɭɡɚɩɿɡɧɟɧɧɹɧɚɞɢɧɚɦɿɤɭ ɜɥɚɫɧɨɝɨ ɪɭɯɭ ɱɚɫɬɨ ɜɢɤɨɪɢɫɬɨɜɭɸɬɶ ɫɢɫɬɟɦɢ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ >@ ɹɤɿ ɡɧɚɱɧɨ ɪɨɡɲɢɪɸɸɬɶ ɡɚɩɚɫ ɫɬɿɣɤɨɫɬɿ ɫɢɫɬɟɦ ɿ ɡɚɛɟɡɩɟɱɭɸɬɶ ʀɯ ɩɪɚɰɟɡɞɚɬɧɿɫɬɶ ɜ ɭɦɨɜɚɯ ɧɟɫɬɚɰɿɨɧɚɪɧɢɯ ɜɥɚɫɬɢɜɨɫɬɟɣ ɨɛ¶ɽɤɬɚ ɤɟɪɭɜɚɧɧɹ Ɍɚɤɨɠ ɩɪɨɝɧɨɡɭɜɚɧɧɹɜɢɤɨɪɢɫɬɨɜɭɽɬɶɫɹɭɫɢɫɬɟɦɚɯɤɟɪɭɜɚɧɧɹɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸModel Predictive Control>@ɜɹɤɢɯ ɤɟɪɭɸɱɢɣ ɜɩɥɢɜ ɧɚ ɤɨɠɧɨɦɭ ɤɪɨɰɿ ɪɨɡɪɚɯɨɜɭɽɬɶɫɹ ɡɚ ɪɚɯɭɧɨɤ ɜɢɪɿɲɟɧɧɹ ɨɩɬɢɦɿɡɚɰɿɣɧɨʀ ɡɚɞɚɱɿ ɧɚ ɨɫɧɨɜɿ ɦɚɬɟɦɚɬɢɱɧɨʀ ɦɨɞɟɥɿ ɨɛ¶ɽɤɬɚ ɤɟɪɭɜɚɧɧɹ ɐɿ ɫɢɫɬɟɦɢ ɬɚɤɨɠ ɡɚɫɬɨɫɨɜɭɸɬɶ ɞɥɹ ɤɟɪɭɜɚɧɧɹ ɨɛ’ɽɤɬɚɦɢ ɬɟɯɧɨɥɨɝɿɱɧɨɝɨ ɬɢɩɭɡɨɤɪɟɦɚɪɟɤɨɦɟɧɞɭɸɬɶɞɨɡɚɫɬɨɫɭɜɚɧɧɹɩɪɢɤɟɪɭɜɚɧɧɿɛɚɝɚɬɨɤɚɧɚɥɶɧɢɦɢɨɛ¶ɽɤɬɚɦɢɤɚɧɚɥɢɹɤɢɯɩɨɜ¶ɹɡɚɧɿɦɿɠ ɫɨɛɨɸɱɟɪɟɡɞɿɸɩɟɪɟɯɪɟɫɧɢɯɡɜ¶ɹɡɤɿɜ ȼɹɤɨɫɬɿɚɥɶɬɟɪɧɚɬɢɜɢɜɤɚɡɚɧɢɦɫɢɫɬɟɦɚɦɡɚɩɪɨɩɨɧɨɜɚɧɚɫɢɫɬɟɦɚɪɟɝɭɥɸɜɚɧɧɹ ɡɩɪɨɝɧɨɡɭɜɚɧɧɹɦɜɢɦɭɲɟɧɨɝɨɪɭɯɭɜɡɚɦɤɧɭɬɢɣɤɨɧɬɭɪɹɤɨʀɜɜɟɞɟɧɨɚɥɝɨɪɢɬɦɩɪɨɝɧɨɡɭɜɚɧɧɹ \Wɜɪɟɚɥɶɧɨɦɭɱɚɫɿ ɧɚ ɨɫɧɨɜɿ ɤɭɛɿɱɧɨɝɨ ɫɩɥɚɣɧɭ ȼ ɡɚɩɪɨɩɨɧɨɜɚɧɿɣ ɋȺɊ ɜɢɤɨɧɚɧɨ ɩɟɪɟɯɿɞ ɜɿɞ ɡɚɝɚɥɶɧɨɩɪɢɣɧɹɬɨɝɨ ɩɪɢɧɰɢɩɭ ɪɟɝɭɥɸɜɚɧɧɹɩɨɩɨɬɨɱɧɨɦɭɡɧɚɱɟɧɧɸɜɢɯɿɞɧɨʀɜɟɥɢɱɢɧɢ\Wɞɨɩɪɢɧɰɢɩɭɪɟɝɭɥɸɜɚɧɧɹɡɚɩɪɨɝɧɨɡɨɦy(t + Wɩɪ)ɐɟ ɨɡɧɚɱɚɽ ɳɨ ɨɛɱɢɫɥɟɧɧɹ ɤɟɪɭɸɱɨɝɨ ɜɩɥɢɜɭ ɩɨɥɨɠɟɧɧɹ ɪɟɝɭɥɸɸɱɨɝɨ ɨɪɝɚɧɭ 8 W ɩɪɨɜɨɞɢɬɶɫɹ ɧɟ ɡɚ ɩɨɬɨɱɧɢɦ ɡɧɚɱɟɧɧɹɦɪɟɝɭɥɶɨɜɚɧɨʀɜɟɥɢɱɢɧɢ\Wɚɩɨɩɪɨɝɧɨɡɨɜɚɧɨɦɭɡɧɚɱɟɧɧɸy(t + Wɩɪ) ɭɦɚɣɛɭɬɧɶɨɦɭ Ⱥɧɚɥɿɡɥɿɬɟɪɚɬɭɪɧɢɯɞɚɧɢɯɿɩɨɫɬɚɧɨɜɤɚɩɪɨɛɥɟɦɢ Ⱥɧɚɥɿɡɥɿɬɟɪɚɬɭɪɧɢɯɞɚɧɢɯɿɩɨɫɬɚɧɨɜɤɚɩɪɨɛɥɟɦɢ Ⱥɧɚɥɿɡɫɭɱɚɫɧɢɯɩɿɞɯɨɞɿɜ>@ɩɿɞɬɜɟɪɞɠɭɽɚɤɬɭɚɥɶɧɿɫɬɶɜɢɤɨɪɢɫɬɚɧɧɹ ɩɪɨɝɧɨɡɭɜɚɧɧɹɜɚɥɝɨɪɢɬɦɚɯɤɟɪɭɜɚɧɧɹ ɫɭɱɚɫɧɢɯ ɋȺɊ ɉɪɨɝɧɨɡɭɜɚɧɧɹ ɜ ɋȺɊ ɦɨɠɟ ɡɞɿɣɫɧɸɜɚɬɢɫɹ ɡɚ ɪɿɡɧɢɦɢ ɦɟɬɨɞɚɦɢ Ⱥɧɚɥɿɡɥɿɬɟɪɚɬɭɪɧɢɯɞɚɧɢɯɿɩɨɫɬɚɧɨɜɤɚɩɪɨɛɥɟɦɢ Ⱥɧɚɥɿɡɫɭɱɚɫɧɢɯɩɿɞɯɨɞɿɜ>@ɩɿɞɬɜɟɪɞɠɭɽɚɤɬɭɚɥɶɧɿɫɬɶɜɢɤɨɪɢɫɬɚɧɧɹ ɩɪɨɝɧɨɡɭɜɚɧɧɹɜɚɥɝɨɪɢɬɦɚɯɤɟɪɭɜɚɧɧɹ ɫɭɱɚɫɧɢɯɋȺɊɉɪɨɝɧɨɡɭɜɚɧɧɹɜɋȺɊɦɨɠɟɡɞɿɣɫɧɸɜɚɬɢɫɹɡɚɪɿɡɧɢɦɢɦɟɬɨɞɚɦɢ Ⱥɧɚɥɿɡɫɭɱɚɫɧɢɯɩɿɞɯɨɞɿɜ>@ɩɿɞɬɜɟɪɞɠɭɽɚɤɬɭɚɥɶɧɿɫɬɶɜɢɤɨɪɢɫɬɚɧɧɹ ɩɪɨɝɧɨɡɭɜɚɧɧɹɜɚɥɝɨ ɱɚɫɧɢɯɋȺɊɉɪɨɝɧɨɡɭɜɚɧɧɹɜɋȺɊɦɨɠɟɡɞɿɣɫɧɸɜɚɬɢɫɹɡɚɪɿɡɧɢɦɢɦɟɬɨɞɚɦɢ Ʉɟɪɭɜɚɧɧɹ ɡ ɩɪɨɝɧɨɡɭɸɱɨɸ ɦɨɞɟɥɥɸ MPɋ ɜɢɤɨɪɢɫɬɨɜɭɽɬɶɫɹ ɭ ɩɟɪɟɪɨɛɧɿɣ ɩɪɨɦɢɫɥɨɜɨɫɬɿ ɧɚ ɯɿɦɿɱɧɢɯ ɬɚ ɧɚɮɬɨɩɟɪɟɪɨɛɧɢɯɡɚɜɨɞɚɯɩɨɱɢɧɚɸɱɢ ɡ 1980-ɯɪɨɤɿɜ. ȼɨɫɬɚɧɧɿɪɨɤɢ ɰɟɣɦɟɬɨɞɤɟɪɭɜɚɧɧɹɬɚɤɨɠɜɢɤɨɪɢɫɬɨɜɭɽɬɶɫɹɜ ɫɢɫɬɟɦɚɯ ɛɚɥɚɧɫɭɜɚɧɧɹ ɟɧɟɪɝɨɫɢɫɬɟɦ ɜ ɫɢɥɨɜɿɣ ɟɥɟɤɬɪɨɧɿɰɿ [2] ɭ ɫɢɫɬɟɦɚɯ ɩɟɪɟɬɜɨɪɟɧɧɹ ɟɧɟɪɝɿʀ ɜɿɬɪɭ [3] ɬɚ ɧɚ ɩɿɞɩɪɢɽɦɫɬɜɚɯ ɩɨ ɨɱɢɳɟɧɧɸ ɫɬɿɱɧɢɯ ɜɨɞ [3] ȼ ɚɥɝɨɪɢɬɦɚɯ MPɋ ɜɢɤɨɪɢɫɬɨɜɭɸɬɶɫɹ ɞɢɧɚɦɿɱɧɿ ɦɨɞɟɥɿ ɩɪɨɰɟɫɿɜ ɱɚɫɬɿɲɟɡɚɜɫɶɨɝɨɥɿɧɿɣɧɿɟɦɩɿɪɢɱɧɿɦɨɞɟɥɿɹɤɿɨɬɪɢɦɚɧɿɲɥɹɯɨɦɿɞɟɧɬɢɮɿɤɚɰɿʀ ɦɨɞɟɥɿɨɛ¶ɽɤɬɚɤɟɪɭɜɚɧɧɹɈɄȾɥɹ ɫɩɪɨɳɟɧɨʀɦɨɞɟɥɿɨɛ ɽɤɬɚɿɩɨɱɚɬɤɨɜɢɯɭɦɨɜɜɢɤɨɧɭɽɬɶɫɹɩɪɨɝɧɨɡɭɜɚɧɧɹɩɨɜɟɞɿɧɤɢɩɿɞɜɩɥɢɜɨɦɤɟɪɭɸɱɨɝɨɫɢɝɧɚɥɭɧɚ ɞɟɹɤɨɦɭɤɿɧɰɟɜɨɦɭɜɿɞɪɿɡɤɭ ɱɚɫɭɡɜɚɧɨɦɭɝɨɪɢɡɨɧɬɩɪɨɝɧɨɡɭ3UHGLFWLRQ+RUL]RQȼɢɤɨɧɭɽɬɶɫɹɨɩɬɢɦɿɡɚɰɿɹɤɟɪɭɸɱɨɝɨ ɫɢɝɧɚɥɭ ɡ ɭɪɚɯɭɜɚɧɧɹɦ ɜɫɶɨɝɨ ɤɨɦɩɥɟɤɫɭ ɨɛɦɟɠɟɧɶ ɧɚɤɥɚɞɟɧɢɯ ɧɚ ɤɟɪɭɸɱɿ ɿ ɪɟɝɭɥɶɨɜɚɧɿ ɡɦɿɧɧɿ ɡɧɚɯɨɞɢɬɶɫɹ ɨɩɬɢɦɚɥɶɧɟɤɟɪɭɜɚɧɧɹɇɚɱɚɫɨɜɨɦɭɜɿɞɪɿɡɤɭɜɢɡɧɚɱɟɧɨɦɭɨɞɧɢɦɤɪɨɤɨɦɨɛɱɢɫɥɟɧɧɹɹɤɢɣɫɬɚɧɨɜɢɬɶɮɿɤɫɨɜɚɧɭɦɚɥɭ ɱɚɫɬɢɧɭ ɝɨɪɢɡɨɧɬɭ ɩɪɨɝɧɨɡɭ ɡɜɚɧɨɦɭ ɝɨɪɢɡɨɧɬɨɦ ɤɟɪɭɜɚɧɧɹ &RQWURO +RUL]RQ ɪɟɚɥɿɡɭɽɬɶɫɹ ɡɧɚɣɞɟɧɟ ɨɩɬɢɦɚɥɶɧɟ ɤɟɪɭɜɚɧɧɹɉɨɡɚɤɿɧɱɟɧɧɸɿɧɬɟɪɜɚɥɭɡɞɿɣɫɧɸɽɬɶɫɹɜɢɦɿɪɮɚɤɬɢɱɧɨɝɨɫɬɚɧɭɹɤɿɩɪɢɣɦɚɸɬɶɫɹɡɚɧɨɜɿɩɨɱɚɬɤɨɜɿɭɦɨɜɢ Ƚɨɪɢɡɨɧɬ ɩɪɨɝɧɨɡɭ ɡɫɭɜɚɽɬɶɫɹ ɧɚ ɤɪɨɤ ɜɩɟɪɟɞ ɿ ɡɧɨɜ ɩɨɜɬɨɪɸɸɬɶɫɹ ɟɬɚɩɢ ɪɨɡɪɚɯɭɧɤɭ ɤɟɪɭɸɱɨʀ ɞɿʀ ȼɢɤɨɪɢɫɬɚɧɧɹ MPɋ ɞɥɹ ɤɟɪɭɜɚɧɧɹ ɩɪɨɦɢɫɥɨɜɢɦɢ ɨɛ¶ɽɤɬɚɦɢ ɩɿɞɬɪɢɦɭɽɬɶɫɹ ɡɨɤɪɟɦɚ ɮɿɪɦɨɸ Siemens ɭ ɛɿɛɥɿɨɬɟɰɿ ɩɿɞɩɪɨɝɪɚɦ SIMATIC PCS7 APC-Library (Advanced Process Library ɹɤɚ ɪɨɡɩɨɜɫɸɞɠɭɽɬɶɫɹ ɞɥɹ ɜɿɞɨɦɨɝɨ ɫɟɪɟɞɨɜɢɳɚ ɪɨɡɪɨɛɤɢ ɩɪɢɤɥɚɞɧɨɝɨ ɩɪɨɝɪɚɦɧɨɝɨ ɡɚɛɟɡɩɟɱɟɧɧɹ - PCS Ɍɚɤɨɠ ɰɿɽɸ ɛɿɛɥɿɨɬɟɤɨɸ ɩɿɞɬɪɢɦɭɽɬɶɫɹ ɪɟɚɥɿɡɚɰɿɹ ɫɢɫɬɟɦ 39 Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ ɪɟɝɭɥɸɜɚɧɧɹ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ ɋɢɫɬɟɦɢ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ ɱɚɫɬɨ ɡɚɫɬɨɫɨɜɭɸɬɶɫɹ ɞɥɹ ɤɟɪɭɜɚɧɧɹ ɧɟɫɬɚɰɿɨɧɚɪɧɢɦɢ ɨɛ¶ɽɤɬɚɦɢ ɯɚɪɱɨɜɨʀ ɩɪɨɦɢɫɥɨɜɨɫɬɿ ɡ ɡɚɩɿɡɧɟɧɧɹɦ ɜ ɤɚɧɚɥɚɯ ɤɟɪɭɜɚɧɧɹ >@ Ɋɨɡɝɥɹɧɭɬɿ ɚɥɝɨɪɢɬɦɢ ɪɨɡɲɢɪɸɸɬɶ ɡɚɩɚɫ ɫɬɿɣɤɨɫɬɿ ɫɢɫɬɟɦ ɿ ɡɚɛɟɡɩɟɱɭɸɬɶ ɡɧɢɠɟɧɧɹ ɩɨɦɢɥɨɤ ɜɿɞɬɜɨɪɟɧɧɹ ɛɿɥɶɲ ɟɮɟɤɬɢɜɧɨ ɧɿɠ ɩɨɦɢɥɨɤ ɫɬɚɛɿɥɿɡɚɰɿʀ Ɋɟɡɭɥɶɬɚɬɢ ɞɨɫɥɿɞɠɟɧɶ [4-7] ɩɨɤɚɡɭɸɬɶ ɳɨ ɞɥɹ ɤɟɪɭɜɚɧɧɹ ɨɛ¶ɽɤɬɚɦɢ ɬɟɯɧɨɥɨɝɿɱɧɨɝɨ ɬɢɩɭ ɦɨɠɧɚ ɬɚɤɨɠ ɡɚɩɪɨɩɨɧɭɜɚɬɢɫɢɫɬɟɦɭɪɟɝɭɥɸɜɚɧɧɹɡɩɪɨɝɧɨɡɭɜɚɧɧɹɦɜɢɦɭɲɟɧɨɝɨɪɭɯɭɜɡɚɦɤɧɭɬɢɣɤɨɧɬɭɪɹɤɨʀɜɜɟɞɟɧɨɚɥɝɨɪɢɬɦ ɩɪɨɝɧɨɡɭɜɚɧɧɹ \WɜɪɟɚɥɶɧɨɦɭɱɚɫɿɊɟɝɭɥɶɨɜɚɧɭɡɦɿɧɧɭ \Wɜɨɛ¶ɽɤɬɚɯɬɟɯɧɨɥɨɝɿɱɧɨɝɨɬɢɩɭɜɿɞɮɿɥɶɬɪɨɜɚɧɭɜɿɞ ɲɭɦɿɜ ɦɨɠɧɚ ɪɨɡɝɥɹɞɚɬɢ ɹɤ ɛɚɝɚɬɨɤɪɚɬɧɨ ɞɢɮɟɪɟɧɰɿɣɨɜɚɧɭ ɮɭɧɤɰɿɸ ɱɚɫɭ ɿ ɜɟɫɬɢ ɩɪɨɝɧɨɡɭɜɚɧɧɹ ʀʀ ɡɧɚɱɟɧɶ ɭ ɪɟɚɥɶɧɨɦɭɱɚɫɿɧɚɨɫɧɨɜɿɧɚɩɪɢɤɥɚɞɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭɌɚɤɢɦɱɢɧɨɦɦɢɨɬɪɢɦɭɽɦɨɋȺɊɜɹɤɿɣɜɢɤɨɧɚɧɨɩɟɪɟɯɿɞɜɿɞ ɡɚɝɚɥɶɧɨɩɪɢɣɧɹɬɨɝɨ ɩɪɢɧɰɢɩɭ ɪɟɝɭɥɸɜɚɧɧɹ ɩɨ ɩɨɬɨɱɧɨɦɭ ɡɧɚɱɟɧɧɸ ɜɢɯɿɞɧɨʀ ɜɟɥɢɱɢɧɢ \ W ɞɨ ɩɪɢɧɰɢɩɭ ɪɟɝɭɥɸɜɚɧɧɹɡɚɩɪɨɝɧɨɡɨɦy(t + Wɩɪ)Ɉɰɿɧɢɬɢɟɮɟɤɬɢɜɧɿɫɬɶɬɚɤɨʀɫɢɫɬɟɦɢɜɩɨɪɿɜɧɹɧɧɿɡɫɢɫɬɟɦɚɦɢɧɚɛɚɡɿɚɥɝɨɪɢɬɦɿɜ MPC ɬɚɭɩɟɪɟɞɠɭɜɚɱɚɋɦɿɬɚɽɡɚɜɞɚɧɧɹɦɞɚɧɨɝɨɞɨɫɥɿɞɠɟɧɧɹ Ɇɟɬɚɬɚɡɚɞɚɱɿɞɨɫɥɿɞɠɟɧɧɹ ɆɟɬɨɸɞɨɫɥɿɞɠɟɧɧɹɽɜɢɡɧɚɱɟɧɧɹɞɢɧɚɦɿɱɧɨʀɬɨɱɧɨɫɬɿɿɡɚɩɚɫɭɫɬɿɣɤɨɫɬɿɋȺɊɹɤɚɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɤɟɪɭɜɚɧɧɹɡɚ ɩɪɨɝɧɨɡɨɦɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭɞɥɹɨɛ¶ɽɤɬɿɜɬɟɯɧɨɥɨɝɿɱɧɨɝɨɬɢɩɭɜɩɨɪɿɜɧɹɧɧɿɡɋȺɊɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚ ɬɚɫɢɫɬɟɦɚɦɢɤɟɪɭɜɚɧɧɹɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸȾɥɹɞɨɫɹɝɧɟɧɧɹɜɢɡɧɚɱɟɧɨʀɦɟɬɢɧɟɨɛɯɿɞɧɨɩɪɨɜɟɫɬɢɫɬɪɭɤɬɭɪɧɢɣɿ ɨɩɬɢɦɚɥɶɧɢɣ ɩɚɪɚɦɟɬɪɢɱɧɢɣ ɫɢɧɬɟɡ ɋȺɊ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ ɿ ɡ ɩɪɨɝɧɨɡɭɸɱɨɸ ɦɨɞɟɥɥɸ ɬɚ ɋȺɊ ɹɤɚ ɪɟɚɥɿɡɭɽ ɩɪɢɧɰɢɩɤɟɪɭɜɚɧɧɹɡɚɩɪɨɝɧɨɡɨɦɉɿɫɥɹɰɶɨɝɨɩɪɨɜɟɫɬɢɩɨɪɿɜɧɹɥɶɧɢɣɚɧɚɥɿɡɰɢɯɫɢɫɬɟɦ Ɇɟɬɨɞɢɿɦɚɬɟɪɿɚɥɢɞɨɫɥɿɞɠɟɧɶ Ⱦɥɹɩɪɨɜɟɞɟɧɧɹɞɨɫɥɿɞɠɟɧɶɜɹɤɨɫɬɿɨɫɧɨɜɧɨɝɨɦɟɬɨɞɭɛɭɞɟɦɨɜɢɤɨɪɢɫɬɨɜɭɜɚɬɢɦɟɬɨɞɿɦɿɬɚɰɿɣɧɨɝɨɦɨɞɟɥɸɜɚɧɧɹɜ ɫɟɪɟɞɨɜɢɳɿ 6LPXOLQN ɫɢɫɬɟɦɢ 0DWODE Ⱦɥɹ ɰɶɨɝɨ ɧɚ ɩɟɪɲɨɦɭ ɟɬɚɩɿ ɪɨɡɪɨɛɢɦɨ ɫɬɪɭɤɬɭɪɧɿ ɫɯɟɦɢ ɋȺɊ ɜɢɡɧɚɱɢɦɨ ɪɟɝɭɥɹɬɨɪ ɜɿɪɬɭɚɥɶɧɢɣ ɬɟɫɬɨɜɢɣ ɈɄ ȼɌɈɄ ɤɪɢɬɟɪɿɣ ɨɰɿɧɤɢ ɹɤɨɫɬɿ ɪɨɛɨɬɢ ɋȺɊ ɦɨɞɟɥɶ ɭɩɟɪɟɞɠɭɜɚɱɚ ɋɦɿɬɚ ɿ ɩɪɨɜɟɞɟɦɨ ɞɥɹ ɦɨɞɟɥɿ ɤɭɛɿɱɧɨɝɨ ɫɩɥɚɣɧɚ ɪɨɡɪɚɯɭɧɨɤ ɫɩɿɜɜɿɞɧɨɲɟɧɶ ɳɨ ɞɨɡɜɨɥɹɸɬɶ ɨɰɿɧɸɜɚɬɢ ɣɨɝɨ ɩɚɪɚɦɟɬɪɢ ɜ ɪɟɚɥɶɧɨɦɭɱɚɫɿɿɜɟɫɬɢɪɨɡɪɚɯɭɧɨɤɩɪɨɝɧɨɡɧɨɝɨɡɧɚɱɟɧɧɹɪɟɝɭɥɶɨɜɚɧɨʀɡɦɿɧɧɨʀɋɬɪɭɤɬɭɪɧɿɫɯɟɦɢɋȺɊɹɤɚɪɟɚɥɿɡɭɽ ɩɪɢɧɰɢɩɤɟɪɭɜɚɧɧɹɡɚɩɪɨɝɧɨɡɨɦɋȺɊɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚɬɚɋȺɊɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɧɚɜɟɞɟɧɿɧɚɪɢɫ Ɋɢɫ– ɋɬɪɭɤɬɭɪɧɿɫɯɟɦɢɋȺɊɡɦɨɞɭɥɟɦɩɪɨɝɧɨɡɭɜɚɧɧɹɜʀʀɤɨɧɬɭɪɿɡɜɨɪɨɬɧɨɝɨɡɜ¶ɹɡɤɭɚ ɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚɛɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɜ (p) Wp - ɩɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɉȱȾɪɟɝɭɥɹɬɨɪɚ (p) Wuy - ɩɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɈɄɡɚɤɚɧɚɥɨɦ ɤɟɪɭɜɚɧɧɹɆɉ– ɦɨɞɭɥɶɩɪɨɝɧɨɡɭɜɚɧɧɹ (p) Wk - ɩɟɪɟɞɚɬɨɱɧɚ ɮɭɧɤɰɿɹɭɩɟɪɟɞɠɭɜɚɱɚɋɦɿɬɚMPC – ɪɟɝɭɥɹɬɨɪɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸ y(t) U(t) f̦(t WP(p) ȴy(t) yʯʪ(t ʺʿ Wuy(p y(t+'W̪̬ ʶʿ y(t) U(t) f̦(t WP(p) ȴy(t) yʯʪ(t Wk(p) Wuy(p 'yk(t ʶʿ y(t) U(t) f̦(t MPC yʯʪ(t Wuy(p ʶʿ ̌) ̍) ɜ Ɋɢɫ– ɋɬɪɭɤɬɭɪɧɿɫɯɟɦɢɋȺɊɡɦɨɞɭɥɟɦɩɪɨɝɧɨɡɭɜɚɧɧɹɜʀʀɤɨɧɬɭɪɿɡɜɨɪɨɬɧɨɝɨɡɜ¶ɹɡɤɭɚ ɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚɛɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɜ (p) Wp - ɩɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɉȱȾɪɟɝɭɥɹɬɨɪɚ (p) Wuy - ɩɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɈɄɡɚɤɚɧɚɥɨɦ ɤɟɪɭɜɚɧɧɹɆɉ– ɦɨɞɭɥɶɩɪɨɝɧɨɡɭɜɚɧɧɹ (p) Wk - ɩɟɪɟɞɚɬɨɱɧɚ ɮɭɧɤɰɿɹɭɩɟɪɟɞɠɭɜɚɱɚɋɦɿɬɚMPC – ɪɟɝɭɥɹɬɨɪɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸ Ɋɢɫ– ɋɬɪɭɤɬɭɪɧɿɫɯɟɦɢɋȺɊɡɦɨɞɭɥɟɦɩɪɨɝɧɨɡɭɜɚɧɧɹɜʀʀɤɨɧɬɭɪɿɡɜɨɪɨɬɧɨɝɨɡɜ¶ɹɡɤɭɚ ɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚɛɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɜ (p) Wp - ɩɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɉȱȾɪɟɝɭɥɹɬɨɪɚ (p) Wuy - ɩɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɈɄɡɚɤɚɧɚɥɨɦ ɤɟɪɭɜɚɧɧɹɆɉ– ɦɨɞɭɥɶɩɪɨɝɧɨɡɭɜɚɧɧɹ (p) Wk - ɩɟɪɟɞɚɬɨɱɧɚ ɮɭɧɤɰɿɹɭɩɟɪɟɞɠɭɜɚɱɚɋɦɿɬɚMPC – ɪɟɝɭɥɹɬɨɪɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸ ȼ ɹɤɨɫɬɿ ɪɟɝɭɥɹɬɨɪɚ ɛɭɞɟɦɨ ɜɢɤɨɪɢɫɬɨɜɭɜɚɬɢ ɉȱȾ - ɪɟɝɭɥɹɬɨɪ ɡ ɪɟɚɥɶɧɢɦ ɞɢɮɟɪɟɧɰɿɚɬɨɪɨɦ Ⱦɥɹ ɩɪɨɜɟɞɟɧɧɹ ɨɩɬɢɦɚɥɶɧɨɝɨ ɩɚɪɚɦɟɬɪɢɱɧɨɝɨ ɫɢɧɬɟɡɭ ɋȺɊ ɿ ɨɰɿɧɤɢ ɹɤɨɫɬɿ ʀʀ ɪɨɛɨɬɢ ɜ ɩɟɪɟɯɿɞɧɢɯ ɪɟɠɢɦɚɯ ɫɤɨɪɢɫɬɚɽɦɨɫɹ ɿɧɬɟɝɪɚɥɶɧɢɦɤɜɚɞɪɚɬɢɱɧɢɦɩɨɤɚɡɧɢɤɨɦɹɤɨɫɬɿɚɨɰɿɧɤɭɹɤɨɫɬɿʀʀɪɨɛɨɬɢɜɫɬɚɥɢɯɪɟɠɢɦɚɯɛɭɞɟɦɨɩɪɨɜɨɞɢɬɢɡɚ ɩɨɤɚɡɧɢɤɨɦ ȼ ɹɤɨɫɬɿ ɪɟɝɭɥɹɬɨɪɚ ɛɭɞɟɦɨ ɜɢɤɨɪɢɫɬɨɜɭɜɚɬɢ ɉȱȾ - ɪɟɝɭɥɹɬɨɪ ɡ ɪɟɚɥɶɧɢɦ ɞɢɮɟɪɟɧɰɿɚɬɨɪɨɦ Ⱦɥɹ ɩɪɨɜɟɞɟɧɧɹ ɨɩɬɢɦɚɥɶɧɨɝɨ ɩɚɪɚɦɟɬɪɢɱɧɨɝɨ ɫɢɧɬɟɡɭ ɋȺɊ ɿ ɨɰɿɧɤɢ ɹɤɨɫɬɿ ʀʀ ɪɨɛɨɬɢ ɜ ɩɟɪɟɯɿɞɧɢɯ ɪɟɠɢɦɚɯ ɫɤɨɪɢɫɬɚɽɦɨɫɹ ɿɧɬɟɝɪɚɥɶɧɢɦɤɜɚɞɪɚɬɢɱɧɢɦɩɨɤɚɡɧɢɤɨɦɹɤɨɫɬɿɚɨɰɿɧɤɭɹɤɨɫɬɿʀʀɪɨɛɨɬɢɜɫɬɚɥɢɯɪɟɠɢɦɚɯɛɭɞɟɦɨɩɪɨɜɨɞɢɬɢɡɚ ɩɨɤɚɡɧɢɤɨɦ 40 Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ dt y(t)) (t) (y I 7ɦɨɞ t t 2 ɡɞ 1 1 1³ , (1) ³ 7ɦɨɞ t t 2 ɡɞ ɦɨɞ 2 1 1 dt y(t)) (t) (y T 1 I , (2) ɞɟɌɦɨɞ – ɱɚɫ ɦɨɞɟɥɸɜɚɧɧɹt1 – ɱɚɫ ɩɨɱɚɬɤɭ ɜɿɞɥɿɤɭ. ɞ ɭ ɭ Ɇɨɞɟɥɶɨɛ ɽɤɬɚɤɟɪɭɜɚɧɧɹɪɨɡɝɥɹɧɟɦɨ ɡɿɫɬɚɬɢɱɧɢɦɢɜɥɚɫɬɢɜɨɫɬɹɦɢɡɚɤɚɧɚɥɨɦɤɟɪɭɜɚɧɧɹɉɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹ ɜɿɪɬɭɚɥɶɧɨɝɨɬɟɫɬɨɜɨɝɨɈɄȼɌɈɄɦɚɽɧɚɫɬɭɩɧɢɣɜɢɝɥɹɞ ɞ ɭ ɭ Ɇɨɞɟɥɶɨɛ ɽɤɬɚɤɟɪɭɜɚɧɧɹɪɨɡɝɥɹɧɟɦɨ ɡɿɫɬɚɬɢɱɧɢɦɢɜɥɚɫɬɢɜɨɫɬɹɦɢɡɚɤɚɧɚɥɨɦɤɟɪɭɜɚɧɧɹɉɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹ ɜɿɪɬɭɚɥɶɧɨɝɨɬɟɫɬɨɜɨɝɨɈɄȼɌɈɄɦɚɽɧɚɫɬɭɩɧɢɣɜɢɝɥɹɞ k 1) (T2p 1) (T1p ko (p) W 6 uy o (3) (3) ɞɟɌ- ɩɨɫɬɿɣɧɚɱɚɫɭɳɨɯɚɪɚɤɬɟɪɢɡɭɽɽɦɧɿɫɧɟɡɚɩɿɡɧɟɧɧɹ7- ɩɨɫɬɿɣɧɚɱɚɫɭNR- ɤɨɟɮɿɰɿɽɧɬɩɟɪɟɞɚɱɿɈɄɉɪɢ ɰɶɨɦɭɩɨɫɬɿɣɧɚɱɚɫɭɌɌȾɥɹɩɪɨɜɟɞɟɧɧɹɞɨɫɥɿɞɠɟɧɶɛɭɥɢɩɪɢɣɧɹɬɿɧɚɫɬɭɩɧɿɛɚɡɨɜɿɡɧɚɱɟɧɧɹɩɚɪɚɦɟɬɪɿɜȼɌɈɍ T1 = 0.5, T2 = 7, ko = 1.2. ɞɟɌ- ɩɨɫɬɿɣɧɚɱɚɫɭɳɨɯɚɪɚɤɬɟɪɢɡɭɽɽɦɧɿɫɧɟɡɚɩɿɡɧɟɧɧɹ7- ɩɨɫɬɿɣɧɚɱɚɫɭNR- ɤɨɟɮɿɰɿɽɧɬɩɟɪɟɞɚɱɿɈɄɉɪɢ ɰɶɨɦɭɩɨɫɬɿɣɧɚɱɚɫɭɌɌȾɥɹɩɪɨɜɟɞɟɧɧɹɞɨɫɥɿɞɠɟɧɶɛɭɥɢɩɪɢɣɧɹɬɿɧɚɫɬɭɩɧɿɛɚɡɨɜɿɡɧɚɱɟɧɧɹɩɚɪɚɦɟɬɪɿɜȼɌɈɍ T1 = 0.5, T2 = 7, ko = 1.2. T1 0.5, T2 7, ko 1.2. ɇɚɪɢɫɩɪɟɞɫɬɚɜɥɟɧɚɧɨɪɦɨɜɚɧɚɩɟɪɟɯɿɞɧɚɯɚɪɚɤɬɟɪɢɫɬɢɤɚȼɌɈɍɩɨɤɚɧɚɥɭɤɟɪɭɜɚɧɧɹɡɿɥɸɫɬɪɚɰɿɽɸɩɪɨɰɟɞɭɪɢ ɩɚɪɚɦɟɬɪɢɱɧɨʀɿɞɟɧɬɢɮɿɤɚɰɿʀɦɨɞɟɥɿɩɟɪɲɨɝɨɩɨɪɹɞɤɭɡɚɦɟɬɨɞɢɤɨɸɆɿɧɿɧɨʀ t, c h(t Ɋɢɫ– ȱɥɸɫɬɪɚɰɿɹɩɪɨɰɟɞɭɪɢɩɚɪɚɦɟɬɪɢɱɧɨʀɿɞɟɧɬɢɮɿɤɚɰɿʀɦɨɞɟɥɿɈɄɦɟɬɨɞɢɤɚɆɿɧɿɧɨɣɡɚɤɚɧɚɥɨɦ ɤɟɪɭɜɚɧɧɹ t, c h(t Ɋɢɫ– ȱɥɸɫɬɪɚɰɿɹɩɪɨɰɟɞɭɪɢɩɚɪɚɦɟɬɪɢɱɧɨʀɿɞɟɧɬɢɮɿɤɚɰɿʀɦɨɞɟɥɿɈɄɦɟɬɨɞɢɤɚɆɿɧɿɧɨɣɡɚɤɚɧɚɥɨɦ ɤɟɪɭɜɚɧɧɹ ɉɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɦɨɞɟɥɿɩɟɪɲɨɝɨɩɨɪɹɞɤɭɩɨɤɚɧɚɥɭɤɟɪɭɜɚɧɧɹ 2 1 1 3 p ɉɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɦɨɞɟɥɿɩɟɪɲɨɝɨɩɨɪɹɞɤɭɩɨɤɚɧɚɥɭɤɟɪɭɜɚɧɧɹ 1 3 p ɉɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɦɨɞɟɥɿɩɟɪɲɨɝɨɩɨɪɹɞɤɭɩɨɤɚɧɚɥɭɤɟɪɭɜɚɧɧɹ 1 3 1 7 2.1 (p) W 1.3 uy p e p (4) (4) ɉɟɪɟɞɚɬɨɱɧɚɮɭɧɤɰɿɹɭɩɟɪɟɞɠɭɜɚɱɚɋɦɿɬɚɩɪɢɣɦɟɧɚɫɬɭɩɧɢɣɜɢɝɥɹɞ 1 3 1 7 ) 1( 2.1 1 ) 1( (p) W 1.3 k p e p T e k p o p o o W (5) (5) Ɇɨɞɭɥɶɩɪɨɝɧɨɡɭɜɚɧɧɹɩɨɤɭɛɿɱɧɨɦɭɫɩɥɚɣɧɭɛɭɞɟɦɨɪɟɚɥɿɡɨɜɭɜɚɬɢɧɚɨɫɧɨɜɿɧɚɫɬɭɩɧɨɝɨɜɢɪɚɡɭ a(t) ) ǻ Ĳ b(t)(t ) ǻ Ĳ c(t)(t ) ǻ Ĳ d(t)(t (t) ) ǻ Ĳ y(t ɩɪ 2 ɩɪ 3 ɩɪ ɩɪ y , (6) Ɇɨɞɭɥɶɩɪɨɝɧɨɡɭɜɚɧɧɹɩɨɤɭɛɿɱɧɨɦɭɫɩɥɚɣɧɭɛɭɞɟɦɨɪɟɚɥɿɡɨɜɭɜɚɬɢɧɚɨɫɧɨɜɿɧɚɫɬɭɩɧɨɝɨɜɢɪɚɡɭ a(t) ) ǻ Ĳ b(t)(t ) ǻ Ĳ c(t)(t ) ǻ Ĳ d(t)(t (t) ) ǻ Ĳ y(t ɩɪ 2 ɩɪ 3 ɩɪ ɩɪ y , (6) Ɇɨɞɭɥɶɩɪɨɝɧɨɡɭɜɚɧɧɹɩɨɤɭɛɿɱɧɨɦɭɫɩɥɚɣɧɭɛɭɞɟɦɨɪɟɚɥɿɡɨɜɭɜɚɬɢɧɚɨɫɧɨɜɿɧɚɫɬɭɩɧɨɝɨɜɢɪɚɡɭ a(t) ) ǻ Ĳ b(t)(t ) ǻ Ĳ c(t)(t ) ǻ Ĳ d(t)(t (t) ) ǻ Ĳ y(t ɩɪ 2 ɩɪ 3 ɩɪ ɩɪ y , (6) 'Wɩɪ [0, Wɩɪ]. Ɍɨɞɿ ɹɤɳɨ ɜɞɚɫɬɶɫɹ ɡɧɚɣɬɢ ɡɧɚɱɟɧɧɹ ɬɪɶɨɯ ɩɨɯɿɞɧɢɯ ɰɢɯ ɨɰɿɧɨɤ ɜ ɦɨɦɟɧɬ ɱɚɫɭ W W0 ɬɨ ɨɰɿɧɤɢ ɡɧɚɱɟɧɶ ɤɨɟɮɿɰɿɽɧɬɿɜɜɜɢɡɧɚɱɚɸɬɶɫɹɡɩɪɨɫɬɢɯɿɡɪɭɱɧɢɯɞɥɹɪɨɡɪɚɯɭɧɤɭɜɪɟɚɥɶɧɨɦɭɱɚɫɿɫɩɿɜɜɿɞɧɨɲɟɧɶ /2 )t (t y )t (t y - ) (t y ) b(t 2 0 0 0 0 0 0 , /6 )t (t y /2 )t (t y )t (t y ) y(t ) (t y ) a(t 3 0 0 2 0 0 0 0 0 0 0 , (7) (7) ɚɩɪɨɝɧɨɡɨɜɚɧɿɧɚWɩɪ ɜɩɟɪɟɞɡɧɚɱɟɧɧɹɧɟɨɛɯɿɞɧɢɯɨɰɿɧɨɤɪɨɡɪɚɯɨɜɭɸɬɶɫɹɡɜɢɪɚɡɿɜ ɚɩɪɨɝɧɨɡɨɜɚɧɿɧɚWɩɪ ɜɩɟɪɟɞɡɧɚɱɟɧɧɹɧɟɨɛɯɿɞɧɢɯɨɰɿɧɨɤɪɨɡɪɚɯɨɜɭɸɬɶɫɹɡɜɢɪɚɡɿɜ 41 Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ ) (t y ) a(t ) Ĳ )(t b(t ) Ĳ )(t c(t ) Ĳ )(t d(t ) Ĳ y(t 0 0 ɩɪ 0 0 2 ɩɪ 0 0 3 ɩɪ 0 0 ɩɪ 0 , (8) ) (t y ) a(t ) Ĳ )(t b(t ) Ĳ )(t c(t ) Ĳ )(t d(t ) Ĳ y(t 0 0 ɩɪ 0 0 2 ɩɪ 0 0 3 ɩɪ 0 0 ɩɪ 0 , (8) ) (t y ) a(t ) Ĳ )(t b(t ) Ĳ )(t c(t ) Ĳ )(t d(t ) Ĳ y(t 0 0 ɩɪ 0 0 2 ɩɪ 0 0 3 ɩɪ 0 0 ɩɪ 0 , (8) ɇɟɨɛɯɿɞɧɨɸ ɭɦɨɜɨɸ ɪɨɛɨɬɢ ɚɥɝɨɪɢɬɦɭ ɽ ɳɨɛ y(t) ɛɭɥɚ ɬɪɢɪɚɡɨɜɨ ɞɢɮɟɪɟɧɰɿɣɨɜɚɧɨɸ ɐɹ ɭɦɨɜɚ ɰɿɥɤɨɦ ɡɞɿɣɫɧɟɧɧɚɬɚɤɹɤɡɦɿɧɧɿɧɚɜɢɯɨɞɿɈɄɜɿɞɮɿɥɶɬɪɨɜɚɧɿɜɿɞɲɭɦɿɜɹɤɩɪɚɜɢɥɨɽɛɚɝɚɬɨɪɚɡɿɜɞɢɮɟɪɟɧɰɿɣɨɜɚɧɢɦɢ ɇɟɨɛɯɿɞɧɨɸ ɭɦɨɜɨɸ ɪɨɛɨɬɢ ɚɥɝɨɪɢɬɦɭ ɽ ɳɨɛ y(t) ɛɭɥɚ ɬɪɢɪɚɡɨɜɨ ɞɢɮɟɪɟɧɰɿɣɨɜɚɧɨɸ ɐɹ ɭɦɨɜɚ ɰɿɥɤɨɦ ɡɞɿɣɫɧɟɧɧɚɬɚɤɹɤɡɦɿɧɧɿɧɚɜɢɯɨɞɿɈɄɜɿɞɮɿɥɶɬɪɨɜɚɧɿɜɿɞɲɭɦɿɜɹɤɩɪɚɜɢɥɨɽɛɚɝɚɬɨɪɚɡɿɜɞɢɮɟɪɟɧɰɿɣɨɜɚɧɢɦɢ ɍɪɨɡɝɥɹɧɭɬɨɦɭɫɩɥɚɣɧɿɪɨɡɪɚɯɭɧɨɤɤɨɟɮɿɰɿɽɧɬɿɜɜɟɞɟɬɶɫɹɜɩɪɢɩɭɳɟɧɧɿɳɨɜɦɨɦɟɧɬW0 ɜɿɞɨɦɿɡɧɚɱɟɧɧɹɫɚɦɨʀ ɡɦɿɧɧɨʀɿɜɫɿɯɧɟɨɛɯɿɞɧɢɯɩɨɯɿɞɧɢɯɍɪɚɡɿɰɢɮɪɨɜɨʀɪɟɚɥɿɡɚɰɿʀɚɥɝɨɪɢɬɦɭɤɟɪɭɜɚɧɧɹɚɫɚɦɟɜɨɧɚɩɟɪɟɞɛɚɱɚɽɬɶɫɹɹɤ ɨɫɧɨɜɧɚɩɨɯɿɞɧɿɨɛɱɢɫɥɸɸɬɶɫɹɡɚɡɧɚɱɟɧɧɹɦɢɪɟɲɿɬɱɚɫɬɨʀɮɭɧɤɰɿʀ ) (nT f ɤɜ k , f ,0 n ɡɤɪɨɤɨɦɤɜɚɧɬɭɜɚɧɧɹɌɤɜ. ɫɥɸɸɬɶɫɹɡɚɡɧɚɱɟɧɧɹɦɢɪɟɲɿɬɱɚɫɬɨʀɮɭɧɤɰɿʀ ) (nT f ɤɜ k , f ,0 n ɡɤɪɨɤɨɦɤɜɚɧɬɭɜɚɧɧɹɌɤɜ. Ⱦɥɹ ɩɪɨɜɟɞɟɧɧɹ ɩɨɪɿɜɧɹɥɶɧɨɝɨ ɚɧɚɥɿɡɭ ɪɨɛɨɬɢ ɋȺɊɜɜɟɞɟɦɨ ɬɚɤɿɩɨɡɧɚɱɟɧɧɹ ɋȺɊɉBɄ - ɋȺɊɪɟɚɥɿɡɭɽ ɩɪɢɧɰɢɩ ɭɩɪɚɜɥɿɧɧɹɡɚɩɪɨɝɧɨɡɨɦɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨ ɫɩɥɚɣɧɭɋȺɊBɍɋ- ɋȺɊɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɭɩɪɚɜɥɿɧɧɹɡɚɫɬɚɧɨɦɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚɋȺɊBɆɊɋ- ɋȺɊɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɭɩɪɚɜɥɿɧɧɹɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸ ɭ ɪ ɭ B ɪ ɭ ɪ ɭ ɪ ɪ ɭ ɇɚ ɩɟɪɲɨɦɭ ɟɬɚɩɿ ɞɨɫɥɿɞɠɟɧɶ ɩɪɨɜɟɞɟɦɨ ɨɩɬɢɦɚɥɶɧɢɣ ɩɚɪɚɦɟɬɪɢɱɧɢɣ ɫɢɧɬɟɡ Ɉɉɋ ɋȺɊ ɡɚ ɤɪɢɬɟɪɿɽɦ ɩɪɢ ɮɿɤɫɨɜɚɧɢɯ ɩɚɪɚɦɟɬɪɚɯ ɦɨɞɟɥɿ Ɉɍ ɛɟɡ ɦɨɞɭɥɿɜ ɩɪɨɝɧɨɡɭɜɚɧɧɹ Ⱥ ɩɨɬɿɦ ɜ ɡɚɦɤɧɭɬɢɣ ɤɨɧɬɭɪ ɨɬɪɢɦɚɧɨʀ ɋȺɊ ɡ ɨɩɬɢɦɚɥɶɧɢɦɢ ɩɚɪɚɦɟɬɪɚɦɢ ɪɟɝɭɥɹɬɨɪɚ ɜɜɟɞɟɦɨ ɦɨɞɭɥɿ ɩɪɨɝɧɨɡɭɜɚɧɧɹ ɡ ɥɿɧɿɣɧɢɦ ɩɪɨɝɧɨɡɨɦ ɿ ɩɪɨɝɧɨɡɨɦ ɩɨ ɤɭɛɿɱɧɨɦɭɫɩɥɚɣɧɭɬɚɭɩɟɪɟɞɠɭɜɚɜɋɦɿɬɚɇɚɥɚɲɬɭɜɚɧɧɹɋȺɊBɆɊɋɩɪɨɜɟɞɟɦɨɿɧɬɟɝɪɭɸɱɢɦɨɞɟɥɶɈɄɜɚɥɝɨɪɢɬɦ ɆɊɋ ɪɟɚɥɿɡɨɜɚɧɢɣ ɜ Simulink. Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ ɋɯɟɦɢ ɦɨɞɟɥɸɜɚɧɧɹ ɋȺɊ ɡ ɉȱȾ ɪɟɝɭɥɹɬɨɪɨɦ ɿ ɪɟɡɭɥɶɬɚɬɢ ɨɩɬɢɦɚɥɶɧɨɝɨ ɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɩɪɟɞɫɬɚɜɥɟɧɿɧɚɪɢɫ Ɋɢɫ 3 – ɋɯɟɦɚɦɨɞɟɥɸɜɚɧɧɹɋȺɊɚɡɪɨɡɤɪɢɬɨɸɩɿɞɫɢɫɬɟɦɨɸɉȱȾɪɟɝɭɥɹɬɨɪɚɛɪɟɡɭɥɶɬɚɬɢʀʀ ɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɜɩɪɢɮɿɤɫɨɜɚɧɢɯɡɧɚɱɟɧɧɹɯɩɚɪɚɦɟɬɪɨɦɦɨɞɟɥɿȼɌɈɄ ̌) ̍) ̏) ̏) ̌) ̍) Ɋɢɫ 3 – ɋɯɟɦɚɦɨɞɟɥɸɜɚɧɧɹɋȺɊɚɡɪɨɡɤɪɢɬɨɸɩɿɞɫɢɫɬɟɦɨɸɉȱȾɪɟɝɭɥɹɬɨɪɚɛɪɟɡɭɥɶɬɚɬɢʀʀ ɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɜɩɪɢɮɿɤɫɨɜɚɧɢɯɡɧɚɱɟɧɧɹɯɩɚɪɚɦɟɬɪɨɦɦɨɞɟɥɿȼɌɈɄ Ⱦɥɹɞɨɫɥɿɞɠɟɧɧɹɪɨɛɨɬɢɫɢɫɬɟɦɭɫɬɚɥɢɯɪɟɠɢɦɚɯɪɨɛɨɬɢɛɭɥɚɫɮɨɪɦɨɜɚɧɚɦɨɞɟɥɶɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶɭ ɜɢɝɥɹɞɿɜɢɩɚɞɤɨɜɨɝɨɩɪɨɰɟɫɭɊɟɚɥɿɡɚɰɿɹɜɢɩɚɞɤɨɜɨɝɨɩɪɨɰɟɫɭɿɣɨɝɨ ɿɦɨɜɿɪɧɿɫɧɿɫɬɚɬɢɱɧɿɬɚɞɢɧɚɦɿɱɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢ ɧɚɜɟɞɟɧɨɧɚɪɢɫ Ɋɢɫ– Ɋɟɚɥɿɡɚɰɿɹɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶɬɚʀɯɿɦɨɜɿɪɧɿɫɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢ Ɋɢɫ– Ɋɟɚɥɿɡɚɰɿɹɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶɬɚʀɯɿɦɨɜɿɪɧɿɫɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢ 42 Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www atbp onaft edu ua/ Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ Ɋɢɫ 5 – ɋɯɟɦɚɦɨɞɟɥɸɜɚɧɧɹɞɥɹɩɪɨɜɟɞɟɧɧɹɩɨɪɿɜɧɹɥɶɧɨɝɨɚɧɚɥɿɡɭɪɨɛɨɬɢɞɨɫɥɿɞɠɭɜɚɧɢɯɋȺɊ Ɋɢɫ 5 – ɋɯɟɦɚɦɨɞɟɥɸɜɚɧɧɹɞɥɹɩɪɨɜɟɞɟɧɧɹɩɨɪɿɜɧɹɥɶɧɨɝɨɚɧɚɥɿɡɭɪɨɛɨɬɢɞɨɫɥɿɞɠɭɜɚɧɢɯɋȺ Ɋɟɡɭɥɶɬɚɬɢɞɨɫɥɿɞɠɟɧɶɬɚʀɯɨɛɝɨɜɨɪɟɧɧɹ ɇɚ ɪɢɫ ɩɪɟɞɫɬɚɜɥɟɧɿ ɩɟɪɟɯɿɞɧɿ ɯɚɪɚɤɬɟɪɢɫɬɢɤɢ ɋȺɊ ɚ ɧɚ ɪɢɫ ɪɟɡɭɥɶɬɚɬɢ ɦɨɞɟɥɸɜɚɧɧɹ ɭ ɫɬɚɥɢɯ ɪɟɠɢɦɚɯ ɪɨɛɨɬɢɩɪɢɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶɜɢɩɚɞɤɨɜɨɝɨɯɚɪɚɤɬɟɪɭɁɚɪɟɡɭɥɶɬɚɬɚɦɢɦɨɞɟɥɸɜɚɧɧɹɛɚɱɢɦɨɳɨɋȺɊɡ ɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɬɚɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚɦɚɸɬɶɫɯɨɠɿɞɢɧɚɦɿɱɧɿɜɥɚɫɬɢɜɨɫɬɿɐɟɨɛɭɦɨɜɥɟɧɨɜɢɤɨɪɢɫɬɚɧɧɹɦɜ ɨɛɨɯ ɚɥɝɨɪɢɬɦɚɯ ɤɟɪɭɜɚɧɧɹ ɨɞɧɚɤɨɜɨʀ ɦɨɞɟɥɿ ɈɄ ɞɥɹ ɩɪɨɝɧɨɡɭɜɚɧɧɹ ɤɟɪɭɸɱɨʀ ɞɿʀ ɋɢɫɬɟɦɚ ɡ ɚɥɝɨɪɢɬɦɨɦ ɩɪɨɝɧɨɡɭɜɚɧɧɹɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭɡɚɛɟɡɩɟɱɭɽɛɿɥɶɲɭɞɢɧɚɦɿɱɧɭɬɨɱɧɿɫɬɶɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ Ɇɚɤɫɢɦɚɥɶɧɟ ɞɢɧɚɦɿɱɧɟ ɜɿɞɯɢɥɟɧɧɹ ɭ ɩɟɪɟɯɿɞɧɢɯ ɪɟɠɢɦɚɯ ɪɨɛɨɬɢ ɬɚ ɫɟɪɟɞɧɶɨɤɜɚɞɪɚɬɢɱɧɟ ɜɿɞɯɢɥɟɧɧɹ ɭ ɫɬɚɥɢɯɪɟɠɢɦɚɯɪɨɛɨɬɢɩɪɢɛɥɢɡɧɨɭ ɞɜɿɱɿɦɟɧɲɟɧɿɠɭɫɢɫɬɟɦɚɯɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɬɚɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚ Ⱥɥɟɡɚɩɚɫɢɫɬɿɣɤɨɫɬɿɡɚɤɨɟɮɿɰɿɽɧɬɨɦɩɟɪɟɞɚɱɿɬɚɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹɭɰɢɯɫɢɫɬɟɦɧɚɛɚɝɚɬɨɛɿɥɶɲɿ y(t 2 1 3 t,c Ɋɢɫ– ɉɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɞɨɫɥɿɞɠɭɜɚɧɢɯɜɚɪɿɚɧɬɿɜɋȺɊ (1 – ɋȺɊ_ɍɋ, 2 - ɋȺɊɉBɄ– ɋȺɊBɆɊɋ). Ɋɢɫ 7 – ɊɟɡɭɥɶɬɚɬɢɦɨɞɟɥɸɜɚɧɧɹɋȺɊɭɫɬɚɥɨɦɭɪɟɠɢɦɿɪɨɛɨɬɢɩɪɢɞɿʀɜɢɩɚɞɤɨɜɢɯɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ (1 – ɋȺɊBɍɋ-ɋȺɊɉBɄ– ɋȺɊBɆɊɋ 1 2 3 2 1 3 y(t) t,c Ɋɢɫ– ɉɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɞɨɫɥɿɞɠɭɜɚɧɢɯɜɚɪɿɚɧɬɿɜɋȺɊ (1 – ɋȺɊ ɍɋ, 2 - ɋȺɊɉ Ʉ – ɋȺɊ ɆɊɋ). 1 2 3 2 1 3 y(t) t,c Ɋɢɫ– ɉɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɞɨɫɥɿɞɠɭɜɚɧɢɯɜɚɪɿɚɧɬɿɜɋȺɊ (1 – ɋȺɊ_ɍɋ, 2 - ɋȺɊɉBɄ– ɋȺɊBɆɊɋ). Ɋɢɫ 7 – ɊɟɡɭɥɶɬɚɬɢɦɨɞɟɥɸɜɚɧɧɹɋȺɊɭɫɬɚɥɨɦɭɪɟɠɢɦɿɪɨɛɨɬɢɩɪɢɞɿʀɜɢɩɚɞɤɨɜɢɯɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ (1 – ɋȺɊBɍɋ-ɋȺɊɉBɄ– ɋȺɊBɆɊɋ 43 Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ ɢɫ 8 – ɑɚɫɬɨɬɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɪɨɡɿɦɤɧɭɬɢɯɋȺɊɡɪɨɡɪɚɯɭɧɤɨɦɡɧɚɱɟɧɶ ʀɯɡɚɩɚɫɿɜɫɬɿɣɤɨɫɬɿɡɚɤɨɟɮɿɰɿɽɧɬɨɦ ɩɟɪɟɞɚɱɿ ɝɭ ɨ k ǻ ɿɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹ ɝɭ ɨ ǻ Ĳ ɜɦɨɞɟɥɿɈɄ-ɋȺɊBɍɋ– ɋȺɊɉBɄ Z-S Z-S Z1 Z1 'ʤʧ˄ 'Mʧ˄ 1 2 2 1 2 1 Z1 Z1 1 2 Ɋɢɫ 8 – ɑɚɫɬɨɬɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɪɨɡɿɦɤɧɭɬɢɯɋȺɊɡɪɨɡɪɚɯɭɧɤɨɦɡɧɚɱɟɧɶ ʀɯɡɚɩɚɫɿɜɫɬɿɣɤɨɫɬɿɡɚɤɨɟɮɿɰɿɽɧɬɨɦ ɩɟɪɟɞɚɱɿ ɝɭ ɨ k ǻ ɿɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹ ɝɭ ɨ ǻ Ĳ ɜɦɨɞɟɥɿɈɄ-ɋȺɊBɍɋ– ɋȺɊɉBɄ Ɋɢɫ 8 – ɑɚɫɬɨɬɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɪɨɡɿɦɤɧɭɬɢɯɋȺɊɡɪɨɡɪɚɯɭɧɤɨɦɡɧɚɱɟɧɶ ʀɯɡɚɩɚɫɿɜɫɬɿɣɤɨɫɬɿɡɚɤɨɟɮɿɰɿɽɧɬɨɦ ɩɟɪɟɞɚɱɿ ɝɭ ɨ k ǻ ɿɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹ ɝɭ ɨ ǻ Ĳ ɜɦɨɞɟɥɿɈɄ-ɋȺɊBɍɋ– ɋȺɊɉBɄ Ɉɬɪɢɦɚɧɿɪɟɡɭɥɶɬɚɬɢɽɩɨɩɟɪɟɞɧɿɦɢɿɧɟɞɨɡɜɨɥɹɸɬɶɡɪɨɛɢɬɢɜɢɫɧɨɜɨɤɩɪɨɩɟɪɟɜɚɝɭɨɞɧɨɝɨɡɜɚɪɿɚɧɬɿɜɫɢɫɬɟɦ Ⱦɥɹ ɛɿɥɶɲ ɬɨɱɧɨʀ ɨɰɿɧɤɢ ɟɮɟɤɬɢɜɧɨɫɬɿ ɪɨɛɨɬɢ ɪɨɡɝɥɹɧɭɬɢɯ ɜɚɪɿɚɧɬɿɜ ɫɢɫɬɟɦ ɩɪɨɜɟɞɟɦɨ ʀɯ Ɉɉɋ ɜ ɨɞɧɚɤɨɜɢɯ ɭɦɨɜɚɯ ɿ ɩɪɨɚɧɚɥɿɡɭɽɦɨ ʀɯ ɪɨɛɨɬɭ Ɉɬɪɢɦɚɧɿɪɟɡɭɥɶɬɚɬɢɽɩɨɩɟɪɟɞɧɿɦɢɿɧɟɞɨɡɜɨɥɹɸɬɶɡɪɨɛɢɬɢɜɢɫɧɨɜɨɤɩɪɨɩɟɪɟɜɚɝɭɨɞɧɨɝɨɡɜɚɪɿɚɧɬɿɜɫɢɫɬɟɦ Ⱦɥɹ ɛɿɥɶɲ ɬɨɱɧɨʀ ɨɰɿɧɤɢ ɟɮɟɤɬɢɜɧɨɫɬɿ ɪɨɛɨɬɢ ɪɨɡɝɥɹɧɭɬɢɯ ɜɚɪɿɚɧɬɿɜ ɫɢɫɬɟɦ ɩɪɨɜɟɞɟɦɨ ʀɯ Ɉɉɋ ɜ ɨɞɧɚɤɨɜɢɯ ɭɦɨɜɚɯ ɿ ɩɪɨɚɧɚɥɿɡɭɽɦɨʀɯɪɨɛɨɬɭ ɇɚɥɚɲɬɭɜɚɧɧɹ ɋȺɊ ɡ ɩɪɨɝɧɨɡɭɸɱɨɸ ɦɨɞɟɥɥɸ ɩɪɨɜɨɞɢɦɨ ɡ ɜɢɤɨɪɢɫɬɚɧɧɹɦ ɿɫɧɭɸɱɨɝɨ ɜ ɫɢɫɬɟɦɿ Matlab ɩɪɨɝɪɚɦɧɨɝɨ ɿɧɫɬɪɭɦɟɧɬɚɥɶɧɨɝɨɡɚɫɨɛɭɪɢɫɇɚɥɚɲɬɭɜɚɧɧɹɋȺɊɮɚɤɬɢɱɧɨɡɜɨɞɢɬɶɫɹɞɨɜɢɛɨɪɭɜɚɝɨɜɢɯɤɨɟɮɿɰɿɽɧɬɿɜ ɭɤɪɢɬɟɪɿʀɹɤɢɣɜɢɤɨɪɢɫɬɨɜɭɽɬɶɫɹɞɥɹɜɢɪɿɲɟɧɧɹɡɚɞɚɱɿɨɩɬɢɦɿɡɚɰɿʀɜɚɥɝɨɪɢɬɦɿɆɊɋ ɇɚɥɚɲɬɭɜɚɧɧɹ ɋȺɊ ɡ ɩɪɨɝɧɨɡɭɸɱɨɸ ɦɨɞɟɥɥɸ ɩɪɨɜɨɞɢɦɨ ɡ ɜɢɤɨɪɢɫɬɚɧɧɹɦ ɿɫɧɭɸɱɨɝɨ ɜ ɫɢɫɬɟɦɿ Matlab ɩɪɨɝɪɚɦɧɨɝɨ ɿɧɫɬɪɭɦɟɧɬɚɥɶɧɨɝɨɡɚɫɨɛɭɪɢɫɇɚɥɚɲɬɭɜɚɧɧɹɋȺɊɮɚɤɬɢɱɧɨɡɜɨɞɢɬɶɫɹɞɨɜɢɛɨɪɭɜɚɝɨɜɢɯɤɨɟɮɿɰɿɽɧɬɿɜ ɭɤɪɢɬɟɪɿʀɹɤɢɣɜɢɤɨɪɢɫɬɨɜɭɽɬɶɫɹɞɥɹɜɢɪɿɲɟɧɧɹɡɚɞɚɱɿɨɩɬɢɦɿɡɚɰɿʀɜɚɥɝɨɪɢɬɦɿɆɊɋ Ɋɢɫ 9 – ȽɨɥɨɜɧɟɜɿɤɧɨɩɪɨɝɪɚɦɢɧɚɥɚɲɬɭɜɚɧɧɹɋȺɊɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɆɊɋ. Ɋɢɫ 9 – ȽɨɥɨɜɧɟɜɿɤɧɨɩɪɨɝɪɚɦɢɧɚɥɚɲɬɭɜɚɧɧɹɋȺɊɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɆɊɋ. ɉɪɨɜɟɞɟɧɟɧɚɥɚɲɬɭɜɚɧɧɹɡɜɢɤɨɪɢɫɬɚɧɧɹɦɲɬɚɬɧɨʀɩɪɨɝɪɚɦɢɧɟɞɨɡɜɨɥɢɥɨɡɧɚɱɧɨɩɨɤɪɚɳɢɬɢɞɢɧɚɦɿɱɧɭɬɨɱɧɿɫɬɶ ɋȺɊɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶɿɡɦɟɧɲɢɬɢɩɨɯɢɛɤɢɫɬɚɛɿɥɿɡɚɰɿʀɁɦɿɧɚɧɚɥɚɲɬɭɜɚɧɶɦɨɠɟɩɪɢɜɨɞɢɬɢ ɞɨɡɧɚɱɧɨɝɨɡɛɿɥɶɲɟɧɧɹɱɚɫɭɩɟɪɟɯɿɞɧɨɝɨɩɪɨɰɟɫɭɪɢɫɚɱɢɧɚɛɥɢɠɭɜɚɬɢɫɢɫɬɟɦɭɞɨɦɟɠɿɫɬɿɣɤɨɫɬɿɪɢɫɛ Ɍɨɦɭɭɩɨɞɚɥɶɲɢɯɞɨɫɥɿɞɠɟɧɧɹɯɧɚɩɪɚɜɥɟɧɢɯɧɚɡɦɟɧɲɟɧɧɹɩɨɯɢɛɨɤɫɬɚɛɿɥɿɡɚɰɿʀɛɭɞɟɦɨɪɨɡɝɥɹɞɚɬɢɬɿɥɶɤɢɫɢɫɬɟɦɢ ɋȺɊBɍɋɬɚɋȺɊɉBɄ 44 Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ y(t) y(t) t,c t,c 1 2 3 3 1 2 ɚ ɛ Ɋɢɫ 10 – ɉɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɞɨɫɥɿɞɠɭɜɚɧɢɯɜɚɪɿɚɧɬɿɜɋȺɊɩɪɢɪɿɡɧɢɯɧɚɥɚɲɬɭɜɚɧɧɹɯɚɥɝɨɪɢɬɦɭɆɊɋ (1 – ɋȺɊBɍɋ- ɋȺɊɉBɄ– ɋȺɊBɆɊɋ Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ y(t) y(t) t,c t,c 1 2 3 3 1 2 ɚ ɛ Ɋɢɫ 10 – ɉɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɞɨɫɥɿɞɠɭɜɚɧɢɯɜɚɪɿɚɧɬɿɜɋȺɊɩɪɢɪɿɡɧɢɯɧɚɥɚɲɬɭɜɚɧɧɹɯɚɥɝɨɪɢɬɦɭɆɊɋ (1 – ɋȺɊBɍɋ- ɋȺɊɉBɄ– ɋȺɊBɆɊɋ y(t) t,c 1 2 3 ɚ y(t) t,c 3 1 2 ɛ ɯɚɪɚɤɬɟɪɢɫɬɢɤɢɞɨɫɥɿɞɠɭɜɚɧɢɯɜɚɪɿɚɧɬɿɜɋȺɊɩɪɢɪɿɡɧɢɯɧɚɥɚɲɬɭɜɚɧɧɹɯɚɥɝɨɪɢɬɦɭɆɊɋ (1 – ɋȺɊBɍɋ- ɋȺɊɉBɄ– ɋȺɊBɆɊɋ Ɉɛ ɽɤɬɢ ɬɟɯɧɨɥɨɝɿɱɧɨɝɨ ɬɢɩɭ ɦɚɸɬɶ ɡɧɚɱɧɢɣ ɪɿɜɧɟɦ ɧɟɜɢɡɧɚɱɟɧɨɫɬɿ ɿ ɧɟɫɬɚɰɿɨɧɚɪɧɨɫɬɿ ʀɯ ɩɚɪɚɦɟɬɪɿɜ Ɍɨɦɭ ɨɩɬɢɦɚɥɶɧɢɣ ɩɚɪɚɦɟɬɪɢɱɧɢɣ ɫɢɧɬɟɡ ɋȺɊ ɛɭɞɟɦɨ ɩɪɨɜɨɞɢɬɢ ɜ ɭɦɨɜɚɯ ɩɚɪɚɦɟɬɪɢɱɧɨʀ ɧɟɜɢɡɧɚɱɟɧɨɫɬɿ ɦɨɞɟɥɿ ɈɄ ȼ ɰɶɨɦɭ ɜɢɩɚɞɤɭ ɩɚɪɚɦɟɬɪɢ ɦɨɞɟɥɿ Ɉɍ ɛɭɞɭɬɶ ɡɚɞɚɧɨ ɧɟ ɮɿɤɫɨɜɚɧɢɦɢ ɡɧɚɱɟɧɧɹɦɢ ɚ ɞɿɚɩɚɡɨɧɚɦɢ ʀɯ ɡɦɿɧɢ ko[0.96…1.44], T1 [0.4…0.6]. Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ Ɉɉɋ ɜ ɭɦɨɜɚɯ ɩɚɪɚɦɟɬɪɢɱɧɨʀ ɧɟɜɢɡɧɚɱɟɧɨɫɬɿ ɦɨɞɟɥɿ ɈɄ ɞɨɡɜɨɥɢɬɶ ɡɚɛɟɡɩɟɱɢɬɢ ɫɬɿɣɤɿɫɬɶ ɋȺɊ ɞɥɹ ɡɚɞɚɧɨɝɨ ɞɿɚɩɚɡɨɧɭ ɡɦɿɧɢ ɩɚɪɚɦɟɬɪɿɜ Ɉɍ ɿ ɩɪɢɛɥɢɡɧɨ ɨɞɧɚɤɨɜɢɣ ɡɚɩɚɫ ɫɬɿɣɤɨɫɬɿ ɋȺɊ ɡɚ ɤɨɟɮɿɰɿɽɧɬɨɦɩɟɪɟɞɚɱɿɿɱɚɫɭɡɚɩɿɡɧɸɜɚɧɧɹɇɚɪɢɫɿɩɪɟɞɫɬɚɜɥɟɧɿɪɟɡɭɥɶɬɚɬɢɈɉɋɋȺɊɉBɄɿɋȺɊBɍɋɚɬɚɤɨɠ ɪɟɡɭɥɶɬɚɬɢɩɟɪɟɜɿɪɤɢɰɢɯɫɢɫɬɟɦɧɚɝɪɭɛɿɫɬɶɳɨɩɿɞɬɜɟɪɞɠɭɸɬɶʀɯɫɬɿɣɤɿɫɬɶɿɩɪɚɰɟɡɞɚɬɧɿɫɬɶɜɡɚɞɚɧɨɦɭɞɿɚɩɚɡɨɧɿ ɡɦɿɧɢɩɚɪɚɦɟɬɪɿɜɈɄ ɡɦɿɧɢɩɚɪɚɦɟɬɪɿɜɈɄ Ɋɢɫ 11 – ɊɟɡɭɥɶɬɚɬɢɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɿɩɟɪɟɜɿɪɤɢɝɪɭɛɨɫɬɿɋȺɊɡɉȱȾɪɟɝɭɥɹɬɨɪɨɦɹɤɚ ɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɭɩɪɚɜɥɿɧɧɹɡɚɩɪɨɝɧɨɡɨɦɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭ Ɋɢɫ 12 – ɊɟɡɭɥɶɬɚɬɢɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɿɩɟɪɟɜɿɪɤɢɝɪɭɛɨɫɬɿɋȺɊɡɉȱȾɪɟɝɭɥɹɬɨɪɨɦɹɤɚ ɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɭɩɪɚɜɥɿɧɧɹɡɚɜɿɞɯɢɥɟɧɧɹɦɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚ ɉɪɨɜɟɞɟɦɨ ɩɨɪɿɜɧɹɥɶɧɢɣ ɚɧɚɥɿɡ ɨɬɪɢɦɚɧɢɯ ɋȺɊ ɩɪɢ ɧɨɦɿɧɚɥɶɧɢɯ ɡɧɚɱɟɧɧɹɯ ɩɚɪɚɦɟɬɪɿɜ ɈɄ ɇɚ ɪɢɫ ɩɪɟɞɫɬɚɜɥɟɧɿɩɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɋȺɊɚɧɚɪɢɫɪɟɡɭɥɶɬɚɬɢɦɨɞɟɥɸɜɚɧɧɹɭɫɬɚɥɢɯɪɟɠɢɦɚɯɪɨɛɨɬɢɩɪɢɞɿʀ ɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɜɢɩɚɞɤɨɜɨɝɨ ɯɚɪɚɤɬɟɪɭ Ɂɚ ɪɟɡɭɥɶɬɚɬɚɦɢ ɦɨɞɟɥɸɜɚɧɧɹ ɛɚɱɢɦɨ ɳɨ ɋȺɊ ɡ ɚɥɝɨɪɢɬɦɨɦ ɩɪɨɝɧɨɡɭɜɚɧɧɹɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭɡɚɛɟɡɩɟɱɭɽɛɿɥɶɲɭɞɢɧɚɦɿɱɧɭɬɨɱɧɿɫɬɶɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɧɿɠ ɋȺɊ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ Ɇɚɤɫɢɦɚɥɶɧɟ ɞɢɧɚɦɿɱɧɟ ɜɿɞɯɢɥɟɧɧɹ ɭ ɩɟɪɟɯɿɞɧɢɯ ɪɟɠɢɦɚɯ ɪɨɛɨɬɢ ɩɪɢɛɥɢɡɧɨɧɚɦɟɧɲɟɚɿɧɬɟɝɪɚɥɶɧɨ-ɤɜɚɞɪɚɬɢɱɧɢɣɤɪɢɬɟɪɿɣI1 ɦɟɧɲɟɦɚɣɠɟɭɬɪɢɪɚɡɢɍɫɬɚɥɢɯɪɟɠɢɦɚɯɪɨɛɨɬɢ ɪ ɪ Ɋɢɫ 11 – ɊɟɡɭɥɶɬɚɬɢɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɿɩɟɪɟɜɿɪɤɢɝɪɭɛɨɫɬɿɋȺɊɡɉȱȾɪɟɝɭɥɹɬɨɪɨɦɹɤɚ ɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɭɩɪɚɜɥɿɧɧɹɡɚɩɪɨɝɧɨɡɨɦɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭ Ɋɢɫ 11 – ɊɟɡɭɥɶɬɚɬɢɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɿɩɟɪɟɜɿɪɤɢɝɪɭɛɨɫɬɿɋȺɊɡɉȱȾɪɟɝɭɥɹɬɨɪɨɦɹɤɚ ɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɭɩɪɚɜɥɿɧɧɹɡɚɩɪɨɝɧɨɡɨɦɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭ Ɋɢɫ 12 – ɊɟɡɭɥɶɬɚɬɢɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɿɩɟɪɟɜɿɪɤɢɝɪɭɛɨɫɬɿɋȺɊɡɉȱȾɪɟɝɭɥɹɬɨɪɨɦɹɤɚ ɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɭɩɪɚɜɥɿɧɧɹɡɚɜɿɞɯɢɥɟɧɧɹɦɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚ ɉɪɨɜɟɞɟɦɨ ɩɨɪɿɜɧɹɥɶɧɢɣ ɚɧɚɥɿɡ ɨɬɪɢɦɚɧɢɯ ɋȺɊ ɩɪɢ ɧɨɦɿɧɚɥɶɧɢɯ ɡɧɚɱɟɧɧɹɯ ɩɚɪɚɦɟɬɪɿɜ ɈɄ ɇɚ ɪɢɫ ɩɪɟɞɫɬɚɜɥɟɧɿɩɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɋȺɊɚɧɚɪɢɫɪɟɡɭɥɶɬɚɬɢɦɨɞɟɥɸɜɚɧɧɹɭɫɬɚɥɢɯɪɟɠɢɦɚɯɪɨɛɨɬɢɩɪɢɞɿʀ ɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɜɢɩɚɞɤɨɜɨɝɨ ɯɚɪɚɤɬɟɪɭ Ɂɚ ɪɟɡɭɥɶɬɚɬɚɦɢ ɦɨɞɟɥɸɜɚɧɧɹ ɛɚɱɢɦɨ ɳɨ ɋȺɊ ɡ ɚɥɝɨɪɢɬɦɨɦ ɩɪɨɝɧɨɡɭɜɚɧɧɹɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭɡɚɛɟɡɩɟɱɭɽɛɿɥɶɲɭɞɢɧɚɦɿɱɧɭɬɨɱɧɿɫɬɶɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɧɿɠ ɋȺɊ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ Ɇɚɤɫɢɦɚɥɶɧɟ ɞɢɧɚɦɿɱɧɟ ɜɿɞɯɢɥɟɧɧɹ ɭ ɩɟɪɟɯɿɞɧɢɯ ɪɟɠɢɦɚɯ ɪɨɛɨɬɢ ɩɪɢɛɥɢɡɧɨ ɧɚ ɦɟɧɲɟ ɚ ɿɧɬɟɝɪɚɥɶɧɨ-ɤɜɚɞɪɚɬɢɱɧɢɣ ɤɪɢɬɟɪɿɣ I1 ɦɟɧɲɟ ɦɚɣɠɟ ɭ ɬɪɢ ɪɚɡɢ ɍ ɫɬɚɥɢɯ ɪɟɠɢɦɚɯ ɪɨɛɨɬɢ Ɋɢɫ 12 – ɊɟɡɭɥɶɬɚɬɢɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨɫɢɧɬɟɡɭɿɩɟɪɟɜɿɪɤɢɝɪɭɛɨɫɬɿɋȺɊɡɉȱȾɪɟɝɭɥɹɬɨɪɨɦɹɤɚ ɪɟɚɥɿɡɭɽɩɪɢɧɰɢɩɭɩɪɚɜɥɿɧɧɹɡɚɜɿɞɯɢɥɟɧɧɹɦɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚ ɉɪɨɜɟɞɟɦɨ ɩɨɪɿɜɧɹɥɶɧɢɣ ɚɧɚɥɿɡ ɨɬɪɢɦɚɧɢɯ ɋȺɊ ɩɪɢ ɧɨɦɿɧɚɥɶɧɢɯ ɡɧɚɱɟɧɧɹɯ ɩɚɪɚɦɟɬɪɿɜ ɈɄ ɇɚ ɪɢɫ ɩɪɟɞɫɬɚɜɥɟɧɿɩɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɋȺɊɚɧɚɪɢɫɪɟɡɭɥɶɬɚɬɢɦɨɞɟɥɸɜɚɧɧɹɭɫɬɚɥɢɯɪɟɠɢɦɚɯɪɨɛɨɬɢɩɪɢɞɿʀ ɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɜɢɩɚɞɤɨɜɨɝɨ ɯɚɪɚɤɬɟɪɭ Ɂɚ ɪɟɡɭɥɶɬɚɬɚɦɢ ɦɨɞɟɥɸɜɚɧɧɹ ɛɚɱɢɦɨ ɳɨ ɋȺɊ ɡ ɚɥɝɨɪɢɬɦɨɦ ɩɪɨɝɧɨɡɭɜɚɧɧɹɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭɡɚɛɟɡɩɟɱɭɽɛɿɥɶɲɭɞɢɧɚɦɿɱɧɭɬɨɱɧɿɫɬɶɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɧɿɠ ɋȺɊ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ Ɇɚɤɫɢɦɚɥɶɧɟ ɞɢɧɚɦɿɱɧɟ ɜɿɞɯɢɥɟɧɧɹ ɭ ɩɟɪɟɯɿɞɧɢɯ ɪɟɠɢɦɚɯ ɪɨɛɨɬɢ ɩɪɢɛɥɢɡɧɨɧɚɦɟɧɲɟɚɿɧɬɟɝɪɚɥɶɧɨ-ɤɜɚɞɪɚɬɢɱɧɢɣɤɪɢɬɟɪɿɣI1 ɦɟɧɲɟɦɚɣɠɟɭɬɪɢɪɚɡɢɍɫɬɚɥɢɯɪɟɠɢɦɚɯɪɨɛɨɬɢ 45 ȼɢɫɧɨɜɤɢ ɋȺɊɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɡɬɢɩɨɜɢɦɚɥɝɨɪɢɬɦɨɦɧɚɥɚɲɬɭɜɚɧɧɹɡɚɛɟɡɩɟɱɭɽɜɢɫɨɤɢɣɡɚɩɚɫɫɬɿɣɤɨɫɬɿɫɢɫɬɟɦɢɿ ɡɦɟɧɲɟɧɧɹɞɢɧɚɦɿɱɧɢɯɩɨɯɢɛɨɤɡɚɤɚɧɚɥɨɦɡɚɜɞɚɧɧɹȾɢɧɚɦɿɱɧɚɬɨɱɧɿɫɬɶɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ ɡɚɥɢɲɚɽɬɶɫɹ ɧɢɡɤɨɸ ȼɜɟɞɟɧɧɹ ɜ ɋȺɊ ɡ ɉȱȾ ɪɟɝɭɥɹɬɨɪɨɦ ɭɩɟɪɟɞɠɭɜɚɱɚ ɋɦɿɬɭ ɛɟɡ ɞɨɞɚɬɤɨɜɢɯ ɧɚɥɚɲɬɭɜɚɧɶ ɬɚɤɨɠ ɡɧɚɱɧɨ ɪɨɡɲɢɪɸɽ ʀʀ ɡɚɩɚɫ ɫɬɿɣɤɨɫɬɿ Ⱦɢɧɚɦɿɱɧɚ ɬɨɱɧɿɫɬɶ ɫɢɫɬɟɦɢ ɩɪɢ ɮɿɤɫɨɜɚɧɢɯ ɡɧɚɱɟɧɧɹɯ ɩɚɪɚɦɟɬɪɿɜ ɈɄ ɡɚ ɤɚɧɚɥɚɦɢ ɡɚɜɞɚɧɧɹ ɬɚ ɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɜɿɞɩɨɜɿɞɚɽ ɞɢɧɚɦɿɱɧɿɣ ɬɨɱɧɨɫɬɿ ɋȺɊ ɡ ɩɪɨɝɧɨɡɭɸɱɨɸ ɦɨɞɟɥɥɸ ȼɜɟɞɟɧɧɹɜɋȺɊɚɥɝɨɪɢɬɦɭɩɪɨɝɧɨɡɭɜɚɧɧɹɦɡɚɤɭɛɿɱɧɢɦɫɩɥɚɣɧɨɦɩɿɞɜɢɳɭɽɞɢɧɚɦɿɱɧɭɬɨɱɧɿɫɬɶɫɢɫɬɟɦɢɡɚɤɚɧɚɥɨɦ ɞɿʀ ɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɿ ɪɨɡɲɢɪɸɽ ɡɚɩɚɫ ɫɬɿɣɤɨɫɬɿ ɫɢɫɬɟɦɢ ɚɥɟ ɡɧɚɱɧɨ ɦɟɧɲɟ ɧɿɠ ɫɢɫɬɟɦɢ ɡ ɩɪɨɝɧɨɡɭɸɱɨɸ ɦɨɞɟɥɥɸ ɬɚ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɭ Ɍɚɤɢɦ ɱɢɧɨɦ ɹɤɳɨ ɰɿɥɥɸ ɽ ɡɧɚɱɧɟ ɩɿɞɜɢɳɟɧɧɹ ɡɚɩɚɫɭ ɫɬɿɣɤɨɫɬɿ ɫɢɫɬɟɦɢ ɬɨ ɩɨɬɪɿɛɧɨɨɛɪɚɬɢɋȺɊɡɩɪɨɝɧɨɡɭɸɱɨɸɦɨɞɟɥɥɸɱɢɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɭəɤɳɨɰɿɥɥɸɽɜɢɫɨɤɚɞɢɧɚɦɿɱɧɚɬɨɱɧɿɫɬɶ ɫɬɚɛɿɥɿɡɚɰɿʀ ɩɪɢ ɨɞɧɨɱɚɫɧɨɦɭ ɡɚɛɟɡɩɟɱɟɧɧɿ ɝɪɭɛɨɫɬɿ ɋȺɊ ɜ ɩɟɜɧɨɦɭ ɞɿɚɩɚɡɨɧɿ ɡɦɿɧ ɩɚɪɚɦɟɬɪɿɜ ɈɄ ɬɨ ɞɨɰɿɥɶɧɨ ɜɢɤɨɪɢɫɬɨɜɭɜɚɬɢɋȺɊɡɚɥɝɨɪɢɬɦɨɦɩɪɨɝɧɨɡɭɜɚɧɧɹɦɡɚɤɭɛɿɱɧɢɦɫɩɥɚɣɧɨɦɊɟɡɭɥɶɬɚɬɢɨɩɬɢɦɚɥɶɧɨɝɨɩɚɪɚɦɟɬɪɢɱɧɨɝɨ ɫɢɧɬɟɡɭ ɩɪɨɜɟɞɟɧɨɝɨ ɜ ɭɦɨɜɚɯ ɩɚɪɚɦɟɬɪɢɱɧɨʀ ɧɟɜɢɡɧɚɱɟɧɨɫɬɿ ɈɄ ɩɿɞɬɜɟɪɞɢɥɢ ɩɟɪɟɜɚɝɭ ɋȺɊ ɡ ɩɪɨɝɧɨɡɭɜɚɧɧɹɦ ɡɚ ɤɭɛɿɱɧɢɦɫɩɥɚɣɧɨɦɩɟɪɟɞɋȺɊɡɭɩɟɪɟɞɠɭɜɚɱɟɦɋɦɿɬɚɜɡɦɟɧɲɟɧɧɿɩɨɯɢɛɨɤɫɬɚɛɿɥɿɡɚɰɿʀɋȺɊɡɩɪɨɝɧɨɡɭɜɚɧɧɹɦɡɚ ɤɭɛɿɱɧɢɦɫɩɥɚɣɧɨɦɡɚɛɟɡɩɟɱɭɽɡɧɢɠɟɧɧɹɦɚɤɫɢɦɚɥɶɧɨɝɨɞɢɧɚɦɿɱɧɨɝɨɜɿɞɯɢɥɟɧɧɹɩɪɢɛɥɢɡɧɨɧɚɬɚɿɧɬɟɝɪɚɥɶɧɨɝɨ ɩɨɤɚɡɧɢɤɚ ɹɤɨɫɬɿ ɭ ɪɚɡɢ ɜ ɩɨɪɿɜɧɹɧɧɿ ɡ ɋȺɊ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ ɉɪɢ ɰɶɨɦɭ ɡɚɩɚɫɢ ɫɬɿɣɤɨɫɬɿ ɰɢɯ ɫɢɫɬɟɦ ɡɚ ɤɨɟɮɿɰɿɽɧɬɨɦɩɟɪɟɞɚɱɿɬɚɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹɛɭɥɢɦɚɣɠɟɨɞɧɚɤɨɜɿ Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ ɋȺɊɡɚɥɝɨɪɢɬɦɨɦɩɪɨɝɧɨɡɭɜɚɧɧɹɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭ ɬɚɤɨɠɡɚɛɟɡɩɟɱɭɽɩɪɢɛɥɢɡɧɨɡɧɢɠɟɧɧɹɡɧɚɱɟɧɧɹ ɤɪɢɬɟɪɿɸ I2 ɜ ɩɨɪɿɜɧɹɧɧɿ ɡ ɫɢɫɬɟɦɨɸ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ ɇɚ ɪɢɫ ɧɚɜɟɞɟɧɿ ɱɚɫɬɨɬɧɿ ɯɚɪɚɤɬɟɪɢɫɬɢɤɢ ɪɨɡɿɦɤɧɟɧɢɯɋȺɊɿɪɨɡɪɚɯɭɧɤɨɜɿɡɧɚɱɟɧɧɹɡɚɩɚɫɿɜɫɬɿɣɤɨɫɬɿɫɢɫɬɟɦɡɚɤɨɟɮɿɰɿɽɧɬɨɦɩɟɪɟɞɚɱɿ ɝɭ ɨ k ǻ ɿɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹ ɝɭ ɨ ǻ Ĳ Ɂɧɚɱɟɧɧɹ ɡɚɩɚɫɿɜ ɫɬɿɣɤɨɫɬɿ ɞɨɫɥɿɞɠɭɜɚɧɢɯ ɜɚɪɿɚɧɬɿɜ ɋȺɊ ɛɥɢɡɶɤɿ ɨɞɢɧ ɞɨ ɨɞɧɨɝɨ Ɍɨɛɬɨ ɋȺɊ ɡ ɚɥɝɨɪɢɬɦɨɦ ɩɪɨɝɧɨɡɭɜɚɧɧɹ ɧɚ ɨɫɧɨɜɿ ɤɭɛɿɱɧɨɝɨ ɫɩɥɚɣɧɭ ɩɪɢ ɨɞɧɚɤɨɜɢɯ ɡɚɩɚɫɚɯ ɫɬɿɣɤɨɫɬɿ ɦɨɠɟ ɡɚɛɟɡɩɟɱɢɬɢ ɛɿɥɶɲ ɜɢɫɨɤɭ ɞɢɧɚɦɿɱɧɭɬɨɱɧɿɫɬɶɫɬɚɛɿɥɿɡɚɰɿʀɪɟɝɭɥɶɨɜɚɧɨʀɡɦɿɧɧɨʀɧɚɡɚɞɚɧɨɦɭɡɧɚɱɟɧɧɿ ɋȺɊɡɚɥɝɨɪɢɬɦɨɦɩɪɨɝɧɨɡɭɜɚɧɧɹɧɚɨɫɧɨɜɿɤɭɛɿɱɧɨɝɨɫɩɥɚɣɧɭ ɬɚɤɨɠɡɚɛɟɡɩɟɱɭɽɩɪɢɛɥɢɡɧɨɡɧɢɠɟɧɧɹɡɧɚɱɟɧɧɹ ɤɪɢɬɟɪɿɸ I2 ɜ ɩɨɪɿɜɧɹɧɧɿ ɡ ɫɢɫɬɟɦɨɸ ɡ ɭɩɟɪɟɞɠɭɜɚɱɟɦ ɋɦɿɬɚ ɇɚ ɪɢɫ ɧɚɜɟɞɟɧɿ ɱɚɫɬɨɬɧɿ ɯɚɪɚɤɬɟɪɢɫɬɢɤɢ ɪɨɡɿɦɤɧɟɧɢɯɋȺɊɿɪɨɡɪɚɯɭɧɤɨɜɿɡɧɚɱɟɧɧɹɡɚɩɚɫɿɜɫɬɿɣɤɨɫɬɿɫɢɫɬɟɦɡɚɤɨɟɮɿɰɿɽɧɬɨɦɩɟɪɟɞɚɱɿ ɝɭ ɨ k ǻ ɿɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹ ɝɭ ɨ ǻ Ĳ Ɂɧɚɱɟɧɧɹ ɡɚɩɚɫɿɜ ɫɬɿɣɤɨɫɬɿ ɞɨɫɥɿɞɠɭɜɚɧɢɯ ɜɚɪɿɚɧɬɿɜ ɋȺɊ ɛɥɢɡɶɤɿ ɨɞɢɧ ɞɨ ɨɞɧɨɝɨ Ɍɨɛɬɨ ɋȺɊ ɡ ɚɥɝɨɪɢɬɦɨɦ ɩɪɨɝɧɨɡɭɜɚɧɧɹ ɧɚ ɨɫɧɨɜɿ ɤɭɛɿɱɧɨɝɨ ɫɩɥɚɣɧɭ ɩɪɢ ɨɞɧɚɤɨɜɢɯ ɡɚɩɚɫɚɯ ɫɬɿɣɤɨɫɬɿ ɦɨɠɟ ɡɚɛɟɡɩɟɱɢɬɢ ɛɿɥɶɲ ɜɢɫɨɤɭ ɞɢɧɚɦɿɱɧɭɬɨɱɧɿɫɬɶɫɬɚɛɿɥɿɡɚɰɿʀɪɟɝɭɥɶɨɜɚɧɨʀɡɦɿɧɧɨʀɧɚɡɚɞɚɧɨɦɭɡɧɚɱɟɧɧɿ Ɋɢɫ 13 – ɉɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɋȺɊɩɪɢɧɨɦɿɧɚɥɶɧɢɯɡɧɚɱɟɧɧɹɯɩɚɪɚɦɟɬɪɿɜɈɄ (1 – ɋȺɊBɍɋ– ɋȺɊɉBɄ Ɋɢɫ 14 – ɊɟɡɭɥɶɬɚɬɢɦɨɞɟɥɸɜɚɧɧɹɋȺɊɭɫɬɚɥɨɦɭɪɟɠɢɦɿɪɨɛɨɬɢɩɪɢɞɿʀɜɢɩɚɞɤɨɜɢɯɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ (1 – ɋȺɊBɍɋ-ɋȺɊɉBɄ 1 2 y(t t, c y(t t, c Ɋɢɫ 13 – ɉɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɋȺɊɩɪɢɧɨɦɿɧɚɥɶɧɢɯɡɧɚɱɟɧɧɹɯɩɚɪɚɦɟɬɪɿɜɈɄ (1 – ɋȺɊBɍɋ– ɋȺɊɉBɄ 1 2 y(t t, c Ɋɢɫ 13 – ɉɟɪɟɯɿɞɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɋȺɊɩɪɢɧɨɦɿɧɚɥɶɧɢɯɡɧɚɱɟɧɧɹɯɩɚɪɚɦɟɬɪɿɜɈɄ (1 – ɋȺɊBɍɋ– ɋȺɊɉBɄ Ɋɢɫ 14 – ɊɟɡɭɥɶɬɚɬɢɦɨɞɟɥɸɜɚɧɧɹɋȺɊɭɫɬɚɥɨɦɭɪɟɠɢɦɿɪɨɛɨɬɢɩɪɢɞɿʀɜɢɩɚɞɤɨɜɢɯɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ (1 – ɋȺɊBɍɋ-ɋȺɊɉBɄ y(t t, c Ɋɢɫ 14 – ɊɟɡɭɥɶɬɚɬɢɦɨɞɟɥɸɜɚɧɧɹɋȺɊɭɫɬɚɥɨɦɭɪɟɠɢɦɿɪɨɛɨɬɢɩɪɢɞɿʀɜɢɩɚɞɤɨɜɢɯɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ (1 – ɋȺɊBɍɋ-ɋȺɊɉBɄ ȿɮɟɤɬɢɜɧɿɫɬɶ ɩɪɢɞɭɲɟɧɧɹ ɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɋȺɊɉBɄ ɩɿɞɬɜɟɪɞɠɭɽɬɶɫɹ ɿ ɱɚɫɬɨɬɧɢɦɢ ɯɚɪɚɤɬɟɪɢɫɬɢɤɚɦɢ ɡɚɦɤɧɭɬɢɯ ɋȺɊ ɪɢɫ ɇɚ ɫɟɪɟɞɧɶɨɤɜɚɞɪɚɬɢɱɧɿ ɱɚɫɬɨɬɿ ɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɡɧɚɱɟɧɧɹ Ⱥɑɏ ɰɿɽʀ ɫɢɫɬɟɦɢ ɡɧɚɱɧɨɦɟɧɲɟɧɿɠɡɜɢɱɚɣɧɨʀɋȺɊɬɚɋȺɊBɍɋ B 1 2 Z1 Z1 1 2 'MȽɍ 'ȺȽɍ Ɋɢɫ 15 – ɑɚɫɬɨɬɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɪɨɡɿɦɤɧɭɬɢɯɋȺɊɡɪɨɡɪɚɯɭɧɤɨɦɡɧɚɱɟɧɶʀɯɡɚɩɚɫɿɜɫɬɿɣɤɨɫɬɿɡɚɤɨɟɮɿɰɿɽɧɬɨɦ ɩɟɪɟɞɚɱɿ ɝɭ ɨ k ǻ ɿɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹ ɝɭ ɨ ǻ Ĳ ɜɦɨɞɟɥɿɈɄ– ɋȺɊBɍɋ-ɋȺɊɉBɄ 1 2 1 2 2 1 Z1 Z1 Z-S Z-S 'MȽɍ 'ȺȽɍ Ɋɢɫ 15 – ɑɚɫɬɨɬɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɪɨɡɿɦɤɧɭɬɢɯɋȺɊɡɪɨɡɪɚɯɭɧɤɨɦɡɧɚɱɟɧɶʀɯɡɚɩɚɫɿɜɫɬɿɣɤɨɫɬɿɡɚɤɨɟɮɿɰɿɽɧɬɨɦ ɩɟɪɟɞɚɱɿ ɝɭ ɨ k ǻ ɿɱɚɫɨɦɡɚɩɿɡɧɟɧɧɹ ɝɭ ɨ ǻ Ĳ ɜɦɨɞɟɥɿɈɄ– ɋȺɊBɍɋ-ɋȺɊɉBɄ 46 Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ Volume 13, Issue 3 /2021 http://www.atbp.onaft.edu.ua/ Ɋɢɫ 16 – ɑɚɫɬɨɬɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɡɚɦɤɧɭɬɢɯɨɩɬɢɦɚɥɶɧɢɯɋȺɊɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ– ɋȺɊ– ɋȺɊBɍɋ– ɋȺɊɉBɄȦɫɤɩ – ɫɟɪɟɞɧɶɨɤɜɚɞɪɚɬɢɱɧɚɱɚɫɬɨɬɚɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ 1 2 3 ʘ̡̭̪=0.19 1 2 3 ʘ̡̭̪=0.19 Ɋɢɫ 16 – ɑɚɫɬɨɬɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɡɚɦɤɧɭɬɢɯɨɩɬɢɦɚɥɶɧɢɯɋȺɊɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ– ɋȺɊ– ɋȺɊBɍɋ– ɋȺɊɉBɄȦɫɤɩ – ɫɟɪɟɞɧɶɨɤɜɚɞɪɚɬɢɱɧɚɱɚɫɬɨɬɚɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ Ɋɢɫ 16 – ɑɚɫɬɨɬɧɿɯɚɪɚɤɬɟɪɢɫɬɢɤɢɡɚɦɤɧɭɬɢɯɨɩɬɢɦɚɥɶɧɢɯɋȺɊɡɚɤɚɧɚɥɨɦɞɿʀɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ– ɋȺɊ– ɋȺɊBɍɋ– ɋȺɊɉBɄȦɫɤɩ – ɫɟɪɟɞɧɶɨɤɜɚɞɪɚɬɢɱɧɚɱɚɫɬɨɬɚɧɟɤɨɧɬɪɨɥɶɨɜɚɧɢɯɡɛɭɪɟɧɶ ȼɢɫɧɨɜɤɢ ɋɩɢɫɨɤɜɢɤɨɪɢɫɬɚɧɢɯɞɠɟɪɟɥ ɋ ɫɨ ɨɪ ɫ ɚ ɞ ɟɪɟɥ [1] O. J. M. Smith. A controller to overcome dead-time. ISA Transactions, 6 (2):28–33, 1959. [1] O. J. M. Smith. A controller to ov [2] J. Rodriguez and P. Cortes. Predictive Control of Power Converters and Electrical Drives. Chichester, UK: Wiley-IEEE Press, 2012. https://doi.org/10.1002/9781119941446 [2] J. Rodriguez and P. Cortes. Predictive Control of Power Converters and Electrical Drives. Chichester, UK: Wiley-IEEE Press, 2012. https://doi.org/10.1002/9781119941446 [2] J. Rodriguez and P. Cortes. Predictive Control of Power Converters and Electrical Drives. Press, 2012. https://doi.org/10.1002/9781119941446 p g [3] Eduardo F. Camacho and Carlos Bordons Alba. Model predictive control. Springer Science & Busine [4] ɋɬɟɩɚɧɨɜ ɆɌ ɏɨɛɢɧ ȼȺ ɉɪɨɝɧɨɡɢɪɨɜɚɧɢɟ ɜɵɧɭɠɞɟɧɧɨɝɨ ɞɜɢɠɟɧɢɹ ɢ ɟɝɨ ɩɪɢɦɟɧɟɧɢɟ ɜ ɫɢɫɬɟɦɚɯ ɝɚɪɚɧɬɢɪɭɸɳɟɝɨɭɩɪɚɜɥɟɧɢɹȺɜɬɨɦɚɬɢɡɚɰɿɹɬɟɯɧɨɥɨɝɿɱɧɢɯɿɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ – Ɉɞɟɫɚ– ʋ-6. – ɋ– 25. [5] ɉɢɤɢɧɚ ȽȺ Ʉɭɡɧɟɰɨɜ Ɇɋ ɋɢɧɬɟɡ ɥɢɧɟɣɧɵɯ ɩɪɨɝɧɨɫɬɢɱɟɫɤɢɯ ɚɥɝɨɪɢɬɦɨɜ ɪɟɝɭɥɢɪɨɜ ɪɨɫɫɢɣɫɤɨɣɷɥɟɤɬɪɨɷɧɟɪɝɟɬɢɤɟʋɋ-44. [6] ɉɢɤɢɧɚȽȺɄɭɡɧɟɰɨɜɆɋɉɪɨɝɧɨɫɬɢɱɟɫɤɢɟɬɢɩɨɜɵɟɚɥɝɨɪɢɬɦɵɪɟɝɭɥɢɪɨɜɚɧɢɹɌɟɩɥɨɷɧɟɪɝɟɬɢɤɚʋ ɋ-66. [6] ɉɢɤɢɧɚȽȺɄɭɡɧɟɰɨɜɆɋɉɪɨɝɧɨɫɬɢɱɟɫɤɢɟɬɢɩɨɜɵɟɚɥɝɨɪɢɬɦɵɪɟɝɭɥɢɪɨɜɚɧɢɹɌɟɩɥɨɷɧɟɪɝɟɬɢɤɚʋ ɋ-66. [7] ɋɬɟɩɚɧɨɜ ɆɌ ɏɨɛɢɧ ȼȺ ɋɢɫɬɟɦɚ ɚɜɬɨɦɚɬɢɱɧɨɝɨ ɪɟɝɭɥɸɜɚɧɧɹ ɿɧɜɚɪɿɚɧɬɧɚ ɞɨ ɤɨɧɬɪɨɥɶɨɜɚɧɢɯ ɡɛɭɪɟɧɶ ɡ ɩɪɨɝɧɨɡɭɜɚɧɧɹɦ ɫɢɝɧɚɥɭ ɤɨɪɟɤɰɿʀ ɩɨ ɤɭɛɿɱɧɨɦɭ ɫɩɥɚɣɧɭ Ⱥɜɬɨɦɚɬɢɡɚɰɿɹ ɬɟɯɧɨɥɨɝɿɱɧɢɯ ɿ ɛɿɡɧɟɫ-ɩɪɨɰɟɫɿɜ. – Ɉɞɟɫɚ– ʋ– Ɍ– ɋ– 70. 47 References References [1] O. J. M. Smith. A controller to overcome dead-time. ISA Transactions, 6 (2):28–33, 1959. References [1] O. J. M. Smith. A controller to overcome dead-time. ISA Transactions, 6 (2):28–33, 1959. [2] J. Rodriguez and P. Cortes. Predictive Control of Power Converters and Electrical Drives. Chichester, UK: Wiley-IEEE Press, 2012. https://doi.org/10.1002/9781119941446 Eduardo F. Camacho and Carlos Bordons Alba. Mo p p g [4] M.T. Stepanov et al. “Prognozirovanie vyinuzhdennogo dvizheniya i ego primenenie v sistemah garantiruyuschego upravleniya”, Avtomatizatsiya tehnologichnih ta biznes-protsesiv”, no.5-6, pp.20-25, 2011. [5] G.A. Pikina et al. “Sintez lineynyih prognosticheskih algoritmov regulirovaniya”, Novoe v rossiyskoy elektroenergetike 2009, vol. 10. pp. 40-44. pp et al. “Prognosticheskie tipovyie algoritmyi regulirovaniya”, Teploenergetika 2011, vol. 4. pp. 61-66. pp [6] G.A. Pikina et al. “Prognosticheskie tipovyie algoritmyi regulirovaniya”, Teploenergetika 2011, vol. [7] M.T. Stepanov et al. “Systema avtomatychnogo regulyuvannya invariantna do kontrolovanyx zburen z prognozuvannyam sygnalu korekciyi po kubichnomu splajnu”, Avtomatizatsiya tehnologichnih ta biznes-protsesiv”, no.1, vol 12, pp.64-70, 2020. Ɉɬɪɢɦɚɧɚɜɪɟɞɚɤɰɿʀ19.08.2021. ɉɪɢɣɧɹɬɚ ɞɨ ɞɪɭɤɭ 27.08.2021. Received 19 August 2021. Approved 27 August 2021. Available in Internet 31 September 2021. ɍȾɄ ȺɇȺɅȱɁɑɂɇɇɂɄȱȼɓɈȼɉɅɂȼȺɘɌɖɇȺ ɍɋɉȱɒɇȱɋɌɖȼȿȻɊȿɋɍɊɋȱȼɉȱȾɊɈɁȾȱɅȱȼɁȺɄɅȺȾȱȼ ȼɂɓɈȲɈɋȼȱɌɂ ɋɟɥɿɜɚɧɨɜɚȺȼ1, ȻɨɞɸɥɈɋ2Ʉɨɪɨɛɨɜȼɋ3 ɋɟɥɿɜɚɧɨɜɚȺȼ1, ȻɨɞɸɥɈɋ2Ʉɨɪɨɛɨɜȼɋ3 1,2,3Ɉɞɟɫɶɤɚɧɚɰɿɨɧɚɥɶɧɚɚɤɚɞɟɦɿɹɯɚɪɱɨɜɢɯɬɟɯɧɨɥɨɝɿɣ ORCID: 10000-0002-3395-1422, 20000-0001-9925-434X E-mail: 1av_selivanova@ukr.net, bodyulolena@ukr.net2, vetkor97@gmail.com3 Copyright © 2021 by author and the journal “Automation of technological and business – processes”. This work is licensed under the Creative Commons Attribution International License (CC BY). http://creativecommons.org/licanses/by/4.0 Copyright © 2021 by author and the journal “Automation of technological and business – processes” This work is licensed under the Creative Commons Attribution International License (CC BY). http://creativecommons.org/licanses/by/4.0 opyright © 2021 by author and the journal “Automation of technological and business – processes”. hi k i li d d h C i C A ib i I i l Li (CC BY) DOI: DOI: Ⱥɧɨɬɚɰɿɹɍɫɬɚɬɬɿɪɨɡɝɥɹɞɚɽɬɶɫɹɩɪɨɰɟɫɜɢɹɜɥɟɧɧɹɬɚɞɨɫɥɿɞɠɟɧɧɹɱɢɧɧɢɤɿɜɳɨɜɩɥɢɜɚɸɬɶɧɚɭɫɩɿɲɧɿɫɬɶɜɟɛ ɪɟɫɭɪɫɿɜɩɿɞɪɨɡɞɿɥɿɜɡɚɤɥɚɞɿɜɜɢɳɨʀɨɫɜɿɬɢɁȼɈȼɩɪɨɰɟɫɿɞɨɫɥɿɞɠɟɧɧɹɩɪɨɜɟɞɟɧɨɚɧɚɥɿɡɪɨɛɿɬɩɪɢɫɜɹɱɟɧɢɯɰɶɨɦɭ ɩɢɬɚɧɧɸ ɩɪɨɜɟɞɟɧɨ ɚɧɚɥɿɡ ɿɫɧɭɸɱɢɯ ɩɨɩɭɥɹɪɧɢɯ ɫɟɪɟɞ ɤɨɪɢɫɬɭɜɚɱɿɜ ɜɟɛ ɪɟɫɭɪɫɿɜ ɜɢɹɜɥɟɧɨ ɤɪɢɬɟɪɿʀ ɨɰɿɧɤɢ ɬɚɤɢɯ ɫɢɫɬɟɦȼɪɨɛɨɬɚɯɿɧɲɢɯɚɜɬɨɪɿɜɹɤɿɩɪɨɜɨɞɢɥɢɚɧɚɥɨɝɿɱɧɿɞɨɫɥɿɞɠɟɧɧɹɚɧɚɥɿɡɭɸɬɶɫɹɱɢɧɧɢɤɢɹɤɿɜɩɥɢɜɚɸɬɶɧɚ ɩɿɞɜɢɳɟɧɧɹ ɹɤɿɫɧɨɝɨ ɪɿɜɧɹ ɜɟɛ-ɫɚɣɬɿɜ ɞɨɜɟɞɟɧɨ ɹɤ ɜɚɠɥɢɜɨ ɩɿɞɜɢɳɭɜɚɬɢ ɤɨɦɭɧɿɤɚɬɢɜɧɭ ɮɭɧɤɰɿɨɧɚɥɶɧɿɫɬɶ ɫɚɣɬɭ ɣɨɝɨɜɿɞɜɿɞɭɜɚɧɿɫɬɶɬɚɜɿɞɝɭɤɚɭɞɢɬɨɪɿʀȾɥɹɞɨɫɹɝɧɟɧɧɹɦɟɬɢɩɨɫɬɚɜɥɟɧɨʀɚɜɬɨɪɚɦɢɞɚɧɨʀɪɨɛɨɬɢɛɭɥɨɩɪɨɜɟɞɟɧɨ ɨɩɢɬɭɜɚɧɧɹɫɟɪɟɞɩɨɬɟɧɰɿɣɧɢɯ ɫɩɨɠɢɜɚɱɿɜɜɟɛɪɟɫɭɪɫɿɜɊɨɡɩɨɞɿɥɪɟɫɩɨɧɞɟɧɬɿɜɡɚɜɿɤɨɦɬɚɮɚɯɨɜɢɦɢɤɚɬɟɝɨɪɿɹɦɢ ɞɨɡɜɨɥɢɥɨ ɜɢɹɜɢɬɢ ɳɨ ɜ ɨɫɧɨɜɧɨɦɭ ɫɩɨɠɢɜɚɱɚɦɢ ɫɚɣɬɿɜ ɤɚɮɟɞɪ ɁȼɈ ɽ ɡɞɨɛɭɜɚɱɿ ɨɫɜɿɬɢ Ⱥɧɚɥɿɡ ɪɟɡɭɥɶɬɚɬɿɜ ɩɪɨɜɟɞɟɧɨɝɨ ɨɩɢɬɭɜɚɧɧɹ ɤɨɪɢɫɬɭɜɚɱɿɜ ɫɚɣɬɿɜ ɁȼɈ ɞɨɩɨɦɿɝ ɜɢɹɜɢɬɢ ɱɢɧɧɢɤɢ ɳɨ ɜɩɥɢɜɚɸɬɶ ɧɚ ʀɯ ɭɫɩɿɲɧɿɫɬɶ ɬɚ ɩɨɛɭɞɭɜɚɬɢ ɦɨɞɟɥɶ ɭɫɩɿɲɧɨɫɬɿ ɜɟɛ ɪɟɫɭɪɫɿɜ ɩɿɞɪɨɡɞɿɥɿɜ ɁȼɈ ɹɤɚ ɫɬɚɥɚ ɜɚɠɥɢɜɢɦ ɿɧɮɨɪɦɚɬɢɜɧɢɦ ɮɚɤɬɨɪɨɦ ɞɥɹ ɩɪɢɣɧɹɬɬɹɪɿɲɟɧɧɹɩɪɨɞɢɡɚɣɧɟɥɟɦɟɧɬɢɬɚɡɦɿɫɬɧɨɜɨʀɜɟɛɫɢɫɬɟɦɢɊɨɡɪɨɛɥɟɧɚɦɨɞɟɥɶɜɩɪɨɜɚɞɠɟɧɚɭɩɪɨɰɟɫɿ ɪɨɡɪɨɛɤɢɧɨɜɨɝɨɜɟɛ-ɪɟɫɭɪɫɭɤɚɮɟɞɪɢȱɧɮɨɪɦɚɰɿɣɧɢɯɬɟɯɧɨɥɨɝɿɣɬɚɤɿɛɟɪɛɟɡɩɟɤɢɈɇȺɏɌɳɨɡɚɛɟɡɩɟɱɢɥɨɩɿɞɜɢɳɟɧɧɹ ɟɮɟɤɬɢɜɧɨɫɬɿ ɩɨɞɚɱɿ ɿɧɮɨɪɦɚɰɿʀ ɩɪɢɜɚɛɥɢɜɿɫɬɶ ɬɚ ɡɚɬɪɟɛɭɜɚɧɿɫɬɶ ɡɚɡɧɚɱɟɧɨɝɨ ɪɟɫɭɪɫɭ Ɉɬɪɢɦɚɧɿ ɜ ɞɨɫɥɿɞɠɟɧɧɿ ɪɟɡɭɥɶɬɚɬɢ ɦɨɠɭɬɶ ɜɢɤɨɪɢɫɬɨɜɭɜɚɬɢɫɹ ɩɪɢ ɩɨɛɭɞɨɜɿ ɪɿɡɧɨɦɚɧɿɬɧɢɯ ɿɧɮɨɪɦɚɰɿɣɧɢɯ ɪɟɫɭɪɫɿɜ ɩɿɞɪɨɡɞɿɥɿɜ ɡɚɤɥɚɞɿɜ ɜɢɳɨʀɨɫɜɿɬɢ ɞɥɹɩɿɞɜɢɳɟɧɧɹʀɯɩɪɢɜɚɛɥɢɜɨɫɬɿɿɧɮɨɪɦɚɬɢɜɧɨɫɬɿɬɚɜɿɞɜɿɞɭɜɚɧɨɫɬɿɩɥɚɧɭɽɬɶɫɹɩɨɞɚɥɶɲɚɪɨɛɨɬɚɡ ɜɞɨɫɤɨɧɚɥɟɧɧɹɪɨɡɪɨɛɥɟɧɨʀɦɨɞɟɥɿɿɜɟɛɫɢɫɬɟɦɢɳɨʀʀɜɢɤɨɪɢɫɬɨɜɭɽ Abstract. The article considers the process of identifying and researching the factors influencing the success of web resources of higher education institutions (HEIs). In the course of the research the analysis of works devoted to this question is carried out, the analysis of the existing popular among users web resources is carried out, the criteria of an estimation of such systems are revealed. The analysis of the results of the survey of consumers of web resources of free economic education 48
https://openalex.org/W2060098285	http://www.journal-imab-bg.org/issue-2013/issue4/JofIMAB2013vol19b4p388-391.pdf	English	null	ORAL SIGNS OF LEUKEMIA AND DENTAL MANAGEMENT – literature data and case report.	Journal of IMAB	2,013	cc-by-sa	2,723	Journal of IMAB - Annual Proceeding (Scientific Papers) 2013, vol. 19, issue 4 Journal of IMAB - Annual Proceeding (Scientific Papers) 2013, vol. 19, issue 4 ISSN: 1312-773X (Online) INTRODUCTION: There are three main groups of hematologic malignancies: leukemia, lymphoma and plasma cell tumors. Leukemia is a hematological disorder which is caused by proliferating white blood cell-forming tissues resulting in a marked increase in circulating immature or abnormal white blood cells. Leukemia arises from a hematopoietic stem cell characterized by a disordered differentiation and proliferation of neoplastic cells. Leukemia results from the proliferation of a clone of abnormal hematopoietic cells with impaired differentiation, regulation, and programmed cell death (apoptosis). Leukemic cell multiplication at the expense of normal hematopoietic cell lines causes marrow failure, depressed blood cell count (cytopenia), and death as a result of infection, bleeding, or both [1, 2]. Classification of Leukemia Leukemia is classified based on clinical behavior (acute or chronic) and the primary hematopoietic cell line affected (myeloid or lymphoid). The four principal diagnostic categories are the following [3, 4]: In the oral cavity local symptoms and findings of leukemia include paleness of the oral mucosa with gingival bleeding that develops into painless gingival hyperplasia, petechiae, hemorrhages, and ulcerative necrotic lesions 1. acute myelogenous leukemia (AML), 2. acute lymphocytic leukemia (ALL), 3. chronic myelogenous leukemia (CML) and 4. chronic lymphocytic leukemia (CLL). Because of their clinical importance, all such lesions deserve the full attention of the dental doctors. ABSTRACT: smoking and exposure to electromagnetic fields also have been proposed to be causative (3). The oral signs and symptoms may be reflect a undetected serious systemic diseases. Depending on the oral manifestation the dentists and physicians make attention and focusing on specific diagnoses. In some cases oral involvement may be frequently herald the onset of the disease which requires the dentists to better knowledge of changes in the oral cavity. Clinical symptoms The aim is to evaluate in detail the oral complications of leukemia at initial presentation and present a clinical case of 54 old female with oral manifestation as initial signs of the disease. Chronic leukemia, with a less pronounced marrow failure, has an indolent course that usually lasts several years. Symptoms are generally flu-like with bone pain, joint pain, or both, caused by malignant marrow expansion [1, 4]. Acute myelogeous leukemia symptoms include fever, fatigue, pallor, mucosal bleeding, petechiae, and local infections; clinical manifestations of acute lymphocytic leukemia are similar to those of acute myelogeous leukemia, but with a high incidence of central nervous system disease [1, 4]. Key words: oral manifestation, leukemia, dentist, management ORAL SIGNS OF LEUKEMIA AND DENTAL MANAGEMENT – literature data and case report Elitsa G. Deliverska1, Assya Krasteva2 1)Department of Oral and Maxillofacial surgery, 2)Department of Oral Imaging and oral diagnostic, Faculty of dental medicine, Medical University, Sofia, Bulgaria Elitsa G. Deliverska1, Assya Krasteva2 1)Department of Oral and Maxillofacial surgery, 2)Department of Oral Imaging and oral diagnostic, Faculty of dental medicine, Medical University, Sofia, Bulgaria / J of IMAB. 2013, vol. 19, issue 4/ http://dx.doi.org/10.5272/jimab.2013194.388 http://dx.doi.org/10.5272/jimab.2013194.388 AIM: 65% of patients with leukemia reviewed in the course of their disease oral signs or symptoms. The aim of this article is to evaluate in detail the oral complications of leukemia at initial presentation and present a clinical case of 54 y old female with oral manifestation as initial signs of the disease. Associations between oral manifestations dental management were also comment. CASE REPORT: Ecchymosis is included in the differential diagnosis, a hemorrhagic diathesis or coagulation disorder. Certainly, patients taking anticoagulant drugs may present with oral ecchymosis, particularly on the buccal mucosa or tongue, either of which can be traumatized while chewing. Ecchymoses of the oral mucosa may also be encountered in patients with liver cirrhosis, and end-stage renal disease undergoing dialysis treatment [3,7]. A 54-year-old woman applied to Medical University, Faculty of Dentistry, Department of Oral and Maxillofacial surgery, with the chief complaint of severe gingival hyperplasia with rapid development in four weeks time. Extra-oral examination revealed swelling and tenderness to palpitation of the cervical lymph nodes. Dental examination showed a prominent generalized gingival hyperplasia without bleeding on the maxilla and mandibula. Gingival hyperplasia was involving the buccal, lingual and palatal aspects, as well. Laboratory finding in leukemia In patients with leukemia the overgrowth of malignant hematopoietic cells in the bone marrow with subsequent spillage into the peripheral blood leads to a reduction in the number of normal circulating blood cells. Patients can present with symptoms related to anemia, neutropenia, and hrombocytopenia [1, 2, 5, 6]. The peripheral granulocyte count is markedly elevated in chronic leukemia but may be increased (with numerous blast forms), decreased, or normal in acute leukemia. The laboratory diagnosis of leukemia is made from the identification of abnormal hematopoietic cells in the peripheral blood and bone marrow. Further characterization is by cytochemical staining (myeloperoxidase, Sudan black B), immunophenotyping (cell surface markers, cytoplasmic immunoglobulin, terminal deoxynucleotide transferase The cause of leukemia remains unknown. Increased risk is associated with large doses of ionizing radiation, certain chemicals (benzene), and infection with specific viruses (e.g., Epstein-Barr virus, human lymphotropic virus. Cigarette / J of IMAB. 2013, vol. 19, issue 4/ http://www.journal-imab-bg.org 388 detection), and cytogenetic analysis of chromosomal abnormalities [3]. affecting both the keratinized and nonkeratinized mucosa and oral colonization by Candida albicans [4, 5, 6, 7, 9]. Gingival bleeding, petechiae, ecchymosis Thrombocytopenia is manifested in the oral cavity of petechiae, ecchymosis, and gingival hemorrhage or gingival bleeding [1, 7, 8, 9, 10]. Gingival Enlargement Gingival enlargement or overgrowth is usually caused by local inflammatory conditions such as poor oral hygiene, food impaction, or mouth breathing. Systemic conditions such as hormonal changes, drug therapy, or tumor infiltrates may also cause or contribute to the severity of gingival enlargement. [3, 11, 12]. Gingival hyperplasia can be detect in von Recklinghausen’s neurofibromatosis (neurofibromatosis 1), Wegener’s granulomatosis, sarcoidosis, Crohn’s disease, primary amyloidosis, Kaposi’s sarcoma, acromegaly, and lymphoma and patients with leukemia [2, 3, 7, 8]. In medical examination, the patient mention of clinical symptoms such as fatigue, nausea, vomit, anorexia, and weight loss in a month. Complete blood count, peripheral blood smear were taken from the patient. Complete blood count displayed lowered hematocrit and hemoglobin levels (anemia); and a low platelet count (thrombocytopenia), white blood cell levels were 99,2õ10/9/ l, and a decrease in neutrophil levels (neutropenia). The results confirmed the diagnosis of acute leukaemia and the patient was immediately admitted to the oncology unit of a general hospital for bone marrow biopsy and further management. Gingival hyperplasia secondary to infiltration of the gingival tissue with leukemia cells is thoroughly described in the literature. It is characterized by progressive enlargement of the interdental papillae as well as the marginal and attached gingival. In the condition’s most pronounced form, the crowns of the teeth may be covered. Gingiva appear swollen, devoid of stippling and pale red to deep purple in colour. Gingival infiltration by leukemic cells will also predispose the patient with leukemia to bleeding [1]. Gingival hiperplasia is more common in acute tan chronic leukemia nonetheless; the development of gingival infiltration is unpredictable in any individual patient. Generally, gingival hyperplasia resolves completely or at least partly with effective leukemia chemotherapy [1, 7, 9]. / J of IMAB. 2013, vol. 19, issue 4/ http://www.journal-imab-bg.org DISCUSSION: The oral manifestation of leukemia depend on the general status of patients. For instance before treatment leukemic infiltrates cause a wide range of oral disease. Typical oral manifestations of acute leukaemias include gingival swelling, oral ulceration, spontaneous gingival bleeding, petechiae, mucosal pallor, herpetic infections and Gingival ulceration and oral infection Oral signs and symptoms in leukemia may consist of paleness of the oral mucosa and also mucosal ulcers. Oral infections often present atypically, for example dental abscesses may present as soft tissue necrosis without swelling, and recrudescent HSV, may present with widespread lesions Fig. 1. / J of IMAB. 2013, vol. 19, issue 4/ 389 http://www.journal-imab-bg.org candidosis [4, 5, 6, 7, 13, 14]. Advanced cases may involve malaise, cervical lymphadenopathy, laryngeal pain and fever [8, 15]. Cervical lymphadenopathy, caused by infiltration of leukemic cells into the regional lymph nodes and hyperplasia of lymphatic issue particulary in Waldeyers ring. Enlarge tonsils and pharyngitis may be the initial complaint. Other uncommon orofacial signs of leukemia are pallor, parotid swelling and palatal pigmentation [16] . Radiographically leukemic infiltrates may produce destructive radiolucencies with loss of the lamina dura and erosion of the crestal alveolar bone [17, 18]. Later once medial therapy has been initiated manifestation of leukemia are often replaced by the effects of chemotherapy and total body irradiation because these agents ate toxic to rapidly dividing cells both cancer call and normal cells. Complications of chemotherapy and irradiation include mucositis, hemorrhage, xerostomia, periodontal inflammation, recurrent herpes simplex virus infection and bacterial and fungial infections [18]. Based on these complication was established protocol for monitoring patients with leukemia. Fig. 2. Figure 1 and figure 2. Initial intra-oral view of the patient. Fig. 2. Figure 1 and figure 2. Initial intra-oral view of the patient. Dental management of patient with hematologic malignancies Considerations in dental management of patients with hematologic malignancies [1] are summarized in table 1. ns in dental treatment of patients with hematologic malignancies by FA Mancheño et coworkers. Tabl. 1 Considerations in dental treatment of patients with hematologic malignancies by FA Mancheño et coworkers. Tabl. 1 Considerations in dental treatment of patients with hematologic malignancies by FA Mancheño et coworkers. Prior to dental treatment 1. Dental treatment should be performed always after consultation with the specialist 2. It is important to carry out a detailed history, a comprehensive oral and dental evaluation and a complete radiographic exam. 3. Dental treatment should be performed before starting the chemo/radiotherapy. 4. Patients in long-term remission can undergo dental treatment, while patients with advanced or relapsed disease with reserved prognosis should receive palliative or urgent treatment only. During dental treatment 1. Bleeding tendency 2. Increased risk of infection. 3. Risk of developing osteonecrosis of the jaw. 4. Anemia 5. Corticosteroids treatment. 6. Secondary malignancies. 7. Specific considerations. secondary adrenal insufficiency. • Secondary malignancies Once the diagnosis has been made, consultation with the patient physician or oncologist is mandatory before commencing dental treatm. REFERENCES: Periodontol. 1984 Oct;55(10):585-588. [PubMed] [CrossRef] Periodontol. 1984 Oct;55(10):585-588. [PubMed] [CrossRef] of acute lymphoblastic leukemia in a young girl. J Indian Soc Pedod Prev Dent. 2012 Apr-Jun;30(2):166-168. [PubMed] [CrossRef] 1. Franch AM, Esteve CG, Perez, GS. Oral manifestations and dental management of patient with leukocyte alterations. J Clin Exp Dent. 2011; 3(1):e53-59. [CrossRef] 14. Orbak R, Orbak Z. Oral condition of patients with leukemia and lymphoma. J Nihon Univ Sch Dent. 1997 Jun;39(2):67-70. [PubMed] 8. Dean AK, Ferguson JW, Marvan ES. Acute leukaemia presenting as oral ulceration to a dental emergency service. Aust Dent J. 2008 Sep;48(3): 195-7. [PubMed] [CrossRef] 2. Demirer S, Ozdemir H, Sencan M, Marakoglu I. Gingival Hyperplasia as an Early Diagnostic Oral Manifes- tation in Acute Monocytic Leukemia: A Case Report. Eur J Dent. 2007 Apr; (2):111–114. [PubMed] 2. Demirer S, Ozdemir H, Sencan M, Marakoglu I. Gingival Hyperplasia as an Early Diagnostic Oral Manifes- tation in Acute Monocytic Leukemia: A Case Report. Eur J Dent. 2007 Apr; (2):111–114. [PubMed] 15. Hou GL, Huang JS, Tsai CC. Analysis of oral manifestations of leukemia: a retrospective study. Oral Dis. 1997 Mar;3(1):31–38. [PubMed] [CrossRef] 9. Guzeldemir E, Toygar HU, Kocer NE, Kizilkilic E. The periodontal management of a patient with acute myelomonocytic leukemia. Nobel Med. 2012 Jan-Apr;8(1):110-113. [Full text] 9. Guzeldemir E, Toygar HU, Kocer NE, Kizilkilic E. The periodontal management of a patient with acute myelomonocytic leukemia. Nobel Med. 2012 Jan-Apr;8(1):110-113. [Full text] 10. Chapple IL, Saxby MS, Murray JA. Gingival hemorrhage, myelo- dysplastic syndromes, and acute myeloid leukemia. A case report. J Periodontol. 1999 Oct;70(10):1247- 1253. [PubMed] [CrossRef] 16. Loudona JA, Colemanb, HJ, Allenc CM, Schifter M, Ng T, Al- Horani G. Rare oral manifestation of chronic myelogenous leukemia and its targeted therapy: A case report and literature review. Univers J Med Dent. 2012 Oct;1(7):79-85. [Full text] 3. Greenberg MS, Glick M, Ship JA. (editors.) Burket’s Oral Medicine. 11th edition. Hamilton. BC Decker inc. 2008; 400-403. 10. Chapple IL, Saxby MS, Murray JA. Gingival hemorrhage, myelo- dysplastic syndromes, and acute myeloid leukemia. A case report. J Periodontol. 1999 Oct;70(10):1247- 1253. [PubMed] [CrossRef] 4. Wu J, Fantasia JE, Kaplan R. Oral manifestations of acute myelomonocytic leukemia: a case report and review of the classification of leukemias. J Periodontol. 2002 Jun; 73(6):664-668. [PubMed] [CrossRef] 17. Albuquerque MA, Migliari DA, Sugaya NN, Kuroishi M, Capuano AC, Sousa SO, et al. CONCLUSIONS: CONCLUSIONS: Oral health care professionals should be aware of the CONCLUSIONS: Oral health care professionals should be aware of the 18. Bricker SL, Langlais RP, Miller CS. Oral Diagnosis, Oral Medicine, and Treatment Planning. 2nd edition. Philadelphia; Lea & Febiger. 1994. / J of IMAB. 2013, vol. 19, issue 4/ of acute lymphoblastic leukemia in a young girl. J Indian Soc Pedod Prev Dent. 2012 Apr-Jun;30(2):166-168. [PubMed] [CrossRef] Specific considerations: The main problems in dental treatment of patients with hematologic malignancies of white cells are: - Patients with renal dysfunction may require modified dosing intervals of medications. • Bleeding tendency - In patients with multiple myeloma, it is important to evaluate for presence of hard/soft tissue masses that could indicate deposition of plasma cells and/or light chain associated amyloid, and biopsy if necessary. Patients with multiple myeloma and significant bone pain, especially in the • Increased risk of infection - odontogenic infections and opportunistic infections: • Risk of developing osteonecrosis of the jaw • Risk of developing osteonecrosis of the jaw • Anemia • Risk of developing osteonecrosis of the jaw • Anemia • Anemia • Anemia • Corticosteroids treatment - may display evidence of / J of IMAB. 2013, vol. 19, issue 4/ 390 importance of recognizing oral manifestations of systemic diseases. The dentist, and mainly the periodontists and oral pathologist, plays a fundamental role in the early diagnosis of leukemia knowing that the first symptoms of the disease occur in the oral cavity with normal or show subtle changes in initial laboratory tests. It is essential for the professional to be able to clearly recognize oral physiological characteristics, and, when identifying a change of normalcy, to fully investigate it requesting additional tests or referring the patient to specialized professional. back, may need frequent breaks and may require frequent repositioning during dental ients undergoing orthodontic treatment, the removal of orthodontic appliances and delivery of retainers is recommended, as well as the postponement of orthodontic treatment until the patient has finished immunosuppressive therapy and the risk of hemato- logic relapse requiring further intervention is reduced [1]. http://www.journal-imab-bg.org Address for correspondence: dr Elitsa Deliverska, Department of Oral and Maxillofacial surgery, Faculty of Dental Medicine, 1, Georgi Sofiyski blvd., 1431 Sofia, Bulgaria; tel.+359 888949740; email: elitsadeliverska@yahoo.com, REFERENCES: Adult T-cell leukemia/lymphoma with predominant bone involvement, initially diagnosed by its oral manifestation: A case report. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2005 Sep;100(3):315– 320. [PubMed] [CrossRef] 11. Cooper CL, Loewen R, Shore T. Gingival hyperplasia complicating acute myelomonocytic leukemia. J Can Dent Assoc. 2000 Feb;66(2):78-9. [PubMed] 11. Cooper CL, Loewen R, Shore T. Gingival hyperplasia complicating acute myelomonocytic leukemia. J Can Dent Assoc. 2000 Feb;66(2):78-9. [PubMed] 5. Weckx LL, Hidal LB, Marcucci G. Oral manifestations of leukemia. Ear Nose Throat J. 1990 May;69(5): 341-2, 345-346. [PubMed] 6. Bruch JM, Nathaniel ST. Clinical Oral Medicine and Pathology. Humana Press. 2010; [CrossRef] 12. Khera P, Zirwas MJ, English JC 3rd. Diffuse gingival enlargement. J Am Acad Dermatol. 2005 Mar;52(3 Pt 1):491-499. [PubMed] [CrossRef] 18. Bricker SL, Langlais RP, Miller CS. Oral Diagnosis, Oral Medicine, and Treatment Planning. 2nd edition. Philadelphia; Lea & Febiger. 1994. ; [ ] 7. Silva BA, Siqueira C, Castro P, Araújo SS, Volpato LE. Oral manifestations leading to the diagnosis 7. Silva BA, Siqueira C, Castro P, Araújo SS, Volpato LE. Oral manifestations leading to the diagnosis ) [ ] [ ] 13. Barrett AP. Gingival lesions in leukemia. A classification. J ) [ ] [ ] 13. Barrett AP. Gingival lesions in leukemia. A classification. J http://www.journal-imab-bg.org Address for correspondence: dr Elitsa Deliverska, Department of Oral and Maxillofacial surgery, Faculty of Dental Medicine, 1, Georgi Sofiyski blvd., 1431 Sofia, Bulgaria; tel.+359 888949740; email: elitsadeliverska@yahoo.com, / J of IMAB. 2013, vol. 19, issue 4/ / J of IMAB. 2013, vol. 19, issue 4/ 391 http://www.journal-imab-bg.org
https://openalex.org/W592514002	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0125212&type=printable	English	null	Usefulness of Midregional Proadrenomedullin to Predict Poor Outcome in Patients with Community Acquired Pneumonia	PloS one	2,015	cc-by	8,373	RESEARCH ARTICLE OPEN ACCESS Citation: Gordo-Remartínez S, Calderón-Moreno M, Fernández-Herranz J, Castuera-Gil A, Gallego- Alonso-Colmenares M, Puertas-López C, et al. (2015) Usefulness of Midregional Proadrenomedullin to Predict Poor Outcome in Patients with Community Acquired Pneumonia. PLoS ONE 10(6): e0125212. doi:10.1371/journal.pone.0125212 Usefulness of Midregional Proadrenomedullin to Predict Poor Outcome in Patients with Community Acquired Pneumonia Susana Gordo-Remartínez1¤, María Calderón-Moreno1☯, Juan Fernández-Herranz1☯, Ana Castuera-Gil1☯, Mar Gallego-Alonso-Colmenares1☯, Carolina Puertas-López2☯, José A. Nuevo-González1☯, Domingo Sánchez-Sendín1☯, Mercedes García-Gámiz1☯, José A. Sevillano-Fernández1‡, Luis A. Álvarez-Sala3,4‡, Juan A. Andueza-Lillo1,4‡, José M. de Miguel-Yanes3‡ a11111 1 Emergency Department, Hospital General Universitario “Gregorio Marañón”, Madrid, Spain, 2 Biochemical Department, Hospital General Universitario “Gregorio Marañón”, Madrid, Spain, 3 Internal Medicine Department, Hospital General Universitario “Gregorio Marañón”, Madrid, Spain, 4 Department of Medicine. Facultad de Medicina. Universidad Complutense de Madrid, Spain ☯These authors contributed equally to this work. ¤ Current address: Servicio de Urgencias, Hospital General Universitario “Gregorio Marañón”, C/. Doctor Esquerdo, 46, Madrid, Spain ‡ These authors also contributed equally to this work. susanagordo79@gmail.com ☯These authors contributed equally to this work. ¤ Current address: Servicio de Urgencias, Hospital General Universitario “Gregorio Marañón”, C/. Doctor Esquerdo, 46, Madrid, Spain ‡ These authors also contributed equally to this work. * susanagordo79@gmail.com ☯These authors contributed equally to this work. ¤ Current address: Servicio de Urgencias, Hospital General Universitario “Gregorio Marañón”, C/. Doctor Esquerdo, 46, Madrid, Spain ‡ These authors also contributed equally to this work. * susanagordo79@gmail.com ¤ Current address: Servicio de Urgencias, Hospital General Universitario “Gregorio Marañón”, C/. Doctor Esquerdo, 46, Madrid, Spain ‡ These authors also contributed equally to this work. * susanagordo79@gmail com Background midregional proadrenomedullin (MR-proADM) is a prognostic biomarker in patients with community-acquired pneumonia (CAP). We sought to confirm whether MR-proADM added to Pneumonia Severity Index (PSI) improves the potential prognostic value of PSI alone, and tested to what extent this combination could be useful in predicting poor outcome of pa- tients with CAP in an Emergency Department (ED). Academic Editor: James D. Chalmers, University of Dundee, UNITED KINGDOM Received: January 11, 2015 Accepted: March 22, 2015 Published: June 1, 2015 Copyright: © 2015 Gordo-Remartínez 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 file. Data Availability Statement: All relevant data are within the paper and its Supporting Information file. Funding: The reagents for pro-adrenomedullin determination in this study were supplied by Thermo Scientific (BRAHMS Iberia S.L.). This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials. This study was performed with no financial support. Methods Consecutive patients diagnosed with CAP were enrolled in this prospective, single-centre, observational study. We analyzed the ability of MR-proADM added to PSI to predict poor outcome using receiver operating characteristic (ROC) curves, logistic regression and risk reclassification and comparing it with the ability of PSI alone. The primary outcome was “poor outcome”, defined as the incidence of an adverse event (ICU admission, hospital re- admission, or mortality at 30 days after CAP diagnosis). Conclusion MR-proADM in combination with PSI may be helpful in individual risk stratification for short- term poor outcome of CAP patients, allowing a better reclassification of patients compared with PSI alone. Proadrenomedullin in Community Pneumonia alone (AUC 0.75 [0.65-0.85, p=0.56]). Ten patients were appropriately reclassified when the combined PSI and MR-proADM model was used as compared with the model of PSI alone. Net reclassification improvement (NRI) index was statistically significant (7.69%, p = 0.03) with an improvement percentage of 3.03% (p = 0.32) for adverse event, and 4.66% (P = 0.02) for no adverse event. (500 € each) for a lecture in one meeting sponsored by Thermo Scientific. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials. Materials and Methods Setting and study population Introduction A better knowledge of those patients with a higher risk of needing intensive care or those with a higher risk of being admitted after discharge could also help in the decision making. Concerning this, six studies have analysed the role of MR- proADM to predict complications in patients with CAP, but only three of them have tested its usefulness as compared with that of PSI, and none of them studied the risk of readmission of those patients [18–23]. The main objective of this study is to confirm whether MR-proADM added to PSI improves the potential prognostic value of PSI alone, and to what extent this combination could be useful in predicting poor outcome of patients with CAP. Results 226 patients were included; 33 patients (14.6%) reached primary outcome. To predict pri- mary outcome the highest area under curve (AUC) was found for PSI (0.74 [0.64-0.85]), which was not significantly higher than for MR-proADM (AUC 0.72 [0.63-0.81, p > 0.05]). The combination of PSI and MR-proADM failed to improve the predictive potential of PSI Competing Interests: Susana Gordo-Remartínez and Juan Andueza-Lillo received modest honoraria Competing Interests: Susana Gordo-Remartínez and Juan Andueza-Lillo received modest honoraria 1 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Introduction Community-acquired pneumonia (CAP) is the third most frequently diagnosed infection in the emergency departments (ED) [1–3]. CAP ranges from a mild affection that can be treated at home to a severe disease. In fact, CAP is one of the leading causes of death in developed countries [4–10]. An early identification of patients with a higher risk for complications could lead to both a reduction of medical costs and deaths. Clinical practice guidelines recommend the use of severity prediction rules to choose among the adequate empirical antibiotic regimen, the intensity of the requested complementary studies, and need of intensive care to finally de- cide whether treatment can be followed on an outpatient basis. All these recommendations fa- cilitate decision making in the ED [11–13]. In this way, the Pneumonia Severity Index (PSI) [14] is the most widely validated score [15] despite having some limitations [16]. The potential prognostic usefulness of several biomarkers for patients with CAP has been explored in recent years. Proadrenomedullin (ProADM) is a peptide with vasodilatory, antimi- crobial and anti-inflammatory properties. Specifically, its midregional fragment (MR- proADM) has been associated with mortality in patients with CAP [17]. Predicting the risk of death is crucial in the ED, but at the same time, the level of care required by patients cannot be assessed only in terms of mortality. A better knowledge of those patients with a higher risk of needing intensive care or those with a higher risk of being admitted after discharge could also help in the decision making. Concerning this, six studies have analysed the role of MR- proADM to predict complications in patients with CAP, but only three of them have tested its usefulness as compared with that of PSI, and none of them studied the risk of readmission of those patients [18–23]. The potential prognostic usefulness of several biomarkers for patients with CAP has been explored in recent years. Proadrenomedullin (ProADM) is a peptide with vasodilatory, antimi- crobial and anti-inflammatory properties. Specifically, its midregional fragment (MR- proADM) has been associated with mortality in patients with CAP [17]. Predicting the risk of death is crucial in the ED, but at the same time, the level of care required by patients cannot be assessed only in terms of mortality. Proadrenomedullin in Community Pneumonia attended 148,000 emergencies last year (pediatric and gynecological emergencies are not included). All consecutive patients diagnosed with CAP in the ED were offered to participate in the study. Pneumonia was defined as the presence of at least one of the following symptoms: cough, expectoration, dyspnea, chest pain, fever, abnormal lung auscultation or leukocytes > 10,000 or < 4,000 cells/μL in combination with a new infiltrate on a chest X-ray [9, 23]. The excluded cases were: pneumonia due to nosocomial origin (more than 48 hours of hospitalization in the last 30 days), patients with health-care associated pneumonia or patients whose baseline medical conditions could have conditioned the prognosis of pneumonia. The process of patients’ selection is summarised in Fig 1. Before the beginning of the study, several informative meetings were conducted among medical assistants and residents of the ED. When a diagnosis of pneumonia was made at the ED, a researcher was contacted 24 hours a day to assess whether the patients met the inclusion criteria. Where it was possible for a conscious and comprehending patient to give informed consent to take part in this study, the study was explained verbally to that patient by the investi- gator or research coordinator and the patient was given the opportunity to read the participant information sheet. After they had had their questions answered and if they were willing to take part in the study, they were asked to sign the consent form. Clinical assessment of the compe- tence of a potential participant to consent for research was undertaken by a treating Emergency clinician. Emergency physicians are experienced at evaluating the competence of their patients to understand their illness and consent for therapeutic interventions. Specific situations that precluded potential participants to sign informed consents by themselves were: previous cogni- tive impairment, a decreased consciousness or medical instability judged to limit their ability to understand the purpose of the intervention or to give an unbiased consent to be part of the study. Where it was not practicable to approach the participant, consent was obtained from the “next of kin”, term that describes the person who is legally allowed to give consent for the pa- tient. When the next of kin could not attend the hospital to receive information and sign the consent form in person, the participant was finally not included in the study. After signing the informed consent, the patient was enrolled in the study. In all cases, patients were treated by the Emergency clinician without knowing the MR-proADM results. Patients’ follow-up was done through their computerised medical records and phone calls after 30 and 90 days of con- sultation to the ED. No therapeutic interventions were considered in the study. It was approved by the Hospital Ethic Committee (CEIC Hospital General Universitario Gregorio Marañón 278/12). PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Setting and study population The NACURG (Neumonía Adquirida en la Comunidad en URGencias; in English, Communi- ty-Acquired Pneumonia at the ED) is an observational, prospective, single-centre cohort study in patients diagnosed with CAP at the ED of the Gregorio Marañón University Hospital, be- tween November 2012 and March 2013. This institution is a university tertiary-care public hos- pital of the Community of Madrid (Spain) with a reference population of 318,000 people that 2 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Outcomes and measurements The main outcome, namely poor outcome of CAP, was measured through a composite variable defined as the existence of an adverse event: ICU admission, hospital readmission, or mortality at 30 days after CAP diagnosis in the ED (short-term mortality). Other secondary outcomes were: mortality due to any cause at 90 days after CAP diagnosis in the ED (mid-term mortality). The independent variables were: • Epidemiological data, Charlson index of comorbidity [24], symptoms of the current disease and previous antibiotic treatment. Exploratory findings and routine analytical and radiologi- cal data were prospectively recorded by the attending physicians of the ED. • Microbiological tests: according to the SEPAR (Spanish Society of Respiratory Diseases) guidelines [11], we requested 3 blood culture samples using bottles with liquid media for each 3 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Proadrenomedullin in Community Pneumonia Fig 1. Patients included in the study. doi:10.1371/journal.pone.0125212.g001 Fig 1. Patients included in the study. doi:10.1371/journal.pone.0125212.g001 doi:10.1371/journal.pone.0125212.g001 hospitalised patient (with or without measurable fever). The following tests were done: fluoro- metric measurements using Bactec system for pneumococcus and Legionella pneumophila through the BinaxNOW immunochromatographic membrane test; in epidemical situation or in cases of a high clinical suspicion, influenza antigens were detected by means of immuno- chromatography for influenza A and B antigens in nasopharyngeal aspirates; finally, GRAM and sputum cultures with direct seeding were also processed. • MR-proADM was determined in plasma through a sandwich immunoassay using time-re- solved amplified cryptate emission technology (TRACE) in a Kryptor, BRAHMS AG. Refer- ence values were P95: 0.52 nmol/L, median: 0.39 nmol/L. Detection limit: 0.05 nmol/L. Reagents were supplied by ThermoScientific (BRAHMS Iberia S.L.). • Procalcitonin (PCT) and N-terminal pro b-type natriuretic peptide (NT-proBNP) were mea- sured in plasma by electrochemiluminescence immunoassays on a COBAS 8000 modular an- alyzer (Roche Diagnostics, Mannheim, Germany). PCT reference values were < 0.05 μg/L with a detection limit < 0.02 ng/mL. NT-proBNP reference values were < 300 ng/L with a detection limit < 5 pg/mL. • C-reactive protein (CRP) was measured in serum or plasma by immunoturbidimetry with latex particles on a COBAS 8000 modular analyzer (Roche Diagnostics, Mannheim, Ger- many). Reference values were < 0.5 mg/dL with a detection limit < 0.03 mg/dL. • Other variables related to CAP worsening such as PSI, calculated according to Fine et al. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Statistical analysis Categorical variables were expressed as percentages and continuous variables as medians with interquartile ranges. The relationship between MR-proADM and risk categories of PSI was as- sessed through the Jonckheere-Terpstra trend test. The correlation coefficient was calculated with the Kendall’s tau-b. The association between continuous independent variables (analytical markers) and the main dependent variable was tested by the Student t test or the non-parametric Mann-Whitney U test when appropriate. For dichotomous categorical variables, the association was assessed using the [χ2] test or the Fisher exact test when categories had a size < 5. The association be- tween independent variables and mortality at 90 days was analysed with a Cox regression sur- vival analysis for independent continuous variables, and a Kaplan-Meyer survival curve with log-rank tests for independent categorical variables. The predictive ability of continuous independent variables (those not included in the PSI that showed statistically significant probabilities in the univariate analysis) for either an adverse event or mortality at 90 days was evaluated through ROC (receiver operating characteristic) curves. The area under the curve (AUC) and its confidence interval was compared with that obtained with the PSI. Multivariate analyses to check for the best predictive model were done with binary logistic regression (with all possible equations) for adverse event, and with Cox re- gression for mortality at 90 days. The analyses included the PSI and those statistically signifi- cant independent variables not included in the PSI with an AUC > 0.70. Indices of goodness of fit were calculated choosing the best models according to the lowest AIC (Akaike Information Criterion). Odds Ratios (ORs) for binary logistic regression, hazard ratios (HRs) for Cox re- gression and 95% confidence intervals (CI95%) were also calculated. Models were calibrated with the Hosmer-Lemeshow test. The predictive ability of MR-proADM combined with PSI was studied by means of a logistic regression adjust model. The obtained probabilities were analysed with ROC curves and com- pared with the predictive ability of MR-proADM and PSI models alone. A net reclassification index (NRI) was calculated to check whether the combined model of MR-proADM and PSI improved patients’ risk classification as compared with the model with PSI alone [25]. The risk limits for reclassification were assessed using the ones proposed by Schuetz et al.[19] that con- sider as low risk the probability of an advent < 5%, high risk > 20%, and two more categories for intermediate risk. Outcomes and measurements [14], presence of bacteremia (significant isolation of a non-contaminant microorganism in hemocultures) or hospital admission on a common ground were also recorded. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 4 / 15 Proadrenomedullin in Community Pneumonia Statistical analysis For those cases with an adverse event, it was considered an improvement of risk category when individuals moved to a higher category when MR-proADM was added, and a worsening of risk category when individuals moved to a lower category. The inverse was considered for cases without adverse event. For mortality at 90 days, the risk limits for reclassi- fication proposed by Courtais et al. [26] were used. A p value of 0.05 was considered statistically significant. According to previous studies, it was estimated that, to predict mortality, a specificity of 80% could be achieved for an optimal cut-off of MR-proADM. With a confidence level of 95%, 246 patients would be needed to esti- mate this specificity with a precision of 5%. Data were introduced in Microsoft Office Access 2007 using a form sheet with restriction for anomalous data. Data were analysed using STATA, version 12 for Windows (StataCorp). Results During the study period, a total of 60,021 medical emergencies (excluding pediatric and gyne- cological emergencies) were registered at the ED of the Gregorio Marañón University Hospital. This means an average of 398 emergencies per day. Five hundred and thirty-two patients were diagnosed with pneumonia (symptoms or signs associated with a new radiological infiltrate). 5 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Proadrenomedullin in Community Pneumonia Of these, 269 patients were diagnosed with CAP in the ED and did not meet any exclusion cri- teria, in 35 patients the diagnosis was not confirmed at discharge, and 8 patients were lost dur- ing the 3 months of follow-up. Thus, a total of 226 patients were included in the NACURG cohort analysed in this study (Fig 1). Median age (IQR) was 75.6 (27.8) years. Of all patients, 55.3% were male, 39.4% had PSI IV or V, 1.8% were admitted to the ICU, and 11.8% were readmitted during the first month after discharge. Mortality was 1.8% and 4.4% at 30 and 90 days, respectively. Microbiological isola- tion was achieved in 32 (17.5%) hospitalised patients. Streptococcus pneumoniae was the most frequently isolated microorganism (68.8%). Main baseline characteristics of patients are re- corded in Table 1. We found a significant increasing trend between MR-proADM levels and severity of CAP as assessed by PSI (Fig 2a). Likewise, MR-proADM levels were significantly higher in hospital- ised patients (median [IQR]: 1.21[0.71] vs 0.58[0.38], p = 0.00), patients with bacteremia (2.42 [0.93] vs 1.11[0.75], p = 0.00), patients with early readmission after discharge (1.39[1.23] vs 1.05[0.77], p = 0.00) and patients with early mortality (4.3[6.21] vs 1.07[0.8], p = 0.01). MR- proADM was predictor of short-term mortality with values similar to PSI (AUC 0.90[0.72– 1.00] vs AUC 0.90[0.80–1.00], p = 0.91), respectively. A cut-off point for ProADM of 1.5 nmol/ L had a sensitivity of 75%, specificity of 73%, likelihood ratio for positive test (LR+) of 2.73 and likelihood ratio for negative test (LR-) of 0.34 for predicting 30-day mortality. In patients ad- mitted to the ICU, MR-proADM levels were higher (2.21 [2.67] vs 1.07[0.81]) but failed to reach statistical significance (Fig 2b). Main outcome: MR-proADM levels and adverse event Thirty-three patients, 14.6% of the cohort, showed at least one adverse event (ICU admission, readmission after discharge, or death at 30 days of diagnosis). Independent variables signifi- cantly related to the combined variable adverse event and their statistical significances are re- corded in Table 1. When a ROC analysis to predict the probability of an adverse event was conducted, sensitiv- ity was assessed by means of those patients that showed any adverse event during the follow-up (n = 33), and specificity by those patients that failed to show any adverse event at the 30 days after follow-up. All the statistically significant independent variables were added to the analysis except lactate that was available only in 122 patients. Although the highest AUC was found for PSI (0.74 [0.64–0.85]), this was not significantly higher than the AUC obtained for MR- proADM, NT-proBNP and PCT (p > 0.05) (Fig 3a). The optimal cut-off point for MR- proADM for predicting adverse event was 0.85 nmol/L, with 97% sensitivity, 36% specificity, likelihood ratio for positive test (LR+) 1.5 and likelihood ratio for negative test (LR-) 0.08. A cut-off point of 1.3 nmol/L had a sensitivity and a specificity of 64 and 65%, respectively. Table 2 shows the different predictive models of adverse event together with several parame- ters of goodness to fit and calibration. The combination of PSI and MR-proADM was the well-calibrated predictive model with a lower AIC, as well as the one that best explained data uncertainty (McFadden´s R2 = 0.17), but failed to improve the predictive potential of PSI alone (AUC 0.75 [0.65–0.85, p = 0.56]) (Fig 3b). Results of reclassification tables are shown in Table 3. Ten patients were correctly classified when the combined PSI and MR-proADM model was used as compared with the model of PSI alone. Net reclassification improvement (NRI) index was statistically significant (7.69%, p = 0.03) with an improvement percentage of 3.03% (p = 0.32) for adverse event, and 4.66% (P = 0.02) for no adverse-event. Results of reclassification tables are shown in Table 3. Ten patients were correctly classified when the combined PSI and MR-proADM model was used as compared with the model of PSI alone. Net reclassification improvement (NRI) index was statistically significant (7.69%, p ( ) y g ( p = 0.03) with an improvement percentage of 3.03% (p = 0.32) for adverse event, and 4.66% (P = 0.02) for no adverse-event. Main outcome: MR-proADM levels and adverse event 6 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Proadrenomedullin in Community Pneumonia Table 1. Baseline characteristics of the NACURG cohort. Relationship between different independent variables and adverse event and 90-day mortality after consulting the Emergency Department. CHARACTERISTICS TOTALS ADVERSE EVENT p* 90-DAY MORTALITY p* WITH WITHOUT WITH WITHOUT Total cohort, count (%) 226 (100%) 33(14.6) 193(85.4) 10(4.4) 216(95.6) Age (years), median (IQR) 75.6(27.8) 80.9(13.9) 74.6(29.1) 0.03 87.1(5.5) 74.2(28.0) 0.00 Male, n (%) 125(55.3) 21(63.6) 104(53.9) 0.3 9(90) 116(53.7) 0.02 Charlson Index 3, n (%) 49(21.7) 14(42.4) 35(18.1) 0.00 6(60) 43(19.9) 0.00 Prior antibiotic treatment, n (%) 60(26.7) 8(24.2) 52(26.9) 0.73 1(10) 59(27.4) 0.23 Clinical sings Confusion, n° (%) 17(7.5) 8(24.2) 9(4.7) 0.00 3(30) 14(6.5) 0.00 Respiratory rate>30 bpm, n (%) 10(4.4) 4(12.1) 6(3.1) 0.02 3(30) 7(3.2) 0.00 Radiological ﬁndings Extension, n° (%) 0.16 0.20 Unilobar 173(76.6) 22(66.7) 151(78.2) 6(60) 167(77.3) Multilobar 53(23.4) 11(33.3) 42(21.8) 4(40) 49(22.7) Effusion on chest X-ray, n (%) 20(8.9) 6(18.2) 14(7.3) 0.04 2(20) 18(8.3) 0.21 Analytical variables Urea(mg/dl), median (IQR) 38(28) 47(43) 36(26) 0.01 69(94) 36(27) 0.00 Sodium (mmol/L), median (IQR) 137(5) 136.5(9.5) 137(5) 0.40 137(7) 137(5) 0.11 Arterial O2 pressure (mmHg), median (IQR) 62(14) 61(16) 62(13) 0.83 58(21) 62(14) 0.85 pH, median (IQR) 7.44(0.07) 7.40(0.15) 7.44(0.07) 0.03 7.34(0.15) 7.44(0.07) 0.01 Lactate** (mmol/L), median (IQR) 1.5(1)* 2(1.6) 1.4(0.9) 0.01 1.7(3) 1.5(1) 0.00 Leucocytes (cells/microL), median (IQR) 12200(8000) 14400(8000) 11900(7700) 0.08 15050(6800) 12100(7800) 0.29 CRP(mg/dl), median (P25-P75) 9.15(15.7) 9.1(14) 9.4(16.1) 0.77 6.75(18.9) 9.25(15.2) 0.66 MR-proADM (nmol/L), median (IQR) 1.08(0.8) 1.56(1.37) 1.05(0.77) 0.00 3.15(2.47) 1.06(0.79) 0.00 PCT (μg/L), median (IQR) 0.13(0.47) 0.22(1.43) 0.12(0.37) 0.02 0.37(2.36) 0.13(0.43) 0.54 NT-proBNP(ng/L), median (IQR) 510(1518) 1621(3231) 384(1236) 0.00 2572(6130) 460(1427) 0.03 CAP severity PSI, median (IQR) 83.5(49) 81(54–100) 122(89–148) 0.00 145(45) 82(47.5) 0.00 PSI, n (%) 0.00 0.00 I,II,III(Mild) 137(60.6) 10(30.3) 127(65.8) 0(0) 137(63.4) VI(Moderate) 58(25.7) 9(27.3) 49(25.4) 3(30) 55(25.5) V(Severe) 31(13.7) 14(42.4) 17(8.8) 7(70) 24(11.1) Bacteremia, n (%) 6(2.7) 2(6.1) 4(2.1) 0.21 0(0) 6(2.8) 0.61 Hospital admission, n (%) 187(81) 28(84.4) 155(80.3) 0.53 10 (100) 0(0) 0.12 Outcome ICU admission, n (%) 4(1,8) 30-day readmission, n (%) 26(11,8)* 30-day mortality, n (%) 4(1,8) Differences between patients who died and those who survived were assessed by Cox regression survival analysis for independent continuous variables, CURG cohort. Relationship between different independent variables and adverse event and 90-day mortality Table 1. Baseline characteristics of the NACURG cohort. Relationship between different independent variables after consulting the Emergency Department. ns out of the total number of patients discharged (221; 4 patients died while in hospital and 1 was still inpatient at Differences between patients who died and those who survived were assessed by Cox regression survival analysis for independent continuous variables, and a Kaplan-Meyer survival curve with log-rank tests for independent categorical variables. Differences between patients with or without adverse event were assessed by the Student t test or the non-parametric Mann-Whitney U test for continuous variables and the [χ2] test or the Fisher exact test for dichotomous categorical variables. p g g Lactate levels only available for 122 patients (54%) and not therefore included in the multivariate analysis. ce. r 122 patients (54%) and not therefore included in the multivariate analysis. Differences between patients who died and those who survived were assessed by Cox regression survival analysis for independent continuous variables, and a Kaplan-Meyer survival curve with log-rank tests for independent categorical variables. Differences between patients with or without adverse event were assessed by the Student t test or the non-parametric Mann-Whitney U test for continuous variables and the [χ2] test or the Fisher exact test for dichotomous categorical variables. ***The percentage of readmissions out of the total number of patients discharged (221; 4 patients died while in hospital and 1 was still inpatient at 30 days) doi:10.1371/journal.pone.0125212.t001 7 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Proadrenomedullin in Community Pneumonia Fig 2. MR-proADM and CAP severity. Fig 2a. Relationship between MR-proADM and severity as established by the PSI. Analysis performed with the Jonckheere-Terpstra trend test. Tau b: Kendall’s rank correlation. p: level of statistical significance. Fig 2b. MR-proADM levels according to hospital admission, bacteremia, ICU admission, hospital readmission and 30-day mortality. doi:10.1371/journal.pone.0125212.g002 Fig 2. MR-proADM and CAP severity. Fig 2a. Relationship between MR-proADM and severity as established by the PSI. Analysis performed with the Jonckheere-Terpstra trend test. Tau b: Kendall’s rank correlation. p: level of statistical significance. Fig 2b. MR-proADM levels according to hospital admission, bacteremia, ICU admission, hospital readmission and 30-day mortality. doi:10.1371/journal.pone.0125212.g002 doi:10.1371/journal.pone.0125212.g002 Secondary outcome: MR-proADM levels and mid-term mortality Proadrenomedullin in Community Pneumonia Table 2. Multivariate predictive models of adverse event and 90-day mortality. ADVERSE EVENT OR CI 95% p value AIC McFadden´s R2 Calibration χ2 (p value) Model 1 (MaxM) 0.00 170.2 0.14 240.9(0.17) PSI 1.02 1.01–1.03 MR-proADM 1.16 0.77–1.75 NT-proBNP 1.00 0.99–1.00 Intercept 0.02 0.01–0.07 Model 2 0.00 167.4 0.12 172.6(0.00) PSI 1.02 1.01–1.03 Intercept 0.02 0.01–0.06 Model 3 0.00 168.5 0.17 240.1(0.17) PSI 1.02 1.01–1.03 MR-proADM 1.16 0.79–1.72 Intercept 0.02 0.01–0.07 90-DAY MORTALITY HR CI 95% p value AIC Atkinson R2 Test of proportional-hazards assumption χ2(p value) Model 1 (MaxM) 0.00 84.3 0.22 2.56(0.47) PSI 1.00 0.97–1.02 MR-proADM 3.14 1.25–7.86 NT-proBNP 1.00 0.99–1.01 Model 2 0.00 80.4 0.26 0.34(0.56) MR-proADM 2.70 1.79–4.05 Model 3 0.00 82.3 0.24 2.67(0.26) PSI 1.00 0.97–1.02 MR-proADM 2.93 1.23–6.97 Model 4 0.00 94.0 0.13 0.07(0.80) PSI 1.03 1.01–1.03 MaxM: Maximum Model: includes signiﬁcant independent variables in the univariate analysis with AUC higher than 0.7. OR: Odds Ratio and HR: Hazard Ratio. CI 95%: conﬁdence interval of 95%. Table 2. Multivariate predictive models of adverse event and 90-day mortality. MaxM: Maximum Model: includes signiﬁcant independent variables in the univariate analysis with AUC higher than 0.7. OR: Odds Ratio and HR: Hazard Ratio. AIC: Akaike Information Criterion (better ﬁt of the model when AIC lower). McFadden´s and Atkinson R2: proportion of uncertainty data explained by the model. Calibration χ2 (p value): Hosmer and Lemeshow test. MR-proADM was the analytical marker not included in the PSI with the highest AUC to predict mortality at 90 days (0.88 [0.79–0.98], showing an AUC similar to that obtained with the PSI (p = 0.90). ROC curve analysis to predict mortality at 90 days is depicted in (Fig 5a). A cut-off point for MR-proADM of 1.5 nmol/L had a sensitivity of 80%, specificity of 74%, likeli- hood ratio for positive test (LR+) of 3.09 and likelihood ratio for negative test (LR-) of 0.27 for predicting mid-term mortality. Both the MR-proADM alone and the combined PSI plus MR- proADM were the predictive models that showed the best adjustment (Table 2), but they did not improve the predictive ability of PSI alone (AUC 0.89 [0.80–0.98], p = 0.95) (Fig 5b). The combined model failed to significantly improve classification of patients to predict mid- term risk mortality, NRI 5.74% (p = 0.83). The improvement percentage was -10% (p = 0.71) for event, and 15.74% (p = 0.00) for no event. Secondary outcome: MR-proADM levels and mid-term mortality Ten patients of the NACURG cohort died within 90 days after diagnosis. Variables associated with mid-term mortality are recorded in Table 1. MR-proADM levels were associated with mortality at 90 days (OR 2.70, 1.79–4.05) (Table 2). Significant differences (p < 0.01) were found among the survival curves according to MR-proADM quartiles (Fig 4). Ten patients of the NACURG cohort died within 90 days after diagnosis. Variables associated with mid-term mortality are recorded in Table 1. MR-proADM levels were associated with mortality at 90 days (OR 2.70, 1.79–4.05) (Table 2). Significant differences (p < 0.01) were found among the survival curves according to MR-proADM quartiles (Fig 4). Fig 3. ROC curves for predicting adverse event. Fig 3a. ROC curves for different biomarkers and PSI. Fig 3b. ROC curves for the PSI & MR-proADM prediction model compared to PSI verse event. Fig 3a. ROC curves for different biomarkers and PSI. Fig 3b. ROC curves for the PSI & MR-proADM s for predicting adverse event. Fig 3a. ROC curves for different biomarkers and PSI. Fig 3b. ROC curves for the PS ompared to PSI Fig 3. ROC curves for predicting adverse event. Fig 3a. ROC curves for different biomarkers and PSI. Fig 3b prediction model compared to PSI Fig 3. ROC curves for predicting adverse event. Fig 3a. ROC curves for different biomarkers and PSI. Fig 3b. ROC curves for the PSI & MR-proADM prediction model compared to PSI d i 10 1371/j l 0125212 003 doi:10.1371/journal.pone.0125212.g003 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 8 / 15 doi:10.1371/journal.pone.0125212.t002 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Proadrenomedullin in Community Pneumonia Table 3. Reclassification table for adverse event in PSI plus MR-proADM compared to PSI alone. Total cohort, N = 226 Adverse Event PSI+MR-proADM Reclassiﬁed PSI <5% 5–10% 10–20% >20% Total Increased risk Decreased risk NET correctly reclassiﬁed Patients with adverse event (n = 33) <5% 2 1 0 0 3 5–10% 0 3 0 0 3 10–20% 0 0 10 0 10 1(3.03%) 0(0%) 1(3.03%) >20% 0 0 0 17 17 TOTAL 2 4 10 17 33 Patients without adverse event (n = 193) <5% 29 3 0 0 32 5–10% 0 59 0 0 59 10–20% 0 5 67 0 72 3(1.55%) 12(6.22%) 9(4.66%) >20% 0 0 7 23 30 TOTAL 29 67 74 23 193 Total 4 12 10 doi:10.1371/journal.pone.0125212.t003 Table 3. Reclassification table for adverse event in PSI plus MR-proADM compared to PSI alone. doi:10.1371/journal.pone.0125212.t003 These and other supplementary results are depicted in (S1 File). 9 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Discussion We found that MR-proADM, measured in the ED, is useful for predicting poor clinical out- come of CAP patients, showing a similar predictive ability than that found for PSI. Even though MR-proADM combined with PSI did not improve prognostic accuracy, it allowed a better reclassification of patients, mostly at the expense of a better classification of patients without adverse event. Adrenomedullin (ADM) is a peptide produced by multiple tissues under stress conditions with vasodilatory, immunomodulatory, metabolic and bactericide attributed properties. MR- Fig 4. Kaplan-Meier survival curves at 90-days mortality according to MR-proADM quartiles. doi:10.1371/journal.pone.0125212.g004 10 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Proadrenomedullin in Community Pneumonia Fig 5. ROC curves for predicting 90-day mortality. Fig 5a. ROC curves for different biomarkers and PSI. Fig 5b. ROC curves for the PSI & MR-proADM prediction model compared to PSI. Fig 5. ROC curves for predicting 90-day mortality. Fig 5a. ROC curves for different biomarkers and PSI. Fig 5b. ROC curves for the PSI & MR-proADM prediction model compared to PSI. -day mortality. Fig 5a. ROC curves for different biomarkers and PSI. Fig 5b. ROC curves for the PSI & MR-proADM Fig 5. ROC curves for predicting 90-day mortality. Fig 5a. ROC curves for different biomarkers and PSI. Fig 5b. R prediction model compared to PSI. Fig 5. ROC curves for predicting 90-day mortality. Fig 5a. ROC curves for different biomarkers and PSI. Fig 5b. ROC curves for the PSI & MR-proADM prediction model compared to PSI. doi:10.1371/journal.pone.0125212.g005 doi:10.1371/journal.pone.0125212.g005 doi:10.1371/journal.pone.0125212.g005 proADM is a more stable and easy-to-measure fragment that directly reflects ADM activity lev- els [27, 28]. Since 2006, several research groups have studied the prognostic role of MR- proADM in CAP patients finding a strong predictive potential of this peptide for short-term mortality [19, 21–23, 26, 29, 30]. Nonetheless, the decisions made in the ED cannot rely solely on the risk of mortality, especially in a sample of low mortality as ours which excludes cases with worse prognosis (Fig 1). It is also useful to know which patients are at increased risk of ICU admission or at early readmission after discharge. Certainly, an increased surveillance of these patients may improve survival. Half of the patients at low risk of mortality according to PSI are hospitalised [31]. The com- bined method of PSI plus MR-proADM increases the number of patients without complica- tions that are classified as at “low risk”. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 [21]. We would like to underline that, although NT-proBNP is also a predictor of poor outcome of CAP and mid-term mortality, it is not better than MR-proADM or PSI. We cannot discard that the substitution of the variable "cardiac insufficiency" included in the PSI by NT-proBNP levels could improve the predictive potential of this score, but this analysis was not included among the objectives of this study. We would like to underline that, although NT-proBNP is also a predictor of poor outcome of CAP and mid-term mortality, it is not better than MR-proADM or PSI. We cannot discard that the substitution of the variable "cardiac insufficiency" included in the PSI by NT-proBNP levels could improve the predictive potential of this score, but this analysis was not included among the objectives of this study. Several limitations of our study should be taken into consideration. Firstly, this is a single- center study, so that its results can be conditioned by both our patients’ profile and the way we work in our hospital and therefore be less generalisable. The endpoint ICU admission is vulner- able to bias by ICU admission policy and highly dependent on individual physician decisions. Moreover, “do not intubate” patients were not excluded. Our endpoint adverse event was mainly driven by a quite high readmission rate; however, ICU admission and 30-day mortality rates were low. Thus, generalisability of the findings to other health care systems is reduced. In addition, the analyses performed to predict short and mid-term mortality should be interpreted with caution because of the low number of adverse events recorded. We have not compared MR-proADM in relation to lactate levels in their prognostic ability because lactate level mea- surements were not available in more than 100 patients. Although the physician responsible for taking care of the patient did not know the results of MR-proADM, he/she did know that the patient could be included in the study, being able to produce an inclusion bias. Finally, even though the microbiological investigation of hospitalised patients was protocolised, it was not systematic, a fact that can explain the ethiological underdiagnosis. even though the microbiological investigation of hospitalised patients was protocolised, it was not systematic, a fact that can explain the ethiological underdiagnosis. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Proadrenomedullin in Community Pneumonia As in the studies mentioned above, we found that MR-proADM is a good predictor of poor outcome of CAP. However, in our cohort we have not included patients a priori not eligible for hospital discharge. These patients were those needing specific antibiotic therapy (immunosup- pressed or institutionalized) or those associated with worse CAP evolution such as patients with non-invasive home mechanical ventilation. Therefore, our cohort has a higher representa- tion of patients at lower risk of mortality, the ones that most benefit from the determination of MR-proADM to decide whether they can receive outpatient treatment. These specific features of our cohort could be the reason for differences in optimal cut-off points compared to other studies. A single clinical trial studied the feasibility and effects of adding MR-proADM to the CURB-65 score on triage decisions and length of stay [34]. There were no differences in ad- verse outcome or readmission but important logistic obstacles were found in this study. More- over, CURB-65 score does not account for patients’ comorbidity, which may influence the CAP outcome. Our study hints at the possibility of designing clinical trials that evaluate the ability of new biomarkers, namely MR-proADM, to classify patients’ risks also accounting for additional clinically relevant endpoints beyond mortality, such as our study points to (i.e., ICU admission or short-term hospital readmission). MR-proADM could better inform clinical- making decisions in patients showing intermediate risk who are currently being admitted, or alternatively being discharged due to low specific mortality risk, albeit with a higher probability of short-term readmission. We found that MR-proADM levels predict mid-term mortality as in previous studies [21, 29] but the combined model of PSI plus MR-proADM does not improve either the predictive potential or the reclassification of patients according to mid-term mortality risk. Our results are similar to that obtained by Huang et al.[29] that analysed a cohort with 60% of CAP pa- tients with low risk (similar to our cohort) but differ from the ones reported by Bello et al. [21]. 29] but the combined model of PSI plus MR-proADM does not improve either the predictive potential or the reclassification of patients according to mid-term mortality risk. Our results are similar to that obtained by Huang et al.[29] that analysed a cohort with 60% of CAP pa- tients with low risk (similar to our cohort) but differ from the ones reported by Bello et al. Discussion This can be useful in the clinical practice to decrease the number of unnecessary hospitalisations, avoiding nosocomial infections and reducing health costs [32]. Other studies had assessed the influence of MR-proADM in combined variables such as ours, although none had included early hospital readmission as a variable to predict poor out- come of CAP. Christ-Crain et al. [23] found that MR-proADM has a predictive ability similar to that of PSI, and that PSI combined with MR-proADM improves the predictive power of therapeutic failure defined as persistence or recurrence of clinical, analytical or radiologic alter- ations, or mortality during the follow-up (6.9 ± 1.9 weeks). Schuetz et al. [19]analysed the Pro- HOSP cohort [33] and concluded that the combined model of PSI improves the predictive potential of adverse event (empyema, abscess or acute respiratory distress syndrome, ICU ad- mission or mortality at 30 days) as compared with PSI alone. Bello et al. [21] found that MR- proADM is a predictor of CAP complications (respiratory, renal and cardiac insufficiency among others) similar to PSI, but they did not examine the predictive power of the combined model. Other authors who also analysed the ProHOSP cohort concluded that MR-proADM plus CURB65 improves the predictive ability of adverse event with regard to CURB65 alone [20]. Additionally they found that combining MR-proADM with the REA-ICU score (Risk of Early Admission of ICU) improved the ability of predicting early severe community-acquired pneumonia (need of mechanical ventilation, vasoactive drugs or mortality at 30 days) in com- parison with the REA-ICU score alone[31]. However, these studies did not analyse the prog- nostic usefulness of PSI. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 11 / 15 Acknowledgments Susana Gordo-Remartínez is the guarantor of the content of the manuscript, including the data and analysis (Original Research). María Calderón-Moreno, Juan Fernández-Herranz, Ana Cas- tuera-Gil, Mar Gallego-Alonso-Colmenares, Carolina Puertas-López, José A. Nuevo-González, Domingo Sánchez-Sendín and Mercedes García-Gámiz 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. José A. Sevillano-Fernández, Luis A. Álvarez-Sala, Juan A. Andueza-Lillo and José M. de Mi- guel-Yanes contributed substantially to the study design, data analysis and interpretation, and the writing of the manuscript. Author Contributions Conceived and designed the experiments: SG MC JAN JAS JAA JMM. Performed the experi- ments: SG MC JF AC MGA CP JAN DS MG. Analyzed the data: SG JAS LAA JAA JMM. Con- tributed reagents/materials/analysis tools: SG MC JF AC MGA CP JAN DS MG JAS JAA. Wrote the paper: SG JAN JAS LAA JAA JMM tributed reagents/materials/analysis tools: SG MC JF AC MGA CP JAN DS MG JAS JAA. Wrote the paper: SG JAN JAS LAA JAA JMM. Wrote the paper: SG JAN JAS LAA JAA JMM. In conclusion, as compared with PSI alone, MR-proADM in combination with PSI may be helpful in individual risk stratification for short-term poor outcome of CAP patients, allowing a better reclassification of patients. Although MR-proADM is a good predictor of mid-term 12 / 15 PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 Proadrenomedullin in Community Pneumonia mortality, similar to PSI, the combined model does not improve the predictive ability, and does not allow a more accurate reclassification of patients according to their mortality risk. mortality, similar to PSI, the combined model does not improve the predictive ability, and does not allow a more accurate reclassification of patients according to their mortality risk. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 References 2004; 32(4):234–8. doi: 10.1007/s15010-004-3107-z PMID: 15293080. 11. Menéndez R T A. Aspa J Capelastegui A Prat C. Rodriguez de Castro F. Neumonía adquirida en la comunidad. Nueva normativa de la Sociedad Española de Neumología y Cirugía Torácica (SEPAR). Archivos de bronconeumologia. 2010; 46(10):543–58. PubMed Central PMCID: PMC20832928 doi: 10.1016/j.arbres.2010.06.014 PMID: 20832928 12. Lim WS, Baudouin SV, George RC, Hill AT, Jamieson C, Le Jeune I, et al. BTS guidelines for the man- agement of community acquired pneumonia in adults: update 2009. Thorax. 2009; 64 Suppl 3:iii1–55. doi: 10.1136/thx.2009.121434 PMID: 19783532. 13. Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, et al. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of communi- ty-acquired pneumonia in adults. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2007; 44 Suppl 2:S27–72. doi: 10.1086/511159 PMID: 17278083. 14. Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, Singer DE, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. The New England journal of medicine. 1997; 336(4):243–50. doi: 10.1056/NEJM199701233360402 PMID: 8995086. 15. Aujesky D, Fine MJ. The pneumonia severity index: a decade after the initial derivation and validation. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2008; 47 Suppl 3:S133–9. doi: 10.1086/591394 PMID: 18986279. 16. Aujesky D, McCausland JB, Whittle J, Obrosky DS, Yealy DM, Fine MJ. Reasons why emergency de- partment providers do not rely on the pneumonia severity index to determine the initial site of treatment for patients with pneumonia. Clinical infectious diseases: an official publication of the Infectious Dis- eases Society of America. 2009; 49(10):e100–8. doi: 10.1086/644741 PMID: 19842971. 17. Seligman R, Ramos-Lima LF, Oliveira Vdo A, Sanvicente C, Pacheco EF, Dalla Rosa K. Biomarkers in community-acquired pneumonia: a state-of-the-art review. Clinics. 2012; 67(11):1321–5. PMID: 23184211; PubMed Central PMCID: PMC3488993. 18. Renaud B, Schuetz P, Claessens YE, Labarere J, Albrich W, Mueller B. Proadrenomedullin improves Risk of Early Admission to ICU score for predicting early severe community-acquired pneumonia. Chest. 2012; 142(6):1447–54. doi: 10.1378/chest.11-2574 PMID: 22661450. 19. Schuetz P, Wolbers M, Christ-Crain M, Thomann R, Falconnier C, Widmer I, et al. Prohormones for pre- diction of adverse medical outcome in community-acquired pneumonia and lower respiratory tract in- fections. Critical care. 2010; 14(3):R106. doi: 10.1186/cc9055 PMID: 20529344; PubMed Central PMCID: PMC2911752. 20. References 1. Bartlett JG, Dowell SF, Mandell LA, File TM Jr, Musher DM, Fine MJ. Practice guidelines for the man- agement of community-acquired pneumonia in adults. Infectious Diseases Society of America. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. 2000; 31 (2):347–82. doi: 10.1086/313954 PMID: 10987697. 2. Gutiérrez Macías A, Martínez Ortiz de Zárate M. Infecciones en urgencias. Nuevos retos para el siglo XXI. Emergencias. 2000; 12:77–8. 3. Mateos Rodríguez A. Grupo para el Estudio de la Infección en Urgencias. Estudio epidemiológico de las infecciones en el Área de Urgencias. Emergencias. 2000; 12:80–9. 4. Almirall J, Bolibar I, Vidal J, Sauca G, Coll P, Niklasson B, et al. Epidemiology of community-acquired pneumonia in adults: a population-based study. The European respiratory journal. 2000; 15(4):757–63. PMID: 10780770. 5. Chalmers JD, Singanayagam A, Akram AR, Mandal P, Short PM, Choudhury G, et al. Severity assess- ment tools for predicting mortality in hospitalised patients with community-acquired pneumonia. Sys- tematic review and meta-analysis. Thorax. 2010; 65(10):878–83. doi: 10.1136/thx.2009.133280 PMID: 20729231. 6. Ewig S, Birkner N, Strauss R, Schaefer E, Pauletzki J, Bischoff H, et al. New perspectives on communi- ty-acquired pneumonia in 388 406 patients. Results from a nationwide mandatory performance mea- surement programme in healthcare quality. Thorax. 2009; 64(12):1062–9. doi: 10.1136/thx.2008. 109785 PMID: 19454409; PubMed Central PMCID: PMC2782114. 7. Ewig S, Torres A. Community-acquired pneumonia as an emergency: time for an aggressive interven- tion to lower mortality. The European respiratory journal. 2011; 38(2):253–60. doi: 10.1183/09031936. 00199810 PMID: 21349916. 8. Mortensen EM, Coley CM, Singer DE, Marrie TJ, Obrosky DS, Kapoor WN, et al. Causes of death for patients with community-acquired pneumonia: results from the Pneumonia Patient Outcomes Re- search Team cohort study. Archives of internal medicine. 2002; 162(9):1059–64. PMID: 11996618. 9. Niederman MS, Mandell LA, Anzueto A, Bass JB, Broughton WA, Campbell GD, et al. Guidelines for the management of adults with community-acquired pneumonia. Diagnosis, assessment of severity, PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 13 / 15 Proadrenomedullin in Community Pneumonia antimicrobial therapy, and prevention. American journal of respiratory and critical care medicine. 200 163(7):1730–54. doi: 10.1164/ajrccm.163.7.at1010 PMID: 11401897. antimicrobial therapy, and prevention. American journal of respiratory and critical care medicine. 2001; 163(7):1730–54. doi: 10.1164/ajrccm.163.7.at1010 PMID: 11401897. antimicrobial therapy, and prevention. American journal of respiratory and critical care medicine. 2001; 163(7):1730–54. doi: 10.1164/ajrccm.163.7.at1010 PMID: 11401897. 10. Welte T, Suttorp N, Marre R. CAPNETZ-community-acquired pneumonia competence network. Infec- tion. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 References Albrich WC, Dusemund F, Ruegger K, Christ-Crain M, Zimmerli W, Bregenzer T, et al. Enhancement of CURB65 score with proadrenomedullin (CURB65-A) for outcome prediction in lower respiratory tract in- fections: derivation of a clinical algorithm. BMC infectious diseases. 2011; 11:112. doi: 10.1186/1471- 2334-11-112 PMID: 21539743; PubMed Central PMCID: PMC3119069. 21. Bello S, Lasierra AB, Minchole E, Fandos S, Ruiz MA, Vera E, et al. Prognostic power of proadrenome- dullin in community-acquired pneumonia is independent of aetiology. The European respiratory journal. 2012; 39(5):1144–55. doi: 10.1183/09031936.00080411 PMID: 22075489. 22. Julian-Jimenez A, Timon Zapata J, Laserna Mendieta EJ, Sicilia-Bravo I, Palomo-de Los Reyes MJ, Cabezas-Martinez A, et al. Poder diagnóstico y pronóstico de los biomarcadores para mejorar el man- ejo de la neumonía adquirida en la comunidad en los servicios de urgencias. Enfermedades infeccio- sas y microbiologia clinica. 2014; 32(4):225–35. doi: 10.1016/j.eimc.2013.04.015 PMID: 24182623. 23. Christ-Crain M, Morgenthaler NG, Struck J, Harbarth S, Bergmann A, Muller B. Mid-regional pro-adre- nomedullin as a prognostic marker in sepsis: an observational study. Critical care. 2005; 9(6):R816–24. doi: 10.1186/cc3885 PMID: 16356231; PubMed Central PMCID: PMC1414007. 24. Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. Journal of chronic diseases. 1987; 40(5):373–83. PMID: 3558716. 25. Pencina MJ, D'Agostino RB, Vasan RS. Statistical methods for assessment of added usefulness of new biomarkers. Clinical chemistry and laboratory medicine: CCLM / FESCC. 2010; 48(12):1703–11. doi: 10.1515/CCLM.2010.340 PMID: 20716010; PubMed Central PMCID: PMC3155999. 26. Courtais C, Kuster N, Dupuy AM, Folschveiller M, Jreige R, Bargnoux AS, et al. Proadrenomedullin, a useful tool for risk stratification in high Pneumonia Severity Index score community acquired pneumo- nia. The American journal of emergency medicine. 2013; 31(1):215–21. doi: 10.1016/j.ajem.2012.07. 017 PMID: 23000327. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 14 / 15 Proadrenomedullin in Community Pneumonia 27. Becker KL, Nylen ES, White JC, Muller B, Snider RH Jr. Clinical review 167: Procalcitonin and the calci- tonin gene family of peptides in inflammation, infection, and sepsis: a journey from calcitonin back to its precursors. The Journal of clinical endocrinology and metabolism. 2004; 89(4):1512–25. doi: 10.1210/ jc.2002-021444 PMID: 15070906. 27. Becker KL, Nylen ES, White JC, Muller B, Snider RH Jr. Clinical review 167: Procalcitonin and the calci- tonin gene family of peptides in inflammation, infection, and sepsis: a journey from calcitonin back to its precursors. The Journal of clinical endocrinology and metabolism. 2004; 89(4):1512–25. PLOS ONE \| DOI:10.1371/journal.pone.0125212 June 1, 2015 References doi: 10.1210/ jc.2002-021444 PMID: 15070906. 28. Linscheid P, Seboek D, Zulewski H, Keller U, Muller B. Autocrine/paracrine role of inflammation-mediat- ed calcitonin gene-related peptide and adrenomedullin expression in human adipose tissue. Endocri- nology. 2005; 146(6):2699–708. doi: 10.1210/en.2004-1424 PMID: 15761041. 29. Huang DT, Angus DC, Kellum JA, Pugh NA, Weissfeld LA, Struck J, et al. Midregional proadrenome- dullin as a prognostic tool in community-acquired pneumonia. Chest. 2009; 136(3):823–31. doi: 10. 1378/chest.08-1981 PMID: 19363212; PubMed Central PMCID: PMC2818411. 30. Kruger S, Ewig S, Giersdorf S, Hartmann O, Suttorp N, Welte T, et al. Cardiovascular and inflammatory biomarkers to predict short- and long-term survival in community-acquired pneumonia: Results from the German Competence Network, CAPNETZ. American journal of respiratory and critical care medi- cine. 2010; 182(11):1426–34. doi: 10.1164/rccm.201003-0415OC PMID: 20639437. 31. Yealy DM, Auble TE, Stone RA, Lave JR, Meehan TP, Graff LG, et al. Effect of increasing the intensity of implementing pneumonia guidelines: a randomized, controlled trial. Annals of internal medicine. 2005; 143(12):881–94. PMID: 16365469. 32. Chalmers JD, Al-Khairalla M, Short PM, Fardon TC, Winter JH. Proposed changes to management of lower respiratory tract infections in response to the Clostridium difficile epidemic. The Journal of antimi- crobial chemotherapy. 2010; 65(4):608–18. doi: 10.1093/jac/dkq038 PMID: 20179023. 33. Schuetz P, Christ-Crain M, Thomann R, Falconnier C, Wolbers M, Widmer I, et al. Effect of procalcito- nin-based guidelines vs standard guidelines on antibiotic use in lower respiratory tract infections: the ProHOSP randomized controlled trial. JAMA: the journal of the American Medical Association. 2009; 302(10):1059–66. doi: 10.1001/jama.2009.1297 PMID: 19738090. 34. Albrich WC, Ruegger K, Dusemund F, Schuetz P, Arici B, Litke A, et al. Biomarker-enhanced triage in respiratory infections: a proof-of-concept feasibility trial. The European respiratory journal. 2013; 42 (4):1064–75. doi: 10.1183/09031936.00113612 PMID: 23349444; PubMed Central PMCID: PMC3787815. 15 / 15
https://openalex.org/W4226343672	https://zenodo.org/records/5824480/files/10.pdf	English	null	GIANT LEFT VENTRICULAR THROMBOSED ANEURYSM	Zenodo (CERN European Organization for Nuclear Research)	2,021	cc-by	1,537	GIANT LEFT VENTRICULAR THROMBOSED ANEURYSM Dr. Mesmoudi B.1, Dr. Hdioud O.1, Dr. Louizi W.1, Dr. Jennane R.1, Dr. Nguadi J.2, Pr. Doghmi N.1 and Pr. Cherti M.1 1. Department of Cardiology B, Maternity Souissi, Ibn Sina Hospital Center, Mohamed V University, Rabat, Morocco. 2. Department of Cardiology , Mohamed V MilitaryHospital, Ibn Sina Hospital Center, Mohamed V University, Rabat, Morocco. …………………………………………………………………………………………………….... Manuscript Info Abstract ……………………. ……………………………………………………………… Manuscript History Received: 10 October 2021 Final Accepted: 14 November 2021 Published: December 2021 Copy Right IJAR 2021 All rights reserved Leftventricular (LV)aneurysms are a frequent complication of acute extensive myocardialinfarction and are mostcommonlylocatedat the ventricular apex. A timelydiagnosisis vital due to the serious complications thatcanoccur, includingheartfailure, thromboembolism, or tachyarrhythmias.1 The benefits of surgicalrepair of leftventricularaneurysm have long been debated. ISSN: 2320-5407 ISSN: 2320-5407 Int. J. Adv. Res. 9(12), 329-332 Journal Homepage: -www.journalijar.com Article DOI:10.21474/IJAR01/13912 DOI URL: http://dx.doi.org/10.21474/IJAR01/13912 ISSN: 2320-5407 Int. J. Adv. Res. 9(12), 329-332 ISSN: 2320-5407 Given the uncertaintysurrounding the differentialdiagnosis, cardiacmagneticresonanceimaging (MRI) wasperformed, showing an aspect of ischemicheartdiseaseat the dilated stage withdisorders of segmental kinetics, a reducedsystolicfunction and an apical aneurysm of the LV measuring 63x75x71 mm and containing a thrombus. It revealed a non-viable necroticsequelae in the akineticterritory of the anteriorinterventricularartery (Figures 3). Figure 3 A Figure 3 B Figures3 A and B: MRI images of the hugethrombosedaneurysm of the apex, measuring 64 x 53 mm. Figure 3 A Figure 3 B Figures3 A and B: MRI images of the hugethrombosedaneurysm of the apex, measuring 64 x 53 mm The patient wasadmitted to the cardiac intensive care units. The heartfailureflarewascontrolled and the patient wasdischarged on medicaltreatment for ischemicheartdisease, diuretics, and anticoagulants.He wasreferred for cardiovascular and heartfailure consultationwith a close follow-up.The control transthoracicechocardiographyat one monthshowed a decrease in the size of the thrombus (from 64 x 53 mm to 64 x 38 mm), and in the leftventricularejection (from 38 to 35%). Case Report: Case Report: We report the case of a 61-year-old man with no cardiovascularriskfactors,presenting to the emergency departmentwith a stage III dyspneaassociatedwith an oedema of the lowerlimbs.During the investigation, welearnedthathepresented an acute chest pain one weekbefore, associatedwithsweating. On admission, hisblood pressure wasat 94/74 mmHg, and hisheart rate at 92 bpm.Physicalexaminationrevealedsigns of congestive heartfailurewithcoarsecracklesat the lung bases and lowerextremityedemaat the ankles.The electrocardiogramshowedaregular sinus rhythmat85 beats per minute and Q wavenecrosisassociatedwith an elevation of the ST segment in the extendedanteriorderivations(Figure1). Corresponding Author:- Dr. Mesmoudi B. Address:- Department of Cardiology B,MaternitySouissi, Ibn Sina Hospital Center, Mohamed V University, Rabat, Morocco. 329 Int. J. Adv. Res. 9(12), 329-332 ISSN: 2320-5407 Figure 1:- Electrocardiogramshowing Q wavenecrosisassociatedwith an elevation of the ST segment in the extendedanteriorderivations. Figure 1:- Electrocardiogramshowing Q wavenecrosisassociatedwith an elevation of the ST segment in the extendedanteriorderivations. Transthoracicechocardiographyrevealed a dilatedleftventriclewithakinesia of the apex and its adjacent segments, akinesia of the septal wall and the middle segments of the inferior and anteriorwallswitha hugethrombosedaneurysmof the apex, measuring64 x 53 mm, and a reducedsystolicleftventricularfunctionat38% withoutsignificant mitral regurgitation (Figure 2 ). Transthoracicechocardiographyrevealed a dilatedleftventriclewithakinesia of the apex and its adjacent segments, akinesia of the septal wall and the middle segments of the inferior and anteriorwallswitha hugethrombosedaneurysmof the apex, measuring64 x 53 mm, and a reducedsystolicleftventricularfunctionat38% withoutsignificant mitral regurgitation (Figure 2 ). Figure 2:- Transthoracicechocardiography imageshowing a dilatedleftventriclewith a hugethrombosedaneurysm of the apex, measuring 64 x 53 mm. Figure 2:- Transthoracicechocardiography imageshowing a dilatedleftventriclewith a hugethrombosedaneurysm of the apex, measuring 64 x 53 mm. 330 Discussion:- Leftventricular (LV)aneurysm formation is a complication of transmural myocardial infarction (MI)that leads to awell-delineatedoutwardbulging of the affected LV wall due to myocardialthinning and scar formation.2A leftventricularaneurysmisa segment of a leftventricularwallthatprotrudesfrom the expecteddiastolicoutlineof the ventricularchamber. The wall motion maybeeitherakinetic or dyskinetic. The wall of the aneurysmisgenerallysmoothwithout the usualtrabecular pattern.Leftventricularaneurysmsappear to develop early in the course of myocardialinfarction. In more than 50% of patients whodevelop an aneurysm, itwillbepresentwithin 24 to 48 hours of the onset of infarction and frequentlywillstillbepresent 3 monthslater. On imagingstudies, a large scar or a smallaneurysmmay have the sameappearance. A functionalaneurysm has a normal diastolic contour but a dyskineticsystolic bulge in the region of a large acontractile segment and maycontaineitherreversibleischemicmyocardium or scar. In contrast, an anatomicaneurysm has an abnormal protrusion duringboth systole and diastole because the wallisentirely a scarthat has stretched.3 Aneurysms more commonlycomplicate anterior thaninferiorinfarction, ofteninvolving the apex.2It typically affects the anteroapicalregion of the LV, because the bloodsupply of the anteriorwallishighlydependent on the anteriordescending artery4.Riskfactors for theseaneurysmsafter acute MI include the following: Femalesex, total occlusion of the leftanteriordescending (LAD) coronaryartery, single-vesselcoronaryarterydisease (CAD) and the 331 Int. J. Adv. Res. 9(12), 329-332 Int. J. Adv. Res. 9(12), 329-332 Int. J. Adv. Res. 9(12), 329-332 Int. J. Adv. Res. 9(12), 329-332 ISSN: 2320-5407 absence of previousanginapectoris.Clinically, ventricularaneurysmsmayberecognizedlate, withsymptoms and signs of heartfailure, recurrentventriculararrhythmias, or recurrent embolization.5 absence of previousanginapectoris.Clinically, ventricularaneurysmsmayberecognizedlate, withsymptoms and signs of heartfailure, recurrentventriculararrhythmias, or recurrent embolization.5 In the currentera ofreperfusiontherapy , a LV aneurysmispresent in approximately 10% to 15% of patients withMI. Mechanically, the aneurysmdoes not contribute to LV ejection, but ratheracts as a deadspace for accommodating more blood in the LV cavityduring systole and therebycompromising stroke volume.The presence of a poorlycontracting, dilated leftventricle promotesstasis of blood and leads to an increasedrisk of thrombus formation.Demonstrating thrombus within the aneurysmisclinically important becausesuch thrombi can lead to embolic complications. Results of studiesdifferwidely, but bothpostmortem and surgicalstudies have demonstratedthatapproximately 50% of aneurysmscontain a thrombus 6and resultsfrom the inflammatoryprocess in the endocardialregionaffected by the MI, beingassociatedwith the hypokinesia and hypercoagulabilityexisting in the infarction, increasing the risk of a thromboemboliceventafter the thirdmonth in patients withventricular aneurysm.4 An earlydiagnosisismandatory to avoid life-threatening complications such as :heartfailure, thromboembolism, or tachyarrhythmias. The benefits of surgicalrepair of leftventricularaneurysm have long been debated. Conclusion:- Ventricularaneurysmis a serious complication of transmuralmyocardialinfarction, being the mostcommonmechanical complication and having a negative impact on quality of life. In the treatment of severerefractory cases, surgeryisindicated, despiteitscontroversialbenefits. Discussion:- Although a large amount of studies have showedthataneurysmectomymightimprove the outcome7, the resultsfrom the STICH trial have questioned the benefit of thistreatment .8Therefore, indication for aneurysmectomydepends on the decision of individual surgeons, and shouldbebased on the assessment of the leftventricular dimensions, mitral valve regurgitationseverity, extent of myocardialscar tissue and viability of the otherregions of the leftventricle, and surgeryshouldbeperformed in centerswith a highsurgicalexperience. Surgicaltreatment of the LV aneurysmsmaybeindicated in the setting of intractableheartfailure or refractory ventriculararrhythmias. The aneurysmmaybeeitherresected and replacedwith a Dacron graft or excluded by creatinga partition between the normallyfunctioning LV cavitywalls and the aneurysm. Echocardiographymaybehelpful in determining the suitability for surgery and the approachused. In order for surgery to befeasible, the basal portions of the LV need to benormallyfunctioningsothatoverallcardiac performance ispreservedaftersurgery. Moreover, if the septum isinvolved in the aneurysmal dilatation, resectionbecomeslessfeasible and exclusion surgerymaybe the preferred choice.2 8. Jones RH, Velazquez EJ, Michler RE, Sopko G, Oh JK, O’ Connor CM, Hill JA, Menicanti L, Sadowski Z, Desvigne- Nickens P, Rouleau JL, Lee KL. Coronarybypasssurgerywith or withoutsurgicalventricular reconstruction. N Engl J Med 2009; 360: 1705-1717 [PMID: 19329820 DOI: 10.1056/ NEJMoa0900559]. 7. Castelvecchio S, Menicanti L, Donato MD. Surgicalventricularrestoration to reverse leftventricularremodeling. CurrCardiolRev 2010; 6: 15-23 [PMID: 21286274 DOI: 10.2174/15734031079023 1626] References:- 1. Jose Alberto de Agustin and al : Giant and thrombosedleftventricularaneurysm : Case report World J Cardiol 2015 July 26; 7(7): 431-433 ISSN 1949-8462 (online) © 2015 BaishidengPublishing Group Inc. 1. Jose Alberto de Agustin and al : Giant and thrombosedleftventricularaneurysm : Case report World J Cardiol 2015 July 26; 7(7): 431-433 ISSN 1949-8462 (online) © 2015 BaishidengPublishing Group Inc. 1. Jose Alberto de Agustin and al : Giant and thrombosedleftventricularaneurysm : Case report World J Cardiol 2015 July 26; 7(7): 431-433 ISSN 1949-8462 (online) © 2015 BaishidengPublishing Group Inc. y ; ( ) ( ) g g p 2. Atlas of CardiacSurgical Techniques (Second Edition), 2019. Elsevier ; HeartleftventricleAneur y ( ) ( ) g g p CardiacSurgical Techniques (Second Edition), 2019. Elsevier ; HeartleftventricleAneurysm g q ( ), ; y 3. IschemicHeartDisease : Stephen Wilmot Miller, Lawrence M. Boxt, in Cardiac Imaging (Third Edition), 2009 Definition and PathologicCorrelation. 4. Kamila Seidel Albuquerque, Joao Mauricio CanaveziIndiani and al : Asymptomatic apical aneurysm of the leftventriclewithintracavitary thrombus: adiagnosismissed by echocardiography.Radiol Bras. 2018 Jul-Aug ; 51(4) : 275-276. 5. A Maziar Zafari, MD, PhD, FACC, FAHA; Chief Editor: Eric H Yang, MD : Leftventricularaneurysm formation MyocardialInfarction Treatment& Management, Updated: May 07, 2019. 6. Management of thrombosis in heartfailure : Ronald S. Freudenberger, Shunichi Homma, in HeartFailure: A Companion to Braunwald'sHeartDisease (Second Edition), 2011. Thrombosis in LeftVentricularAneurysms in Patients withHeartFailure 6. Management of thrombosis in heartfailure : Ronald S. Freudenberger, Shunichi Homma, in HeartFailure: A Companion to Braunwald'sHeartDisease (Second Edition), 2011. Thrombosis in LeftVentricularAneurysms in Patients withHeartFailure 7. Castelvecchio S, Menicanti L, Donato MD. Surgicalventricularrestoration to reverse leftventricularremodeling. CurrCardiolRev 2010; 6: 15-23 [PMID: 21286274 DOI: 10.2174/15734031079023 1626] 7. Castelvecchio S, Menicanti L, Donato MD. Surgicalventricularrestoration to reverse leftventricularremodeling. CurrCardiolRev 2010; 6: 15-23 [PMID: 21286274 DOI: 10.2174/15734031079023 1626] 8. Jones RH, Velazquez EJ, Michler RE, Sopko G, Oh JK, O’ Connor CM, Hill JA, Menicanti L, Sadowski Z, Desvigne- Nickens P, Rouleau JL, Lee KL. Coronarybypasssurgerywith or withoutsurgicalventricular reconstruction. N Engl J Med 2009; 360: 1705-1717 [PMID: 19329820 DOI: 10.1056/ NEJMoa0900559]. 8. Jones RH, Velazquez EJ, Michler RE, Sopko G, Oh JK, O’ Connor CM, Hill JA, Menicanti L, Sadowski Z, Desvigne- Nickens P, Rouleau JL, Lee KL. Coronarybypasssurgerywith or withoutsurgicalventricular reconstruction. N Engl J Med 2009; 360: 1705-1717 [PMID: 19329820 DOI: 10.1056/ NEJMoa0900559]. 332
https://openalex.org/W4360608573	https://www.researchsquare.com/article/rs-2703361/latest.pdf	English	null	Bias reduction of high return levels for extreme hazard modelling	Research Square (Research Square)	2,023	cc-by	7,332	Bias reduction of high return levels for extreme hazard modelling Chi-Hsiang Wang (  Chi-hsiang.Wang@csiro.au ) Abstract Many existing extremal data span only a few decades, often resulting in large bias and uncertainty in the estimated shape parameter of the extreme hazard model. This in turn leads to unreliable predicted extreme values at high average recurrence intervals (ARI’s). This paper illustrates a statistical method that provides a mechanism to obtain a hazard model that produces return levels at high ARI’s with reduced bias. The method makes use of the maximum recorded values of extremal data independently recorded from a number of observational sites. The logarithmically transformed probability of the maximum recorded value at a site is shown to follow the Gumbel (Type I extreme-value) distribution, therefore multiple, say m, sites provide a sample of size m transformed probabilities of extreme values, each from a distinct site. The sample can be treated as being drawn from a Gumbel distribution, irrespective of the underlying hazard-generating mechanisms or the statistical hazard models. The method is demonstrated by an analysis of the extreme wind gust data collected from automatic weather stations in South Australia. The results are compared to the specifications in the Australian standard AS/NZS 1170.2:2021 and indicates that the standard may have overestimated the wind gust hazard, hence the specified design wind speeds may fall on the conservative side for South Australia. Research Article DOI: https://doi.org/10.21203/rs.3.rs-2703361/v1 Version of Record: A version of this preprint was published at Natural Hazards on November 20th, 2023. See the published version at https://doi.org/10.1007/s11069-023-06286-2. Page 1/23 Page 1/23 1. Introduction However, the specific values were chosen based either on the consensus of judgment or heuristic averaging of the set of shape parameters derived from analysing the wind gust data (Buishand 1991; Holmes 2002). Either way, there is a lack of objective criteria or theoretical basis to derive the best shape parameter value. A wide range of goodness-of-fit test methods such as the Anderson-Darling statistic, the Kolmogorov- Smirnov test, and graph-based tests exist (Palutikof et al. 1999), but they are applicable only for testing the within-data fitness of a dataset from one individual station (or one ‘super-station’). In addition, insufficient record length of observational data poses a serious challenge to derive a shape parameter value with high confidence for extrapolation to hundreds of years beyond the record length. 0.1 for all of the four wind regions specified, and to avoid underestimating the extreme wind speeds the ASCE 7 Standards Committee on Loads has used 0 (effectively the Type I extreme-value distribution) for extreme non-hurricane speeds (Lombardo, Main, and Simiu 2009; Simiu and Yeo 2019). However, the specific values were chosen based either on the consensus of judgment or heuristic averaging of the set of shape parameters derived from analysing the wind gust data (Buishand 1991; Holmes 2002). Either way, there is a lack of objective criteria or theoretical basis to derive the best shape parameter value. A wide range of goodness-of-fit test methods such as the Anderson-Darling statistic, the Kolmogorov- Smirnov test, and graph-based tests exist (Palutikof et al. 1999), but they are applicable only for testing the within-data fitness of a dataset from one individual station (or one ‘super-station’). In addition, insufficient record length of observational data poses a serious challenge to derive a shape parameter value with high confidence for extrapolation to hundreds of years beyond the record length. Van Den Brink and Können (2011) introduced a concept in terms of return period for modelling the probabilities of occurrences of the maximum values from each record of an ensemble of independently collected records and found that the logarithmically transformed return period of a maximum value is approximately a standard Gumbel variate. They demonstrated that this concept can be employed to check the appropriateness of the probability distribution and its parameter values used for modelling the observed phenomena (Van Den Brink and Können 2008, 2011). 1. Introduction Instead of return period, this paper derives the concept by means of the relationship between the exceedance probability and ARI, and shows that the log-transformed ARI follows exactly the standard Gumbel distribution. The use of ARI assumes that the occurrence of extreme events follows stochastic continuous processes, in contrast to using return period which assumes that the occurrence of events follows discrete event processes (Wang and Holmes 2020). In addition, the use of ARI along with exceedance probability enables more flexible choice of extreme-value distributions between the two most commonly employed: the generalised extreme-value distribution (GEV) and the generalised Pareto distribution (GPD). Extreme natural hazard data were typically extracted through the block maxima (BM) (usually annual extremes) method or the peaks-over-threshold (POT) method, two of the most widely used methods for processing extremal data. For large-scale synoptic wind-induced gusts, for example, extracting annual extremes is straightforward with high-quality data, whereas for less frequent, non-synoptic wind events such as thunderstorms, tornadoes, downbursts, and tropical cyclones, which may not occur every year, the wind gusts over a sufficiently high threshold from independent events may be taken for analysis. In this context, the GEV is conceptually a BM method, whereas the GPD conforms to the POT method. Despite the conceptual distinction in data extraction, the GEV and GPD possess a duality relationship that admits the parameters of one distribution to be converted to that of another (Wang and Holmes 2020). As a result, either of the two distributions can be employed for analysis, whether the dataset is processed via block maxima or peaks-over-threshold as long as the ARI’s are estimated through the rates of exceedance rather than the probabilities of exceedance, allowing the choice between the distributions to be based on the preference of the analyst. Extreme natural hazard data were typically extracted through the block maxima (BM) (usually annual extremes) method or the peaks-over-threshold (POT) method, two of the most widely used methods for processing extremal data. For large-scale synoptic wind-induced gusts, for example, extracting annual extremes is straightforward with high-quality data, whereas for less frequent, non-synoptic wind events such as thunderstorms, tornadoes, downbursts, and tropical cyclones, which may not occur every year, the wind gusts over a sufficiently high threshold from independent events may be taken for analysis. In this context, the GEV is conceptually a BM method, whereas the GPD conforms to the POT method. 1. Introduction Estimates of extremal return levels at high average recurrence intervals (ARI’s) are strongly dependent on the shape parameter of the statistical model. Especially with only a few tens of years of observed data, high bias and uncertainty invariably exist for parameter estimation based on data from an individual observational stations. For instance, with a short record of extreme wind gust data, say around 20 years, the unreliably estimated shape parameter could lead in some cases to the prediction errors of 1000-year wind speeds up to a few hundred percent (Simiu and Filliben 1975). A common approach to ameliorate this shortcoming is the ‘super-station’ (or station-year) approach (Buishand 1991; Peterka 1992; Wang, Wang, and Khoo 2013) which takes advantage of having multiple weather stations with valid records in a climatologically uniform region. This approach commingles all the data of the valid records from independent events into a single record with the years of the record being the summed years of all the original records. A ‘super-station’ extends the length of record and should reduce the uncertainty in high ARI’s, hence has been demonstrated to be a sensible approach for regions in which all the data could be commingled; however, the problem of potentially high bias remains for return levels beyond the accumulated record length of the super-station. With regards to natural hazards, for instance, if multiple regions of different climatic conditions are considered, the hazard in each region is analysed based on the data collected in the region, hence each region has its own hazard model parameters, and in such cases the model estimation processes would most likely give different values of the shape parameters for different regions. In practices, because of consideration for convenience of engineering applications or consensus of expert judgment, it may be decided to use a specific shape parameter value across all the regions. For instance, the Australian design standard for wind actions (Australian / New Zealand Standard 2021) uses a shape parameter of Page 2/23 0.1 for all of the four wind regions specified, and to avoid underestimating the extreme wind speeds the ASCE 7 Standards Committee on Loads has used 0 (effectively the Type I extreme-value distribution) for extreme non-hurricane speeds (Lombardo, Main, and Simiu 2009; Simiu and Yeo 2019). 1. Introduction Despite the conceptual distinction in data extraction, the GEV and GPD possess a duality relationship that admits the parameters of one distribution to be converted to that of another (Wang and Holmes 2020). As a result, either of the two distributions can be employed for analysis, whether the dataset is processed via block maxima or peaks-over-threshold as long as the ARI’s are estimated through the rates of exceedance rather than the probabilities of exceedance, allowing the choice between the distributions to be based on the preference of the analyst. Page 3/23 In the following, the theory which describes the log-transformed ARI’s of maximum values from an ensemble of records of different hazard-generating mechanisms constitute a sample drawn from the standard Gumbel distribution is first derived, followed by an application of the theory to determine the best shape parameter using the wind gust records in South Australia. Since the GEV requires one less (i.e the rate of threshold exceedance) model parameters than the GPD, it is employed for gust hazard analysis. This study demonstrates a theoretical basis to derive the best shape parameter value for one or multiple regions of different climatological conditions, with which the derived hazard model is safeguarded to produce extrapolated wind gust speed at high ARI’s with reduced bias. This method promises to be a useful tool in cases where one shape parameter value is applied across regions of various hazards, as of the case in Australian / New Zealand Standard (2021). 2. Method Y A U FY (y) FA (a) Let be the corresponding ARI (in a reference time interval of years) of , and write An n Yn P{An ≤a} = [P{A ≤a}]n = e−n/a = e−e−(lna−lnn) (6) (6) If we further define ΔLA = lnA −lnn(7) ΔLA = lnA −lnn(7) then is also a random variable with the probability distribution, ΔLA then is also a random variable with the probability distribution, ΔLA P{ΔLA ≤ΔˆLA} = e−e−ΔˆLA (8) which is the standard Gumbel distribution. which is the standard Gumbel distribution. In theory, if is perfectly known, any given and the corresponding will be known. In this case, if we choose, e.g. , will be zero. In practical situations, however, is not known, an observed quantity would correspond to an unknown , which effectively constitutes a random sampling problem with interest of obtaining an estimated or equivalently . FY (y) ya a a = n ΔˆLA FY (y) ya a a ΔˆLA Note that the distribution of depends on neither the underlying distribution nor its distribution parameters. Suppose that there is a sample of size , , from stations, where is the maximum among the values recorded at station , then provided that the ’s are independent, the extreme-value distribution functions used to model each of the records are allowed to be different. Consequently, records from different climatological regions may be combined to gauge the conformance of ’s to the standard Gumbel distribution. In other words, this property allows different records of extremes to ‘learn’ from the experience of others by pooling the ’s as if they were a sample drawn from a standard Gumbel variate. ΔLA FY (y) ΔˆLAi m i = 1, . . . , m m ΔˆLAi Ni i ΔˆLAi FYni (yai) m ΔˆLAi ΔˆLAi Even though is independent of the underlying ’s, in practical situations for extreme hazard analysis, ’s are typically unavailable and still needs to be substituted by an empirically determined distribution with the distribution parameters being estimated from observational data, which are invariably plagued by sampling errors. That is, if the extreme values of are arranged in ascending order, , then based on is obtained. ΔLA FYi (yi) FYi (yi) ~ F Yi (yi) Ni Yi y(1i) ≤y(2i) ≤. . . 2. Method The method described herein is applicable to an ensemble of extremal data records that may be produced by different mechanisms from different geographical regions as it involves only the maximum value of each record. In addition, it applies whether the data be processed by the BM or the POT method. For derivation, we will concentrate on the POT method. If the occurrence of extremes exceeding a high threshold follows a Poisson process, the relationship between return period and average recurrence interval can be shown to be (Wang and Holmes 2020) R A 1/R = 1 −e−1/A (1) in which is conventionally defined as the inverse of the probability of exceedance. Assuming that the probability distribution of annual extreme is , then the probability of less than , the value at -year ARI, is R Y FY (y) = P{Y ≤y} Y ya y a FY (y) = P{Y ≤ya} = e−1/a (2) Let be the extreme value of in an -year time interval, then Yn Y n Let be the extreme value of in an -year time interval, then Yn Y n P{Yn ≤ya} = [P{Y ≤ya}]n = e−n/a (3) If is continuous, for every in , there is a unique in , a bijective function exists that maps to . Therefore is also a random variable. With bijection, the function first maps through the function to the uniform random variable on [0, 1], then maps to by Y y Y a A g : Y →A Y A A g (Y ) Y FY U U A A = −(lnU)−1 (4) A = −(lnU)−1 (4) If is the distribution function of , we have FA (a) A FA (a) = P{A ≤a} = e−1/a (5) Page 4/23 Page 4/23 which is an inverted exponential distribution (Lin, Duran, and Lewis 1989), the same as the right-hand side of Eq. 2. This is not surprising as both and are mapped bijectively to through and , respectively. Y A U FY (y) FA (a) which is an inverted exponential distribution (Lin, Duran, and Lewis 1989), the same as the right-hand side of Eq. 2. This is not surprising as both and are mapped bijectively to through and , respectively. 2. Method 2007), method of moments (Ang and Tang 2007), probability weighted moments, maximum likelihood method, principle of maximum entropy, elemental quantile method, or Bayesian approaches (de Zea Bermudez and Kotz 2010). Except the method of moments and the maximum likelihood method, all other methods require an estimate of empirical cumulative distribution function (ECDF) for parameter estimation. Since the data to be analysed herein were extracted by the POT method, as to be described in the next section, the rate of exceedance was used to obtain the ECDF. For a given wind type (e.g. non-synoptic) at a station, suppose there are extreme gust speeds exceeding a specified threshold in years and the occurrence of exceedance obeys a Poisson process, then an unbiased estimate of the rate of exceedance, , with respect to the j-th smallest gust speed may be estimated by (Ang and Tang 2007) N n λj, j = 1, . . . , N λj = (11) N −j + 1 n n n Because the ARI , Eq. 11 can be used to obtain the ECDF for hazard model fitting. aj = 1/λj , Eq. 11 can be used to obtain the ECDF for hazard model fitting. j = 1/λj Because the ARI , Eq. 11 can be used to obtain the ECDF for hazard model fitting. aj = 1/λj For simplicity and without loss of generality, in the following the least-squares linear (for cases with fixed shape parameter) and nonlinear (for cases with free shape parameter) regression techniques for the wind gust speed on ARI were used for model parameter estimation. 2. Method ≤y(Ni) ΔˆLAi ~ F Yi (y(Ni)) ˆ As is determined by the largest value from station , special attention should be paid to ensure that all the largest values are contributed by independent extreme events. If one event contributes to multiple ’s, the value which represents the highest ARI among them is kept in the analysis but all others triggered by the same event should be discarded. ΔˆLAi i m ΔˆLAi ΔˆLAi Page 5/23 The duality of the GEV and GPD ensures that they exhibit the same tail behaviour and have the same shape parameter (Wang and Holmes 2020) when applied to the same set of data. The GEV is used in this study as it does not depend on the rate of exceedance, even though the wind gust data used in this study (described in the next section) were chosen by the POT method. Because of the duality, the outcomes and conclusion drawn for the GEV should be equally applicable to the GPD. The GEV may be expressed as The GEV may be expressed as The GEV may be expressed as P{Y ≤ya} = e −[1−k( )] 1/k (9) ya−η σ where , , and are the location, scale, and shape parameters, respectively, of the distribution. can be related to its corresponding as follows, η σ k ya a ya = {η + (1 −a−k) , if k ≠0; η + σlna, otherwise. (10) σ k For extreme hazard analysis, the analysts exercise their own decisions for the type of extreme value distributions. The distribution parameters are then estimated by a model-fitting method such as the least- squares regression (Press et al. 2007), method of moments (Ang and Tang 2007), probability weighted moments, maximum likelihood method, principle of maximum entropy, elemental quantile method, or Bayesian approaches (de Zea Bermudez and Kotz 2010). Except the method of moments and the maximum likelihood method, all other methods require an estimate of empirical cumulative distribution function (ECDF) for parameter estimation. Since the data to be analysed herein were extracted by the POT method, as to be described in the next section, the rate of exceedance was used to obtain the ECDF. For extreme hazard analysis, the analysts exercise their own decisions for the type of extreme value distributions. The distribution parameters are then estimated by a model-fitting method such as the least- squares regression (Press et al. 3. Data Page 6/23 The first Dines pressure-tube/float anemometer in South Australia, managed by the Bureau of Meteorology, Australia, became operational around 1956. Three datasets of 3-second wind gust speeds, recorded at 10 meters high, were acquired: half-hourly data (up to January 2015) from 64 stations, daily data (up to May 2017) from 76 stations, and one-minute data (up to May 2017) from 69 stations. After data screening, some of the stations were eliminated because of a high percentage of missing data, suspect recordings, or complicated topographical surroundings that make highly doubtful a gust speed could be corrected to terrain category 2 (i.e. open terrain) exposure as specified by Australian / New Zealand Standard (2021). This gives the longest record length of around 30 years among all the stations. Because of the short record lengths of the datasets, insufficient number of each of the convective windstorms such as downbursts, thunderstorms, and tornadoes were recorded at a station, they were hence grouped as non-synoptic wind events. Other non-convective, large-scale events were grouped as synoptic wind events. Synoptic and non-synoptic winds were considered separately and only the records with data length 10 years were kept for analysis. This leaves 13 stations for synoptic and 12 stations for non-synoptic winds, as shown in Fig. 1. Ten of the stations had both synoptic and non-synoptic wind records. Even though not a large number of stations left for analysis, it fulfils the purpose of illustrating the use of the method introduced in Section 2 to gauge the accuracy of the probability distribution and distribution parameters for extreme wind gust modelling. ≥ Application of probability distributions for analysis of observed data typically requires independence of the data points. For extreme wind gust analysis, this requires different recorded gust speeds be generated by different storm events. To reduce the inadvertent inclusion of multiple peak gust speeds from the same wind event, minimum separation intervals of 4 days for synoptic and of 12 hours for non-synoptic wind gusts (Lombardo, Main, and Simiu 2009) were specified. In addition, the gust speeds were corrected as follows: The instrumented anemometers were changed from the Dines anemometers to the three-cup anemometers around 1991. The wind gust speeds recorded by the two anemometer types were somewhat incompatible, hence required correction as suggested by Holmes and Ginger (2012). 3. Data The recorded 3-second gust speeds were corrected for the effects of terrain, topography and of shielding by nearby plantation and construction in the cardinal and inter-cardinal directions around each station in accordance with Australian / New Zealand Standard (2021). The 3-second gust speeds were then converted to 0.2-second gust speeds (Holmes and Ginger 2012). A storm-type separation algorithm (Holmes 2019) was used to split the gust events into synoptic and non-synoptic wind types. A storm-type separation algorithm (Holmes 2019) was used to split the g non-synoptic wind types. For the wind gust hazard modelling of both wind types in the next section, only those exceeding a threshold of 25 m/s were retained for analysis, as shown in Fig. 2. For the wind gust hazard modelling of both wind types in the next section, only those exceeding a threshold of 25 m/s were retained for analysis, as shown in Fig. 2. 4.1 Shape parameters for component wind hazards Different shape parameter values of a GEV distribution fitted to a wind gust record led to different estimates of . This section illustrates the computation of ’s given data series of a wind hazard type (synoptic, non-synoptic, or combined wind hazard) and the determination of a shape parameter. The resulting shape parameter gives the best fit of to the standard Gumbel distribution. ΔˆLA ΔLA ΔˆLAi m ΔˆLA The independent ’s are plotted against the theoretical quantiles of standard Gumbel variate determined by a plotting position formula (Cunnane 1978). If the plotted data points fall closely along the diagonal line, the chosen hazard model and its assumed parameters are consistent with that implied in Eq. 8. However, if the data points fall below (above) the diagonal line, it means the model underestimate (overestimate) the ARI value; i.e. overestimate (underestimate) the hazard. If the data points form a linear trend that crosses the diagonal line with slope < 1, then it means the hazard model may have too many parameters and hence may be inappropriate for predicting the extreme values of ARI’s beyond the record length (Van Den Brink and Können 2008). m ΔˆLAi m A range of shape parameter values was used to fit the 13 synoptic and 12 non-synoptic wind records. The root mean squared errors (RMSE’s) between and its idealised counterpart from the standard Gumbel variate were computed. The best value was chosen based on the minimisation of RMSE. Figure 3 shows the RMSE values versus values, in which 0.2 for synoptic and 0.25 for non-synoptic (shown as star-shaped points) were revealed to be optimal. The hazard curves determined using the optimal values and the observed wind gusts are plotted in Fig. 4. k ΔˆLA k k k = k = k The Gumbel quantile-quantile (Q-Q) plots in Fig. 5 (a) shows that the GEV models with fixed 0.2 for synoptic and 0.25 for non-synoptic approximately follow the diagonal line, hence are in agreement with the theory implied in Eq. 8. The commingled synoptic and non-synoptic winds (red connected points), representing the sample of maximum recorded data points from stations of two different generating mechanisms, follow also the standard Gumbel distribution, as asserted in Section 2. k = k = Instead of fixing the values, if all three GEV distribution parameters were determined by nonlinear regression for each of the data records, Fig. 4.1 Shape parameters for component wind hazards 5 (b) shows that the lines connecting values cross the diagonal line, meaning that they are overestimated (above the diagonal line) in the lower-value range but underestimated (below the diagonal line) in the higher-value range, and hence do not follow the standard Gumbel. This implies that the fitted models may be biased and hence inappropriate for extrapolation to high ARI levels. The conundrum may be of a consequence that, with free shape parameter, the GEV has too many parameters such that the fitted models exhibit unacceptably high extrapolation bias to high return levels, which manifests as an underestimated standard deviation of k m ΔˆLA . In this regard, fixing the value, as of the case in the Australian standard (Australian / New Zealand Standard 2021), would avoid such unfavourable bias, and hence a sensible decision for more reliably determining the design wind speeds at ARI’s beyond the available data lengths. If indeed ΔˆLA k Page 8/23 Page 8/23 individual value for each station is preferred, then in addition to the standard goodness-of-fit tests for interpolation, Eq. 8 can serve as a safeguard for extrapolation of the estimated hazard models. k The values for the abscissa in Fig. 5 represent theoretical quantile values (denoted by ). They can be computed by inverting the standard Gumbel distribution and using a plotting position formula to estimate the ECDF as follows, m ΔG ΔGi = −ln (−ln ( )) (12) i −c m + 1 −2c where depends on the plotting position used (Cunnane 1978). For small sample sizes (as of the cases in this study), may need to be carefully chosen depending on the objective of study since different choice may lead to unacceptable difference in results. (Weibull plotting position) was used for the results shown in Fig. 5 and the rest of the paper as it produces comparatively conservative results. As an illustration for the extent of difference by using different values, (Hazen plotting position) was tested and the resulting values for synoptic and non-synoptic events, respectively, were 0.217 and 0.276. These represent about 8% and 10% differences, respectively, from that with . Among the most commonly used plotting positions ( ), Weibull and Hazen typically give rise to the most and least, respectively, conservative hazard modelling results (Folland and Anderson 2002). c c c = 0 c c = 0.5 k c = 0 c ≤0.5 4.2 Shape parameter for combined wind hazard 8 shows the Gumbel Q-Q plot of by the GEV models with 0.16, which indicates that the simulated wind hazards agree with the theory, whereas with the hazard models overestimate the hazard (i.e. underestimate the ARI). The simulated maximum wind speed of 1,000 years among the 10 stations is 46.2 m/s at Port Augusta Aero. For 0.16 with 2.33, is predicted (by Eq. 7) to have an ARI of 10,229 years (close to 10,000 years inferred by the 10 stations simulated independently for 1000 years), whereas for with 0.22, it is predicted to have an ARI of 1,250 years. Incidentally, in AS/NZS 1170.2:2021 (Australian / New Zealand Standard 2021), is used for all the four wind regions, the regional wind speed of ARI years for Region A (where the studied area is located) is 46 m/s. This comparison shows that the computed results agree well with the Australian standard and implies that the standard may have overestimated the wind gust hazard for South Australia. Incidentally, a recent study (El Rafei et al. 2023) on the wind gust hazard in New South Wales, Australia, using high-resolution Australian regional reanalysis found that using overestimates the 500-year ARI gust speeds, when compared to that using variable values, by approximately 4% for non-synoptic and 2.5% for synoptic events. ΔˆLA k = k = 0.10 Vmax = k = ΔˆLA = Vmax k = 0.10 ΔˆLA = k = 0.10 = 1000 k = 0.10 k Figure 9 illustrates the fitted combined hazard models with and 0.16 along with the simulated annual extreme gust speeds (i.e. same as the read lines in Fig. 6) for the 10 locations. Compared with , the curves with 0.16 provide closer fit to the data points in most locations and, as expected, result in lower gust speeds at high ARI years. On average, as shown in Fig. 10, the models with give about 2.9% and 3.5% higher gust speed estimates than that with 0.16 for ARI’s of 500 and 1000 years, respectively. That is, the Australian standard-specified design wind speeds for South Australia generally fall on the conservative side with respect to the design for wind actions of structures specified as of importance levels 2 (domestic housing and structures under normal operations) and 3 (construction designed to contain a large number of people) in the 2019 National Construction Code of Australia (Australian Building Codes Board 2019). 4.2 Shape parameter for combined wind hazard As shown in Fig. 4, the synoptic and non-synoptic wind events at a location pose different extent of threats as they are induced by different climatic mechanisms. The hazards posed by both mechanisms need to be taken into account for design of structures. One way for estimating the combined hazard is to commingle all extreme gust speeds from all mechanisms for wind hazard analysis. However, hazard models derived from commingled datasets tend to underestimate wind hazard at higher ARI’s (Gomes and Vickery 1978; Lombardo, Main, and Simiu 2009). An alternative is to combine the hazards of different, independent mechanisms by probability theory (Wang and Holmes 2020; Holmes and Bekele 2021). However, no closed-form probability distribution is readily available for probabilistic combination of hazards with parent GEV distributions, and no simple expression exists for the shape parameter of the combined hazard model. As a result, Monte-Carlo simulation was conducted to generate annual gust speeds for 1000 years from the best models (i.e. 0.2 for synoptic and 0.25 for non-synoptic) for each of the 10 locations where the records of the two wind types were available. For a given year at a specific site, the maximum of the two generated gust speeds was taken as the extreme speed of the year. Figure 6 shows the generated combined annual extremes (red lines) up to 1000-year ARI and the best fitted hazard curves of the non-synoptic (green lines) and synoptic (blue lines) winds for the 10 locations. k = k = As an approximation and for comparison with the Australian Standard, the generated combined gust speeds were fitted to a GEV distribution. Similar to the model parameter estimation described in Section 4.1, we computed the RMSE between estimated and the standard Gumbel quantiles, and plotted it over a range of values, as shown in Fig. 7. It shows that the optimal value is about 0.16 for the ΔˆLA k k Page 9/23 Page 9/23 combined hazard. Incidentally, this value is close to obtained by Holmes and Moriarty (1999) for thunderstorm downbursts at Moree, New South Wales, Australia, which is located in wind hazard region A, the same region as the locations studied herein. k k = 0.161 For the combined hazards, Fig. 4.2 Shape parameter for combined wind hazard Nevertheless, dependent upon the balance of the benefits gained versus the costs incurred due to the more conservative design wind speed, the resulting higher cost but more conservative construction may be justified if the additional benefits gained are deemed to outweigh the extra costs incurred. k = 0.10 k = k = 0.10 k = k = 0.10 k = 5. Conclusion Because many currently available extreme data of climatic events such as observational extreme wind gusts span only a few tens of years, resulting in high bias and uncertainty of distribution parameter values estimated based on the data, and hence unreliable predicted extreme values when extrapolated to high ARI’s beyond the range of data length. For the quality of model fitting, the typical goodness-of-fit tests allows assessment within the record length, but unable to test the fitness for higher ARI’s. The Page 10/23 approach used in this study serves to gauge whether the fitted model is appropriate and unbiased for extrapolation, providing a mechanism to safeguard the accuracy of fitted values longer than the record length or at high ARI’s, which is what’s typically needed for engineering design and reliability assessment. approach used in this study serves to gauge whether the fitted model is appropriate and unbiased for extrapolation, providing a mechanism to safeguard the accuracy of fitted values longer than the record length or at high ARI’s, which is what’s typically needed for engineering design and reliability assessment. The ARI is proved to follow an inverted exponential distribution and the log-transformed ARI (i.e. ) becomes Gumbel distributed and is often used for visualising the estimated hazard. Moreover, the ability of the method in pooling the ARI’s of maximum recorded data points from all observational stations, even when they are from regions of different hazard-generating mechanisms, may be regarded as a generalisation of the ‘super-station’ approach that has been used to commingle all the extreme wind gust or rainfall data from a climatologically uniform region. Therefore, the method is useful for cases such as the wind gust speed specification in the Australian standard in which a shape parameter value is applied across the four wind regions. In such cases, the estimated values of for the records from all regions may be combined and the value that fits best the to the standard Gumbel distribution can be chosen to apply to all regions. ΔˆLA ΔˆLA k ΔˆLA In the Australian context, although it often occurs that non-synoptic wind gusts dominate the extreme wind climate, particularly at larger ARI’s, consideration of both synoptic and non-synoptic is necessary as the combined wind hazard tends to have a smaller shape parameter value that, compared to a larger one, typically leads to higher wind gust values at high ARI’s. 5. Conclusion In addition, synoptic wind gusts at some locations dominate at smaller ARI’s, which is important for construction of temporary and secondary structures such as formwork, circus tents, and farm shelters that are intended to be in services for only a short period of time. The analysis of wind gust data from South Australia indicates that the shape parameter value of 0.1 used in the Australian standard, AS/NZS 1170.2:2021, may be too low that it appears to lead to overestimate the wind hazard, and hence fall on the conservative side, in South Australia. As with typical experimental and observational studies, the accuracy of estimation by the method used in this paper clearly depends on the accuracy of measurement, the quality of the data collection and processing, and the classification of the right hazard-generating mechanism when heterogeneous mechanisms are concerned. In addition, the method assists only in selecting the hazard model and its fitted parameters for bias reduction of prediction at high ARI’s but does not provide uncertainty estimates of the prediction. If preferred, uncertainty quantification may be conducted by the bootstrap (Efron and Tibshirani 1993) or Bayesian analysis (Gelman, Hill, and Vehtari 2021). On the other hand, the method can be used in conjunction with uncertainty minimization techniques as well as within-data goodness-of- fit tests to help obtain a model of minimised extrapolation bias and variance. Funding This work was supported by research project: EE_BEE_NatHERS data and communication, funded by CSIRO, Australia. This work was supported by research project: EE_BEE_NatHERS data and communication, funded by CSIRO, Australia. Page 11/23 Page 11/23 Page 11/23 Author Contributions Study conception and design, material preparation, data collection and analysis, and draft of the manuscript were performed by Chi-Hsiang Wang. The author has no relevant financial or non-financial interests to disclose. The author has no relevant financial or non-financial interests to disclose. Acknowledgments The author thanks Dr. John D. Holmes, Director of JDH Consulting, for kindly providing the correction factors for wind gust speeds related to the effects of terrain, topography, and shielding of buildings surrounding the anemometer stations, and providing insightful comments for the manuscript. References D., and W. W. Moriarty. 1999. “Application of the generalized Pareto distribution to extreme value analysis in wind engineering.” Journal of Wind Engineering and Industrial Aerodynamics 83 (1- 3): 1–10. https://doi.org/10.1016/S0167-6105(99)00056-2. 16. Holmes, J. D., and W. W. Moriarty. 1999. “Application of the generalized Pareto distribution to extreme value analysis in wind engineering.” Journal of Wind Engineering and Industrial Aerodynamics 83 (1- 3): 1–10. https://doi.org/10.1016/S0167-6105(99)00056-2. 17. Lin, C. T., B. S. Duran, and T. O. Lewis. 1989. “Inverted gamma as a life distribution.” Microelectronics Reliability 29 (4): 619–26. https://doi.org/10.1016/0026-2714(89)90352-1. 17. Lin, C. T., B. S. Duran, and T. O. Lewis. 1989. “Inverted gamma as a life distribution.” Microelectronics Reliability 29 (4): 619–26. https://doi.org/10.1016/0026-2714(89)90352-1. 18. Lombardo, Franklin T., Joseph A. Main, and Emil Simiu. 2009. “Automated extraction and classification of thunderstorm and non-thunderstorm wind data for extreme-value analysis.” Journal of Wind Engineering and Industrial Aerodynamics 97 (3-4): 120–31. https://doi.org/10.1016/j.jweia.2009.03.001. 19. Palutikof, J. P., B. B. Brabson, D. H. Lister, and S. T. Adcock. 1999. “A review of methods to calculate extreme wind speeds.” Meteorological Applications 6 (2): 119–32. https://doi.org/10.1017/S1350482799001103. 19. Palutikof, J. P., B. B. Brabson, D. H. Lister, and S. T. Adcock. 1999. “A review of methods to calculate extreme wind speeds.” Meteorological Applications 6 (2): 119–32. https://doi.org/10.1017/S1350482799001103. 20. Peterka, J. A. 1992. “Improved extreme wind prediction for the United States.” Journal of Wind Engineering and Industrial Aerodynamics 41 (1-3): 533–41. https://doi.org/10.1016/0167- 6105(92)90459-N. 20. Peterka, J. A. 1992. “Improved extreme wind prediction for the United States.” Journal of Wind Engineering and Industrial Aerodynamics 41 (1-3): 533–41. https://doi.org/10.1016/0167- 6105(92)90459-N. 21. Press, William H, Saul A Teukolsky, William T Vetterling, and Brian P Flannery. 2007. Numerical Recipes: The Art of Scientific Computing. 3rd ed. Vol. 1. Cambridge University Press. 21. Press, William H, Saul A Teukolsky, William T Vetterling, and Brian P Flannery. 2007. Numerical Recipes: The Art of Scientific Computing. 3rd ed. Vol. 1. Cambridge University Press. 22. Simiu, Emil, and James Filliben. 1975. “Statistical Analysis of Extreme Winds.” National Bureau of Standards. https://www.govinfo.gov/content/pkg/GOVPUB-C13- 415fc91b46e5b7d06ec2ee0b8e936ed9/pdf/GOVPUB-C13- 415fc91b46e5b7d06ec2ee0b8e936ed9.pdf. 22. Simiu, Emil, and James Filliben. 1975. “Statistical Analysis of Extreme Winds.” National Bureau of Standards. https://www.govinfo.gov/content/pkg/GOVPUB-C13- 415fc91b46e5b7d06ec2ee0b8e936ed9/pdf/GOVPUB-C13- 415fc91b46e5b7d06ec2ee0b8e936ed9.pdf. 23. Simiu, Emil, and DongHun Yeo. 2019. Wind Effects on Structures. Chichester, UK: John Wiley & Sons, Ltd. https://doi.org/10.1002/9781119375890. 23. Simiu, Emil, and DongHun Yeo. 2019. Wind Effects on Structures. Chichester, UK: John Wiley & Sons, Ltd. https://doi.org/10.1002/9781119375890. 24. Van Den Brink, H. W., and G. References 1. Ang, Alfredo H-S., and Wilson H. Tang. 2007. Probability concepts in engineering : emphasis on applications in civil & environmental engineering. 2nd ed. John Wiley & Sons, Inc. 2. Australian / New Zealand Standard. 2021. “Structural design actions Pa 2. Australian / New Zealand Standard. 2021. “Structural design actions Part 2: Wind actions.” 3. Australian Building Codes Board. 2019. “National Construction Code Volume 1.” https://ncc.abcb.gov.au/editions-national-construction-code. 4. Buishand, T. A. 1991. “Extreme rainfall estimation by combining data from several sites.” Hydrological Sciences Journal 36 (4): 345–65. https://doi.org/10.1080/02626669109492519. 5. Cunnane, C. 1978. “Unbiased plotting positions — A review.” Journal of Hydrology 37 (3-4): 205–22. https://doi.org/10.1016/0022-1694(78)90017-3. 5. Cunnane, C. 1978. “Unbiased plotting positions — A review.” Journal of Hydrology 37 (3-4): 205–22. https://doi.org/10.1016/0022-1694(78)90017-3. 6. de Zea Bermudez, P., and Samuel Kotz. 2010. “Parameter estimation of the generalized Pareto distribution-Part I.” Journal of Statistical Planning and Inference 140 (6): 1353–73. https://doi.org/10.1016/j.jspi.2008.11.019. 6. de Zea Bermudez, P., and Samuel Kotz. 2010. “Parameter estimation of the generalized Pareto distribution-Part I.” Journal of Statistical Planning and Inference 140 (6): 1353–73. https://doi.org/10.1016/j.jspi.2008.11.019. 7. Efron, Bradley., and Robert J Tibshirani. 1993. An introduction to the bootstrap. Chapman & Hall. 7. Efron, Bradley., and Robert J Tibshirani. 1993. An introduction to the bootstrap. Chapman & Hall. 8. El Rafei, Moutassem, Steven Sherwood, Jason P. Evans, and Fei Ji. 2023. “Analysis of extreme wind gusts using a high-resolution Australian Regional Reanalysis.” Weather and Climate Extremes 39 (November 2022): 100537. https://doi.org/10.1016/j.wace.2022.100537. 9. Folland, Chris, and Clive Anderson. 2002. “Estimating Changing Extremes Using Empirical Ranking Methods.” Journal of Climate 15 (20): 2954–60. https://doi.org/10.1175/1520- 0442(2002)015<2954:ECEUER>2.0.CO;2. 10. Gelman, Andrew, Jennifer Hill, and Aki Vehtari. 2021. Regression and other stories. Cambridge University Press. 11. Gomes, L., and B. J. Vickery. 1978. “Extreme wind speeds in mixed wind climates.” Journal of Wind Engineering and Industrial Aerodynamics 2 (4): 331–44. https://doi.org/10.1016/0167- Page 12/23 Page 12/23 6105(78)90018-1. 12. Holmes, J. D. 2002. “A Re-analysis of Recorded Extreme Wind Speeds in Region A.” Australian Journal of Structural Engineering 4 (1): 29–40. https://doi.org/10.1080/13287982.2002.11464905. 13. ———. 2019. “Extreme Wind Prediction – The Australian Experience.” Lecture Notes in Civil Engineering 27: 365–75. https://doi.org/10.1007/978-3-030-12815-9_29/FIGURES/9. 14. Holmes, J. D., and S. Bekele. 2021. Wind Loading of Structures. 4th ed. CRC Press, Taylor & Francis Group. 15. Holmes, J. D., and J. D. Ginger. 2012. “The gust wind speed duration in AS/NZS 1170.2.” Australian Journal of Structural Engineering 13 (3): 207–16. https://doi.org/10.7158/S12-017.2012.13.3. 16. Holmes, J. References P. Können. 2008. “The statistical distribution of meteorological outliers.” Geophysical Research Letters 35 (23): 1–5. https://doi.org/10.1029/2008GL035967. 24. Van Den Brink, H. W., and G. P. Können. 2008. “The statistical distribution of meteorological outliers.” Geophysical Research Letters 35 (23): 1–5. https://doi.org/10.1029/2008GL035967. 25. ———. 2011. “Estimating 10000-year return values from short time series.” International Journal of Climatology 31 (1): 115–26. https://doi.org/10.1002/joc.2047. 25. ———. 2011. “Estimating 10000-year return values from short time series.” International Journal of Climatology 31 (1): 115–26. https://doi.org/10.1002/joc.2047. Page 13/23 26. Wang, C.-H., and J. D. Holmes. 2020. “Exceedance rate, exceedance probability, and the duality of GEV and GPD for extreme hazard analysis.” Natural Hazards 102 (3): 1305–21. https://doi.org/10.1007/s11069-020-03968-z. 26. Wang, C.-H., and J. D. Holmes. 2020. “Exceedance rate, exceedance probability, and the duality of GEV and GPD for extreme hazard analysis.” Natural Hazards 102 (3): 1305–21. https://doi.org/10.1007/s11069-020-03968-z. 26. Wang, C.-H., and J. D. Holmes. 2020. “Exceedance rate, exceedance probability, and the duality of GEV and GPD for extreme hazard analysis.” Natural Hazards 102 (3): 1305–21. https://doi.org/10.1007/s11069-020-03968-z. 27. Wang, C.-H., X. Wang, and Y. B. Khoo. 2013. “Extreme wind gust hazard in Australia and its sensitivity to climate change.” Natural Hazards 67 (2). https://doi.org/10.1007/s11069-013-0582-5. 27. Wang, C.-H., X. Wang, and Y. B. Khoo. 2013. “Extreme wind gust hazard in Australia and its sensitivity to climate change.” Natural Hazards 67 (2). https://doi.org/10.1007/s11069-013-0582-5. Figures Figure 1 Locations of anemometer stations Figures Figures Figure 1 Locations of anemometer stations Locations of anemometer stations Page 14/23 Figure 2 3-second wind gust speeds exceeding 25 m/s Figure 2 3-second wind gust speeds exceeding 25 m/s Page 15/23 Figure 3 See image above for figure legend. See image above for figure legend. Page 16/23 Page 16/23 Figure 4 Fitted hazard curves and the observed synoptic and non-synoptic wind gust speeds Figure 4 Figure 4 Fitted hazard curves and the observed synoptic and non-synoptic wind gust speeds Fitted hazard curves and the observed synoptic and non-synoptic wind gust speeds Page 17/23 Figure 5 See image above for figure legend. See image above for figure legend. Page 18/23 Page 18/23 Figure 6 Hazard curves of the optimal synoptic and non-synoptic hazard models and the simulated combined hazards up to 1000-year ARI Figure 6 Figure 6 Hazard curves of the optimal synoptic and non-synoptic hazard models and the simulated combined hazards up to 1000-year ARI Page 19/23 Page 19/23 Figure 7 See image above for figure legend. Page 20/23 Figure 8 See image above for figure legend. Page 21/23 Figure 9 Simulated versus fitted gust speeds for combined gust hazards Figure 9 Figure 10 Percent error between gust speeds by models with k = 0.16 and k = 0.10 Figure 9 Simulated versus fitted gust speeds for combined gust hazards Simulated versus fitted gust speeds for combined gust hazards Page 22/23 Figure 10 Percent error between gust speeds by models with k = 0.16 and k = 0.10 Supplementary Files This is a list of supplementary files associated with this preprint Click to download Supplementary Files This is a list of supplementary files associated with this preprint. Click to download. Stationinformation.csv WindtypesAdelaide.xlsx WindtypesEast.xlsx WindtypesLowerEyre.xlsx WindtypesMidNorth.xlsx WindtypesSouth.xlsx Page 23/23
https://openalex.org/W2889326210	https://dro.deakin.edu.au/articles/journal_contribution/Internet_use_by_older_adults_with_bipolar_disorder_international_survey_results/20793772/1/files/37046587.pdf	English	null	Internet use by older adults with bipolar disorder: international survey results	International journal of bipolar disorders	2,018	cc-by	6,641	Internet use by older adults with bipolar disorder: international survey results Citation: Bauer, Rita, Glenn, Tasha, Strejilevich, Sergio et al. 2018, Internet use by older adults with bipolar disorder: international survey results, International journal of bipolar disorders, vol. 6: 20, pp. 1-7. DOI: http://www.dx.doi.org/10.1186/s40345-018-0127-7 DOI: http://www.dx.doi.org/10.1186/s40345-018-0127-7 ©2018, The Authors ©2018, The Authors Reproduced by Deakin University under the terms of the Creative Commons Attribution Licence Downloaded from DRO: http://hdl.handle.net/10536/DRO/DU:30113646 Downloaded from DRO: http://hdl.handle.net/10536/DRO/DU:30113646 Reproduced by Deakin University under the terms of the Creative Commons Attribution Licence Abstract Background: The world population is aging and the number of older adults with bipolar disorder is increasing. Digi‑ tal technologies are viewed as a framework to improve care of older adults with bipolar disorder. This analysis quanti‑ fies Internet use by older adults with bipolar disorder as part of a larger survey project about information seeking. Background: The world population is aging and the number of older adults with bipolar disorder is increasing. Digi‑ tal technologies are viewed as a framework to improve care of older adults with bipolar disorder. This analysis quanti‑ fies Internet use by older adults with bipolar disorder as part of a larger survey project about information seeking. Methods: A paper-based survey about information seeking by patients with bipolar disorder was developed and translated into 12 languages. The survey was anonymous and completed between March 2014 and January 2016 by 1222 patients in 17 countries. All patients were diagnosed by a psychiatrist. General estimating equations were used to account for correlated data. Results: Overall, 47% of older adults (age 60 years or older) used the Internet versus 87% of younger adults (less than 60 years). More education and having symptoms that interfered with regular activities increased the odds of using the Internet, while being age 60 years or older decreased the odds. Data from 187 older adults and 1021 younger adults were included in the analysis excluding missing values. Conclusions: Older adults with bipolar disorder use the Internet much less frequently than younger adults. Many older adults do not use the Internet, and technology tools are suitable for some but not all older adults. As more health services are only available online, and more digital tools are developed, there is concern about growing health disparities based on age. Mental health experts should participate in determining the appropriate role for digital tools for older adults with bipolar disorder. © The Author(s) 2018. 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. Correspondence: michael.bauer@uniklinikum‑dresden.de 1 Department of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany Mark Zetin: Deceased in 2017 Full list of author information is available at the end of the article Internet use by older adults with bipolar disorder: international survey results Rita Bauer1, Tasha Glenn2, Sergio Strejilevich3, Jörn Conell1,4, Martin Alda5, Raffaella Ardau6, Bernhard T. Baune7, Michael Berk8,9,10,11,12, Yuly Bersudsky13, Amy Bilderbeck14, Alberto Bocchetta15, Angela M. Paredes Castro8,9, Eric Y. W. Cheung16, Caterina Chillotti6, Sabine Choppin17, Alessandro Cuomo18, Maria Del Zompo15, Rodrigo Dias19, Seetal Dodd8,9,10, Anne Duffy20, Bruno Etain21, Andrea Fagiolini18, Miryam Fernández Hernandez22, Julie Garnham5, John Geddes14, Jonas Gildebro23, Michael J. Gitlin24, Ana Gonzalez‑Pinto22, Guy M. Goodwin14, Paul Grof25,26, Hirohiko Harima27, Stefanie Hassel28, Chantal Henry21,29, Diego Hidalgo‑Mazzei30, Anne Hvenegaard Lund23, Vaisnvy Kapur31, Girish Kunigiri32, Beny Lafer19, Erik R. Larsen33,34, Ute Lewitzka1, Rasmus W. Licht35,36, Blazej Misiak37, Patryk Piotrowski37, Ângela Miranda‑Scippa38, Scott Monteith39, Rodrigo Munoz40, Takako Nakanotani41, René E. Nielsen35, Claire O’Donovan5, Yasushi Okamura27, Yamima Osher13, Andreas Reif42, Philipp Ritter1, Janusz K. Rybakowski43, Kemal Sagduyu44, Brett Sawchuk20, Elon Schwartz45, Claire Slaney5, Ahmad H. Sulaiman46, Kirsi Suominen47, Aleksandra Suwalska43, Peter Tam48, Yoshitaka Tatebayashi41, Leonardo Tondo49,50, Julia Veeh42, Eduard Vieta30, Maj Vinberg51, Biju Viswanath52, Mark Zetin53, Peter C. Whybrow24 and Michael Bauer1 Abstract Downloaded from DRO: DRO Deakin Research Online, Deakin University’s Research Repository Deakin University CRICOS Provider Code: 00113B Deakin University CRICOS Provider Code: 00113B Bauer et al. Int J Bipolar Disord (2018) 6:20 https://doi.org/10.1186/s40345-018-0127-7 RESEARCH Internet use by older adults with bipolar disorder: international survey results Rita Bauer1, Tasha Glenn2, Sergio Strejilevich3, Jörn Conell1,4, Martin Alda5, Raffaella Ardau6, Bernhard T. Baune7, Michael Berk8,9,10,11,12, Yuly Bersudsky13, Amy Bilderbeck14, Alberto Bocchetta15, Angela M. Paredes Castro8,9, Eric Y. W. Cheung16, Caterina Chillotti6, Sabine Choppin17, Alessandro Cuomo18, Maria Del Zompo15, Rodrigo Dias19, Seetal Dodd8,9,10, Anne Duffy20, Bruno Etain21, Andrea Fagiolini18, Miryam Fernández Hernandez22, Julie Garnham5, John Geddes14, Jonas Gildebro23, Michael J. Gitlin24, Ana Gonzalez‑Pinto22, Guy M. Goodwin14, Paul Grof25,26, Hirohiko Harima27, Stefanie Hassel28, Chantal Henry21,29, Diego Hidalgo‑Mazzei30, Anne Hvenegaard Lund23, Vaisnvy Kapur31, Girish Kunigiri32, Beny Lafer19, Erik R. Larsen33,34, Ute Lewitzka1, Rasmus W. Licht35,36, Blazej Misiak37, Patryk Piotrowski37, Ângela Miranda‑Scippa38, Scott Monteith39, Rodrigo Munoz40, Takako Nakanotani41, René E. Nielsen35, Claire O’Donovan5, Yasushi Okamura27, Yamima Osher13, Andreas Reif42, Philipp Ritter1, Janusz K. Rybakowski43, Kemal Sagduyu44, Brett Sawchuk20, Elon Schwartz45, Claire Slaney5, Ahmad H. Sulaiman46, Kirsi Suominen47, Aleksandra Suwalska43, Peter Tam48, Yoshitaka Tatebayashi41, Leonardo Tondo49,50, Julia Veeh42, Eduard Vieta30, Maj Vinberg51, Biju Viswanath52, Mark Zetin53, Peter C. Whybrow24 and Michael Bauer1* Open Access Bauer et al. Int J Bipolar Disord (2018) 6:20 https://doi.org/10.1186/s40345-018-0127-7 Open Access Backgroundh was approved by institutional review boards according to local requirements. The patients who completed the survey resided in 17 countries. The survey was translated into 12 local languages: Chinese, Danish, Finnish, French, German, Hebrew, Italian, Japanese, Polish, Portuguese, Spanish, and English (versions for US/Canada, UK and Australia). The 1222 surveys were received from patients in Australia (N = 22), Brazil (N = 100), Canada (N = 109), Denmark (N = 209), Finland (N = 16), France (N = 50), Germany (N = 82), Hong Kong (N = 91), India (N = 30), Israel (N = 46), Italy (N = 80), Japan (N = 35), Malaysia (N = 25), Poland (N = 125), Spain (N = 82), UK (N = 50), and the US (N = 70).h The world’s population is living longer, with the percent- age of people over age 60 years expected to nearly dou- ble from 12 to 22% between 2015 and 2050 (WHO 2015). Today, up to 25% of the population with bipolar disorder is age 60 years or older (Sajatovic et al. 2015). Older adults with bipolar disorder differ in the disease onset and clini- cal course, and most have multiple medical comorbidities especially endocrine, respiratory and cardiovascular con- ditions (Lala and Sajatovic 2012). Digital technology pro- vides a framework to improve care for older adults with bipolar disorder by enabling remote visits, online psy- chological interventions, health monitoring, information seeking, peer support groups and self-management tools (Gliddon et al. 2017; Hidalgo-Mazzei et al. 2015; Torous et al. 2016). The survey questions and methodology were published previously (Bauer et al. 2016; Bauer R et al. 2017; Conell et al. 2016). Since paper-based surveys were used, dupli- cate data entry was performed to minimize data entry errors. A model to evaluate the differences in Internet use by those age 60 years and older was estimated using the generalized estimating equation (GEE) statistical technique to accommodate imbalances in the number of responses from collection sites, and correlation in survey responses within collection sites. Variables significant at the 0.05 level in univariate analyses were included in the multivariate model estimates. SPSS version 24.0 was used for all analyses. In addition to providing help with bipolar disor- der, Internet use by older adults in the community may decrease loneliness, and increase social support (Fors- man and Nordmyr 2017; Heo et al. 2015). Bauer et al. Int J Bipolar Disord (2018) 6:20 Page 2 of 7 Page 2 of 7 Results 1222 patients completed the survey. The patients were 62% female, had a mean age of 44 years (SD 13.8) rang- ing between 17 and 86 years, and completed 14 (SD 3.2) years of education. The demographic characteristics are shown in Table 1. Of the 1222 patients, 81% used the Internet (976 of 1212 valid responses) (Bauer et al. 2016). There were 1208 valid responses to both the ques- tions on age and “Do you use the Internet?” Of the 1208 patients, 187 were 60 years or older and of these 88 (47%) used the Internet. Of the 1021 younger adults, 884 (87%) used the Internet. Table 2 shows the best fitting model to assess differences in Inter- net use between the older adults and younger adults. The model includes variables for age 60 years or older, years of education, and if bipolar disorder sometimes or frequently interfered with regular activities. The estimated coefficients suggest that if age was 60 years or older, the odds of using the Internet will decrease by 86%, a 1 year increase in education will increase the odds of using the Internet by 30%, and if bipolar disor- der interferes with regular activities, the odds of using the Internet will increase by 76%. 1222 patients completed the survey. The patients were 62% female, had a mean age of 44 years (SD 13.8) rang- ing between 17 and 86 years, and completed 14 (SD 3.2) years of education. The demographic characteristics are shown in Table 1. Of the 1222 patients, 81% used the Internet (976 of 1212 valid responses) (Bauer et al. 2016). To gain insight into online information seeking by patients with bipolar disorder, we previously surveyed 1222 adult outpatients with bipolar disorder living in 17 countries between March 2014 and January 2016 (Bauer et al. 2016; Conell et al. 2016). Of the patients in the sur- vey, 81% used the Internet, a percentage similar to that of the general public (Bauer et al. 2016). The purpose of this analysis was to compare Internet use between the older adults, defined as 60 years or older, and younger adults less than 60 years, who completed this survey. Backgroundh Internet use may also contribute to maintaining health literacy, or the ability to read, understand and act on health informa- tion (Kobayashi et al. 2015; Andrus and Roth 2002). Gov- ernment and health care providers increasingly use the Internet as the primary form of communication about health and social services (Chang et al. 2015). Digital technologies are viewed as a means to maximize inde- pendence and facilitate aging in place, including for those with disabilities (Agree 2014; Reeder et al. 2013; Schulz et al. 2015), and to provide cost-effective care for the growing elderly population (Deloitte 2015). Most studies of Internet use involve community dwelling older adults and do not focus on mental illness. As the role of online services and monitoring technologies increases, more understanding of Internet use by older adults with bipo- lar disorder is needed. Methodsh The 39-question survey was anonymous, and took about 20 min to complete. The survey was paper based to maxi- mize participation including of those who do not use the Internet. All participants were recruited locally by their psychiatrist with no online recruitment. The study Of those who used the Internet, 689/880 (78%) of younger adults and 59/88 (67%) of older adults looked for information on bipolar disorder. While this appears Bauer et al. Int J Bipolar Disord (2018) 6:20 Page 3 of 7 Table 1 Patient demographics (N = 1222)a Table 1 Patient demographics (N = 1222)a Table 1 Patient demographics (N = 1222)a a 14 patients were missing responses to questions on age or “Do you use the Internet”? Methodsh All missing values were excluded Age 60 or older (N = 187) Age 59 or younger (N = 1021) All ages (N = 1208) N % N % N % Diagnosis BP I 107 58 657 65 764 63 BP II 70 38 308 30 378 32 BP NOS 8 4 48 5 56 5 Gender Female 120 64 637 62 757 62 Male 68 36 390 38 458 38 Employment status Full-time 31 17 529 52 560 47 Not full-time 156 83 482 48 638 53 Marital status Married or living with partner 112 60 478 47 590 49 Not married 76 40 543 53 619 51 Income group Upper income 14 8 66 7 80 7 Middle income 105 57 487 48 592 49 Lower income 65 35 468 45 533 44 Live alone Yes 54 29 245 24 299 25 No 131 71 777 76 908 75 Mood in last 6 months Mostly normal 118 63 460 45 578 48 Mostly not normal 69 37 561 55 630 52 BP disorder interfered with regular activities Frequently or sometimes 89 47 676 66 765 63 Rarely or never 99 53 347 34 446 37 Confident managing living Very confident 89 48 363 36 452 38 Not very confident 98 52 659 64 754 62 Confident knowing when to see physician Very confident 117 62 578 57 695 57 Not very confident 71 38 445 43 516 43 N Mean (SD) N Mean (SD) N Mean (SD) Years of education 184 13 (3.6) 1010 14 (3.1) 1194 14 (3.2) Age of onset 186 36 (13.6) 1011 25 (9.4) 1197 27 (10.9) research on older adults not specific to bipolar disor- der. Internet use by community dwelling older adults is increasing, with studies reporting percentages between 36 and 67%, but remains considerably lower than for younger adults (Levine et al. 2016; Friemel 2016; Yu et al. 2016; Anderson and Perrin 2017; Chang et al. 2015). As in prior research, more education and expe- riencing symptoms were associated with increased similar, insufficient data were available on the older adults for a more detailed statistical analysis. Discussion Older adults with bipolar disorder used the Internet much less frequently than younger adults. Overall, 47% of older adults used the Internet versus 87% of younger adults. This finding is consistent with prior Bauer et al. Int J Bipolar Disord (2018) 6:20 Page 4 of 7 Table 2 Explanatory model based on responses from the patients who use the Internet (N = 1208) a Patients with missing values were not included b 187 patients were age 60 years or older at time of study Independent variablesa Parameter Significance OR 95% CI Intercept < 0.001 0.136 0.053, 0.349 Age 60 years or olderb < 0.001 0.141 0.082, 0.241 Bipolar disorder sometimes or frequently interferes with regular activities 0.001 1.764 1.246, 2.497 Years of education < 0.001 1.302 1.229, 1.380 Table 2 Explanatory model based on responses from the patients who use the Internet (N = 1208) psychiatrist (Hallett et al. 2013). Some older adults do not trust the Internet as a source of health information (Sbaffi and Rowley 2017; Zulman et al. 2011). In studies of patients with a mean age ≥ 50 years, those with a strong therapeutic relationship with a physician were less likely to search for health information on the Internet (Hou and Shim 2010), and more likely to defer decision mak- ing to the physician (Park et al. 2014). In this survey, the primary reason why Internet users of all ages did not seek information about bipolar disorder was because they pre- fer to rely on information from a physician (Bauer et al. 2016). Regardless of age, most patients who did not use the Internet in this survey lacked technical skills (Bauer et al. 2016). One option to increase technology use by older adults is to provide training, but there are many serious concerns with novice Internet users. The elderly are such frequent targets for financial fraud that it is considered a public health problem in the US (Burnes et al. 2017; CDC 2015), and the scams targeting older adults have moved online (FBI 2014; Carlson 2007). In 2016, in the US, adults over age 60 were the largest group of victims of Internet crime, and suffered the largest monetary losses (FBI 2016). Factors that increase vulnerability to online fraud include low technical skills, individual traits, cognitive impairment, and depression in older adults (Monteith and Glenn 2016; Lichtenberg et al. 2016). Discussion Many older adults are not knowledgeable about Inter- net security hazards and measures to protect privacy (Grimes et al. 2010; Home Instead 2017; Holtfreter et al. 2015; White et al. 2017). Furthermore, many people of all ages have little understanding of privacy issues related to digital technology. For example, in this survey, 43% of patients of all ages searched the Internet for information about bipolar disorder because they mistakenly thought they were anonymous online (Conell et al. 2016). Internet use (Yu et al. 2016; Gell et al. 2015; Powell and Clarke 2006; Gallagher and Doherty 2009; Flynn et al. 2006), although symptoms of depression, and cogni- tive decline may decrease use in older adults (Choi and Dinitto 2013; Levine et al. 2018). Smartphone use by community dwelling older adults is even lower than Internet use, at about 40% (Anderson and Perrin 2017). In this survey, considering Internet users of all ages, 89% accessed the Internet to find information about bipolar disorder from a computer compared with 11% from a smartphone or tablet (Conell et al. 2016). Older adults are diverse, differing in age, education, income, living situation, employment, and experience with technology. Notably, in a US survey of 567 adults age 60 or older, those who used the Internet were often com- fortable doing so, and may have a job requiring computer use (Chang et al. 2015). In an international survey, infor- mation technology professionals over age 50 experienced less trouble working with multiple devices than younger workers (Patrizio 2016). However, as in this study, many older adults with or without bipolar disorder do not use the Internet. The reasons are complex and include dif- ficulty learning technical skills, high costs of comput- ers, mobile devices and broadband services, attitudes towards technology, increasing age, language issues for immigrants, cognitive decline, preference for traditional media, low health literacy, and relocation to a nursing home (Kuerbis et al. 2017; Fischer et al. 2014; Levine et al. 2018; Levy et al. 2015; Nimrod 2017; Chang et al. 2015). Some older adults are concerned that technology use will reduce face-to-face interactions, including contact with health care providers, and increase isolation (Kang et al. 2010; Kuerbis et al. 2017). In addition to financial fraud, older adults may fall vic- tim to risky online medical activities. Author details There are some limitations to this report. The survey was not designed to study technology habits of older adults. The study participants do not reflect the demo- graphic composition of the countries. People with bipo- lar disorder who did not seek professional help did not participate. People who did not understand the local language may not have participated. All data were self- reported and there was no follow-up discussion of responses. Many issues related to digital technology use by older adults were not discussed. These include the complex ethical challenges, quality of web sites and valid- ity of digital tools for bipolar disorder (Bauer M et al. 2017), physiological effects of blue light exposure from digital devices (Bauer et al. 2018), and the potential for digital assistive tools to erode skills, decrease motivation, and promote a false sense of security (Schulz et al. 2015). 1 Department of Psychiatry and Psychotherapy, University Hospital Carl Gustav Carus, Technische Universität Dresden, Dresden, Germany. 2 ChronoRecord Association, Fullerton, CA, USA. 3 Bipolar Disorder Program, Neuroscience Insti‑ tute, Favaloro University, Buenos Aires, Argentina. 4 AMEOS Klinika Holstein, Neustadt, Germany. 5 Department of Psychiatry, Dalhousie University, Halifax, NS, Canada. 6 Unit of Clinical Pharmacology, University Hospital of Cagli‑ ari, Cagliari, Italy. 7 Discipline of Psychiatry, School of Medicine, University of Adelaide, Adelaide, SA, Australia. 8 School of Medicine, IMPACT Strategic Research Centre, Deakin University, Geelong, VIC, Australia. 9 University Hospi‑ tal Geelong, Barwon Health, Geelong, VIC, Australia. 10 Department of Psy‑ chiatry, The University of Melbourne, Parkville, VIC, Australia. 11 Florey Institute of Neuroscience and Mental Health, Parkville, VIC, Australia. 12 Orygen Youth Health Research Centre and the National Centre of Excellence in Youth Mental Health, Parkville, VIC, Australia. 13 Department of Psychiatry, Faculty of Health Sciences, Ben Gurion University of the Negev; Beer Sheva Mental Health Center, Beer Sheva, Israel. 14 Department of Psychiatry, University of Oxford, Warneford Hospital, Oxford, UK. 15 Section of Neurosciences and Clinical Pharmacology, Department of Biomedical Sciences, University of Cagliari, Cagliari, Sardinia, Italy. 16 Department of General Adult Psychiatry, Castle Peak Hospital, Hong Kong, China. 17 AP–HP, Hôpitaux Universitaires Henri-Mondor, Créteil, France. 18 Department of Molecular Medicine and Department of Mental Health (DAI), University of Siena and University of Siena Medical Center (AOUS), Siena, Italy. 19 Bipolar Disorder Research Program, Depart‑ ment of Psychiatry, University of São Paulo Medical School, São Paulo, Brazil. Author details 20 Department of Psychiatry, University of Calgary, Calgary, Canada. 21 AP–HP, Hôpitaux Universitaires Henri‑Mondor, INSERM U955 (IMRB), Université Paris Est, Créteil, France. 22 Department of Psychiatry, University Hospital of Alava, University of the Basque Country, CIBERSAM, Vitoria, Spain. 23 Department of Affective Disorders, Q, Mood Disorders Research Unit, Aarhus University Hospital, Aarhus, Denmark. 24 Department of Psychiatry and Biobehavioral Sciences, Semel Institute for Neuroscience and Human Behavior, University of California Los Angeles (UCLA), Los Angeles, CA, USA. 25 Mood Disorders Center of Ottawa, Ottawa, Canada. 26 Department of Psychiatry, Univer‑ sity of Toronto, Toronto, ON, Canada. 27 Department of Psychiatry, Tokyo Metropolitan Matsuzawa Hospital, Setagaya, Tokyo, Japan. 28 Department of Psychiatry, Cumming School of Medicine, University of Calgary, Calgary, Canada. 29 Unité Perception et Mémoire, Institut Pasteur, F‑75015 Paris, France. 30 Bipolar Disorders Program, Hospital Clinic, University of Barcelona, IDIBAPS, CIBERSAM, Barcelona, Catalonia, Spain. 31 Department of Clinical Psychology, NIMHANS, Bangalore 560029, India. 32 Leicestershire Partnership NHS Trust, Leicester, UK. 33 Institute of Clinical Research, Research Unit of Psychiatry, Uni‑ versity of Southern Denmark, Odense, Denmark. 34 Department of Psychiatry, Psychiatry in the Region of Southern Denmark, Odense, Denmark. 35 Aalborg University Hospital, Psychiatry, Aalborg, Denmark. 36 Department of Clinical Medicine, Aalborg University, Aalborg, Denmark. 37 Department of Psychiatry, Wroclaw Medical University, Wroclaw, Poland. 38 Department of Neuroscience and Mental Health, Federal University of Bahia, Salvador, Brazil. 39 Michigan State University College of Human Medicine, Traverse City Campus, Traverse City, MI, USA. 40 Department of Psychiatry, University of California San Diego, San Diego, CA, USA. 41 Affective Disorders Research Project, Tokyo It is important to remember that technology will keep evolving (Arthur 2010). There will be disparities in the adaption of the new products and services, leaving digi- tal equality a continuously moving target (Hilbert 2014, 2016). Young adults of today who are very comfortable using smartphones will continue to use smartphones as they age, and struggle with the new technologies avail- able when they are seniors. The need to respect genera- tional differences in the preferred means to access health information, minimize the burden of new technologies on older adults, and implement programs that expand access and lessen the negative impacts of digital inequali- ties will remain in the future. Authors’ contributions JC RB TG d MB d i JC, RB, TG and MBa designed the study. MA, RA, RB, MB, YB, AB, APC, EYWC, CC, SC, JC, AC, RD, SD, AD, BE, AF, JG, JG, JG, MJG, AG, GMG, PG, HH, SH, CH, DH-M, VK, GK, BL, CL, ERL, UL, RW, AHL, ÂM-S, BM, SM, RM, TN, REN, CO’D, YO, YO, PP, AR, PR, JKR, KS, BS, ES, CS, SS, AHS, KS, AS, PT, YT, LT, JV, EV, MV, BV, and MZ were involved with data collection. TG provided data analysis. RB, JC, TG, PCW and MBa were involved in the draft manuscript and initial review. All authors read and approved the final manuscript. Discussion Older adults who are seeking to save money by purchasing expensive pre- scription drugs online will primarily be presented with rogue pharmacies that do not require a prescription (Monteith and Glenn 2017; Monteith et al. 2016). Some of the risks of using rogue pharmacies include counter- feit drugs, low-quality drugs, unapproved drugs, substi- tutions of strengths and formulations, drug interactions, adverse reactions, and financial fraud (Mackey and Nay- yar 2016; Mackey and Liang 2011; GAO 2014). Another problem area involves the online advertising of unneces- sary or inappropriate medical screening tests that are not included in evidence-based guidelines (Lovett et al. 2012; Lovett and Mackey 2013). Regardless of Internet use, older adults view health care professionals as the primary and most trusted source of information (Hall et al. 2015; Medlock et al. 2015). Patients of all ages would prefer to learn about a serious mental illness by direct conversation with their Older adults with physical limitations use the Inter- net less frequently than healthier older adults (Gell et al. Bauer et al. Int J Bipolar Disord (2018) 6:20 Page 5 of 7 2015; Levine et al. 2018). Many older adults have vision, hearing, and dexterity impairments. Assistive tech- nologies offer innovative options to get connected such as low-vision software for oversized monitors, speech amplification phones using landlines, and tremor stabi- lizing mouse controls (BT 2013; Watanabe et al. 2015; Fischer et al. 2014). More emphasis is needed on finding the optimal individualized approach for older adults to use digital technology rather than focusing on standard mobile devices (Fischer et al. 2014; Kuerbis et al. 2017). Additionally, technology approaches that combine data from those with and without Internet access, such as interactive voice response (IVR), should be considered for projects involving older adults (Verma et al. 2014; Pie- tte et al. 2013).hh remember that many older adults do not use the Inter- net. As the population is aging and more health services are only available online, there is concern about growing health disparities for older adults with bipolar disorder. Mental health experts should contribute to defining the appropriate role for technologies in the care of older adults with bipolar disorder. Competing interests Michael Berk is supported by a National Health and Medical Research Council (NHMRC) Senior Principal Research Fellowship (Grant Number 1059660). René E. Nielsen has received research grants from H. Lundbeck and Otsuka Pharmaceuticals for clinical trials, received speaking fees from Bristol- Myers Squibb, Astra Zeneca, Janssen and Cilag, Lundbeck, Servier, Otsuka Pharmaceuticals, and Eli Lilly and has acted as advisor to Astra Zeneca, Eli Lilly, Lundbeck, Otsuka Pharmaceuticals, Takeda, and Medivir. Michael Berk is supported by a National Health and Medical Research Council (NHMRC) Senior Principal Research Fellowship (Grant Number 1059660). Burnes D, Henderson CR Jr, Sheppard C, Zhao R, Pillemer K, Lachs MS. Reva‑ lence of financial fraud and scams among older adults in the United States: a systematic review and meta-analysis. Am J Public Health. 2017;107:e13–21. Carlson EL. Phishing for elderly victims: as the elderly migrate to the Internet fraudulent schemes targeting them follow. Elder Law J. 2007;14:423–52. CDC. CDC Brief addressing financial exploitation among people living with cognitive impairment and their caregivers: role of the public health and aging services networks. 2015. https://www.cdc.gov/aging/pdf/exploitati on-cognitive-impairment-brief-july-2015.pdf. Accessed 30 Mar 2018. Rasmus W. Licht has received research grant from Glaxo Smith Kline, honoraria for lecturing from Pfizer, Glaxo Smith Kline, Eli Lilly, Astra-Zeneca, Bristol-Myers Squibb, Janssen Cilag, Lundbeck, Otsuka, Servier and honoraria from advisory board activity from Glaxo Smith Kline, Eli Lilly, Astra-Zeneca, Bristol-Myers Squibb, Janssen Cilag, and Sunovion. The other authors declare that they have no competing interests. Chang J, McAllister C, McCaslin R. Correlates of, and barriers to, Internet use among older adults. J Gerontol Soc Work. 2015;58:66–85. Choi NG, Dinitto DM. Internet use among older adults: association with health needs, psychological capital, and social capital. J Med Internet Res. 2013;15:e97. Acknowledgements British Telecommunications. Communication choices for deaf or hard of hear‑ ing people. 2013. https://btplc.com/inclusion/HelpAndSupport/Docum entsandDownloads/Communicationchoices/Fordeaforhardofhearingpeo ple/Communication_Choices_Deaf_Hard_Hearing.pdf. Accessed 30 Mar 2018. We thank Daniela Jany for preparing submission files and assisting with manuscript submission. We thank Daniela Jany for preparing submission files and assisting with manuscript submission. Funding We acknowledge support by the Open Access Publication Funds of the SLUB/ TU Dresden (Grant No. IN-1502335). Flynn KE, Smith MA, Freese J. When do older adults turn to the Internet for health information? Findings from the Wisconsin longitudinal study. J Gen Intern Med. 2006;21(12):1295–301. Availability of data and materials Conell J, Bauer R, Glenn T, Alda M, Ardau R, Baune BT, et al. Online information seeking by patients with bipolar disorder: results from an international multisite survey. Int J Bipolar Disord. 2016;4:17. The survey questionnaire was published. Sharing of data beyond the study was not approved by ethics boards or participants. The survey questionnaire was published. Sharing of data beyond the study was not approved by ethics boards or participants. Deloitte UK. Connected health. How digital technology is transforming health and social care. 2015. https://www2.deloitte.com/content/dam/Deloitte/ uk/Documents/life-sciences-health-care/deloitte-uk-connected-healt h.pdf. Accessed 30 Mar 2018.f Author details In conclusion, the finding that many older adults with bipolar disorder do not use the Internet confirms the need for further investigation of technology habits, the efficacy of digital tools, and how best to determine who will use these tools appropriately and safely. Today, tech- nology based tools and treatments are suitable for some but not all older adults with bipolar disorder. With gov- ernment and health care providers increasingly rely- ing on electronic communication, it is important to Bauer et al. Int J Bipolar Disord (2018) 6:20 Page 6 of 7 Bauer et al. Int J Bipolar Disord (2018) 6:20 Metropolitan Institute of Medical Science, Setagaya, Tokyo, Japan. 42 Depart‑ ment of Psychiatry, Psychosomatic Medicine and Psychotherapy, University Hospital Frankfurt, Goethe-University Frankfurt am Main, Frankfurt, Germany. 43 Department of Adult Psychiatry, Poznan University of Medical Sciences, Poznan, Poland. 44 Department of Psychiatry, University of Missouri Kansas City School of Medicine, Kansas City, MO, USA. 45 Croton on Hudson, NY, USA. 46 Department of Psychological Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia. 47 City of Helsinki, Department of Social Services and Health Care, Psychiatry, Helsinki, Finland. 48 Department of Psy‑ chiatry, Department of Medicine, University of Hong Kong, Hong Kong, China. 49 McLean Hospital and Harvard Medical School, Boston, MA, USA. 50 Lucio Bini Center, Cagliari, Rome, Italy. 51 Psychiatric Center Copenhagen, Copenhagen, Denmark. 52 Department of Psychiatry, NIMHANS, Bangalore 560029, India. 53 Department of Psychology, Chapman University, Orange, CA, USA. Arthur WB. What is technology and how does it evolve? New York Academy of Sciences Magazine. 2010. https://www.nyas.org/magazines/winter-2010/ what-is-technology-and-how-does-it-evolve/. Accessed 6 July 2018. Bauer M, Glenn T, Monteith S, Bauer R, Whybrow PC, Geddes J. Ethical perspec‑ tives on recommending digital technology for patients with mental illness. Int J Bipolar Disord. 2017;5:6. Bauer M, Glenn T, Monteith S, Gottlieb JF, Ritter PS, Geddes J, Whybrow PC. The potential influence of LED lighting on mental illness. World J Biol Psychiatry. 2018;19:59–73. Bauer R, Conell J, Glenn T, Alda M, Ardau R, Baune BT, et al. Internet use by patients with bipolar disorder: results from an international multisite survey. Psychiatry Res. 2016;242:388–94. Bauer R, Conell J, Glenn T, Alda M, Ardau R, Baune BT, et al. International multi- site survey on the use of online support groups in bipolar disorder. Nord J Psychiatry. 2017;71:473–6. Ethics approval and consent to participate The study was approved by institutional review boards according to local requirements. FBI Internet Crime Center. 2016 Internet crime report. 2016. https://pdf.ic3. gov/2016_IC3Report.pdf. Accessed 30 Mar 2018. Fischer SH, David D, Crotty BH, Dierks M, Safran C. Acceptance and use of health information technology by community-dwelling elders. Int J Med Inform. 2014;83:624–35. Consent for publication All authors contributed to and approved the final manuscript, and gave their consent for publication. All authors contributed to and approved the final manuscript, and gave their consent for publication. FBI. FBI efforts to combat elder fraud. 2014. https://www.fbi.gov/news/testi mony/fbi-efforts-to-combat-elder-fraud. Accessed 30 Mar 2018. Publisher’s Note Forsman AK, Nordmyr J. Psychosocial links between Internet use and mental health in later life: a systematic review of quantitative and qualitative evidence. J Appl Gerontol. 2017;36:1471–518. Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Friemel TN. The digital divide has grown old: determinants of a digital divide among seniors. New Media Soc. 2016;18:313–31. Received: 14 May 2018 Accepted: 1 August 2018 Received: 14 May 2018 Accepted: 1 August 2018 Gallagher S, Doherty DT. Searching for health information online: characteris‑ tics of online health seekers. J Evid Based Med. 2009;2:99–106. GAO (United States Government Accountability Office). Internet pharma‑ cies. Most rogue sites operate from abroad, and many sell counterfeit drugs. GAO-14-386T. 2014. http://www.gao.gov/assets/670/661177.pdf. Accessed 30 Mar 2018. References The bad news is that the digital access divide is here to stay: domestically installed bandwidths among 172 countries for 1986–2014. Telecomm Policy. 2016;40:567–81. Park SG, Derman M, Dixon LB, Brown CH, Klingaman EA, Fang LJ, et al. Factors associated with shared decision-making preferences among veterans with serious mental illness. Psychiatr Serv. 2014;65:1409–13. Hilbert M. Technological information inequality as an incessantly moving target: the redistribution of information and communication capacities between 1986 and 2010. J Assoc Inf Sci Technol. 2014;65:821–35. Patrizio A. Think older workers struggle with technology? Think again, CIO. 2016. https://www.cio.com/article/3103893/it-industry/think-older -workers-struggle-with-technology-think-again.html. Accessed 30 Mar 2018. Holtfreter K, Reisig MD, Pratt TC, Holtfreter RE. Risky remote purchasing and identity theft victimization among older Internet users. Psychol Crime Law. 2015;21:681–98. Piette JD, Rosland AM, Marinec NS, Striplin D, Bernstein SJ, Silveira MJ. Engage‑ ment with automated patient monitoring and self-management support calls: experience with a thousand chronically ill patients. Med Care. 2013;51:216–23. Home Instead. Survey: Two-thirds of seniors have been scammed online. HomeInstead. 2017. https://www.homeinstead.com/370/blog/surve y-two-thirds-of-seniors-have-been-scammed-online. Accessed 30 Mar 2018. Powell J, Clarke A. Internet information-seeking in mental health: population survey. Br J Psychiatry. 2006;189:273–7. Hou J, Shim M. The role of provider-patient communication and trust in online sources in Internet use for health-related activities. J Health Commun. 2010;15(Suppl 3):186–99. Reeder B, Meyer E, Lazar A, Chaudhuri S, Thompson HJ, Demiris G. Framing the evidence for health smart homes and home-based consumer health technologies as a public health intervention for independent aging: a systematic review. Int J Med Inform. 2013;82:565–79. Kang HG, Mahoney DF, Hoenig H, Hirth VA, Bonato P, Hajjar I, et al. In situ moni‑ toring of health in older adults: technologies and issues. J Am Geriatr Soc. 2010;58:1579–86. Sajatovic M, Strejilevich SA, Gildengers AG, Dols A, Al Jurdi RK, Forester BP, et al. A report on older-age bipolar disorder from the international society for bipolar disorders task force. Bipolar Disord. 2015;17:689–704. Kobayashi LC, Wardle J, von Wagner C. Internet use, social engagement and health literacy decline during ageing in a longitudinal cohort of older English adults. J Epidemiol Community Health. 2015;69:278–83. Sbaffi L, Rowley J. Trust and credibility in web-based health information: a review and agenda for future research. J Med Internet Res. 2017;19:e218. Kuerbis A, Mulliken A, Muench F, Moore AA, Gardner D. Older adults and mobile technology: factors that enhance and inhibit utilization in the context of behavioral health. Ment Health Addict Res. 2017. References https://doi. org/10.15761/MHAR.1000136. Schulz R, Wahl HW, Matthews JT, De Vito Dabbs A, Beach SR, Czaja SJ. Advanc‑ ing the aging and technology agenda in gerontology. Gerontologist. 2015;55:724–34. Torous J, Summergrad P, Nassir Ghaemi S. Bipolar disorder in the digital age: new tools for the same illness. Int J Bipolar Disord. 2016;4:25. Lala SV, Sajatovic M. Medical and psychiatric comorbidities among elderly individuals with bipolar disorder: a literature review. J Geriatr Psychiatry Neurol. 2012;25:20–5. Verma SK, Courtney TK, Lombardi DA, Chang WR, Huang YH, Brennan MJ, et al. Internet and telephonic IVR mixed-mode survey for longitudinal studies: choice, retention, and data equivalency. Ann Epidemiol. 2014;24:72–4. Levine DM, Lipsitz SR, Linder JA. Trends in seniors’ use of digital health technol‑ ogy in the United States, 2011–2014. JAMA. 2016;316:538–40. Watanabe T, Yamaguchi T, Minatani K. Advantages and drawbacks of smart‑ phones and tablets for visually impaired people: analysis of ICT user survey results. IEICE Trans Inform Syst. 2015;98:922–9. Levine DM, Lipsitz SR, Linder JA. Changes in everyday and digital health technology use among seniors in declining health. J Gerontol A Biol Sci Med Sci. 2018;73:552–9. survey results. IEICE Trans Inform Syst. 2015;98:922–9. Levy H, Janke AT, Langa KM. Health literacy and the digital divide among older Americans. J Gen Intern Med. 2015;30:284–9. White CM, Gummerum M, Wood S, Hanoch Y. Internet safety and the silver surfer: the relationship between gist reasoning and adults’ risky online behavior. J Behav Decis Making. 2017;30:819–27. Lichtenberg PA, Sugarman MA, Paulson D, Ficker LJ, Rahman-Filipiak A. Psycho‑ logical and functional vulnerability predicts fraud cases in older adults: results of a longitudinal study. Clin Gerontol. 2016;39:48–63. g WHO. Ageing and health. Fact Sheet 404. 2015. http://www.who.int/media centre/factsheets/fs404/en/. Accessed 30 Mar 2018. Lovett KM, Mackey TK. Online threats to senior safety: the direct-to-consumer medical marketplace and elder abuse. NAELA J. 2013;9:91–113. Yu RP, Ellison NB, McCammon RJ, Langa KM. Mapping the two levels of digital divide: Internet access and social network site adoption among older adults in the USA. Inf Commun Soc. 2016;19:1445–64. Lovett KM, Mackey TK, Liang BA. Evaluating the evidence: direct-to-consumer screening tests advertised online. J Med Screen. 2012;19:141–53. Lovett KM, Mackey TK, Liang BA. Evaluating the evidence: direct-to-consumer screening tests advertised online. J Med Screen. 2012;19:141–53. Mackey TK, Liang BA. The global counterfeit drug trade: patient safety and public health risks. J Pharm Sci. 2011;100:4571–9. Zulman DM, Kirch M, Zheng K, An LC. References Gell NM, Rosenberg DE, Demiris G, LaCroix AZ, Patel KV. Patterns of technol‑ ogy use among older adults with and without disabilities. Gerontologist. 2015;55:412–21. Gliddon E, Barnes SJ, Murray G, Michalak EE. Online and mobile technologies for self-management in bipolar disorder: a systematic review. Psychiatr Rehabil J. 2017;40:309–19. Grimes GA, Hough MG, Mazur E, Signorella ML. Older adults’ knowledge of Internet hazards. Educ Gerontol. 2010;36:173–92. Page 7 of 7 Bauer et al. Int J Bipolar Disord (2018) 6:20 Hall AK, Bernhardt JM, Dodd V. Older adults’ use of online and offline sources of health information and constructs of reliance and self-efficacy for medical decision making. J Health Commun. 2015;20:751–8. Medlock S, Eslami S, Askari M, Arts DL, Sent D, de Rooij SE, et al. Health infor‑ mation—seeking behavior of seniors who use the Internet: a survey. J Med Internet Res. 2015;17:e10. Hallett C, Gupta S, Priebe S. What do outpatients with schizophrenia and mood disorders want to learn about their illness? Psychiatr Serv. 2013;64:764–9. Monteith S, Glenn T. Automated decision-making and big data: concerns for people with mental illness. Curr Psychiatry Rep. 2016;18:112. Monteith S, Glenn T, Bauer R, Conell J, Bauer M. Availability of prescrip‑ tion drugs for bipolar disorder at online pharmacies. J Affect Disord. 2016;193:59–65. Heo J, Chun S, Lee S, Lee KH, Kim J. Internet use and well-being in older adults. Cyberpsychol Behav Soc Netw. 2015;18:268–72. Hidalgo-Mazzei D, Mateu A, Reinares M, Matic A, Vieta E, Colom F. Internet- based psychological interventions for bipolar disorder: review of the present and insights into the future. J Affect Disord. 2015;188:1–13. Hilbert M. The bad news is that the digital access divide is here to stay: d ti ll i t ll d b d idth 172 t i f 1986 2014 Hidalgo-Mazzei D, Mateu A, Reinares M, Matic A, Vieta E, Colom F. Internet- based psychological interventions for bipolar disorder: review of the Monteith S, Glenn T. Searching online to buy commonly prescribed psychiatric drugs. Psychiatry Res. 2017;260:248–54. Hidalgo-Mazzei D, Mateu A, Reinares M, Matic A, Vieta E, Colom F. Internet- based psychological interventions for bipolar disorder: review of the present and insights into the future. J Affect Disord. 2015;188:1–13. p y g p present and insights into the future. J Affect Disord. 2015;188:1–13. Nimrod G. Older audiences in the digital media environment. Inf Commun Soc. 2017;20:233–49. Hilbert M. References Trust in the Internet as a health resource among older adults: analysis of data from a nationally representative survey. J Med Internet Res. 2011;13:e19. Mackey TK, Liang BA. The global counterfeit drug trade: patient safety and public health risks. J Pharm Sci. 2011;100:4571–9. Mackey TK, Nayyar G. Digital danger: a review of the global public health, patient safety and cybersecurity threats posed by illicit online pharma‑ cies. Br Med Bull. 2016;118:110–26.
https://openalex.org/W4320342602	https://zenodo.org/records/7425286/files/4.9._%D0%90%D0%B1%D1%80%D0%B0%D0%BC%D0%BE%D0%B2%20%D0%90.%D0%95._14.12.pdf	Russian	null	INVESTMENT RISKS MANAGEMENT OF PRIVATE INVESTORS IN THE RUSSIAN STOCK MARKET	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	13,907	Научно-исследовательская работа выполнена в соответствии с Государственным заданием РАНХиГС на 2021 год Научно-исследовательская работа выполнена в соответствии с Государственным заданием РАНХиГС на 2021 год Абрамов А. Е., к. э. н., заведующий научно-исследовательской лабораторией анализа институтов и финансовых рынков ИПЭИ Российской академии народного хозяйства и государственной службы при Президенте РФ, 0000-0003-4285-9115, abramov-ae@ranepa.ru. Радыгин А. Д., д. э. н., проф., декан экономического факультета Российской академии народного хозяйства и государственной службы при Президенте РФ, член совета директоров Института экономической политики имени Е. Т. Гайдара (Москва), 0000-0003-2242-9994, arad@ranepa.ru. Чернова М. И., научный сотрудник лаборатории анализа институтов и финансовых рынков ИПЭИ Российской академии народного хозяйства и государственной службы при Президенте РФ, 0000-0003-0144-1820, chernova- mi@ranepa.ru Москва, 2021 FEDERAL STATE BUDGETARY EDUCATIONAL INSTITUTION OF HIGHER EDUCATION " Russian Presidential Academy of National Economy and Public Administration" (RANEPA) INVESTMENT RISKS MANAGEMENT OF PRIVATE INVESTORS IN THE RUSSIAN STOCK MARKET The study was prepared as part of the state research order of the RANEPA for 2021 Alexander E. Abramov – Head of Department for Analysis of Institutions and Financial Markets of the Russian Presidential Academy of National Economy and Public Administration, Candidate of Economic Sciences (Moscow, Russia), 0000-0003-4285-9115, abramov-ae@ranepa.ru. Alexander D. Radygin – Chairman of the Scientific Council of the Gaidar Institute for Economic Policy; Director of the Institute of Economics, Mathematics and Information Technology, Russian Presidential Academy of National Economy and Public Administration, Doctor of Economic Sciences, Professor (Moscow, Russia), 0000-0003- 2242-9994, arad@ranepa.ru. Maria I. Chernova – Researcher of the Russian Presidential Academy of National Economy and Public Administration (Moscow, Russia), 0000-0003-0144-1820, chernova- mi@ranepa.ru. Moscow, 2021 Moscow, 2021 2 2 Abstract The goal of this study is to develop a system of measures to increase investment activity and limit the risks of the private investors in the stock market. The object of research is the behavior of private investors in the formation of their individual portfolios. The research method includes analysis of Bloomberg’s historical data series on the return and risk of financial instruments, various methods for quantifying the investor's risk profile. The work features a comparative analysis of various approaches to assessing the individual risk profile of private investors based on analyzing the international experience. The current financial market regulation system is improving slowly and in many ways lags behind the changes in the stock market. The legal prerequisites for the accumulation of long-term savings by the population have not yet been created, the measures proposed by the regulator to limit investments and reduce risk exposure for certain categories of citizens cause serious debate in society. The relevance of the study is explained by a lack of theoretical and empirical studies of the specifics of individual investments in the domestic stock market, the behavior of private investors and the possibilities of risk management in asset allocation. The novelty of the study lies in adapting the best risk profiling practices for Russian private investors, identifying recommendations on asset allocation across all available time horizons, and demonstrating the benefits of diversification not only in the domestic but also in the global market. The main result of the work is the development of methodological approaches necessary to ensure that the generated individual portfolios of private investors match their risk profile. In addition, the main results of the work include the identified advantages of the portfolio approach, in terms of both reducing risks and increasing returns. The goal of this study is to develop a system of measures to increase investment activity and limit the risks of the private investors in the stock market. The object of research is the behavior of private investors in the formation of their individual portfolios. The research method includes analysis of Bloomberg’s historical data series on the return and risk of financial instruments, various methods for quantifying the investor's risk profile. The work features a comparative analysis of various approaches to assessing the individual risk profile of private investors based on analyzing the international experience. Аннотация Целью данного исследования является разработка системы мер, позволяющих повысить инвестиционную активность и ограничить риски населения на фондовом рынке. В качестве объекта исследования выступают отношения, связанные с поведением частных инвесторов при формировании их индивидуальных портфелей. В качестве метода исследования – анализ исторических рядов данных доходности и риска финансовых инструментов, различные методы количественной оценки риск- профиля инвестора на основе данных Блумберг. В работе проводился сравнительный анализ подходов к оценке индивидуального риск-профиля частных инвесторов на основе анализа международного опыта. Действующая система регулирования финансового рынка совершенствуется медленно и во многом отстает от тех изменений, которые происходят на фондовом рынке. До сих пор не созданы необходимые правовые условия для накопления долгосрочных сбережений граждан, предлагаемые регулятором меры по ограничениям инвестиций отдельных категорий граждан и их риск-профилированию вызывают серьезные дискуссии в обществе. Актуальность исследования обусловлена дефицитом теоретических и эмпирических исследований специфики индивидуальных инвестиций на внутреннем фондовом рынке, особенностей поведения частных инвесторов и возможностей управления рисками при формировании индивидуальных портфелей граждан. Новизна исследования заключается в адаптации лучших практик риск-профилирования для российских частных инвесторов, выявлении рекомендаций о распределении активов на всех доступных временных горизонтах и демонстрации преимуществ диверсификации не только на внутреннем, но и на глобальном рынке. Основным результатом работы является разработка методологических подходов, необходимых для того, чтобы формируемые индивидуальные портфели ценных бумаг частных инвесторов более точно соответствовали их персональному риск-профилю. Кроме того, к основным результатам работы можно отнести выявленные преимущества портфельного подхода как с точки зрения снижения рисков, так и с точки зрения повышения доходности. Составлена карта рисков инструментов доступных частному инвестору. А также выявлены основные правила и закономерности для инвестирования на российском рынке в долгосрочном периоде. Основным выводом работы можно считать то, что для российского инвестора крайне важным остается как 3 использование факторных и широко диверсифицированных портфелей, так и глобальная диверсификация вне зависимости от профиля риска и горизонта. В дальнейшем развитие нашей работы, направленное на изучение частного инвестирования в различных странах, должно способствовать развитию института частного инвестирования в России. Ключевые слова: риск-профилирование, частные инвесторы, частные сбережения, инвестирование, поведение инвесторов, распределение активов, долгосрочное инвестирование, глобальная диверсификация. JEL: G11, G23, G41 Abstract The current financial market regulation system is improving slowly and in many ways lags behind the changes in the stock market. The legal prerequisites for the accumulation of long-term savings by the population have not yet been created, the measures proposed by the regulator to limit investments and reduce risk exposure for certain categories of citizens cause serious debate in society. The relevance of the study is explained by a lack of theoretical and empirical studies of the specifics of individual investments in the domestic stock market, the behavior of private investors and the possibilities of risk management in asset allocation. The novelty of the study lies in adapting the best risk profiling practices for Russian private investors, identifying recommendations on asset allocation across all available time horizons, and demonstrating the benefits of diversification not only in the domestic but also in the global market. The main result of the work is the development of methodological approaches necessary to ensure that the generated individual portfolios of private investors match their risk profile. In addition, the main results of the work include the identified advantages of the portfolio approach, in terms of both reducing risks and increasing returns. 4 A risk map of instruments available to a private investor has been compiled. We identified the basic rules and patterns for investing in the Russian market in the long term. The main conclusion of the work is that it is extremely important for the Russian investor to use both factor-based and widely diversified portfolios, as well as global diversification regardless of the risk profile and horizon. In the future, the development of our work aimed at studying private investment in various countries will contribute to the development of the institution of private investment in Russia. Keywords: risk profiling, private investors, private savings, investing, investor behavior, asset allocation, long-term investment, global diversification JEL: G11, G23, G41 Введение Современная модель экономического роста предполагает активное использование потенциала внутренних инвестиций, основанных не только на государственных, но и частных сбережений. За последние два года в силу разных причин наблюдается процесс прихода на внутренний фондовый рынок массового инвестора и перераспределения финансовых активов домашних хозяйств от банковских депозитов в пользу вложений в акции и облигации. Однако действующая система регулирования финансового рынка совершенствуется медленно и во многом отстает от тех изменений, которые происходят на фондовом рынке. До сих пор не созданы необходимые правовые условия для накопления долгосрочных сбережений граждан, предлагаемые регулятором меры по ограничениям инвестиций отдельных категорий граждан и их риск-профилированию вызывают серьезные дискуссии в обществе. Во многом это обусловлено дефицитом теоретических и эмпирических исследований специфики индивидуальных инвестиций на внутреннем фондовом рынке, особенностей поведения частных инвесторов и возможностей управления рисками при формировании индивидуальных портфелей граждан. Стабильность данных сбережений во многом зависит от того, насколько параметры доходности- риска финансовых инструментов, приобретаемых инвестором, будут соответствовать 5 его ожиданиям и склонности к риску. Актуальность исследования заключается в разработке методологических подходов, необходимых для того, чтобы формируемые индивидуальные портфели ценных бумаг частных инвесторов более точно соответствовали их персональному риск-профилю. Актуальность работы заключается в том, что в последнее время широкую популярность начинают приобретать различные виды частного инвестирования. В первую очередь это связано с развитием мобильных каналов инвестирования (приложения банков и т.п.) в том числе с минимизированием бюрократической составляющей процедуры торговли, так и со снижением доходов от депозитов. Новизна заключается в разработке методологических подходов, необходимых для того, чтобы формируемые индивидуальные портфели ценных бумаг частных инвесторов более точно соответствовали их персональному риск-профилю. Также предполагается оценить действующую практику оценки индивидуальных профилей инвесторов российскими и зарубежными финансовыми организациями. В работе применяются методологические подходы к классификации финансовых инструментов и оценке индивидуального профиля риска частных инвесторов, регрессионные модели анализа факторов, влияющих на склонность к сбережениям частных инвесторов, регрессионные модели факторов, влияющих поведение частных инвесторов. Проводится сравнительный анализ подходов к оценке индивидуального профиля инвесторов, применяемых в компаниях в России и в других странах. При написании работы нами сформулированы предложения по регулированию структуры активов индивидуальных инвесторов. Предложены меры по совершенствованию правовой базы в сфере защиты прав частных инвесторов. Кроме того, нами разработаны методики оценки риск-профиля инвесторов для совершения сделок на внутреннем фондовом рынке России. В работе показаны преимущества портфельного подхода как с точки зрения снижения рисков, так и с точки зрения повышения доходности и снижения эффекта ошибки выжившего на рынке акций 6 6 риску Истоки теории неприятия риска инвесторами и эффекты различного уровня толерантности риска на финансовые решения домохозяйств с учетом изменения предпочтений во времени и поведенческих аспектов описаны в достаточно большом кластере научных исследований. Работа Финке и Жульмет [1] является одним и примеров качественного обзора подобного рода литературы. Авторы помимо теоретического обзора описывают способы измерения предпочтений риска инвесторами и предлагают ряд важных рекомендаций к составлению анкет для выявления профиля риска инвестора. Измерение коэффициента неприятия риска для домохозяйств на данный момент не имеет общепринятого алгоритма и методологии. Одним из подходов являются расчеты на основе модели ценообразования активов (CAPM) и ее модификации с учетом потребление (Consumption CAPM). Например, в работе Hansen and Singleton [2] показано с помощью обобщенного метода моментов, что относительное неприятие риска достаточно мало, а Hall [3] и Neely et al. [4] указывают на то, что с помощью такого метода достаточно сложно получить устойчивые оценки из-за невозможности предсказать ожидаемые доходности активов и рос потребления. Другие исследователи (например, Layard et al. [5]) используют итерационную процедуру метода максимального правдоподобия и оценивают функцию полезности и постоянный относительный коэффициент неприятия риска (constant relative risk aversion, CRRA). Согласно этому подходу, предельная полезность дохода соответствует коэффициенту неприятия риска. Многофакторный анализ неприятия риска с учетом таких контрольных переменных, как доход, образование и прочих, подтверждает, что неприятие риска меньше у индивидов с большим доходом, а высшее образование положительно влияет на толерантность к риску (Sung и Hanna, [6]; Grable [7]). Одним из фактором, определяющих толерантность к риску, является финансовое образование и опыт. Значимое влияние образования на толерантность к риску выявляется эмпирическими исследованиями, хотя не подкреплено теорией. Лица, обладающие финансовой грамотностью, более охотно используют рисковые альтернативы. Причиной этого может быть наличие больших знаний о механизме 7 ценообразования, основных принципах финансовых рисков или понимание выгод больших рисков на долгосрочном периоде. Напротив, недостаток финансовой грамотности или опыта приводит к меньшей уверенности относительно риска инвестирования и неприятию неопределенности, к которой склонны люди. Van Rooij, Lusardi and Alessi [8] показывают, что финансовая грамотность и знание принципов диверсификации являются одним из наиболее ключевых предикторов наличия акций в портфеле индивида. Инвесторы склонны к изменениям своих ожиданий относительно выгод риска в зависимости от текущих рыночных условий. Так, финансовые консультанты отмечают рост толерантности к риску у клиентов в годы роста рыночного портфеля и роста неприязни риска в годы падения рыночных индексов (Yao и Curl [9]). риску Таким образом, при отсутствии единых количественных моделей, способных оценить толерантность к риску для инвестора, инвестиционные консультанты вынуждены прибегать к анкетам, которые через серию вопросов выявляли бы косвенным образом все основные параметры для составления профиля инвестора. Большой проблемой в этом случае является определение состава такой анкеты, которая смогла бы объективно и наиболее полно учесть все зависимости, выявляемые тоерией. профилированию в США Современные рекомендации по формированию риск-профиля клиента финансовыми консультантами написаны авторами CFA institute – Hubble, Grable, Dannhauser [10]. Они утверждают, что на текущий момент у каждого финансового консультанта есть свое представление о том, как формировать профиль риска клиента, а у регулятора – только основные пожелания и рекомендации о регулярности подобной оценки. Не существует стандартов или мер оценки эффективности текущих алгоритмов как с точки зрения отрасли, так и регулятора. Регулятором чаще публикуются некоторые общие правила, которые могут быть по-разному интерпретированы. По этим причинам риск-профилирование стало формальной процедурой, чаще ориентированной на определение краткосрочной толерантности к риску с помощью наиболее простых и удобных анкет и дополненной сбором типовых факторов: возраст 8 и финансовое положение клиента. При таком подходе в отрасли сформировалась среда, в которой инструменты риск-профилирования сильно отличаются между консультантами, как и последующие рекомендации относительно инвестиционной стратегии. Между тем устойчивые оценки профиля риска позволят консультантам подбирать наиболее подходящие под цели клиентов портфели. Наилучшей практикой является оценка риск-профиля, состоящего из трех обязательных компонент: потребности риска, способности принимать на себя риск и поведенческие аспекты толерантности к риску. Потребность в риске должна быть оценена исходя из необходимого уровня доходности (required rate of return, RoR) инвестиционного портфели, при которой достигаются цели клиента как по планируемому будущему стилю жизни, так и по другим сферам, например благотворительности или наследства. На основе ожидаемых индикаторов доходности рынков капитала и оцененной для клиента необходимой ставки доходности можно предложить ему стратегию распределения капитала, которая соответствует рыночному риску. Способность инвестора принимать на себя риск основана на оценке его временного горизонта, потенциальной потребности в ликвидности средств и степени приемлемого для него риска. Эти факторы определяют финансовую способность инвестора выдерживать снижения стоимости портфеля. Эта компонента профиля риска чаще является ограничивающим риск фактором при формировании портфеля согласно потребности в риске. Толерантность инвестора к риску может быть еще более ограничивающим фактором, который потенциально может сделать неприемлемой наиболее тщательно разработанную стратегию в соответствии с предыдущими двумя пунктами. Лучшей практикой для оценки этого фактора являются анкеты или психометрические тесты, которые достаточно надежны и обладают доказанной эмпирически прогнозной силой для эмоциональной и поведенческой реакции клиента на потери и соблюдение инвестиционной дисциплины. Особенно важный компонент – толерантность к риску. На практике у финансовых консультантов существует большое количество анкет и опросников, определяющих ее, однако важно учитывать основные принципы и правила их 9 составления. Так, любая анкета должна обладать двумя свойствами: обоснованность и надежность. Обоснованность – это оценка того, насколько хорошо предлагаемый инструмент измеряет нужный показатель. Существует ряд маркеров, который говорит о некорректности анкеты. профилированию в США Она не должна содержать вопросы о толерантности к риску вне контекста инвестирования, так как предпочтения риска специфичны для каждой конкретной жизненной ситуации. Пример неудачного вопроса: любит ли инвестор заниматься рисковыми видами спорта. Вопросы, которые направлены на попытку заставить инвестора оценить будущие действия, также не корректны, так как они не объективны, а инвесторы зачастую не могут корректно оценить свои навыки и предсказать поведение. Пример неудачного вопроса: если рынок в течение 6 месяцев упадет на 30%, что будет делать инвестор. Вопросы, которые предлагают выбор между альтернативами с известным распределением вероятностей, не корректны, так как на практике распределение вероятностей неизвестно, а список возможных альтернатив бесконечен. Пример неудачного вопроса: выбрать между пенсионным планом, который принесет аналогичный текущему доход, и планом, который с вероятностью 50% удвоит доход, а с вероятностью 50% уменьшит его вдвое. Некорректны вопросы, которые смещают мнением инвестора тем, что в них встроено некоторое понятие правильного ответа. Пример неудачного вопроса: много экспертов утверждает, что оптимальный вес акций равен разнице между 100% и возрастом инвестора, а какой для вас наиболее комфортный вес акций? Вопросы анкеты не должны иметь сложных или профессиональных понятий, должны быть выражены простым языком. Ответы на вопросы не могут быть полярными, то есть не корректно задавать вопрос с вариантом ответа «согласен – не согласен», нужны промежуточные варианты ответа. Второе главное свойство анкет – надежность, то есть способность генерировать один и тот же ответ при прочих равных для одного и того же инвестора. Оценка надежности основана на статистическом тесте, чаще на согласованность вопросов. Размер анкеты на выявление толерантности к риску также имеет значение. Финансовые консультанты предпочитают наиболее короткие анкеты для удобства клиентов, однако исследователи утверждают, что только при достаточно большом количестве вопросов можно перепроверить согласованность ответов и оценить все аспекты риск-профиля. Оптимальным считается количество вопросов от 7 до 30 [11]. 10 Важно не включать в такую анкету объективные вопросы, например, временной горизонт, возраст или потребность в ликвидности. Эти факторы важны сами по себе и не должны участвовать в некотором средневзвешенном балле для всей анкеты толерантности к риску. Когда оценены все три компоненты риск-профиля: потребность в риске, способность принимать риск и толерантность к риску, консультант может переходить к этапу анализа полученных значений и формировать рекомендации по составу портфеля. Преимущество создания риск-профиля заключается еще и в том, что в ходе оценки могут выявляться конфликтующие аспекты индивида. Примеры: 1. профилированию в США Потребность в риске не может превышать способность принимать риск для некоторой конкретной цели, а если это случается, то консультант должен вернуться к обсуждению цели с инвестором и либо уменьшить целевые значения, либо увеличить нормы сбережения. 2. Низкое значение потребности в риске может быть частично проигнорировано, если способность принимать риск и толерантность к риску высокие, а портфель может быть собран более рисковый. 3. Высокая способность принимать риск может быть проигнорирована, если потребность в риске и толерантность к риску низкие 4. Высокая толерантность к риску может быть проигнорирована, если потребность в риске и способность принимать риск низкие, а консультант может быть вынужден объяснить клиенту необходимость более сдержанно относиться к риску, преодолевая свою высокую толерантность к нему, которая не отвечает его потребностям и способностям. 5. Низкая толерантность к риску никогда не может быть проигнорирована, даже при высоких потребностях и способностях принимать риск, однако консультант может обучать и способствовать снижению неприятия риска для того, что подготовить клиента к более легкому восприятию рискового портфеля, который необходим для удовлетворения высоких потребности и способности. 5. Низкая толерантность к риску никогда не может быть проигнорирована, даже при высоких потребностях и способностях принимать риск, однако консультант может обучать и способствовать снижению неприятия риска для того, что подготовить клиента к более легкому восприятию рискового портфеля, который необходим для удовлетворения высоких потребности и способности. 6. Финансовый консультант не должен рекомендовать распределение активов, которое превышает способность инвестора принимать риск. Способность принимать риск задает верхнюю границу риска портфеля, 6. Финансовый консультант не должен рекомендовать распределение активов, которое превышает способность инвестора принимать риск. Способность принимать риск задает верхнюю границу риска портфеля, 6. Финансовый консультант не должен рекомендовать распределение активов, которое превышает способность инвестора принимать риск. Способность принимать риск задает верхнюю границу риска портфеля, 11 однако этот индикатор часто меняется с течением времени, поэтому должен пересматриваться регулярно. однако этот индикатор часто меняется с течением времени, поэтому должен пересматриваться регулярно. Эксперты CFA Institute предлагают рассматривать построенные профили в виде светофора для дальнейшей калибровки распределения активов и составления портфелей. Профили, относящиеся к «зеленому свету светофора», не требуют дальнейшей калибровки или работы с инвестором, все три компоненты согласованы и можно приступать к формированию рекомендаций по распределению активов. Так, инвесторам с высокими потребностями, способностями и толерантностью к риску можно предлагать портфель с высоким риском, в котором более 70% активов размещается в активы роста (акции или схожие альтернативы). Умеренный портфель будет только на 30–70% состоять из активов роста, а консервативный иметь на более 30% таких активов. профилированию в США Профили, относящиеся к «желтому свету светофора», подразумевают, что толерантность к риску у инвестора ниже, чем его потребность или способность принимать риск. Это означает, что инвестору можно предложить дополнительное обучение с целью повысить его знания и, соответственно, терпимость к риску. Тогда ему можно было бы предложить более рисковый портфель, который отвечал бы его потребностям и способностям. В ином случае, если инвестор отказывается от обучения или оно не способствовало росту толерантности к риску, инвестору придется пересмотреть свои цели, так как более консервативные портфели не смогут удовлетворить потребности. По статистике, 55% инвесторов понимают, что они должны принимать на себя больший риск для достижения намеченной цели, однако 52% предпочитают отказаться от достижения цели вместо повышения риска сверх своего предела толерантности [12]. В этом случае инвестор должен обсудить с консультантом последствия недостижения цели и то, насколько это приемлемо. Профили, относящиеся к «желтому свету светофора», подразумевают, что способность инвестора принимать риск ниже, чем его потребность. Причинами может служить короткий временной горизонт, ограничения по ликвидности или низкая допустимая степень риска для поддержания уровня жизни. В этом случае требуется повторное обсуждение целей инвестора с их корректировкой для приведения в соответствие со способностями. Распределение активов может быть составлено только, если инвестор снизит потребность, то есть необходимую ставку доходности, 12 например, откладыванием выхода на пенсию, увеличением сбережений или снижением целевых показателей. Таким образом, процесс формирование риск-профиля инвестора является неотъемлемой частью процесса принятия инвестиционных решений. Проведение тестирования или анкетирования может только в общих чертах дать размытое представление об одном или нескольких аспектах риск-профиля. При отсутствии корректной оценки риск-профиля финансовые консультанты рискуют сформировать слишком агрессивный или слишком консервативный портфель для клиента, при котором тот не достигнет своих намеченных целей или не сможет долго его поддерживать. Риск-профилирование не должно быть основано на одной лишь анкете или формировании единственной оценки, а должно стать более комплексным процессом с несколькими независимо оцениваемыми компонентами. 3 Формирование рекомендаций о распределении активов: разработка оптимизационного алгоритма на примере американских инвесторов 3 Формирование рекомендаций о распределении активов: разработка оптимизационного алгоритма на примере американских инвесторов Определение риск-профиля инвестора должно заканчиваться логическим итогом: рекомендацией оптимального или наиболее подходящего портфеля для него. Практика финансовых консультантов в развитых странах показывает, что чаще используется градация результатов риск-профилирования на 3–9 профилей, для каждого из которых рекомендуется некоторое распределение активов между укрупненными классами. Например, Vanguard предлагает 9 возможных портфелей в зависимости от балла, набранного по итогам анкеты из 11 вопросов. Причем все предлагаемые портфели состоят только из трех классов активов: внутренних и международных акций и облигаций, а отличаются только весом. Анкета Charles Schwab закачивается оценкой риск-профиля из 5 возможных вариантов от консервативного до агрессивного, а распределение активов предлагается между пятью классами активов: акциями крупных и малых компаний, международными акциями, облигациями и денежными средствами. Подход Fidelity предполагает 4 возможных риск профиля и четыре соответствующих распределения активов между акциями, международными акциями, облигациями и краткосрочными инструментами. Анкета Merrill Edge, Bank 13 of America corporation заканчивается пятью возможными риск-профилями и портфелем из акций, облигаций и денежных средств. Опросник Goldman Sachs приводит к пяти возможным профилям, но итоговые портфели состоят только из акций и облигаций. Анализ профилей (таблица 1) показывает, что при переходе от наиболее консервативного (1) к наиболее агрессивному (4 или 5 или 9 в зависимости от шкалы) растет доля акций, падает доля облигаций и почти обнуляется доля денежных средств. В ряде анкет рост толерантности к риску ведет не только к росту национальных акций, но и росту наиболее агрессивных подкатегорий: акций малых компаний или международных акций. Проблема существующего подхода заключается в малой эмпирической обоснованности того или иного веса актива в общем портфеле. Принцип построения портфеля для определенного профиля риска основан на предпосылке о том, что акции рискованнее облигаций, а те рискованнее денежных эквивалентов. И хотя в большинстве случаев это так, закономерности или правила определения конкретных весов в портфеле остаются неопределенными. Все это приводит к широкому разнообразию «оптимальных» весов даже у пятерки лидеров инвестиционного консультирования в США и еще большей неоднозначности для развивающихся рынков. Основной задачей является составление набора эмпирических фактов, на основе которых можно было бы построить простой и прозрачный алгоритм, обосновывающий вес конкретного актива в портфеле для конкретного уровня риска. 3 Формирование рекомендаций о распределении активов: разработка оптимизационного алгоритма на примере американских инвесторов Рекомендуемое распределение активов для разных риск-профилей инвесторов пятью крупнейшими инвестиционными компаниями США, % Charles Schwab Fidelity Merrill Edge 1 2 3 4 5 1 2 3 4 1 2 3 4 5 Акции крупных компаний 15 25 35 45 50 14 35 49 60 20 40 60 70 80 Акции малых компаний 5 10 15 20 Международные акции 5 10 15 20 25 6 15 21 25 Инструменты с фиксированным доходом 50 50 35 15 50 40 25 15 55 50 35 25 15 14 Денежные средства 30 10 5 5 5 30 10 5 25 10 5 5 5 Goldman Sachs Vanguard 1 2 3 4 5 1 2 3 4 5 6 7 8 9 Акции крупных компаний 40 60 80 100 15 25 30 40 50 55 65 80 Акции малых компаний 5 5 10 10 10 15 15 20 Международные акции Инструменты с фиксированным доходом 60 40 20 100 80 70 60 50 40 30 20 Денежные средства 100 Примечание: наиболее консервативный профиль - (1), наиболее агрессивный - (4 или 5 или 9 в зависимости от шкалы) Денежные средства 30 10 5 5 5 30 10 5 25 10 5 5 5 Goldman Sachs Vanguard 1 2 3 4 5 1 2 3 4 5 6 7 8 9 Акции крупных компаний 40 60 80 100 15 25 30 40 50 55 65 80 Акции малых компаний 5 5 10 10 10 15 15 20 Международные акции Инструменты с фиксированным доходом 60 40 20 100 80 70 60 50 40 30 20 Денежные средства 100 Примечание: наиболее консервативный профиль - (1), наиболее агрессивный - (4 или 5 или 9 в зависимости от шкалы) Примечание: наиболее консервативный профиль - (1), наиболее агрессивный - (4 или 5 или 9 в зависимости от шкалы) Источник: сайты соответствующих компаний Источник: сайты соответствующих компаний Для определения алгоритма подбора портфеля и эмпирического исследования различных по риску портфелей выбраны пять активов в соответствии с методикой Charles Schwab. Подобный подход может быть расширен и на больший круг активов, однако именно портфели этого инвестиционного консультанта наиболее диверсифицированы. Круг активов для анализа портфельного множества в США включает в себя: - Индекс внутренних акций крупных компаний – индекс общей доходности S&P 500; - Индекс малых компаний – индекс общей доходности малых компаний S&P; - Индекс международных акций – индекс общей доходности мировых развитых рынков MSCI WORLD; - Индекс корпоративных облигаций – индекс Bloomberg US Corporate Total Return Value Unhedged USD (LUACTRUU); - Индекс корпоративных облигаций – индекс Bloomberg US Corporate Total Return Value Unhedged USD (LUACTRUU); Денежные эквиваленты – безрисковая ставка в виде доходности к погашению 10-летних государственных облигаций США Денежные эквиваленты – безрисковая ставка в виде доходности к погашению 10-летних государственных облигаций США Данные по США собраны с 1994 года по 2020 год. Все индексы выражены в долларах США. Безрисковая ставка определяется как 10-летная доходность к погашению в момент начала составления портфеля, то есть определяет зафиксированную доходность инвестора. Так как используются достаточно длинные интервалы времени (от 5 лет и более), то долгосрочная ставка является подходящим 15 прокси. При более точном подходе требуется использование ставки по тем государственным облигациям, чей срок до погашения соответствует горизонту инвестирования, который моделируется. Основные этапы алгоритма включают в себя подготовку базовых классов активов, данных по их индексным прокси и оценку их мер риска и доходности, моделирование всего портфельного множества, разбиение на портфели по профилям риска и определение оптимальных портфелей. Первым шагом является оценка основных четырех классов активов (безрисковая ставка неизменна и исключается из этого этапа) по риску и доходности. Так как в рекомендациях чаще всего не указывается рекомендация распределения, которая зависела бы не только от риск-профиля, но и от горизонта, то рассмотрен весь возможный период. Шаг задан в один год. Например, рассмотрено 27 однолетних периодов, 18 десятилетних периодов и один двадцатисемилетний период. После оценки показателей доходности и риска базовых индексов показатели агрегированы путем оценки медианного значения. На рисунке (рисунок 1) показаны среднегодовые доходности каждого индекса на каждом горизонте времени. Красной линией отмечено медианное значение. На коротких сроках для акций наблюдается наибольший разброс, однако уже с пятилетнего горизонта акции малых компаний не падают в стоимости ниже уровня первоначальных инвестиций для любого портфеля. Для индекса крупнейших компаний это наступает только с 10-летнего горизонта и выше, аналогично и для мировых акций. Источник: сайты соответствующих компаний Корпоративные облигации приносили отрицательную доходность только на малом количестве наиболее коротких периодов (1–2 года). Однако медианная доходность корпоративных облигаций ниже, чем у акций (рисунок 2). Акции малых компаний наиболее привлекательны и имеют наивысшую медианную доходность. Акции крупных компаний США оказываются более привлекательны, чем глобальные акции развитых стран. 16 Рисунок 1. Доходность различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. Источник: расчеты авторов по данным Блумберг. Рисунок 1. Доходность различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. Рисунок 1. Доходность различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. 17 Рисунок 2. Медианная доходность различных рисковых классов активов на разных 0 2 4 6 8 10 12 14 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Эффективная доходность CORBOND EQ SMALL WORLD 0 2 4 6 8 10 12 14 16 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Эффективная доходность CORBOND EQ SMALL WORLD сунок 2. Медианная доходность различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. горизонтах инвестирования, %, 1994–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Оценка рисков этих четырех классов активов представлена на рисунке (рисунок 3). Медианное стандартное отклонение характеризует меру риска, выше которой 50% портфелей на выбранном горизонте инвестирования. Акции малых компаний обладают наибольшим риском, а корпоративные облигации – наименьшим. Акции крупнейших компаний США и компаний развитого рынка крайне схожи. Кроме того, риски этих классов активов не уменьшаются с ростом горизонта инвестирования и не сближаются. 18 Рисунок 3. Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. 0 5 10 15 20 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Стандартное отклонение CORBOND EQ SMALL WORLD Рисунок 3. Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. Источник: сайты соответствующих компаний 0 5 10 15 20 25 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Стандартное отклонение CORBOND EQ SMALL WORLD 25 Рисунок 3. Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. Рисунок 3. Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. Рисунок 3. Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Еще одной мерой риска является максимальная просадка стоимости портфеля или минимальная стоимость активов по отношению к размеру первоначальных инвестиций в течение всего периода (рисунок 4). Эта мера риска отражает вопрос в анкете о толерантности к риску, связанный с тем, способен ли инвестор выдержать временное уменьшение стоимости своего портфеля в течение срока инвестирования, не меняя стратегию и не паникуя. Данные показывают, что наибольшая медианная просадка портфеля наблюдается у акций, хотя этот индикатор сильно меняется от портфеля к портфелю. Корпоративные облигации допускают просадку не более 5%. 19 Рисунок 4. Медианная максимальная просадка различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. -30 -25 -20 -15 -10 -5 0 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Максимальная просадка CORBOND EQ SMALL WORLD Рисунок 4. Медианная максимальная просадка различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. -30 -25 -20 -15 -10 -5 0 5 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Максимальная просадка CORBOND EQ SMALL WORLD Рисунок 4. Медианная максимальная просадка различных рисковых классов активов на разных горизонтах инвестирования, %, 1994–2020 гг. Источник: расчеты авторов по данным Блумберг Последним индикатором риска выступает показатель Value-at-Risk (VaR), который отмечает, какие максимальные потери наступают в 95% случаях для портфеля (рисунок 5). Этот индикатор предполагает измерение амплитуды ежемесячных доходностей портфеля. Медианное его значение по всем портфелям по разным горизонтам оказывается наименьшим для корпоративных облигаций. Ранжирование акций по этому индикатору провести довольно сложно, риски схожи. Источник: сайты соответствующих компаний Для коротких периодов (до 5 лет) акции малых компаний оказываются наиболее рисковыми по этому критерию, однако с 6-летнего до 21-летнего горизонта наибольшим риском обладали глобальные акции. 20 Рисунок 5. Медианный показатель VaR(95%) различных рисковых классов активов на разных горизонтах инвестирования, %, 1994-2020 гг. -35 -30 -25 -20 -15 -10 -5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 VaR CORBOND EQ SMALL WORLD Рисунок 5. Медианный показатель VaR(95%) различных рисковых классов активов на разных горизонтах инвестирования, %, 1994-2020 гг. Источник: расчеты авторов по данным Блумберг По итогам анализа базовых активов можно сделать вывод о том, что с ростом горизонта инвестирования некорректно увеличивать долю акций инвестору, которые не толерантен к риску и на краткосрочных горизонтах. Более того, анализ максимальных просадок портфеля показал, что наиболее уязвимым является среднесрочный горизонт от 9 до 15 лет, когда высокая доля акций может привести к панике инвестора и требованию скорректировать портфель. Это показательный результат, так как на выбранном историческом промежутке наблюдалось сразу три кризиса на американском и глобальном рынке акций: кризис начала 2000-х, кризис 2007–2008 года и кризис 2020 года. Это означает, что в абсолютно каждом промежутке от 9 до 15 лет встречался хотя бы один кризис, который и обеспечивал просадку портфеля. От горизонта времени зависит не только максимальная просадка, но и доходность инвестора. Она немного увеличивается с ростом горизонта инвестирования для всех типов акций. Только этот аргумент может привести к росту 21 доли акций в портфеле инвестора в долгосрочном горизонте и только, если это обусловлено его потребностью в доходности. На следующем этапе аналогичный анализ проведен для портфелей, которые рекомендуются консультантом Charles Schwab (таблица 2). Эти портфели считаются оптимальными для каждой из пяти категорий риска, причем по итогам анкеты подходящие портфели меняются в зависимости от горизонта инвестирования (рисунок 6). Минимальный горизонт инвестирования равен трем годам, поэтому более короткие интервалы не рассматриваются. Таблица 2 Таблица 2 Рекомендуемое распределение активов для разных риск-профилей инвесторов Charles Schwab, %. Charles Schwab Cons ModCons Moderate ModAggr Aggr Акции крупных компаний 15 25 35 45 50 Акции малых компаний 0 5 10 15 20 Международные акции 5 10 15 20 25 Инструменты с фиксированным доходом 50 50 35 15 0 Денежные средства 30 10 5 5 5 Источник: https://www.schwab.com/ Рисунок 6. Рекомендуемый портфель активов для разных риск-профилей инвесторов с учетом горизонта инвестирования Charles Schwab. Источник: https://www.schwab.com/system/file/P-12585786. Источник: сайты соответствующих компаний ц Рекомендуемое распределение активов для разных риск-профилей инвесторов Charles Schwab, %. Charles Schwab Cons ModCons Moderate ModAggr Aggr Акции крупных компаний 15 25 35 45 50 Акции малых компаний 0 5 10 15 20 Международные акции 5 10 15 20 25 Инструменты с фиксированным доходом 50 50 35 15 0 Денежные средства 30 10 5 5 5 Источник: https://www.schwab.com/ мендуемое распределение активов для разных риск-профилей инвесторов Charles wab, %. p Рисунок 6. Рекомендуемый портфель активов для разных риск-профилей инвесторов с учетом горизонта инвестирования Charles Schwab. Источник: https://www.schwab.com/system/file/P-12585786. Рисунок 6. Рекомендуемый портфель активов для разных риск-профилей инвесторов с учетом горизонта инвестирования Charles Schwab. Источник: https://www.schwab.com/system/file/P-12585786. Источник: https://www.schwab.com/system/file/P-12585786. 22 Для портфелей, подобранных Schwab под каждый профиль риска, рассчитаны те же меры, что и для базовых активов: среднегодовая доходность, стандартное отклонение, максимальная просадка и VaR 95%. Для каждого горизонта инвестирования рассчитано максимальное количество портфелей с шагом в 1 год, как и для базовых активов. Далее рассчитано медианное значение каждой меры по всем портфелям (рисунок 7). 23 23 Рисунок 7. Меры доходности и риска для рекомендуемых портфелей активов для ф й Ch l S h b 0 2 4 6 8 10 12 14 3 5 7 9 11 13 15 17 19 21 23 25 27 Эффективная доходность 0 2 4 6 8 10 12 14 16 3 5 7 9 11 13 15 17 19 21 23 25 27 Стандартное отклонение -25 -20 -15 -10 -5 0 5 3 5 7 9 11 13 15 17 19 21 23 25 27 Максимальная просадка -30 -25 -20 -15 -10 -5 0 3 5 7 9 11 13 15 17 19 21 23 25 27 VaR 0 2 4 6 8 10 12 14 3 5 7 9 11 13 15 17 19 21 23 25 27 Эффективная доходность 0 2 4 6 8 10 12 14 16 3 5 7 9 11 13 15 17 19 21 23 25 27 Стандартное отклонение 3 5 7 9 11 13 15 17 19 21 23 25 27 -25 -20 -15 -10 -5 0 5 3 5 7 9 11 13 15 17 19 21 23 25 27 Максимальная просадка Рисунок 7. Источник: сайты соответствующих компаний Меры доходности и риска для рекомендуемых портфелей активов для -25 -20 -15 -10 -5 0 5 3 5 7 9 11 13 15 17 19 21 23 25 27 Максимальная просадка -30 -25 -20 -15 -10 -5 0 3 5 7 9 11 13 15 17 19 21 23 25 27 VaR 0 3 5 7 9 11 13 15 17 19 21 23 25 27 Рисунок 7. Меры доходности и риска для рекомендуемых портфелей активов для разных риск-профилей инвесторов Charles Schwab Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Подход Schwab подразумевает, что с ростом горизонта инвестирования инвестор может принимать на себя все большие риски при одной и той толерантности к риску. Однако эмпирические данные не показывают тенденции к сближению мер риска с ростом горизонта. Разрыв между портфелями по VaR и стандартному отклонению остается примерно постоянным. Некоторое сближение портфелей характерно по мере максимальной просадки стоимости инвестиций, однако это 24 сближение происходит на горизонтах выше 17–18 лет. Единственная выгода в этом для инвестора – несколько более высокая доходность, однако оно не существенное по сравнению с изменением риска. Так, на примере медианного 10-летнего портфеля можно показать, что рекомендуемый по анкете Schwab переход от консервативного к умеренно консервативному портфелю для человека с низкой толерантностью к риску принесет ему рост доходности с 5,89% до 6,63%, а увеличение риска с 4,78% до 7,47%. В расчете на одну единицу риска такой инвестор проиграет достаточно значительно. Таким образом, данные по США не поддерживают гипотезу о возможности увеличения общего риска портфеля с ростом горизонта при прочих равных. Следующий этап заключается в формировании портфельного множества из всех возможных комбинаций активов и поиске оптимального портфеля. Далее описана процедура для одного из 10-летних периодов для США, оцененного на основе исторических данных с 2011 по 2020 год. Для этого сгенерированы веса портфелей с шагом в 5% из всех возможных комбинаций по 3, 4 и 5 активов из базовых: акций крупных компаний (EQ), акций малых компаний (SMALL), международных акций (WORLD), корпоративных облигаций (CORBOND) и денежных эквивалентов (RF). Максимальный вес одного актива в портфеле выбран 85% для обеспечения хотя бы минимальной диверсификации, минимальный вес равен 0. В качестве безрисковой ставки или ставки денежных эквивалентов использована доходность к погашению десятилетних государственных облигаций США по состоянию на начало 2011 года, то есть на начало периода формированию смоделированного портфеля. Источник: сайты соответствующих компаний Это значение составило 3,37%, что означает, что при 10- летнем горизонте инвестор без риска мог бы вложить все свои активы в государственные облигации, получив гарантированно 3,37% доходности. Часть средств, размещенных в денежных эквивалентах, считается, что также размещается в этот актив. Портфельное множество из перечисленных пяти активов получилось вытянутым вдоль диагонали риск-доходность (рисунок 8). На рисунок нанесены портфели со 100% весом каждого из активов, а также портфели, соответствующие 5 профилям риска по Schwab. Это базовые бенчмарки, используемые для сопоставления и поиска соотношений. 25 Вытянутая форма облака обусловлена тем, что в период с 2011 по 2020 гг. акции как внутренние, так и глобальные росли и восстанавливались активно после кризиса 2007–2008 гг. Такая форма облака говорит о малых возможностях диверсификации, так как все активы, связанные с акциями оказались похожи по характеристикам доходности и риска. Этот вывод подтверждается и матрицей корреляций ежемесячных доходностей индексов в таблице (таблица 3). Примечание: круглыми точками обозначены портфели по разным риск- профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный). Рисунок 8. Портфельное множество из 5 базовых активов, а также смоделированные портфели Schwab, %, 2011–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Матрица корреляций базовых активов США, 2011–2020 гг. 26 Матрица корреляций базовых активов США, 2011–2020 гг. CORBOND EQ SMALL WORLD CORBOND 1 0.4 0.33 0.44 26 EQ 0.4 1 0.88 0.98 SMALL 0.33 0.88 1 0.85 WORLD 0.44 0.98 0.85 1 Источник: расчеты авторов по данным Блумберг. EQ Полученное множество портфелей мы делим на 5 групп по разным индикаторам по квантилям на равные группы. Первый – это среднегодовая доходность (рисунок 9а). Второй – стандартное отклонение (рисунок 9б), а также VaR 95% (рисунок 9в) и максимальная просадка (рисунок 9г). Все группы ранжировались от одного (минимальная доходность или риск) до пяти (максимальные значения). Все рисунки сделаны в координатах риск-доходность вне зависимости от сортировки портфелей. Серым на картинке выделены наименее эффективные портфели, для которых группа по риску была выше, чем по доходности. Это очень упрощенная, но наглядная версия критерия Шарпа. Источник: сайты соответствующих компаний Примечание: by.r – классификация по доходности, by.s – по стандартному отклонению, by.var, по VaR 95% (наибольшие потери отнесены в пятую, красную группу), by.w – по максимальной просадке; круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до Примечание: by.r – классификация по доходности, by.s – по стандартному отклонению, by.var, по VaR 95% (наибольшие потери отнесены в пятую, красную группу), by.w – по максимальной просадке; круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до 27 агрессивного (красный); серыми – портфели, для которых группа по риску была выше, чем по доходности. Рисунок 9. Портфельное множество и способы группировки портфелей, а также смоделированные портфели Schwab, %, 2011–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Основным выводом является то, что различные группировки по показателям риска являются достаточно согласованными между собой. Чаще всего, если портфель относится к повышенному риску, то это будет замечено для всех трех индикаторов риска. Изучая полученные классификации портфелей, мы можем сделать вывод о том, что по стандартному отклонению и VaR портфели, предлагаемые Schwab, соответствуют группировкам. Однако по доходности портфель для умеренно агрессивного профиля соответствует более агрессивному подмножеству. Сразу два портфеля относятся к более низкой по риску группе по показателю максимальной просадки: портфель для умеренно консервативного соответствует консервативной группе, а портфель средний – группе умеренно консервативной. В целом, перечисленные наблюдения не выявляют ошибку или несоответствие портфелей, так как риски по ним относятся к более низкой группе, чем то, как выбирались бы оптимальные портфели внутри каждой из групп. Последний этап заключался в усреднении классификации риска по трем полученным группам. Так, если некоторый портфель вошел в группу номер 5 по стандартному отклонению и по VaR и в группу номер 4 по максимальной просадке, то средняя группа для него будет 5 (округленное значение 4,67). Итоговый вид классификации портфелей по всем трем критериям риска показан на рисунке (рисунок 10). 28 Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную группу, круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный); серыми – портфели, для которых группа по риску была выше, чем по доходности. Рисунок 10. Портфельное множество и итоговая классификация портфелей по уровню риска, а также смоделированные портфели Schwab, %, 2011–2020 гг. Рисунок 10. Портфельное множество и итоговая классификация портфелей по уровню риска, а также смоделированные портфели Schwab, %, 2011–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: сайты соответствующих компаний Среди составленных портфелей в каждой из пяти групп можно выбрать наилучший по соотношению риск-доходность. Для этого использовался критерий Шарпа: отношение премии портфеля по сравнению с безрисковой ставкой и риска портфеля, измеренного стандартным отклонением. Для каждой группы определен портфель с наибольшим значением критерия Шарпа (рисунок 11). Все выбранные портфели лежат на эффективной границе портфельного множества, в то время как рекомендуемые портфели Schwab в середине по риску и доходности. Более того, подобранные портфели позволяют не только увеличить доходность, но и снизить риски для наиболее агрессивных профилей. 29 Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную группу, круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный); квадратными – наиболее эффективные по соотношению риск-доходность; серыми – портфели, для которых группа по риску была выше, чем по доходности. Рисунок 11. Портфельное множество и итоговая классификация портфелей по уровню риска, а также смоделированные портфели Schwab, %, 2011–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Статистика и сопоставление по двум типам портфелей – по рекомендациям инвестиционных консультантов и по найденным оптимальным решениям – приведена в таблице (таблица 4). Алгоритм оптимизации выбирает преимущественно портфели с наибольшим весом акций американских компаний, полностью игнорируя менее выгодные по соотношению доходность-риск акции малых компаний и глобальный индекс. Доля корпоративных облигаций подбирается соответствующая или немногим ниже рекомендуемых портфелей. Отличие составляет лишь агрессивный портфель: для него в целях диверсификации включено 10% корпоративных облигаций, что 30 отличается от рекомендуемой агрессивной стратегии. Доминирование актива с наибольшим соотношением доходности и риска позволил улучшить параметры портфелей Schwab, снижая риски и\или увеличивая доходность. Так, в агрессивном распределении на американские акции приходится 50%, а в оптимальном портфеле 85%. Таблица 4 Таблица 4 Веса и параметры риска и доходности портфелей Schwab и оптимальных портфелей, %, 2011–2020 гг. Schwab Оптимальные портфели Уровень риска 1 2 3 4 5 1 2 3 4 5 CORBOND 50 50 35 15 0 5 30 30 15 10 EQ 15 25 35 45 50 10 50 65 80 85 SMALL 0 5 10 15 20 0 0 0 0 0 WORLD 5 10 15 20 25 0 0 0 0 5 RF 30 10 5 5 5 85 20 5 5 0 Доходность 6.56 8.46 9.98 11.33 12.23 4.63 9.46 11.02 12.22 12.94 Риск 4.36 6.95 9.29 11.68 13.60 1.47 7.46 9.46 11.13 12.39 Коэффициент Шарпа 0.72 0.73 0.71 0.68 0.65 0.82 0.81 0.80 0.79 0.77 Источник: расчеты авторов по данным Блумберг. Источник: сайты соответствующих компаний и параметры риска и доходности портфелей Schwab и оптимальных портфелей, 011–2020 гг. Проблемой проведенного анализа является то, что в конкретной точке времени и на конкретном временном интервале подобранные портфели могут, действительно, быть эффективнее, однако создание рекомендация для разных профилей риска является более комплексной задачей. Так, рекомендуемые портфели должны быть достаточно универсальными и оптимальными для всех периодов времени. С этой целью процедура поиска оптимального портфеля проведена для всех доступных 10-летних отрезках на периоде с 1994 по 2020 гг. (всего 18 периодов с шагом в один год). Дополнительно наложено ограничение размера денежных эквивалентов не более 30%. Это не соответствует, например, подходу Goldman Sachs, который советует консервативному профилю инвестора 100% вложение в денежные эквиваленты. Однако поддерживает принцип того, что инвестор приходи на рынок для инвестирования на нем и использования преимущественно рыночных инструментов. Следовательно, его портфель должен преимущественно быть распределен между акциями и облигациями, а также иными рыночными рисковыми инструментами. 31 31 Основным выводом из рассмотрения всех 10-летних интервалов является тот факт, что оптимальный портфель не постоянен. Его веса и параметры риска и доходности меняются достаточно сильно. Поэтому для поиска некоторого универсального распределения полученные результаты необходимо усреднить и получить некоторый средний оптимальный портфель, единый для всех периодов для каждой группы риска. Второй вывод заключается в том, что есть ряд отрезков времени, когда веса рекомендуемого Schwab и оптимального портфелей значительно отличаются. Это происходит, например, в такие периоды как с 1999 по 2008 гг., с 2000 по 2009 гг., 2001–2010 гг. и 2002–2011 гг. В каждом из этих периодов рынок падал, особенно рынок крупнейших акций, который показывал наихудшую среди всех активов доходность. Наиболее агрессивные по риску портфели оказывались среди наименее эффективных, а выбор оптимального из них становился бессмысленным. В некоторой степени это отражает суть риска агрессивных стратегий, которые при росте рынка растут быстрее всего, а в периоды снижения – падают сильнее всех. С точки зрения поиска оптимального алгоритма такие периоды имеет смысл выкинуть из усреднения, так как выбранные алгоритмом портфели отражают, скорее, защитную стратегию, чем поиск универсального оптимума. Веса активов в таких портфелях значительно искажаются и отличаются от средних, так как там максимально, насколько это возможно с учетом ограничений заданного уровня риска, сокращен вес акций и увеличен вес облигаций и безрискового актива. Таким образом, оптимальные портфели определяются как взвешенные средние среди всех 10-летних оптимальных портфелей, рассчитанных в моменты спокойного состояния рынка. Это предполагает корректное с точки зрения теории расположение активов по риску и доходности: например, риск и доходность акций выше, чем риск и доходность облигаций. Источник: сайты соответствующих компаний Наибольший вес имеет портфель, рассчитанный на последних данных, наименьший вес – наиболее далекие в исторической ретроспективе портфели. В этом случае алгоритмы поиска работают наиболее корректно. На рисунках (рисунок 12 и 13) показаны портфельные множества для каждого из периодов, подтверждающие сделанные выводы, а также отражающие единый и универсальный оптимизированный портфель для каждого профиля риска. 32 Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную группу, круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный); квадратными – универсальный оптимизированный портфель для каждого профиля риска; серыми – портфели, для которых группа по риску была выше, чем по доходности. Рисунок 12. Портфельные множества и итоговая классификация портфелей по Рисунок 12. Портфельные множества и итоговая классификация портфелей по Рисунок 12. Портфельные множества и итоговая классификация портфелей по уровню риска, а также смоделированные портфели Schwab и универсальный оптимальный портфель для первой половины 10-летних периодов, %, 1994–2011 гг. Источник: расчеты авторов по данным Блумберг. уровню риска, а также смоделированные портфели Schwab и универсальный оптимальный портфель для первой половины 10-летних периодов, %, 1994–2011 гг. уровню риска, а также смоделированные портфели Schwab и универсальный оптимальный портфель для первой половины 10-летних периодов, %, 1994–2011 гг. Источник: расчеты авторов по данным Блумберг. 33 Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную группу, круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный); квадратными – универсальный оптимизированный портфель для каждого профиля риска; серыми – портфели, для которых группа по риску была выше, чем по доходности. Рисунок 13. Портфельные множества и итоговая классификация портфелей по Рисунок 13. Портфельные множества и итоговая классификация портфелей по у р ф ф ц р ф уровню риска, а также смоделированные портфели Schwab и универсальный оптимальный портфель для второй половины 10-летних периодов, %, 2003–2020 гг. Источник: расчеты авторов по данным Блумберг. уровню риска, а также смоделированные портфели Schwab и универсальный оптимальный портфель для второй половины 10-летних периодов, %, 2003–2020 гг. Источник: расчеты авторов по данным Блумберг. уровню риска, а также смоделированные портфели Schwab и универсальный оптимальный портфель для второй половины 10-летних периодов, %, 2003–2020 гг. Выбранный алгоритмом портфель для каждого профиля риска для 10-летнего горизонта существенно отличается от рекомендаций Schwab (таблица 5). Источник: сайты соответствующих компаний В нем 34 практически не используется глобальная диверсификация, а вес малых компаний гораздо выше. Причиной первого, вероятно, является использование индекса MSCI WORLD в качестве бенчмарка глобальных акций, в котором 68% занимают те же американские акции. Это приводит к высокой корреляции этих двух активов. Остальные же акции в составе глобального индекса, по всей видимости, имели худшую результативность, поэтому индекс всегда оказывался в нижней правой части графиков для любого 10-летнего периода. Это говорит о том, что на долгосрочных горизонтах глобальная диверсификация в развитые страны для инвестора в США не приносит дополнительных выгод. Таблица 5 Веса портфелей Schwab и универсальный оптимизированный портфель для каждого профиля риска, %, 1994–2020 гг. Schwab Оптимальные портфели Уровень риска 1 2 3 4 5 1 2 3 4 5 CORBOND 50 50 35 15 0 42 42 30 23 14 EQ 15 25 35 45 50 14 22 30 36 38 SMALL 0 5 10 15 20 15 26 33 38 45 WORLD 5 10 15 20 25 0 0 0 0 2 RF 30 10 5 5 5 29 9 6 3 2 Источник: расчеты авторов по данным Блумберг. Веса портфелей Schwab и универсальный оптимизированный портфель для каждого профиля риска, %, 1994–2020 гг. Schwab Оптимальные портфели Уровень риска 1 2 3 4 5 1 2 3 4 5 CORBOND 50 50 35 15 0 42 42 30 23 14 EQ 15 25 35 45 50 14 22 30 36 38 SMALL 0 5 10 15 20 15 26 33 38 45 Веса портфелей Schwab и универсальный оптимизированный портфель для каждого профиля риска, %, 1994–2020 гг. Schwab Оптимальные портфели Уровень риска 1 2 3 4 5 1 2 3 4 5 Веса портфелей Schwab и универсальный оптимизированный портфель для каж профиля риска, %, 1994–2020 гг. Веса портфелей Schwab и универсальный оптимизированный портфель для каждого профиля риска, %, 1994–2020 гг. Причиной высокого веса акций малых компаний, по всей видимости, стал долгосрочный период в 10 лет, на котором фундаментальные факторы, включая размер, все еще работают и приносят инвестору высокую избыточную доходность. Исключением стал лишь 2020 год, а именно последний 10-летний интервал с 2011 по 2020 гг., когда акции малых компаний показатели значительно худшую результативность по сравнению с крупными. Именно поэтому для последнего интервала оптимальный портфель не смог превзойти по соотношению риск- доходность рекомендуемые Schwab портфели. Вес корпоративных облигаций соответствует рекомендациям, кроме агрессивного портфеля. Риск агрессивного портфеля смягчен 14% корпоративных облигаций, что позволяет получать высокую доходность, контролируя риск. Источник: сайты соответствующих компаний Вес денежных эквивалентов полностью соответствует рекомендациям. Вес акций крупнейших компаний тоже соответствует, кроме агрессивного портфеля, в котором часть доли перешла на корпоративные облигации. Вес малых 35 компаний соответствует сумме весов малых и глобальных компаний в портфеле Schwab. Таким образом, для американского инвестора разработан алгоритм, который позволяет определять оптимальные портфели для разных профилей риска и обосновывать корректность предлагаемых инвестиционными консультантами рекомендаций. Показано, что для 10-летнего периода оптимальный портфель преимущественно сход с рекомендуемым, кроме инвестиций в глобальные компании, которые не приносят свой вклад на долгосрочном периоде из-за высокой корреляции с внутренним рынком. Для других периодов проведены аналогичные расчеты и составлены оптимальные портфели (таблица 6). Универсальный оптимизированный портфель для каждого профиля риска и временного горизонта, %, 1994–2020 гг. р р р ф р ф р временного горизонта, %, 1994–2020 гг. Оптимальные портфели Уровень риска 1 2 3 4 5 3 года CORBOND 45 35 20 11 7 EQ 24 39 54 59 47 SMALL 4 7 10 14 27 WORLD 7 9 11 14 20 RF 20 10 6 2 0 5 лет CORBOND 44 35 25 14 8 EQ 26 45 57 65 58 SMALL 5 5 8 11 19 WORLD 4 4 6 7 12 RF 21 11 5 2 3 10 лет CORBOND 42 42 30 23 14 EQ 14 22 30 36 38 SMALL 15 26 33 38 45 WORLD 0 0 0 0 2 RF 29 9 6 3 2 15 лет CORBOND 38 32 27 21 10 EQ 14 19 25 29 34 SMALL 19 28 37 43 48 WORLD 0 0 0 0 3 36 RF 29 21 11 7 4 20 лет CORBOND 38 32 23 13 3 EQ 10 1 0 2 19 SMALL 22 42 56 67 64 WORLD 0 0 0 0 0 RF 30 24 21 19 14 Источник: расчеты авторов по данным Блумберг Источник: расчеты авторов по данным Блумберг. По итогам расчетов показано, что акции малых компаний наиболее эффективны на долгосрочных горизонтах, что приводит к росту их доли с увеличением срока инвестирования. Для агрессивного профиля для трехлетнего горизонта алгоритм рекомендует 27% акций малых компаний, а для 20-летнего горизонта – уже 64%. Пропорционально уменьшается доля акций крупнейших компаний в портфеле. Глобальная диверсификация для американского инвестора имеет смысл только на краткосрочных горизонтах, а в долгосрочном аспекте результативность инвестирования в другие рынки значительно хуже, а за счет большого веса американских акций возможности диверсификации также низки. Источник: сайты соответствующих компаний Это приводит к нулевому весу глобальных акций в составе портфелей с 10-летнего горизонта. Примечательно, что чем выше горизонт инвестирования, тем выше доля безрискового актива в составе портфеля. Это связано с тем, что при росте горизонта падает премия за риск для рыночных активов, как следствие сужения различий в доходности между активами. Таким образом, консервативному профилю для трехлетнего горизонта рекомендована доля 20% безрисковых активов, а агрессивному – 0%, а для 20-летнего горизонта 30% и 14%, соответственно. Примечательно также и то, что доля корпоративных облигаций в составе портфелей относительно стабильна и почти не зависит от горизонта инвестирования. Она не нулевая для любого профиля и горизонта. Таким образом, инвестирование в корпоративные облигации крайне важно для американского инвестора с любым профилем, позволяет диверсифицировать портфель и контролировать риски. 37 4 Рекомендации по распределению активов в зависимости от риск-профиля и временного горизонта для российских инвесторов 4 Рекомендации по распределению активов в зависимости от риск-профиля и временного горизонта для российских инвесторов Аналогичный алгоритм применен и для поиска портфелей для разных профилей риска в России. Для определения алгоритма подбора портфеля и эмпирического исследования различных по риску портфелей выбраны пять активов. Круг активов для анализа портфельного множества в России включает в себя - Индекс внутренних акций крупных компаний – индекс общей доходности Московской биржи; - Индекс внутренних акций крупных компаний – индекс общей доходности Московской биржи; - Индекс малых компаний – индекс общей доходности малых компаний Конструктор CAPM-Ru; - Индекс малых компаний – индекс общей доходности малых компаний Конструктор CAPM-Ru; - Индекс международных акций – индекс общей доходности мировых развитых рынков MSCI WORLD; - Индекс корпоративных облигаций – индекс IFX-Cbonds; - Денежные эквиваленты – безрисковая ставка в виде доходности к погашению 10-летних государственных облигаций России по данным Блумберг и расчетам автором. Данные по России собраны с 2003 года по 2020 год. Все индексы выражены в российских рублях. Безрисковая ставка определяется как 10-летная доходность к погашению в момент начала составления портфеля, то есть определяет зафиксированную доходность инвестора. Так как используются достаточно длинные интервалы времени (от 5 лет и более), то долгосрочная ставка является подходящим прокси. При более точном подходе требуется использование ставки по тем государственным облигациям, чей срок до погашения соответствует горизонту инвестирования, который моделируется. На интервала ранее 2007 года по нашим расчетам и наиболее полной выборке государственных облигаций подобраны ставки для тех ценных бумаг, чей срок до погашения был наиболее близок к 10 годам. По всем индексам, связанным с облигациями, данные ограничиваются максимум 2003 годом. Отсутствие длинных рядов данных сильно ограничивает проведенное исследование и не позволяет определять долгосрочные портфели. 38 Основные этапы алгоритма включают в себя подготовку базовых классов активов, данных по их индексным прокси и оценку их мер риска и доходности, моделирование всего портфельного множества, разбиение на портфели по профилям риска и определение оптимальных портфелей. На рисунке (рисунок 14) показаны среднегодовые доходности каждого индекса на каждом горизонте времени. Красной линией отмечено медианное значение. На коротких сроках для акций наблюдается большой разброс, однако уже с пятилетнего горизонта акции не падают в стоимости ниже уровня первоначальных инвестиций для любого портфеля. Индекс малых компаний показывает огромные приросты на краткосрочных периодах, что связано, в первую очередь, с началом периода и бурным ростом российского рынка и не отражает современные реалии. Корпоративные облигации приносили отрицательную доходность только на малом количестве наиболее коротких периодов (1–2 года). Однако медианная доходность корпоративных облигаций ниже, чем у акций (рисунок 15). Акции малых компаний наиболее привлекательны и имеют наивысшую медианную доходность. 4 Рекомендации по распределению активов в зависимости от риск-профиля и временного горизонта для российских инвесторов Акции крупных компаний России оказываются более привлекательны, чем глобальные акции развитых стран на краткосрочных и долгосрочных периодах. В среднесрочном периоде мировые акции приносят большую медианную доходность. 39 39 Рисунок 14. Доходность различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. Источник: расчеты авторов по данным Блумберг. Рисунок 14. Доходность различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. 40 Рисунок 15. Медианная доходность различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. 0 5 10 15 20 25 30 35 40 45 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Эффективная доходность CORBOND EQ SMALL WORLD Рисунок 15 Медианная доходность различных рисковых классов активов на разны 0 5 10 15 20 25 30 35 40 45 50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Эффективная доходность CORBOND EQ SMALL WORLD Рисунок 15. Медианная доходность различных рисковых классов активов на разных Рисунок 15. Медианная доходность различных рисковых классов активов на разных горизонтах инвестирования в России % 2003–2020 гг горизонтах инвестирования в России, %, 2003–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Оценка рисков этих четырех классов активов представлена на рисунке (рисунок 16). Медианное стандартное отклонение характеризует меру риска, выше которой 50% портфелей на выбранном горизонте инвестирования. Акции малых компаний обладают сопоставимым риском с акциями крупнейших компаний, входящих в индекс Московской биржи. Причина этого заключается в большей диверсификации индекса малых акций, для построения которого использована половина рынка (более 100 акций в портфеле с капитализацией менее медианной), а для индекса Московской биржи лишь 50 компаний, часть из которых коррелирована, а концентрация эмитентов внутри индекса высока. Акции глобальных компаний обладают средним риском, даже с учетом валютных рисков при пересчете значений индекса в рубли. Корпоративные облигации ожидаемо обладают наименьшим риском. Риски этих классов активов не уменьшаются с ростом горизонта инвестирования и не сближаются. 41 Рисунок 16. Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. 0 5 10 15 20 25 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Стандартное отклонение CORBOND EQ SMALL WORLD Рисунок 16. 4 Рекомендации по распределению активов в зависимости от риск-профиля и временного горизонта для российских инвесторов Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. 0 5 10 15 20 25 30 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Стандартное отклонение CORBOND EQ SMALL WORLD Рисунок 16. Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. Рисунок 16. Медианное стандартное отклонение различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Еще одной мерой риска является максимальная просадка стоимости портфеля или минимальная стоимость активов по отношению к размеру первоначальных инвестиций в течение всего периода (рисунок 17). Эта мера риска отражает вопрос в анкете о толерантности к риску, связанный с тем, способен ли инвестор выдержать временное уменьшение стоимости своего портфеля в течение срока инвестирования, не меняя стратегию и не паникуя. Данные показывают, что наибольшая медианная просадка портфеля наблюдается у глобальных акций развитых рынков на среднесрочном и долгосрочном периоде. Корпоративные облигации допускают просадку ниже 0%. Акции малых компаний преимущественно растут, медианный портфель также не допускает просадки своей стоимости. 42 Рисунок 17. Медианная максимальная просадка различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. -25 -20 -15 -10 -5 0 5 10 15 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Максимальная просадка CORBOND EQ SMALL WORLD Рисунок 17. Медианная максимальная просадка различных рисковых классов -25 -20 -15 -10 -5 0 5 10 15 20 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Максимальная просадка CORBOND EQ SMALL WORLD Рисунок 17. Медианная максимальная просадка различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Последним индикатором риска выступает показатель Value-at-Risk (VaR), который отмечает, какие максимальные потери наступают в 95% случаях для портфеля (рисунок 18). Этот индикатор предполагает измерение амплитуды ежемесячных доходностей портфеля. Медианное его значение по всем портфелям по разным горизонтам оказывается наименьшим для корпоративных облигаций. Ранжирование акций по этому индикатору для России оказывается достаточно неожиданным. Наибольшим риском обладает портфель из акций крупнейших компаний, меньшим риском – акции малых компаний, еще меньшим – глобальные акции. Причем с ростом горизонта растет и этот индикатор риска. 43 Рисунок 18. 4 Рекомендации по распределению активов в зависимости от риск-профиля и временного горизонта для российских инвесторов Медианный показатель VaR(95%) различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003-2020 гг. -50 -40 -30 -20 -10 0 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 VaR CORBOND EQ SMALL WORLD Рисунок 18. Медианный показатель VaR(95%) различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003-2020 гг. -50 -40 -30 -20 -10 0 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 VaR CORBOND EQ SMALL WORLD Рисунок 18. Медианный показатель VaR(95%) различных рисковых классов активов унок 18. Медианный показатель VaR(95%) различных рисковых классов активов на разных горизонтах инвестирования в России, %, 2003-2020 гг. Источник: расчеты авторов по данным Блумберг Источник: расчеты авторов по данным Блумберг По итогам анализа базовых активов можно сделать вывод о том, в России достаточно сложно корректно ранжировать доступные частному инвестору инструменты, особенно внутри класса активов акций. Даже с учетом валютного риска по двум мерам риска из трех глобальные акции наименее рисковые. Однако акции малых компаний показывают быстрый и стабильный рост за счет диверсификации внутри портфеля, что делает их практически лидером по соотношению риск- доходность, а для большинства периодов и лидером по наименьшему содержанию риска. Это некоторый нонсенс для развитых рынков, на которых акции малых компаний являются одними из самых рисковых. Стоит отметить, что портфель из акций малых компаний должен быть максимально диверсифицирован, иначе подобного успеха нельзя будет добиться. Осуществление такой факторной стратегии возможно только через формы коллективных инвестиций, например, через формирование паевых фондов с соответствующими факторными стратегиями. Следующий этап заключается в формировании портфельного множества из всех возможных комбинаций активов и поиске оптимального портфеля. Далее описана процедура для одного из 10-летних периодов для России, оцененного на основе исторических данных с 2011 по 2020 год. 44 Для этого сгенерированы веса портфелей с шагом в 5% из всех возможных комбинаций по 3, 4 и 5 активов из базовых: акций крупных компаний (EQ), акций малых компаний (SMALL), международных акций (WORLD), корпоративных облигаций (CORBOND) и денежных эквивалентов (RF). Максимальный вес одного актива в портфеле выбран 85% для обеспечения хотя бы минимальной диверсификации, минимальный вес равен 0. Ограничение на использование малых компаний в портфеле составило 30%, а безрисковой ставки 90%. В качестве безрисковой ставки или ставки денежных эквивалентов использована доходность к погашению десятилетних государственных облигаций России по состоянию на начало 2011 года, то есть на начало периода формированию смоделированного портфеля. 4 Рекомендации по распределению активов в зависимости от риск-профиля и временного горизонта для российских инвесторов Это значение составило 8,017%, что означает, что при 10-летнем горизонте инвестор без риска мог бы вложить все свои активы в государственные облигации, получив гарантированно 8,017% доходности. Часть средств, размещенных в денежных эквивалентах, считается, что также размещается в этот актив. Портфельное множество из перечисленных пяти активов состоит из примерно двух частей. В первой диверсификация между корпоративными облигациями, безрисковой ставкой и внутренними акциями, а во второй с глобальными акциями и акциями малых компаний в России (рисунок 19). На рисунок нанесены портфели со 100% весом каждого из активов, а также портфели, соответствующие 5 профилям риска по Schwab. Это базовые бенчмарки, используемые для сопоставления и поиска соотношений. В отличие от США в России есть выгода от диверсификации между внутренним и глобальным рынком, так как корреляции достаточно низкие, а показатели риска и доходности для глобальных акций выше. (таблица 7). Это верно и для корпоративных облигаций по сравнению с акциями. Таким образом, на российском рынке диверсификация играет важную роль. 45 Примечание: круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный) Рисунок 19. Портфельное множество из 5 базовых активов, а также смоделированные портфели Schwab в России, %, 2011–2020 гг. Источник: расчеты авторов по данным Блумберг. Таблица 7 Матрица корреляций базовых активов России, 2011-2020 гг. CORBOND EQ SMALL WORLD CORBOND 1 0.34 0.41 -0.37 EQ 0.34 1 0.57 0.19 SMALL 0.41 0.57 1 0.02 Примечание: круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный) римечание: круглыми точками обозначены портфели по разным риск-профилям рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный) Рисунок 19. Портфельное множество из 5 базовых активов, а также смоделированные портфели Schwab в России, %, 2011–2020 гг. Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Матрица корреляций базовых активов России, 2011-2020 гг. CORBOND EQ SMALL WORLD CORBOND 1 0.34 0.41 -0.37 EQ 0.34 1 0.57 0.19 SMALL 0.41 0.57 1 0.02 WORLD -0.37 0.19 0.02 1 б рица корреляций базовых активов России, 2011-2020 гг. Матрица корреляций базовых активов России, 2011-2020 гг. Источник: расчеты авторов по данным Блумберг. Как и для США, все портфели были отсортированы на 5 групп по трем мерам риска: стандартному отклонению, VaR 95% и максимальной просадке. Затем эти три 46 классификации были усреднены для получения финальной классификации портфелей по уровню риска. Среди составленных портфелей в каждой из пяти групп можно выбрать наилучший по соотношению риск-доходность. 4 Рекомендации по распределению активов в зависимости от риск-профиля и временного горизонта для российских инвесторов Для этого использовался критерий Шарпа: отношение премии портфеля по сравнению с безрисковой ставкой и риска портфеля, измеренного стандартным отклонением. Для каждой группы определен портфель с наибольшим значением критерия Шарпа (рисунок 20). Все выбранные портфели лежат на эффективной границе портфельного множества, в то время как рекомендуемые портфели Schwab в середине по риску и доходности или даже оказываются среди неэффективных портфелей. Более того, подобранные портфели позволяют не только увеличить доходность, но и снизить риски для наиболее агрессивных профилей. Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную группу, круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный); 47 квадратными – наиболее эффективные по соотношению риск-доходность; серыми – портфели, для которых группа по риску была выше, чем по доходности Рисунок 20. Портфельное множество и итоговая классификация портфелей по уровню риска, а также смоделированные портфели Schwab в России, %, 2011–2020 гг. квадратными – наиболее эффективные по соотношению риск-доходность; серыми – портфели, для которых группа по риску была выше, чем по доходности Рисунок 20. Портфельное множество и итоговая классификация портфелей по уровню риска, а также смоделированные портфели Schwab в России, %, 2011–2020 гг. гг. Источник: расчеты авторов по данным Блумберг. Статистика и сопоставление по двум типам портфелей – по рекомендациям инвестиционных консультантов и по найденным оптимальным решениям – приведена в таблице (таблица 8). Алгоритм оптимизации для 2011–2020 горизонта инвестирования выбирает преимущественно портфели без безрискового актива и без внутренних акций крупнейших компаний. Эти два актива оказались наименее эффективными на этом конкретном периоде. Доля корпоративных облигаций подбирается соответствующая или немногим выше рекомендуемых портфелей. Отличия между портфелями соответствуют теоретическому распределению активов: чем консервативнее инвестор, тем больше доля облигаций в составе его портфеля. Остальная часть портфеля делится между акциями малых компаний и глобальными акциями, причем у вторых доля в портфеле в 1,5–2 раза больше для любого уровня риска. Так, в агрессивном распределении на глобальные акции приходится 60%, а на внутренние акции малых компаний 30%. Таблица 8 Веса и параметры риска и доходности портфелей Schwab и оптимальных портфелей для российского инвестора, %, 2011-2020 гг. Schwab Оптимальные портфели Уровень риска 1 2 3 4 5 1 2 3 4 5 CORBOND 50 50 35 15 0 70 55 40 25 10 EQ 15 25 35 45 50 0 0 0 0 0 SMALL 0 5 10 15 20 10 20 25 30 30 WORLD 5 10 15 20 25 20 25 35 45 60 RF 30 10 5 5 5 0 0 0 0 0 Доходность 10.3 12.2 13.7 15.2 16.5 13.1 14.9 16.7 18.5 20.2 Риск 3.3 5.7 8.0 10.3 11.9 3.7 5.2 6.7 8.4 10.4 Коэффициент Шарпа 0.6 0.7 0.7 0.7 0.7 1.3 1.3 1.2 1.2 1.1 а и параметры риска и доходности портфелей Schwab и оптимальных портфелей российского инвестора, %, 2011-2020 гг. Источник: расчеты авторов по данным Блумберг. Проблемой проведенного анализа является то, что в конкретной точке времени и на конкретном временном интервале подобранные портфели могут, действительно, быть эффективнее, однако создание рекомендация для разных профилей риска 48 является более комплексной задачей. Так, рекомендуемые портфели должны быть достаточно универсальными и оптимальными для всех периодов времени. является более комплексной задачей. Так, рекомендуемые портфели должны быть достаточно универсальными и оптимальными для всех периодов времени. С этой целью процедура поиска оптимального портфеля проведена для всех доступных 10-летних отрезках на периоде с 2003 по 2020 гг. (всего 9 периодов с шагом в один год). Дополнительно наложено ограничение размера денежных эквивалентов не более 90%, а акций малых компаний 30%. Основным выводом из рассмотрения всех 10-летних интервалов является тот факт, что оптимальный портфель не постоянен. Его веса и параметры риска и доходности меняются достаточно сильно. Поэтому для поиска некоторого универсального распределения полученные результаты необходимо усреднить и получить некоторый средний оптимальный портфель, единый для всех периодов для каждой группы риска. Таким образом, оптимальные портфели определяются как взвешенные средние среди всех 10-летних оптимальных портфелей, рассчитанных в моменты спокойного состояния рынка. Это предполагает корректное с точки зрения теории расположение активов по риску и доходности: например, риск и доходность акций выше, чем риск и доходность облигаций. Наибольший вес имеет портфель, рассчитанный на последних данных, наименьший вес – наиболее далекие в исторической ретроспективе портфели. На рисунке (рисунок 21) показаны портфельные множества для каждого из периодов, подтверждающие сделанные выводы, а также отражающие единый и универсальный оптимизированный портфель для каждого профиля риска. Таблица 8 49 Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную Примечание: наибольший риск имеют портфели, отнесенные в пятую, красную группу, круглыми точками обозначены портфели по разным риск-профилям, рекомендуемые Schwab от консервативного (зеленый) до агрессивного (красный); квадратными – универсальный оптимизированный портфель для каждого профиля риска; серыми – портфели, для которых группа по риску была выше, чем по доходности Рисунок 21. Портфельные множества и итоговая классификация портфелей по уровню риска, а также смоделированные портфели Schwab и универсальный Рисунок 21. Портфельные множества и итоговая классификация портфелей по уровню риска, а также смоделированные портфели Schwab и универсальный оптимальный портфель для второй половины 10-летних периодов в России, %, 2003– 2020 гг. уровню риска, а также смоделированные портфели Schwab и универсальный оптимальный портфель для второй половины 10-летних периодов в России, %, 2003– 2020 гг. Источник: расчеты авторов по данным Блумберг. Выбранный алгоритмом портфель для каждого профиля риска для 10-летнего горизонта существенно отличается от рекомендаций Schwab (таблица 9). В нем 50 гораздо больший вес приходится на безрисковый актив для всех типов инвесторов, кроме агрессивного. Причем, чем меньше толерантность к риску, тем в большей степени отличается подобранный портфель от международных рекомендаций. Доля корпоративных облигаций пропорционально уменьшается так, что общая доля государственных и корпоративных облигаций в портфеле остается примерно равной международной рекомендации. Внутренний рынок акций практически не задействован, что является следствием его крайней нестабильности и не способности приносить инвестору выгодное для него соотношение доходности и риска. Причем использован индекс общей доходности, с учетом дивидендов, что должно было бы повысить его привлекательность в следствие достаточно высокой по международным меркам дивидендной доходности крупнейших российских компаний. Однако глобальные акции и акции малых компаний остаются более выгодной инвестицией. Глобальная диверсификация, в отличие от аналогичных расчетов для США, используется очень активно и рекомендуется всем типам инвесторов в следствие высоких рисков внутренних рыночных активов. Таким образом, глобальная диверсификация крайне важна для российского инвестора и иностранные активы в виде максимально диверсифицированного портфеля позволят улучшить характеристики портфеля даже с учетом налагаемых валютных рисков. Таблица 9 Веса портфелей Schwab и универсальный оптимизированный портфель для каждого профиля риска для российского инвестора и 10-летнего горизонта инвестирования в России % 2003 2020 гг Веса портфелей Schwab и универсальный оптимизированный портфель для каждого профиля риска для российского инвестора и 10-летнего горизонта инвестирования в России % 2003–2020 гг России, %, 2003 2020 гг. Таблица 8 Schwab Оптимальные портфели Уровень риска 1 2 3 4 5 1 2 3 4 5 CORBOND 50 50 35 15 0 17 22 20 13 3 EQ 15 25 35 45 50 0 0 0 1 6 SMALL 0 5 10 15 20 13 23 29 30 28 WORLD 5 10 15 20 25 11 18 26 45 59 RF 30 10 5 5 5 60 38 24 11 4 Источник: расчеты авторов по данным Блумберг. Источник: расчеты авторов по данным Блумберг. Причиной высокого веса акций малых компаний, по всей видимости, стал долгосрочный период в 10 лет, на котором фундаментальные факторы, включая размер, работают и приносят инвестору высокую избыточную доходность. А также высокая диверсификация портфеля акций малых компаний и быстрый их рост. 51 Таким образом, для российского инвестора разработан алгоритм, который позволяет определять оптимальные портфели для разных профилей риска и обосновывать корректность предлагаемых инвестиционными консультантами рекомендаций. Показано, что для 10-летнего периода оптимальный портфель должен иметь глобальную диверсификацию, а также достаточно высокий вес безрискового актива, что является следствием высоких ставок и низкой рыночной премии в России. Для других периодов проведены аналогичные расчеты и составлены оптимальные портфели (таблица 10). Таблица 10 Универсальный оптимизированный портфель для каждого профиля риска и временного горизонта для российского инвестора, %, 2003–2020 гг. Оптимальные портфели Уровень риска 1 2 3 4 5 3 года CORBOND 33 30 30 21 9 EQ 4 3 7 18 25 SMALL 13 20 27 29 24 WORLD 11 11 17 23 34 RF 39 37 19 9 8 5 лет CORBOND 27 27 22 12 9 EQ 2 1 6 12 22 SMALL 14 19 27 28 25 WORLD 10 12 20 31 37 RF 48 41 25 17 7 7 лет CORBOND 38 33 26 16 6 EQ 0 0 2 8 14 SMALL 14 21 29 30 28 WORLD 13 18 28 38 49 RF 35 28 15 9 3 10 лет CORBOND 17 22 20 13 3 EQ 0 0 0 1 6 SMALL 13 23 29 30 28 WORLD 11 18 26 45 59 RF 60 38 24 11 4 20 лет CORBOND 0 0 18 15 0 EQ 0 0 0 0 7 SMALL 16 30 30 30 28 Универсальный оптимизированный портфель для каждого профиля риск временного горизонта для российского инвестора, %, 2003–2020 гг. 52 WORLD 5 12 32 50 65 RF 78 58 20 5 0 Источник: расчеты авторов по данным Блумберг. WORLD 5 12 32 50 65 RF 78 58 20 5 0 Источник: расчеты авторов по данным Блумберг. По итогам расчетов показано, что у российских инвесторов с низкой и умеренной толерантностью к риску в портфеле большая часть активов должна быть распределена в безрисковый актив. Причем при увеличении горизонта, вопреки ожиданиям, доля безрискового актива только увеличивается. Это происходит из-за того, что на долгосрочных горизонтах снижается эффективность внутренних рисковых инструментов, как и премия за риск. Примечательно также, что даже для агрессивного типа профиля сохраняется ненулевая доля безрискового актива, кроме самых длинных горизонтов. Аналогичная тенденция верна и для корпоративных облигаций. С ростом горизонта их доля в портфеле уменьшается для всех профилей риска. Причем на длинных горизонтах доля корпоративных облигаций переходит в безрисковую для консервативных типов инвесторов, а для агрессивных – в глобальные акции. Акции малых компаний присутствуют в каждом портфеле с весом не менее 10% даже для консервативных инвесторов. Это подчеркивает важность диверсификации портфелей и выбор таких индексных стратегий внутри заданного класса, который включал бы максимальное число активов с малой концентрацией эмитентов внутри. Это подтверждается и практически нулевым весом внутренних акций крупнейших компаний, которые представлены индексом общей доходности (с учетом дивидендов) Московской биржи. Глобальная диверсификация для российского инвестора крайне важна и присутствует во всех оптимальных распределениях активов. Причем, чем длиннее горизонт инвестирования, тем больше средств должно быть размещено в диверсифицированный портфель иностранных активов. 53 Заключение Определение риск-профиля инвестора должно заканчиваться рекомендацией наиболее подходящего распределения активов. Проблема существующего подхода заключается в малой эмпирической обоснованности того или иного веса актива в общем портфеле. Принцип построения портфеля для определенного профиля риска основан на предпосылке о том, что акции рискованнее облигаций, а те рискованнее денежных эквивалентов. И хотя в большинстве случаев это так, закономерности или правила определения конкретных весов в портфеле остаются неопределенными. Все это приводит к широкому разнообразию «оптимальных» весов даже у пятерки лидеров инвестиционного консультирования в США и еще большей неоднозначности для развивающихся рынков. Основной задачей является составление набора эмпирических фактов, на основе которых можно было бы построить простой и прозрачный алгоритм, обосновывающий вес конкретного актива в портфеле для конкретного уровня риска. Для определения алгоритма подбора портфеля и эмпирического исследования различных по риску портфелей предложен алгоритм, позволяющий находить оптимальное распределение активов для разных профилей риска с учетом горизонта инвестирования. Показано, что у российских инвесторов с низкой и умеренной толерантностью к риску в портфеле большая часть активов должна быть распределена в безрисковый актив. Причем при увеличении горизонта доля безрискового актива только увеличивается, так как снижается премия за риск внутренних рисковых инструментов. Для корпоративных облигаций с ростом горизонта доля в портфеле уменьшается для всех профилей риска. Причем на длинных горизонтах доля корпоративных облигаций переходит в безрисковую для консервативных типов инвесторов, а для агрессивных – в глобальные акции. Акции малых компаний присутствуют в каждом портфеле с весом не менее 10% даже для консервативных инвесторов. Это подчеркивает важность диверсификации портфелей и выбор таких индексных стратегий внутри заданного класса, который включал бы максимальное число активов с малой концентрацией эмитентов внутри. Это подтверждается и практически нулевым весом внутренних 54 акций крупнейших компаний, которые представлены индексом общей доходности (с учетом дивидендов) Московской биржи. Глобальная диверсификация для российского инвестора крайне важна и присутствует во всех оптимальных распределениях активов. Причем, чем длиннее горизонт инвестирования, тем больше средств должно быть размещено в диверсифицированный портфель иностранных активов. Таким образом, с помощью предложенного алгоритма обоснованы оптимальные распределения, протестированы существующие рекомендации ведущих инвестиционных консультантов в развитых странах, а также выявлены основные правила и закономерности для инвестирования на российском рынке в долгосрочном периоде. 55 55 12. Blake D., Haig A. How Do Savers Think About and Respond To Risk? // Evidence from a Population Survey and Lessons for the Investment Industry. London: The Pensions Institute, -2014. Список источников 1. Finke M., Guillemette M. Measuring risk tolerance: A review of literature // Journal of Personal Finance. - 2016. Vol. 15, Is. 1, PP. 63-76. 2. Hansen L., Singleton K. J. Generalized instrumental variables estimation of nonlinear rational expectations models // Econometrica. – 1982. Vol. 50, PP. 1269- 1286. 3. Hall R. Intertemporal substitution in consumption // Journal of Political Economy. – 1988. Vol. 96, PP. 339-357. 4. Neely C. J.; Roy A., Whiteman C. H. Risk aversion versus intertemporal substitution: a case study of identification failure in the intertemporal consumption capital asset pricing model // Journal of Business & Economic Statistics – 2001. Vol. 19(4), PP. 395-403. 5. Layard R., Mayraz G., Stephen N. J. The Marginal Utility of Income // Journal of Public Economics. – 2008. Vol. 92(8-9), PP. 1846-57. 6. Sung J., Hanna S. Factors related to risk tolerance // Financial Counseling and Planning. – 1996. Vol. 7(1), PP. 11–20. 7. Grable J. E. Financial risk tolerance and additional factors that affect risk taking in everyday money matters // Journal of Business and Psychology. – 2000. Vol. 14(4), PP. 625–630. 8. Van Rooij M., Lusardi A., Alessie R. Financial literacy and stock market participation // Journal of Financial Economics. – 2012. Vol. 101(2), PP. 449–472. 9. Yao R., Curl A. L. Do market returns influence risk tolerance? Evidence from panel data // Journal of Family and Economic Issues. – 2011. Vol. 32(3), PP. 532–544. 10. Hubble A., Grable J., Dannhauser R. W. Investment risk profiling a guide for financial advisors // Report CFA Institute. – 2020. Available at https://www.cfainstitute.org/-/media/documents/survey/investment-risk- profiling.pdf 11. Roszkowski M. J. How to Assess an Investor’s Financial Risk Tolerance: The Basics // Personal Finance Risk Tolerance, Bryn Mawr, PA: The American College, 1992. 56 12. Blake D., Haig A. How Do Savers Think About and Respond To Risk? // Evidence from a Population Survey and Lessons for the Investment Industry. London: The Pensions Institute, -2014. 57 57
https://openalex.org/W2580730749	https://europepmc.org/articles/pmc5292202?pdf=render	English	null	Hyperforin/HP-<i>β</i>-Cyclodextrin Enhances Mechanosensitive Ca<sup>2+</sup> Signaling in HaCaT Keratinocytes and in Atopic Skin Ex Vivo Which Accelerates Wound Healing	BioMed research international	2,017	cc-by	8,961	Hindawi Publishing Corporation BioMed Research International Volume 2017, Article ID 8701801, 9 pages http://dx.doi.org/10.1155/2017/8701801 Hindawi Publishing Corporation BioMed Research International Volume 2017, Article ID 8701801, 9 pages http://dx.doi.org/10.1155/2017/8701801 Hindawi Publishing Corporation BioMed Research International Volume 2017, Article ID 8701801, 9 pages http://dx.doi.org/10.1155/2017/8701801 Hiroya Takada,1,2 Jun Yonekawa,1 Masami Matsumoto,2 Kishio Furuya,3 and Masahiro Sokabe3 1Department of Physiology, Nagoya University Graduate School of Medicine, 65 Tsurumai, Nagoya 466-8550, Japan 2Pixy Central Research Institute, 3-7-1 Kamitsuchidananaka, Ayase, Kanagawa 252-1113, Japan 3Mechanobiology Laboratory, Nagoya University Graduate School of Medicine, 65 Tsurumai, Nagoya 466-8550, Japan 1Department of Physiology, Nagoya University Graduate School of Medicine, 65 Tsurumai, Nagoya 466-8550, Japan 2Pixy Central Research Institute, 3-7-1 Kamitsuchidananaka, Ayase, Kanagawa 252-1113, Japan 3Mechanobiology Laboratory, Nagoya University Graduate School of Medicine, 65 Tsurumai, Nagoya 466-8550, Japan Correspondence should be addressed to Kishio Furuya; furuya@med.nagoya-u.ac.jp Received 16 September 2016; Accepted 29 November 2016; Published 22 January 2017 Academic Editor: Adam Reich Academic Editor: Adam Reich Copyright © 2017 Hiroya Takada et al. This is an open access article distributed under the Creative Commons A which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is p Copyright © 2017 Hiroya Takada 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. Cutaneous wound healing is accelerated by mechanical stretching, and treatment with hyperforin, a major component of a traditional herbal medicine and a known TRPC6 activator, further enhances the acceleration. We recently revealed that this was due to the enhancement of ATP-Ca2+ signaling in keratinocytes by hyperforin treatment. However, the low aqueous solubility and easy photodegradation impede the topical application of hyperforin for therapeutic purposes. We designed a compound hydroxypropyl- 𝛽-cyclodextrin- (HP-𝛽-CD-) tetracapped hyperforin, which had increased aqueous solubility and improved photoprotection. We assessed the physiological effects of hyperforin/HP-𝛽-CD on wound healing in HaCaT keratinocytes using live imaging to observe the ATP release and the intracellular Ca2+ increase. In response to stretching (20%), ATP was released only from the foremost cells at the wound edge; it then diffused to the cells behind the wound edge and activated the P2Y receptors, which caused propagating Ca2+ waves via TRPC6. This process might facilitate wound closure, because the Ca2+ response and wound healing were inhibited in parallel by various inhibitors of ATP-Ca2+ signaling. We also applied hyperforin/HP-𝛽-CD on an ex vivo skin model of atopic dermatitis and found that hyperforin/HP-𝛽-CD treatment for 24 h improved the stretch-induced Ca2+ responses and oscillations which failed in atopic skin. 1. Introduction known to be a TRPC6 activator, further accelerated wound closure [2]. We revealed that the facilitation of wound closure by mechanical stretching and hyperforin occurs due to the release of ATP via mechanosensitive hemichannels at the wound edge and the P2Y receptor-mediated Ca2+influx via TRPC6 in the cells located behind the wound edge using real- time ATP luminescence imaging and Ca2+ fluorescence mea- surement [2]. The influx of Ca2+ through TRPC6 channels was also reported to be essential for wound healing in vivo in TRPC6 knockout mice [3]. Epidermal keratinocytes are located at the surface of the skin and are exposed to various environmental stimuli including mechanical and physical stimuli and are susceptible to these stimuli. During the wound healing process, these exogenous stimuli and the endogenous stimuli, such as the tension and traction forces generated between the migration of the foremost cells and the cells that are located behind them, may affect the rate of wound closure. Our earlier study demon- strated that mechanical stretching facilitated wound closure in bovine aortic endothelial cells [1]. We recently reported that wound healing in HaCaT keratinocytes was accelerated by stretching and treatment with hyperforin, which is a major component of a traditional herbal medicine and which is Hyperforin is a major active constituent of St. John’s wort (Hypericum perforatum L.) extract, which is widely used in traditional herbal medicines, to promote wound healing [4–9]. The use of hyperforin-rich cream as a topical 2 BioMed Research International in a freezer until use. All of the procedures were performed under light-shielded conditions. medication for atopic dermatitis was recently reported [10– 14]. In spite of its potential therapeutic activities, the extreme sensitivity of hyperforin to photodegradation has impeded its topical application. The complexation of St. John’s wort extract with 𝛽- and 𝛾-cyclodextrin (CD) was reported to enhance the photoprotection and solubility of hyperforin in aqueous solutions [15–17]. In the present study, we aimed to develop a novel formation of encapsulated hyperforin with hydroxypropyl-𝛽-cyclodextrin (HP-𝛽-CD) to improve its aqueous solubility and photostability, because HP-𝛽-CD has been shown to possess the highest solubility not only in water but also in ethanol among several of the CD compounds that are commonly used. We also assessed the effects of the compound on the wound healing and ATP-Ca2+ signaling in HaCaT keratinocytes.l Stoichiometry of the reaction between hyperforin and HP-𝛽-CD was spectroscopically determined. 1. Introduction The concentra- tion of hyperforin in these studies was 4.66 × 10−4 M whereas the HP-𝛽-CD concentration was used in the range of 0– 8.0 equivalents. The UV spectra of hyperforin were recorded using a UV/VIS scanning spectrophotometer (Gene Spec III, Hitachi Naka Instruments, Hitachinaka, Japan). The changes in the absorbance of hyperforin following the addition of various concentrations of the HP-𝛽-CD complexing agent were measured at 𝜆max 281 ± 7 nm. 2.3. The Analysis of Irradiated Hyperforin Solution by HPLC. An irradiation test was performed using a 6-watt LED light bulb (total luminous flux 480 lm, color temperature: 6700 K, Panasonic, Osaka, Japan) that was placed 14 cm above the samples. Irradiation was conducted in a dark room under temperature control (25∘C). Aliquots of 40 𝜇L were taken every 30 min for the analysis. All of the quantitative measure- ments were conducted using a Hitachi LaChrom Elite HPLC system (Hitachi High-Technologies, Tokyo, Japan) equipped with a quaternary pump (L-2130), an autosampler (L-2200), a column oven (L-2300), and a diode array detector (DAD/L- 2450). Separation was performed using a TSKgel ODS-100Z reversed phase column (4.6 mm × 250 mm, 5 𝜇m, Tosho, Tokyo, Japan) with a mobile phase composed of acetonitrile- water-methanol-trifluoroacetic acid (72 : 18 : 10 : 0.5, v/v/v/v). The flow rate was 1.6 mL/min. The UV detector was set at 270 nm. Curve fitting was performed using Excel (MS Office 2013) to minimize the 𝑅2 value. Atopic dermatitis is a chronic inflammatory skin disease that develops due to various factors that are associated with epidermal barrier dysfunction [18]. It is known that the Ca2+ gradient in the epidermis is necessary for maintaining the barrier function; however, the Ca2+ dynamics of atopic skin remain to be elucidated. We herein measured the stretch- induced Ca2+ responses ex vivo in atopic skin using a confocal microscope. We found that the Ca2+ responses were impaired in the atopic epidermis and that the responses recovered after the application of hyperforin/HP-𝛽-CD. The data suggested that hyperforin/HP-𝛽-CD is a potent targeted therapeutic agent that can be used to promote epidermal wound healing and treat atopic dermatitis. 2. Material and Methods 2.1. Reagents. Hyperfolin/hydroxypropyl-𝛽-cyclodextrin was prepared by the complexation of hyperforin (Cayman Chemical, Ann Arbor, MI) and hydroxypropyl-𝛽-cyclo- dextrin (HP-𝛽-CD; CycloChem, Tokyo, Japan) as described below. The other chemicals and reagents were as follows: carbenoxolone disodium salt (CBX), apyrase (from potato), GdCl3, U73122, and Cremophor EL (Sigma-Aldrich, St. Louis, MO); ionomycin (Calbiochem, San Diego, CA); suramin hexasodium (RBI, Natick, MA); GsMTx-4 (Peptide Institute, Osaka, Japan); diC8-PIP2 (Echelon Biosciences, Salt Lake City, UT); dispase (Godo Shusei, Tokyo, Japan); Fluo-8 AM (AAT Bioquest, Sunnyvale, CA); Cellmatrix type IA (Nitta Gelatin, Osaka, Japan); Lipofectamine (18324, Invi- trogen, Carlsbad, CA); DME/F12 (D9785; Sigma-Aldrich, St. Louis, MO); FBS (12483; Gibco, Carlsbad, CA). 2.4. Cell Culture. HaCaT human keratinocyte cells [19] at passages 36 and 37 were purchased from Cell Lines Services (CLS, Heidelberg, Germany) and were grown in DME/F12 (0.07 mM Ca2+) supplemented with 2% FBS at 37∘C in a humidified atmosphere of 5% CO2. The growth medium was prepared from DME/F12 (D9785; Sigma-Aldrich) by adding 0.07 mM Ca2+, 365 mg/L L-glutamine, 59.05 mg/L L-leucine, 91.25 mg/L L-lysine⋅HCl, 61.2 mg/L MgCl2⋅6H2O, 48.84 mg/L MgSO4 (anhydrous), 17.2 mg/L L-methionine, and 1.2 g/L NaHCO3 and was adjusted to pH 7.4 with 1 mM NaOH. For the experiments, the cells were seeded on a collagen- coated (Cellmatrix type IA) silicone stretch chamber (see the following) or 15 mm round glass coverslips (Matsunami, Osaka, Japan) and cultured in DME/F12 (1.05 mM Ca2+) supplemented with 10% FBS to allow cell attachment. After 1 day, the medium was replaced with DME/F12 (0.07 mM Ca2+) supplemented with 2% FBS, and the cells were further incubated for 1 day to achieve confluence.h 2.2. The Preparation of Hydrophilic and Stable Hyperforin/HP- 𝛽-CD. Solutions of 4.66 × 10−4 M hyperforin (in 1 mL methanol) and 1.86 × 10−3 M HP-𝛽-CD (in 1 mL ethanol) were mixed and then stirred for 30 min. The solvents were then removed in vacuo with a centrifugal evaporator (0.1 Mpa, 2800 rpm, 90 min, WKN-PV-1200, Wakenyaku, Kyoto, Japan) at ambient temperature. The obtained white solid was dissolved in Milli-Q water by ultrasonication for at least 10 min. The resulting aqueous solution of hyperforin/HP-𝛽- CD was syringe-filtered with a 0.20 𝜇m pore size and kept l The physiological experiments were performed as de- scribed previously [2]. A brief explanation follows. 2.5. Cell Stretch Experiments and Wound Closure Assay. 2. Material and Methods HaCaT cells were seeded on collagen-coated silicone stretch chambers or 15 mm round glass coverslips at 3 × 105 cells/cm2 and grown to confluence. A narrow cell- free gap (about 250 𝜇m) was created in a fully confluent monolayer by removing a silicone strip that was attached to the bottom of the stretch chamber during cell seeding. The wound closure process was monitored every 3 h after making the scratch using an inverted microscope (IX-70 Olympus) with a 4x (UPlanFL N, 0.13) objective. The wound closure speed was defined as the percentage of the wound closure area, which was calculated from the ratio of the final migrated area to the initial cell-free area. 2.6. Intracellular Ca2+ Measurement and Real-Time Imaging of the Released ATP. At 3 h after making a scratch, the HaCaT cells in the stretch chamber were loaded with 1 𝜇M Fluo-8 AM using 0.1–0.2% of Cremophor EL (Sigma-Aldrich) for 40– 60 min in an incubator at 37∘C. After washing away the dye with DME/F12 containing 2.0 mM Ca2+, the chamber with the cells was attached to the stretching device on the stage of an inverted microscope with 4x (UPlanFL N, 0.13) or 10x (UPlanFL 0.30) objectives. Time-lapse Fluo-8 fluorescence images were acquired at 0.5 s intervals using MetaMorph software (v6.3 and 7.5, Molecular Devices, Downingtown, PA).h 3. Results 3.1. Preparation of Stable Hydrophilic Hyperforin Encapsulated in HP-𝛽-CD and Its Effect on Wound Closure. To improve the photostability and aqueous solubility of hyperforin, it was molecularly encapsulated in cyclodextrin. Hyperforin was complexed with hydroxypropyl-𝛽-cyclodextrin (HP-𝛽- CD) at different molar ratios using the solvent evaporation method. The hyperforin/HP-𝛽-CD complexes were investi- gated with UV/Vis spectroscopy in aqueous solution. The molar ratio method was used to determine the stoichiometry of the inclusion complex formed by hyperforin and HP- 𝛽-CD. ΔA, the difference in the absorbance of hyperforin with and without HP-𝛽-CD, was plotted against the molar ratio of HP-𝛽-CD to hyperforin at 280 nm (Figure 1(a)). The curve for hyperforin/HP-𝛽-CD showed an inflexion point at a ratio of 1 : 4, suggesting that the inclusion complex formed HP-𝛽-CD tetracapped hyperforin at hemiterpene terminal moieties (Figure 1(b)). Next, we checked the light stability of a 1 : 4 complex of hyperforin/HP-𝛽-CD using HPLC. Fig- ure 1(c) shows the visible light-induced degradation curves of hyperforin and the hyperforin/HP-𝛽-CD complex. The apparent half-life of hyperforin was 30 min, while that of the hyperforin/HP-𝛽-CD complex was prolonged to 180 min. A curve fitting analysis by single exponential decay with a baseline showed a large baseline (44%) in the curve for hyperforin/HP-𝛽-CD, suggesting the existence of a nonde- graded (photoprotected) form in the complex (Figure 1(c) fitting line). The stretch-induced release of ATP was measured in real-time using the imaging system, as described previously [20]. Briefly, the luciferin-luciferase ATP bioluminescence was detected using a high-sensitivity camera system simul- taneously with infrared DIC imaging to monitor exact cell locations and extension during stretching. At 3 h after making a scratch, the cells in the stretching chamber were attached to the stretching device on the stage of an upright microscope (BX51WI, Olympus) with a 4x objective (340 Fluor XL, 0.28) and the medium was replaced with DME/F12 medium (2.0 mM Ca2+ and 10 mM HEPES, pH 7.4) containing high- sensitivity luciferin-luciferase solution (60315; Lucifer HS Set, Kikkoman Biochemifa, Tokyo, Japan). Images were acquired using the MetaMorph software with a stream acquisition mode (exposure time 100 ms). 2.7. The Knockdown of TRPC6 by shRNA. TRPC6 shRNA plasmids that coexpressed RFP (TF308626; OriGene Technologies, Rockville, MD) were used. The shTRPC6 targeting sequence was 59-AAGCAGGACATCTCAAGT- CTCCGCTATGA-39. A scrambled noneffective plasmid with the same nucleotide content was used as a negative control. 2. Material and Methods The cells were cultured in a stretch chamber molded out of Silpot 184 W/C silicone elastomers (Dow Corning Toray, Tokyo, Japan). A chamber with cultured cells was attached to a stretching machine (NS-600W or ST-600W, STREX, Osaka, 3 BioMed Research International Pixy Central Institute Ethics Committee, 2009. The sample of skin tissue was placed in PBS prior to treatment with dispase. After overnight digestion with 500 U/mL dispase in serum-free F12/DME with or without hyperforin/HP-𝛽- CD at 4∘C, the epidermis was peeled off from the dermis with forceps. The detached pieces of epidermis were fixed with intradermal needles on an elastic silicone chamber and were further incubated at 37∘C in a humidified 5% CO2 atmosphere for 12 h. Epidermis tissue was loaded with 1 𝜇M Fluo-8 AM using 0.2% of Cremophor EL in culture media for 1 h at 37∘C. After washing away the dye with DME/F12 con- taining 2.0 mM Ca2+, the chamber containing the cells was attached to a pulse-motor-driven stretching machine (NS- 600W or ST-600W, STREX) mounted on the stage of an inverted laser scanning confocal microscope (LSM510 with a 10x lens, Carl Zeiss, Jena, Germany). Time-lapse Fluo-8 fluorescence and Nomarski differential interference contrast images were acquired at 1 s intervals. The Ca2+ imaging experiments were performed at room temperature (24±2∘C). Japan) mounted on the stage of an inverted microscope (IX- 70, Olympus, Tokyo, Japan) for intracellular Ca2+ imaging or an upright microscope (BX51WI, Olympus) for extracellular ATP imaging. HaCaT cells were seeded on collagen-coated silicone stretch chambers or 15 mm round glass coverslips at 3 × 105 cells/cm2 and grown to confluence. A narrow cell- free gap (about 250 𝜇m) was created in a fully confluent monolayer by removing a silicone strip that was attached to the bottom of the stretch chamber during cell seeding. The wound closure process was monitored every 3 h after making the scratch using an inverted microscope (IX-70 Olympus) with a 4x (UPlanFL N, 0.13) objective. The wound closure speed was defined as the percentage of the wound closure area, which was calculated from the ratio of the final migrated area to the initial cell-free area. Japan) mounted on the stage of an inverted microscope (IX- 70, Olympus, Tokyo, Japan) for intracellular Ca2+ imaging or an upright microscope (BX51WI, Olympus) for extracellular ATP imaging. 3. Results Each shRNA at a concentration of 45 nM was transfected into HaCaT cells using Lipofectamine reagent, according to the manufacturer’s instructions. 2.7. The Knockdown of TRPC6 by shRNA. TRPC6 shRNA plasmids that coexpressed RFP (TF308626; OriGene Technologies, Rockville, MD) were used. The shTRPC6 targeting sequence was 59-AAGCAGGACATCTCAAGT- CTCCGCTATGA-39. A scrambled noneffective plasmid with the same nucleotide content was used as a negative control. Each shRNA at a concentration of 45 nM was transfected into HaCaT cells using Lipofectamine reagent, according to the manufacturer’s instructions. i We previously demonstrated that the wound closure of keratinocytes was accelerated by stretching and that hyperforin treatment further enhanced the effect [2] (Fig- ure 1(d)). In the present study, we examined the effect of hyperforin/HP-𝛽-CD on wound closure. A confluent monolayer of HaCaT cells cultured on silicone membrane was linearly scratched to create a cell-free gap of ∼250 𝜇m width, and the wound was allowed to heal under various conditions. Figure 1(d) shows representative wound closing 2.8. Ex Vivo Skin Preparation and Live Ca2+ Imaging of the Epidermis. Biopsies were taken from the outer forearm of a volunteer with atopic skin. Written informed consent was obtained from the volunteer. The study was approved by the BioMed Research International 4 0 0.2 0.4 0.6 0.8 1 1.2 1.4 0 1 2 3 4 5 6 7 8 Molar ratio of HP-𝛽-CD to hyperforin ΔA 0 20 40 60 80 100 Wound closure area (%) Stretch + hyperforin Stretch + hyperforin HP-𝛽-CD Stretch Control (a) 0 h 200 𝜇m 0 0.2 0.4 0.6 0.8 1 1.2 0 1 2 3 4 5 6 7 8 (b) O O O HO Hyperforin HP-𝛽-CD (c) (d) 20% stretch 20% stretch 20% stretch 6 h 20% stretch 20% stretch 20% stretch Molar ratio of HP-𝛽-CD to hyperforin ΔA ∗ ∗∗ ∗∗ ∗∗∗ (n = 5–8, ∗p < 0.05, ∗∗p < 0.5, ∗∗∗p < 0.1) Remaining hyperforin (%) 100 80 60 40 20 0 Time (min) 0 30 60 90 120 150 180 Hyperforin/HP-𝛽-CD Hyperforin Figure 1: The complexation of hyperforin with hydroxypropyl-𝛽-cyclodextrin (HP-𝛽-CD) and its effects on wound closure in HaCaT cells. (a) The molar ratio graph obtained by UV/Vis spectra measurements of the inclusion complex formed by hyperforin (4.66 × 10−4 M) and HP-𝛽- CD (0–8.0 equivalents) at 25∘C. (b) A possible model of the 1 : 4 complex of hyperforin/HP-𝛽-CD. 3. Results (c) The photodegradation of hyperforin/HP- 𝛽-CD in aqueous solution and hyperforin in methanol induced by LED light exposure. Curve fitting was performed by single exponential decay with a baseline. The baselines obtained for hyperforin/HP-𝛽-CD and hyperforin were 44% and 1.4%, respectively. (d) The effects of hyperforin and hyperforin/HP-𝛽-CD treatments on wound closure in keratinocytes under sustained stretching. Stretch stimulation (20%) facilitated wound closure in HaCaT keratinocytes (stretch). Treatment with hyperforin (1 𝜇M) further accelerated the wound closure (stretch + hyperforin) and the wound gap was nearly closed at 6 h after scratching. Hyperforin/HP-𝛽-CD (1 𝜇M as hyperforin) showed equal or greater efficacy to hyperforin (stretch + hyperforin/HP-𝛽-CD) in promoting wound closure. The data are shown as representative DIC images (upper pictures at 0 h and 6 h) and by the averages of the calculated percentage of the wound closure areas (lower graph). The quantitative data in (c) and (d) are shown as the mean ± SEM. 0 20 40 60 80 100 Wound closure area (%) Stretch + hyperforin Stretch + hyperforin HP-𝛽-CD Stretch Control 0 h 200 𝜇m (d) 20% stretch 20% stretch 20% stretch 6 h 20% stretch 20% stretch 20% stretch ∗ ∗∗ ∗∗ ∗∗∗ (n = 5–8, ∗p < 0.05, ∗∗p < 0.5, ∗∗∗p < 0.1) l-𝛽-cyclodextrin (HP-𝛽-CD) and its effects on wound closure in HaCaT cells. (a ments of the inclusion complex formed by hyperforin (4.66 × 10−4 M) and HP-𝛽 4 complex of hyperforin/HP-𝛽-CD. (c) The photodegradation of hyperforin/HP 0 h 200 𝜇m 20% stretch 20% stretch 20% stretch 6 h 20% stretch 20% stretch 20% stretch ∗ ∗∗ ∗∗∗ 20% stretch 20% stretch (a) (b) O O O HO Hyperforin HP-𝛽-CD 200 𝜇m 20% stretch Remaining hyperforin (%) (c) Remaining hyperforin (%) 100 80 60 40 20 0 Time (min) 0 30 60 90 120 150 180 Hyperforin/HP-𝛽-CD Hyperforinh Remaining hyperforin (%) 100 80 60 40 20 0 Time (min) 0 30 60 90 120 150 180 0 20 40 60 80 100 Wound closure area (%) Stretch + hyperforin Stretch + hyperforin HP-𝛽-CD Stretch Control (d) ∗∗ (n = 5–8, ∗p < 0.05, ∗∗p < 0.5, ∗∗∗p < 0.1) Time (min) (c) (d) Figure 1: The complexation of hyperforin with hydroxypropyl-𝛽-cyclodextrin (HP-𝛽-CD) and its effects on wound closure in HaCaT cells. 3. Results (a) The molar ratio graph obtained by UV/Vis spectra measurements of the inclusion complex formed by hyperforin (4.66 × 10−4 M) and HP-𝛽- CD (0–8.0 equivalents) at 25∘C. (b) A possible model of the 1 : 4 complex of hyperforin/HP-𝛽-CD. (c) The photodegradation of hyperforin/HP- 𝛽-CD in aqueous solution and hyperforin in methanol induced by LED light exposure. Curve fitting was performed by single exponential decay with a baseline. The baselines obtained for hyperforin/HP-𝛽-CD and hyperforin were 44% and 1.4%, respectively. (d) The effects of hyperforin and hyperforin/HP-𝛽-CD treatments on wound closure in keratinocytes under sustained stretching. Stretch stimulation (20%) facilitated wound closure in HaCaT keratinocytes (stretch). Treatment with hyperforin (1 𝜇M) further accelerated the wound closure (stretch + hyperforin) and the wound gap was nearly closed at 6 h after scratching. Hyperforin/HP-𝛽-CD (1 𝜇M as hyperforin) showed equal or greater efficacy to hyperforin (stretch + hyperforin/HP-𝛽-CD) in promoting wound closure. The data are shown as representative DIC images (upper pictures at 0 h and 6 h) and by the averages of the calculated percentage of the wound closure areas (lower graph). The quantitative data in (c) and (d) are shown as the mean ± SEM. initiated from the leading cells on the wound edge and that this occurred due to the release of ATP from the leading cells and the activation of TRPC6 on the cells behind the leading edge through the activation of P2Y with the spread ATP [2]. We assessed whether hyperforin/HP-𝛽-CD also has the same effects on HaCaT keratinocytes. At 3 h after making a narrow scar on the confluent monolayer of hyperforin/HP-𝛽-CD-treated cells, stretching (20% for 1 s, perpendicular to the linear gap) induced an increase in the intracellular Ca2+ in almost all of the leading cells on the wound edge and the Ca2+ increase propagated towards the rear cells behind the edge in a wave-like pattern images captured at 0 and 6 h after scratching and the average of the calculated percentage of the wound closure area. A 20% sustained stretch facilitated wound closure in comparison to nonstretched cells (control) and hyperforin (1 𝜇M) treatment further enhanced the effect of stretching, as shown previously. Hyperforin/HP-𝛽-CD (1 𝜇M) was equally (or more) effective in facilitating wound closure. 3.2. The Stretch-Induced ATP Release and the Initiation of Ca2+ Waves from the Leading Cells on the Wound Gap. 3.2. The Stretch-Induced ATP Release and the Initiation of Ca2+ Waves from the Leading Cells on the Wound Gap. It was reported that stretch stimulation induced intercellular Ca2+ waves in hyperforin-treated HaCaT cells, which were 3. Results Following stretching, the release of ATP was only observed in the cells at the leading edge. The released ATP diffused into the entire area and remained at a high concentration for several minutes (Movie S2 online). rate of decay was obviously faster, especially at the distant regions (Figure 3(b)). The Ca2+ responses were similarly measured under various conditions and inhibitors and were evaluated by the peak response in an averaged trace of the responses at different distances (Figure 3(c)). The suppression observed in Ca2+-free medium, in hyperforin/HP-𝛽-CD- untreated cells and in shTRPC6-treated cells, suggested the involvement of the influx of Ca2+via TRPC6. The inhibition by the treatments with suramin (P2-receptor antagonist, 100 𝜇M), apyrase (ATP-hydrolyzing enzyme, 20 U/mL), and CBX (hemichannel blocker, 100 𝜇M) suggested the contribu- tion of ATP signaling in this process. The reduction by each treatment with U73122 (PLC inhibitor, 10 𝜇M) and diC8-PIP2 (a water-soluble PIP2 analog that suppresses the activity of PLC by competing with PIP2, 10 𝜇M) suggested that the P2Y receptor-Gq-PLC-DAG-mediated signaling cascade was involved in the activation of TRPC6. These results were the same as those obtained by treatment with hyperforin (nonencapsulate) and stretch stimulation [2]. This suggests the involvement of the release of ATP via hemichannels in the leading cells and that the activation of P2Y in the cells behind the wound edge prolonged the influx of Ca2+ via TRPC6 through the Gq-PLC-DAG cascade.i (Figure 2(a); Movie S1) (see Supplementary Material avail- able online at https://doi.org/10.1155/2017/8701801). The Ca2+ waves occurred due to the release of ATP from the leading cells and its diffusion to the surrounding cells behind the edge, as shown in Figure 2(b) (Movie S2). The stretch applied parallel to the linear gap had essentially the same effect on ATP-Ca2+ signaling and wound healing in HaCaT cells [2]. g g g One notable advantage of hyperforin/HP-𝛽-CD was that the effects on the stretch-induced ATP and Ca2+ signal- ing were more reproducible than those obtained simple hyperforin. This may be attributed to the improvement of photostability and the aqueous solubility of hyperforin/HP- 𝛽-CD. 3.3. The Pharmacological Analysis of the Stretch-Induced Ca2+ Responses and Wound Closure in Hyperforin/HP-𝛽-CD- Treated Cells. To analyze the characteristics of the stretch- induced Ca2+response, the effects of various inhibitors on the Ca2+response were evaluated in hyperforin/HP-𝛽-CD- treated HaCaT cells. 3. Results It was reported that stretch stimulation induced intercellular Ca2+ waves in hyperforin-treated HaCaT cells, which were BioMed Research International 5 30 s 3 s 10 s 60 s 0 s 20% stretch 0.5 s 20% stretch 1 s −1 s 200 𝜇m (a) −1 s 0 s 20% stretch 0.5 s 20% stretch 1 s 10 s 3 s 30 s 60 s (b) Figure 2: Stretch-induced Ca2+ wave propagation from the wound edge and ATP release from the leading cells. Hyperforin/HP-𝛽-CD-treated HaCaT cells at 3 h after scratching were subjected to a single stretch (20% for 1 s), which was applied perpendicular to the linear gap. (a) The intracellular Ca2+ responses were measured using the Ca2+ fluorescence indicator, Fluo-8. In response to the stretch, the cells at the leading edge exhibited a remarkably long-lasting increase in intracellular Ca2+, and the Ca2+ increase subsequently propagated to the cells located behind the edge (Movie S1 online). (b) The release of ATP was visualized using a real-time luciferin-luciferase bioluminescence imaging system. Representative overlay images of the ATP-dependent luminescence (red) and infrared DIC images (green) are shown. Following stretching, the release of ATP was only observed in the cells at the leading edge. The released ATP diffused into the entire area and remained at a high concentration for several minutes (Movie S2 online). 1 s 60 s 0.5 s 20% stretch 30 s ) 30 s 3 s 10 s 0 s 20% stretch 0.5 s 20% stretch −1 s (a) 60 s 1 s 200 𝜇m 1 s 0 s 20% stretch 0.5 s 20% stretch 1 s 10 s s 30 s 60 s (b) (a) (b) Figure 2: Stretch-induced Ca2+ wave propagation from the wound edge and ATP release from the leading cells. Hyperforin/HP-𝛽-CD-treated HaCaT cells at 3 h after scratching were subjected to a single stretch (20% for 1 s), which was applied perpendicular to the linear gap. (a) The intracellular Ca2+ responses were measured using the Ca2+ fluorescence indicator, Fluo-8. In response to the stretch, the cells at the leading edge exhibited a remarkably long-lasting increase in intracellular Ca2+, and the Ca2+ increase subsequently propagated to the cells located behind the edge (Movie S1 online). (b) The release of ATP was visualized using a real-time luciferin-luciferase bioluminescence imaging system. Representative overlay images of the ATP-dependent luminescence (red) and infrared DIC images (green) are shown. 3. Results Hyperforin/HP-𝛽-CD −10 0 10 20 30 40 50 60 70 80 90 Time (s) Ca2+ response (%) 200 𝜇m (𝜇m) 15 s −10 0 10 20 30 40 50 60 70 80 90 Time (s) 0 20 40 60 80 100 Ca2+ response (%) 10 𝜇M Gd3+ 200 𝜇m 0 60 120 180 240 (𝜇m) 15 s −10 0 10 20 30 40 50 60 70 80 90 Time (s) 0 20 40 60 80 100 Ca2+ response (%) 10 𝜇M Gd3+ 200 𝜇m 0 60 120 180 240 (𝜇m) (b) (a) (b) free GsMTx-4 U73122 suramin apyrase CBX Hyperforin HP-𝛽-CD- HP-𝛽-CD- untreated 0 20 40 60 80 100 Ca2+ response (%) Gd3+ shTRPC6 20 unit/mL 100 𝜇M 100 𝜇M 10 𝜇M 10 𝜇M 5 𝜇M 5 𝜇M Cont. Hyperforin Ca2+- treated diC8-PIP2 (n = 3–6, ∗p < 0.05, ∗∗p < 0.001) ∗ ∗ ∗ ∗ ∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ (b) free GsMTx-4 U73122 suramin apyrase CBX Hyperforin HP-𝛽-CD- HP-𝛽-CD- untreated 0 20 40 60 80 100 Ca2+ response (% Gd3+ shTRPC6 20 unit/mL 100 𝜇M 100 𝜇M 10 𝜇M 10 𝜇M 5 𝜇M 5 𝜇M Cont. Hyperforin Ca2+- treated diC8-PIP2 (n = 3–6, ∗p < 0.05, ∗∗p < 0.001) ∗ ∗ ∗ ∗ ∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ (c) 0 20 40 60 80 100 Nominally shTRPC6 GsMTx-4 suramin 20 unit/mL apyrase CBX area (%) 6 h 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch Gd3+ 200 𝜇m 100 𝜇M 100 𝜇M 10 𝜇M 5 𝜇M Hyperforin HP-𝛽-CD- Cont. treated (n = 3–8, ∗p < 0.001) ∗ ∗ ∗ ∗ ∗ ∗ ∗ Ca2+-free Wound closure (d) (c) (c) 0 20 40 60 80 100 Nominally shTRPC6 GsMTx-4 suramin 20 unit/mL apyrase CBX area (%) 6 h 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch Gd3+ 200 𝜇m 100 𝜇M 100 𝜇M 10 𝜇M 5 𝜇M Hyperforin HP-𝛽-CD- Cont. treated (n = 3–8, ∗p < 0.001) ∗ ∗ ∗ ∗ ∗ ∗ ∗ Ca2+-free Wound closure (d) (c) (d) Figure 3: The effects of various inhibitors on the stretch-induced Ca2+ responses and wound closure. (a) The time course of changes in the fluorescence intensity of Fluo-8 due to a transient 20% stretch in hyperforin/HP-𝛽-CD-treated HaCaT cells (control). 3. Results The time course of the intracellular Ca2+response induced by a 20% stretch was measured at different distances (0–240 𝜇m) from the scar (Figure 3(a), control; hyperforin/HP-𝛽-CD-treated cells). At 0 𝜇m (wound edge), the Ca2+response was evoked immediately after stretching and it was prolonged by several min. When the distance from the edge was increased, a longer time lag was found before the onset of the activity; however, the amplitudes of the plateau phase were nearly the same. These results were consistent with the idea that Ca2+ waves caused by the simple diffusion of ATP released from the leading cells at wound edge and the activation of P2Y in the surrounding cells behind the wound edge. When Gd3+ (10 𝜇M), an inhibitor of the stretch-activated channel, was applied, the Ca2+ response in the peak was reduced and the g q To confirm whether hyperforin/HP-𝛽-CD facilitates wound closure by amplifying ATP-Ca2+ signaling, we assessed the effects of the various inhibitors that were used above on the wound closure during sustained stretching. The wound gap was almost closed at approximately 6 h (Figure 3(d), control) after scratching, while treatment with Ca2+ depletion (nominally Ca2+-free), CBX (100 𝜇M), apyrase (20 U/mL), suramin (100 𝜇M), Gd3+(10 𝜇M), GsMTx-4 (5 𝜇M), and shTRPC6 treatment delayed wound closure (Figure 3(d)). This suggested that the wound closure process required ATP-Ca2+ signaling, especially the influx of Ca2+ through TRPC6. BioMed Research International 0 60 120 180 240 15 s 0 20 40 60 80 100 Cont. Hyperforin/HP-𝛽-CD −10 0 10 20 30 40 50 60 70 80 90 Time (s) Ca2+ response (%) 200 𝜇m (𝜇m) (a) 15 s −10 0 10 20 30 40 50 60 70 80 90 Time (s) 0 20 40 60 80 100 Ca2+ response (%) 10 𝜇M Gd3+ 200 𝜇m 0 60 120 180 240 (𝜇m) (b) free GsMTx-4 U73122 suramin apyrase CBX Hyperforin HP-𝛽-CD- HP-𝛽-CD- untreated 0 20 40 60 80 100 Ca2+ response (%) Gd3+ shTRPC6 20 unit/mL 100 𝜇M 100 𝜇M 10 𝜇M 10 𝜇M 5 𝜇M 5 𝜇M Cont. 3. Results Hyperforin Ca2+- treated diC8-PIP2 (n = 3–6, ∗p < 0.05, ∗∗p < 0.001) ∗ ∗ ∗ ∗ ∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ (c) 0 20 40 60 80 100 Nominally shTRPC6 GsMTx-4 suramin 20 unit/mL apyrase CBX area (%) 6 h 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch Gd3+ 200 𝜇m 100 𝜇M 100 𝜇M 10 𝜇M 5 𝜇M Hyperforin HP-𝛽-CD- Cont. treated (n = 3–8, ∗p < 0.001) ∗ ∗ ∗ ∗ ∗ ∗ ∗ Ca2+-free Wound closure (d) gure 3: The effects of various inhibitors on the stretch-induced Ca2+ responses and wound closure. (a) The time course of changes in e fluorescence intensity of Fluo-8 due to a transient 20% stretch in hyperforin/HP-𝛽-CD-treated HaCaT cells (control). Each color trace dicated the data at different distances of 0, 60, 120, 180, and 240 from the wound edge (inset image). The intensity was normalized to e peak value obtained with ionomycin treatment at the end of each experiment. (b) The effects of Gd3+ on the stretch-induced Ca2+ sponse as a typical example of the blocking effects of the inhibitors. Gd3+ (10 𝜇M) was applied at 10 min before the application of a 20% retch. (c) The effects of various inhibitors on 20% transient stretch-induced Ca2+ responses in hyperforin/HP-𝛽-CD-treated HaCaT cells. The intensity traces at each distance from the wound edge were averaged and normalized to the peak intensity obtained with ionomycin eatment. The data show the average of the peak values obtained in 3–6 separate experiments. Various inhibitors, including CBX (100 𝜇M), yrase (20 Unit/mL), suramin (100 𝜇M), U73122 (10 𝜇M), diC8-PIP2 (5 𝜇M), Gd3+ (10 𝜇M), and GsMTx-4 (5 𝜇M), were applied at 10 min fore the stretch stimulation. A Ca2+-free condition was achieved by changing the medium to Ca2+-free medium that contained 0.5 𝜇M GTA. All of the quantitative data are shown as the mean (±SEM). (d) The effects of various inhibitors on the stretch facilitated wound osure in hyperforin/HP-𝛽-CD-treated HaCaT cells. Confluent cell cultures were scratched and allowed to migrate for 6 h under a sustained % stretch in a medium that contained various inhibitors, including CBX (100 𝜇M), apyrase (20 Unit/mL), suramin (100 𝜇M), Gd3+ (10 𝜇M), d GsMTx-4 (5 𝜇M) or in nominally Ca2+-free medium. shTRPC6 was applied to the cells for 3 h; the cells were then grown to confluence. 3. Results Each color trace indicated the data at different distances of 0, 60, 120, 180, and 240 from the wound edge (inset image). The intensity was normalized to the peak value obtained with ionomycin treatment at the end of each experiment. (b) The effects of Gd3+ on the stretch-induced Ca2+ response as a typical example of the blocking effects of the inhibitors. Gd3+ (10 𝜇M) was applied at 10 min before the application of a 20% stretch. (c) The effects of various inhibitors on 20% transient stretch-induced Ca2+ responses in hyperforin/HP-𝛽-CD-treated HaCaT cells. The intensity traces at each distance from the wound edge were averaged and normalized to the peak intensity obtained with ionomycin treatment. The data show the average of the peak values obtained in 3–6 separate experiments. Various inhibitors, including CBX (100 𝜇M), apyrase (20 Unit/mL), suramin (100 𝜇M), U73122 (10 𝜇M), diC8-PIP2 (5 𝜇M), Gd3+ (10 𝜇M), and GsMTx-4 (5 𝜇M), were applied at 10 min before the stretch stimulation. A Ca2+-free condition was achieved by changing the medium to Ca2+-free medium that contained 0.5 𝜇M EGTA. All of the quantitative data are shown as the mean (±SEM). (d) The effects of various inhibitors on the stretch facilitated wound closure in hyperforin/HP-𝛽-CD-treated HaCaT cells. Confluent cell cultures were scratched and allowed to migrate for 6 h under a sustained 20% stretch in a medium that contained various inhibitors, including CBX (100 𝜇M), apyrase (20 Unit/mL), suramin (100 𝜇M), Gd3+ (10 𝜇M), and GsMTx-4 (5 𝜇M) or in nominally Ca2+-free medium. shTRPC6 was applied to the cells for 3 h; the cells were then grown to confluence. Representative DIC images (upper panel) and the means of 3–8 wound closure experiments at 6 h after scratching (lower panel) are shown. All of the quantitative data are shown as the mean (±SEM). 3.4. The Effects of Hyperforin/HP-𝛽-CD Treatment on the Ca2+ Responses in the Ex Vivo Skin of Atopic Dermatitis. Next, we assessed the effects of hyperforin/HP-𝛽-CD treatment and stretch mechanical stimulation on an ex vivo epidermis of atopic dermatitis. The epidermis, which was detached from the dermis after overnight treatment with dispase, was loaded with Fluo-8AM and observed with a laser confocal 3.4. The Effects of Hyperforin/HP-𝛽-CD Treatment on the Ca2+ Responses in the Ex Vivo Skin of Atopic Dermatitis. Next, we assessed the effects of hyperforin/HP-𝛽-CD treatment and stretch mechanical stimulation on an ex vivo epidermis of atopic dermatitis. 3. Results epresentative DIC images (upper panel) and the means of 3–8 wound closure experiments at 6 h after scratching (lower panel) are shown. l of the quantitative data are shown as the mean (±SEM). 6 BioMed Research International 6 0 60 120 180 240 15 s 0 20 40 60 80 100 Cont. Hyperforin/HP-𝛽-CD −10 0 10 20 30 40 50 60 70 80 90 Time (s) Ca2+ response (%) 200 𝜇m (𝜇m) (a) 15 s −10 0 10 20 30 40 50 60 70 80 90 Time (s) 0 20 40 60 80 100 Ca2+ response (%) 10 𝜇M Gd3+ 200 𝜇m 0 60 120 180 240 (𝜇m) (b) free GsMTx-4 U73122 suramin apyrase CBX Hyperforin HP-𝛽-CD- HP-𝛽-CD- untreated 0 20 40 60 80 100 Ca2+ response (%) Gd3+ shTRPC6 20 unit/mL 100 𝜇M 100 𝜇M 10 𝜇M 10 𝜇M 5 𝜇M 5 𝜇M Cont. Hyperforin Ca2+- treated diC8-PIP2 (n = 3–6, ∗p < 0.05, ∗∗p < 0.001) ∗ ∗ ∗ ∗ ∗ ∗∗ ∗∗ ∗∗ ∗∗ ∗∗ (c) 0 20 40 60 80 100 Nominally shTRPC6 GsMTx-4 suramin 20 unit/mL apyrase CBX area (%) 6 h 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch 20% stretch Gd3+ 200 𝜇m 100 𝜇M 100 𝜇M 10 𝜇M 5 𝜇M Hyperforin HP-𝛽-CD- Cont. treated (n = 3–8, ∗p < 0.001) ∗ ∗ ∗ ∗ ∗ ∗ ∗ Ca2+-free Wound closure (d) 3 Th ff f i i hibi h h i d d C 2+ d d l ( ) Th i f h 0 60 120 180 240 15 s 0 20 40 60 80 100 Cont. 3. Results Normal skin 20% 1s stretch (a) Intensity (a.u.) 4000 3000 2000 1000 0 Time (s) 0 20 40 60 80 100 120 (a) 20% 1s stretch (b) (a) Atopic skin, nontreated Time (s) Intensity (a.u.) 2000 1500 1000 500 0 0 50 100 150 200 250 Time (s) (b) Atopic skin, hyperforin/HP-𝛽-CD treated Atopic skin, hyperforin/HP-𝛽-CD trea 2 Intensity (a.u.) 2000 1500 1000 500 0 Time (s) 0 50 100 150 200 250 200 𝜇m 20% 1s stretch (c) 200 𝜇m (c) Intensity (a.u.) (c) Figure 4: The effects of hyperforin/HP-𝛽-CD treatment on the Ca2+dynamics in the epidermis of atopic skin ex vivo. The epidermis was detached from the dermis by dispase treatment and fixed on a stretch chamber with intradermal needles. The fluorescence of the Ca2+ indicator, Fluo-8AM, was observed with a laser confocal microscope (left image panels) and the changes in intensity in several cells were plotted in the right panels. (a) The Ca2+ oscillation and the Ca2+ response to stretch stimulation (20%, 1 s, transient) in the epidermis of normal skin ex vivo. Frequent Ca2+ oscillation and a large Ca2+ response to stretching and subsequent Ca2+ waves were prominent (see also Movie S3). (b) In the atopic epidermis, little Ca2+ oscillation and a very weak Ca2+ response to stretching were observed (see also Movie S4). (c) The 24 h treatment of atopic skin with hyperforin/HP-𝛽-CD drastically induced autonomous Ca2+ oscillation and led to a transient, long-lasting Ca2+ increase induced by stretching (see also Movie S5). Ca2+ through TRPC6 [11]. Our previous studies showed that hyperforin-treated HaCaT keratinocytes could accelerate wound closure in conjunction with exogenous and endoge- nous mechanical stretching through the facilitation of the ATP-Ca2+signaling cascade [2]. The impact of hyperforin on mechanosensitivity remains unclear, but hyperforin certainly amplifies ATP-Ca2+signaling and facilitates reepithelializa- tion during wound healing. However, due to the photoin- stability of hyperforin, daylight initiates its facile oxidative degradation [21]. Prenyl side chains (hemiterpene moieties) containing conjugated double bonds are generally prone to oxidation. In order to enhance the stability of hyperforin and exert its topical therapeutic potential, hyperforin was encap- sulated by forming a supramolecular complexation with HP- 𝛽-CD. 3. Results The solubility of HP-𝛽-CD is highest in ethanol as well as water among the CD compounds, 𝛼-CD, 𝛽-CD, that had been treated with hyperforin/HP-𝛽-CD for 24 h exhibited autonomous Ca2+ oscillation, and a transient long- lasting increase in Ca2+ and more frequent Ca2+oscillations following stretch stimulation (Figure 4(c), Movie S5). The application of hyperforin/HP-𝛽-CD-treatment to atopic skin for 24 h led to the recovery of the mechanosensitive ATP-Ca2+ signaling, which was dysfunctional in the untreated atopic epidermis. that had been treated with hyperforin/HP-𝛽-CD for 24 h exhibited autonomous Ca2+ oscillation, and a transient long- lasting increase in Ca2+ and more frequent Ca2+oscillations following stretch stimulation (Figure 4(c), Movie S5). The application of hyperforin/HP-𝛽-CD-treatment to atopic skin for 24 h led to the recovery of the mechanosensitive ATP-Ca2+ signaling, which was dysfunctional in the untreated atopic epidermis. 3. Results The epidermis, which was detached from the dermis after overnight treatment with dispase, was loaded with Fluo-8AM and observed with a laser confocal microscope. Normal skin exhibited frequent spontaneous Ca2+ oscillations and a large Ca2+ response to stretch stim- ulation (1 s single) and subsequent Ca2+ waves with the long-lasting elevation of Ca2+ (Figure 4(a), Movie S3). In contrast, the epidermis of atopic dermatitis showed few oscillations and only a small response to stretching without any waves (Figure 4(b), Movie S4). In contrast, atopic skin 7 BioMed Research International Normal skin Intensity (a.u.) 4000 3000 2000 1000 0 Time (s) 0 20 40 60 80 100 120 20% 1s stretch (a) Atopic skin, nontreated Time (s) Intensity (a.u.) 2000 1500 1000 500 0 0 50 100 150 200 250 20% 1s stretch (b) Atopic skin, hyperforin/HP-𝛽-CD treated Intensity (a.u.) 2000 1500 1000 500 0 Time (s) 0 50 100 150 200 250 200 𝜇m 20% 1s stretch (c) Figure 4: The effects of hyperforin/HP-𝛽-CD treatment on the Ca2+dynamics in the epidermis of atopic skin ex vivo. The epidermis was detached from the dermis by dispase treatment and fixed on a stretch chamber with intradermal needles. The fluorescence of the Ca2+ indicator, Fluo-8AM, was observed with a laser confocal microscope (left image panels) and the changes in intensity in several cells were plotted in the right panels. (a) The Ca2+ oscillation and the Ca2+ response to stretch stimulation (20%, 1 s, transient) in the epidermis of normal skin ex vivo. Frequent Ca2+ oscillation and a large Ca2+ response to stretching and subsequent Ca2+ waves were prominent (see also Movie S3). (b) In the atopic epidermis, little Ca2+ oscillation and a very weak Ca2+ response to stretching were observed (see also Movie S4). (c) The 24 h treatment of atopic skin with hyperforin/HP-𝛽-CD drastically induced autonomous Ca2+ oscillation and led to a transient long-lasting Ca2+ increase induced by stretching (see also Movie S5). 4. Discussion The topical application of hyperforin, which is a traditional folk remedy, has anti-inflammatory, antioxidative, antibac- terial, antinociceptive, and wound healing effects. Recently, accumulating evidence indicates that hyperforin facilitates the keratinocyte differentiation caused by the uptake of 8 BioMed Research International BioMed Research International methylated-𝛽-CD, sulfobutyl ethyl-𝛽-CD, 𝛾-CD, and HP-𝛽- CD. This amphipathic property was a major advantage when making the inclusion complex with hydrophobic hyperforin. The molar ratio method indicated that the optimal ratio of the hyperforin/HP-𝛽-CD complex was 1 : 4. This meant the formation of HP-𝛽-CD-tetracapped hyperforin, where the hyperforin was encapped with HP-𝛽-CD at each hemiterpene moiety [22] as shown in Figure 1(b). The novel inclusion complex showed obvious photostability in comparison to hyperforin (Figure 1(c)). The curve fitting of the decay time course of hyperforin/HP-𝛽-CD indicated the existence of a large nondecayed component that corresponded to photostable hyperforin. This modification can contribute to both pharmaceutical application and topical medication. In our experimental design, HaCaT keratinocytes were cultured under low extracellular Ca2+ conditions (0.07 mM), which mimicked the extracellular Ca2+ environment for barrier-perturbed epidermis, such as the environment that would result from skin stripping or the use of surfactants. Atopic dermatitis is also a skin barrier dysfunction. Topical medication of hyperforin-rich St. John’s wort cream has been shown to be effective in patients with atopic dermatitis [10– 14]. The analysis of the laser scanning microscopy images has shown that the hyperforin-rich cream reduces the skin surface dryness and improves the moisture level of the stratum corneum [14]. However, the mechanism underly- ing the improvement of symptoms in atopic skin remains unclear. Our present study is the first to demonstrate how the Ca2+dynamics of atopic skin behave under mechanical environments such as wound healing and reepithelialization. We observed the Ca2+ dynamics induced by the stretching of epidermis ex vivo using a confocal microscope (Figure 4). In atopic epidermis, there was a remarkable decrease in the Ca2+ responses and oscillations induced by stretching (Fig- ure 4(b)). Treatment with hyperforin/HP-𝛽-CD for 24 h restored the Ca2+ responses and oscillations, even in atopic skin (Figure 4(c)). These results suggest that the pathogenesis of atopic dermatitis is related to ATP-Ca2+ signaling and that hyperforin/HP-𝛽-CD may have therapeutic application in the treatment of atopic dermatitis. We assessed the effects of hyperforin/HP-𝛽-CD on wound healing and ATP-Ca2+ signaling in keratinocytes. 5. Conclusions Cutaneous wound healing is accelerated by mechanical stress both exogenously and endogenously, and treatment with hyperforin enhances the acceleration through the facilitation of ATP-Ca2+ signaling in keratinocytes. We succeeded in making HP-𝛽-CD-tetracapped hyperforin (hyperforin/HP- 𝛽-CD), which possessed increased aqueous solubility and improved photoprotection. Treatment with hyperforin/HP- 𝛽-CD enhanced the mechanically induced ATP-Ca2+ signal- ing and accelerated wound closure in HaCaT keratinocytes with equal (or greater) efficacy to hyperforin. We also applied hyperforin/HP-𝛽-CD on atopic skin ex vivo and found that hyperforin/HP-𝛽-CD treatment for 24 h improved the stretch-induced Ca2+ responses and oscillations, which reduced in atopic skin. The data suggest that hyperforin/HP- 𝛽-CD is a potent targeted therapeutic agent that can be used to promote epidermal wound healing and to treat atopic dermatitis. Interestingly, the reagents that blocked the increase in Ca2+ also suppressed the acceleration of wound closure in response to stretching in hyperforin/HP-𝛽-CD-treated cells (Figure 3(d)). Thus, the hyperforin/HP-𝛽-CD complex shows a similar efficiency to hyperforin in inducing mechanosensi- tive ATP-Ca2+ signaling and wound closure in keratinocytes. In fact, hyperforin/HP-𝛽-CD seems to be superior due to the reproducibility of the data concerning stretch-induced ATP and Ca2+ signaling, which may be attributed to the photostability and aqueous solubility of hyperforin/HP-𝛽- CD. 4. Discussion Hyperforin/HP-𝛽-CD enhanced the acceleration of wound closure by stretching with a similar efficiency to hyperforin (Figure 1(d)). In hyperforin/HP-𝛽-CD-treated keratinocytes, stretching induced a conspicuous increase in the Ca2+ levels in the leading cells facing the wound edge and the Ca2+ waves slowly propagated to the cells behind the wound edge (Figure 2(a)). These propagating Ca2+ waves were entirely due to the release of ATP from the leading cells (Figure 2(b)). The pathway of ATP release was CBX sensitive (Figure 3(c)) and presumably pannexin hemichannels from our previous study [2]. The migrating cells at the wound edge represented morphological changes that were similar to those observed at the epithelial-to-mesenchymal transition and might be more susceptible to endogenous and exogenous mechanical stress [2, 23]. The increase in Ca2+ in the cells behind the wound edge was dependent on the influx of Ca2+ via the TRPC6 channels, which were activated by the activation of P2Y through the Gq-PLC-DAG-mediated signaling cascade (Fig- ure 3(c)) [2] and lasted for a relatively long period. The concentration and duration of the Ca2+ increase were depen- dent on the distance from the wound edge, making a Ca2+ gradient from the leading cells to the following cells. This Ca2+ gradient may be essential for organized wound healing, including cell migration, molecular relocation, and gene expression. The cell traction of the cells located behind the edge by migrating leading cells is also an important mechanical cue for wound healing that is controlled by Ca2+ dependent cell-cell interaction molecules such as E- cadherin. This Ca2+ signaling is enhanced by treatment with hyperforin/HP-𝛽-CD. Competing Interests The authors declare that there is no conflict of interests regarding the publication of this paper. References [1] T. Tanaka, K. Naruse, and M. Sokabe, “Effects of mechanical stresses on the migrating behavior of endothelial cells,” in Biomechanics at Micro- and Nanoscale Levels, H. Wada, Ed., vol. I, pp. 75–87, World Scientific, Singapore, 2005. [17] J. Hatanaka, Y. Shinme, K. Kuriyama et al., “In vitro and in vivo characterization of new formulations of st. john’s wort extract with improved pharmacokinetics and anti-nociceptive effect,” Drug Metabolism and Pharmacokinetics, vol. 26, no. 6, pp. 551– 558, 2011. [2] H. Takada, K. Furuya, and M. Sokabe, “Mechanosensitive ATP release from hemichannels and Ca2+ influx through TRPC6 accelerate wound closure in keratinocytes,” Journal of Cell Science, vol. 127, no. 19, pp. 4159–4171, 2014. [18] M. J. Cork, S. G. Danby, Y. Vasilopoulos et al., “Epidermal barrier dysfunction in atopic dermatitis,” Journal of Investigative Dermatology, vol. 129, no. 8, pp. 1892–1908, 2009. [3] J. Davis, A. R. Burr, G. F. Davis, L. Birnbaumer, and J. D. Molkentin, “A TRPC6-dependent pathway for myofibroblast transdifferentiation and wound healing in vivo,” Developmental Cell, vol. 23, no. 4, pp. 705–715, 2012. [19] P. Boukamp, R. T. Petrussevska, D. Breitkreutz, J. Hornung, A. Markham, and N. E. Fusenig, “Normal keratinization in a spontaneously immortalized aneuploid human keratinocyte cell line,” Journal of Cell Biology, vol. 106, no. 3, pp. 761–771, 1988. [4] S. Samadi, T. Khadivzadeh, A. Emami, N. S. Moosavi, M. Tafaghodi, and H. R. Behnam, “The effect of hypericum perforatum on the wound healing and scar of cesarean,” Journal of Alternative and Complementary Medicine, vol. 16, no. 1, pp. 113–117, 2010. [20] K. Furuya, M. Sokabe, and R. Grygorczyk, “Real-time lumines- cence imaging of cellular ATP release,” Methods, vol. 66, no. 2, pp. 330–344, 2014. [5] Z. Saddiqe, I. Naeem, and A. Maimoona, “A review of the antibacterial activity of Hypericum perforatum L.,” Journal of Ethnopharmacology, vol. 131, no. 3, pp. 511–521, 2010. [21] C. Y. W. Ang, L. Hu, T. M. Heinze et al., “Instability of St. John’s wort (Hypericum perforatum L.) and degradation of hyperforin in aqueous solutions and functional beverage,” Journal of Agricultural and Food Chemistry, vol. 52, no. 20, pp. 6156–6164, 2004. [6] I. S¨untar, E. K. Akkol, H. Keles¸, A. Oktem, K. H. C. Bas¸er, and E. Yes¸ilada, “A novel wound healing ointment: a formulation of Hypericum perforatum oil and sage and oregano essential oils based on traditional Turkish knowledge,” Journal of Ethnophar- macology, vol. 134, no. 1, pp. 89–96, 2011. Acknowledgments The authors thank K. Uekama (Professor Emeritus, Kumamoto University) for his kind help in characterizing the hyperforin/HP-𝛽-CD complex. This work was supported by funds from the Japan Society for the Promotion of Science 9 BioMed Research International (JSPS) Institutional Program for Young Researcher Overseas Visits (to Hiroya Takada); by a Medical Research Grant of Kyousaidan (to Hiroya Takada); by JSPS KAKENHI Grant nos. JP24590274 and JP15K09174 (to Kishio Furuya). scanning microscopy,” Journal of Biomedical Optics, vol. 20, no. 5, Article ID 051013, 2015. scanning microscopy,” Journal of Biomedical Optics, vol. 20, no. 5, Article ID 051013, 2015. [15] H. C. J. Orth and P. C. Schmidt, “Stability and stabilization of hyperforin,” Pharmazeutische Industrie, vol. 62, no. 1, pp. 60–63, 2000. [16] B. Isacchi, N. Galeotti, M. C. Bergonzi, C. Ghelardini, A. R. Bilia, and F. F. Vincieri, “Pharmacological in vivo test to evaluate the bioavailability of some St John’s wort innovative oral preparations,” Phytotherapy Research, vol. 23, no. 2, pp. 197– 205, 2009. References [22] I. Clarot, D. Cl´edat, S. Battu, and P. J. P. Cardot, “Chromato- graphic study of terpene derivatives on porous graphitic carbon stationary phase with 𝛽-cyclodextrin as mobile phase modifier,” Journal of Chromatography A, vol. 903, no. 1-2, pp. 67–76, 2000. [7] M. Dikmen, Y. ¨Ozt¨urk, G. Sagratini, M. Ricciutelli, S. Vittori, and F. Maggi, “Evaluation of the wound healing potentials of two subspecies of hypericum perforatum on cultured NIH3T3 fibroblasts,” Phytotherapy Research, vol. 25, no. 2, pp. 208–214, 2011. [23] N. D. Evans, R. O. C. Oreffo, E. Healy, P. J. Thurner, and Y. H. Man, “Epithelial mechanobiology, skin wound healing, and the stem cell niche,” Journal of the Mechanical Behavior of Biomedical Materials, vol. 28, pp. 397–409, 2013. [8] U. W¨olfle, G. Seelinger, and C. M. Schempp, “Topical applica- tion of St. John’s Wort (Hypericum perforatum),” Planta Medica, vol. 80, no. 2-3, pp. 109–120, 2014. [9] M. Marrelli, G. Statti, F. Conforti, and F. Menichini, “New potential pharmaceutical applications of hypericum species,” Mini-Reviews in Medicinal Chemistry, vol. 16, no. 9, pp. 710–720, 2016. [10] C. M. Schempp, T. Windeck, S. Hezel, and J. C. Simon, “Topical treatment of atopic dermatitis with St. John’s wort cream—a randomized, placebo controlled, double blind half- side comparison,” Phytomedicine, vol. 10, no. 4, pp. 31–37, 2003. [11] M. M¨uller, K. Essin, K. Hill et al., “Specific TRPC6 channel activation, a novel approach to stimulate keratinocyte differ- entiation,” Journal of Biological Chemistry, vol. 283, no. 49, pp. 33942–33954, 2008. [12] S. Arndt, S. F. Haag, A. Kleemann, J. Lademann, and M. C. Meinke, “Radical protection in the visible and infrared by a hyperforin-rich cream—in vivo versus ex vivo methods,” Experimental Dermatology, vol. 22, no. 5, pp. 354–357, 2013. [13] U. W¨olfle, G. Seelinger, and C. M. Schempp, “Topical applica- tion of St. John’s wort (Hypericum perforatum),” Planta Medica, vol. 80, no. 2-3, pp. 109–120, 2014. [14] M. C. Meinke, H. Richter, A. Kleemann et al., “Characterization of atopic skin and the effect of a hyperforin-rich cream by laser
https://openalex.org/W2787490993	https://hal.archives-ouvertes.fr/hal-02051513/file/ris00001078.pdf	English	null	Occupational exposure to organic solvents and risk of male breast cancer: a European multicenter case-control study	Scandinavian journal of work, environment & health	2,018	cc-by	9,005	Occupational exposure to organic solvents and risk of male breast cancer: a European multicenter case-control study Nasser Laoulali, Corinne Pilorget, Diane Cyr, Monica Neri, Linda Kaerlev, Svend Sabroe, Giuseppe Gorini, Lorenzo Richiardi, Maria Morales-Suarez-Varela, Agustin Llopis-Gonzalez, et al. To cite this version: Nasser Laoulali, Corinne Pilorget, Diane Cyr, Monica Neri, Linda Kaerlev, et al.. Occupational exposure to organic solvents and risk of male breast cancer: a European multicenter case-control study. Scandinavian Journal of Work, Environment and Health, 2018, 44 (3), pp. 310-322. 10.5271/sjweh.3717. hal-02051513 Occupational exposure to organic solvents and risk of male breast cancer: a European multicenter case-control study Distributed under a Creative Commons Attribution 4.0 International License doi:10.5271/sjweh.3717 doi:10.5271/sjweh.3717 Occupational exposure to organic solvents and risk of male breast cancer: a European multicenter case-control study by Laouali N, Pilorget C, Cyr D, Neri M, Kaerlev L, Sabroe S, Gorini G, Richiardi L, Morales-Suárez-Varela M, Llopis-Gonzalez A, Ahrens W, Jöckel K-H, Afonso N, Eriksson M, Merletti F, Olsen J, Lynge E, Guénel P Exposure to organic solvents is suspected to increase breast cancer risk, but previous epidemiological studies have often been restricted to women who are generally less exposed than men). In our data, high occupational exposure to trichloroethylene was associated with a doubling of odds ratio of male breast cancer and a dose-response trend. A possible role for benzene and ethylene glycol was also suggested. Affiliation: Center for research in Epidemiology and Population Health (CESP), Inserm U1018, 16 avenue Paul Vaillant-Couturier, 94807 Villejuif Cedex, France. pascal.guenel@inserm.fr Refers to the following text of the Journal: 1999;25(3):0 Key terms: alcoholic solvent; benzene; breast cancer; case-control study; case-control study; chlorinated solvent; ethylene glycol; European; JEM; job-exposure matrix; male breast cancer; multicenter case-control study; occupational exposure; organic solvent; petroleum solvent; trichloroethylene HAL Id: hal-02051513 https://hal.science/hal-02051513v1 Submitted on 20 May 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 Downloaded from www.sjweh.fi on February 06, 2018 Downloaded from www.sjweh.fi on February 06, 2018 Original article Scand J Work Environ Health Online-first -articl Scand J Work Environ Health Online-first -article Original article cand J Work Environ Health – online first. doi:10.5271/sjweh.3717 Scand J Work Environ Health – online first. doi:10.5271/sjweh.3717 Additional material Please note that there is additional material available belonging to this article on the Scandinavian Journal of Work, Environment & Health -website. Print ISSN: 0355-3140 Electronic ISSN: 1795-990X Copyright (c) Scandinavian Journal of Work, Environmen Occupational exposure to organic solvents and risk of male breast cancer: a European multicenter case–control study by Nasser Laouali, MSc,1 Corinne Pilorget, PhD,2, 3 Diane Cyr, MSc,4 Monica Neri, PhD,1 Linda Kaerlev, PhD,5, 6 Svend Sabroe, PhD,7 Giuseppe Gorini, PhD,8 Lorenzo Richiardi, PhD,9 Maria Morales-Suárez- Varela, PhD,10, 11 Agustin Llopis-Gonzalez, PhD,10, 11 Wolfgang Ahrens, PhD,12 Karl-Heinz Jöckel, PhD,13 Noemia Afonso, MD, PhD,14 Mikael Eriksson, PhD,15 Franco Merletti, PhD,9 Jørn Olsen, PhD,7 Elsebeth Lynge, PhD,16 Pascal Guénel, MD, PhD 1, 2 Laouali N, Pilorget C, Cyr D, Neri M, Kaerlev L, Sabroe S, Gorini G, Richiardi L, Morales-Suárez-Varela M, Llopis- Gonzalez A, Ahrens W, Jöckel K-H, Afonso N, Eriksson M, Merletti F, Olsen J, Lynge E, Guénel P. Occupational exposure to organic solvents and risk of male breast cancer: a European multicenter case–control study. Scand J Work Environ Health – online first. doi:10.5271/sjweh.3717 Objectives The etiology of male breast cancer (MBC) is largely unknown but a causal role of exposure to organic solvents has been suggested. Previous studies on occupational risk factors of breast cancer were often restricted to women who are frequently exposed to lower levels and at a lower frequency than men. We inves- tigated the association between MBC and occupational exposure to petroleum and oxygenated and chlorinated solvents in a multicenter case–control study of rare cancers in Europe. Methods The study included 104 MBC cases and 1901 controls. Detailed lifetime work history was obtained during interviews, together with sociodemographic characteristics, medical history and lifestyle factors. Occu- pational exposures to solvents were estimated from a job-exposure matrix. Odds ratios (OR) and their 95% confidence intervals (CI) were calculated using unconditional logistic regression models. Results Lifetime cumulative exposure to trichloroethylene >23.9 ppm years was associated with an increased MBC risk, compared to non-exposure [OR (95% CI): 2.1 (1.2–4.0); P trend <0.01). This increase in risk persisted when only exposures that occurred ≥10 years before diagnosis were considered. In addition, a possible role for benzene and ethylene glycol in MBC risk was suggested, but no exposure–response trend was observed. Conclusions These findings add to the evidence of an increased risk of breast cancer among men professionally exposed to trichloroethylene and possibly to benzene or ethylene glycol. Further studies should be conducted in populations with high level of exposure to confirm our results. Key terms alcoholic solvent; benzene; chlorinated solvent; ethylene glycol; JEM; job-exposure matrix; petro- leum solvent; trichloroethylene. Organic solvents and male breast cancer Organic solvents and male breast cancer Breast cancer is the most frequent cancer among women, with over 1.6 million new cases in 2012 across the world (1), but it is a rare disease among men, accounting for <1% of all breast cancers (2). Risk factors related to hormones and reproduction are well-established causes of female breast cancer, but the etiology of male breast cancer (MBC) is largely unknown (3). However, the clinical features of MBC are often similar to those of the late-onset type of female breast tumors (4), suggesting that these two conditions may share some risk factors. The risk factors for MBC that have been investigated so far include genetics (family history of breast cancer, mutations in BRCA2 or CHEK2), conditions associated with an abnormal estrogen-to-androgen ratio (Klinefel- ter’s syndrome, obesity, orchitis, infertility, exogenous estrogen or testosterone use), and lifestyle (lack of physical activity, alcohol consumption) (3, 5, 6). A limited number of epidemiological studies have investigated the role of organic solvents in female breast cancer, and the increases were either small or inconsistent (15–22). Only two studies have considered MBC in relation to occupational exposure to solvents (23, 24). Because occupational exposures to many chemicals, including solvents, are usually present in jobs mostly held by men (eg, mechanics or painters), studies among men with higher prevalence of exposure than women and less competing risk factors (eg, hormonal and reproductive factors) may facilitate the detection of an association between exposure to solvents and breast cancer, despite the rarity of the disease in men. In a previous paper based on data from a European case–control study on rare cancers including 104 cases of MBC (25), we reported that motor vehicle mechanics and painters with probable exposure to organic solvents had a two- to threefold increased risk of MBC. To further evaluate the hypothesis that organic solvents increases the incidence of MBC, we specifically assessed lifetime occupational exposure to organic solvents by solvent subtype using a detailed job exposure matrix (JEM) (26). Environmental and occupational factors are also suspected to play a role in the etiology of breast cancer in both sexes (7, 8). Solvents are ubiquitous chemicals in occupational settings. They have retained particular attention because they are highly lipophilic compounds that can accumulate in the adipose tissue of the breast and initiate or promote carcinogenesis through geno- toxic mechanisms (9). Organic solvents and male breast cancer Because of the high prolifera- tive activity of epithelial cells of the mammary gland and susceptibility to chemical carcinogens, mammary terminal duct lobular units are likely target tissues for tumorigenesis (10). Animal studies have provided strong evidence for an association between organic solvents and breast cancer (11), and the International Agency for Research on Cancer (IARC) has recognized solvents such as benzene and trichloroethylene (TCE) as known human carcinogens. However, there are limited data on solvents as human breast carcinogens (12, 13). Methods We conducted a European multi-center retrospective case–control study on occupational risk factors of seven rare cancer sites (gallbladder and extra-hepatic bile ducts, small intestine, bone, eye melanoma, mycosis fungoides, and male breast). Cases and controls were recruited from selected areas of eight European countries (Denmark, France, Germany, Italy, Latvia, Portugal, Spain and Sweden) representing a source population of 37 million people. The study design and the procedures of data col- lection have been described in detail earlier (25, 27) and are summarized below. The local ethics committees in each participating country approved the study. The major families of solvents used in the work places include petroleum, chlorinated and oxygenated solvents. Petroleum solvents consist of fuel and organic solvents produced by oil refining. The main petroleum solvent, benzene, was initially used as industrial solvent, eg, to degrease metals, besides being a constituent of gasoline. The toxicity of benzene has largely been proven in leuke- mia risk (14). Its use has now been restricted as an inter- mediate for the synthesis of other chemicals. Chlorinated solvents including TCE, perchloroethylene or methylene chloride, are usually used as solvents in paints, paint removers or resines, chemical intermediates for pesticide synthesis, in dry-cleaning and in the steel industry. Most of them have been used as solvents in place of benzene. Oxygenated solvents include alcohols, ketones, esters, ethylene glycol, or tetrahydrofurans and are widely used in the paint, ink, pharmaceutical, fragrance, adhesive, cosmetic, detergent, or food industries. Ethylene glycol, in particular, is used as a sterilizing agent for medical equipment and supplies. Occupational exposure to organic solvents and risk of male breast cancer: a European multicenter case–control study 1 Center for research in Epidemiology and Population Health (CESP) – Cancer and Environment Team – Inserm UMR 1018 – Université Paris-Sud – UVSQ - Université Paris-Saclay – Villejuif, France. 2 1 Center for research in Epidemiology and Population Health (CESP) – Cancer and Environment Team – Inserm UMR 1018 – Université Paris-Sud – UVSQ - Université Paris-Saclay – Villejuif, France. 1 Center for research in Epidemiology and Population Health (CESP) – Cancer and Environment Team – Inserm UMR 1018 – Université Paris-Sud – UVSQ - Université Paris-Saclay – Villejuif, France. 2 nté Publique France, National Agency for Public Health, Occupational Health Department, Saint-Maurice, France. q g y p p 3 Claude Bernard Lyon University, Epidemiological research and surveillance unit in transport, occupation and environment, Lyon, France. 4 Population-based Epidemiologic Cohorts Unit, Inserm, Villejuif, France. ation-based Epidemiologic Cohorts Unit, Inserm, Villejuif, Fran p p g j 5 Research Unit of Clinical Epidemiology, Institute of Clinical Research, University of Southern Denmark, Odense, Denmark. 6 Center for Clinical Epidemiology, Odense University Hospital, Odense, Denmark. 7 Department of Public Health, Section for Epidemiology, Aarhus University, Aarhus, Denmark. 8 Occupational & Environmental Epidemiology Section, Cancer Research & Prevention Institute (ISPO), Florence, Italy. 9 p y y 10 Unit of Public Health and Environmental Care, Department of Preventive Medicine, University of Valencia, Burjassot, Va 11 12 Institute for Prevention Research and Epidemiology; BIPS-Institute for Epidemiology and Prevention Research GmbH, Bremen, Germany. 13 Institute for Medical Informatics, Biometry and Epidemiology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany. 14 Centro Hospitalar do Porto Porto Portugal p gy; p gy , , y 13 Institute for Medical Informatics, Biometry and Epidemiology, University Hospital Essen, University of Duisburg-Essen, Essen, Germany. 14 4 Centro Hospitalar do Porto, Porto, Portugal. 15 Department of Oncology, Skane University Hospital, Lund, Sweden. 16 16 Department of Public Health, University of Copenhagen, Copenhagen, Denmark. Correspondence to: Pascal Guénel, MD, PhD, CESP Inserm U1018, 16 avenue Paul Vaillant-Couturier, 94807 Villejuif Cedex, France. [E- mail: pascal.guenel@inserm.fr] Recruitment of cases and controls Men living in the study areas who were diagnosed with a breast cancer between 1995 and 1997, aged 35–70 years at diagnosis were eligible for inclusion in the study. This age range was based on the assumption that occupational exposures are not likely to be the cause of the disease in younger patients. Case ascertainment was based on regular contacts with clinical and pathology departments and/or cancer registers in each study area. For each case, an expert pathologist reviewed the pathology report and, if pos- sible, a representative histological slide of the tumor. In total, 122 MBC patients were eligible for the study from all the countries. Eighteen cases could not be included 2 Scand J Work Environ Health – online first Laouali et al (28). Briefly, occupation was coded according to the International Standard Classification of Occupations of the International Labor Office (ISCO), 1968 revision, while industry was coded according to the Classifica- tion of Activities in the European Community (NACE), 1996 revision. because the doctor did not give permission to contact the patient or the patient refused to participate. Thus, 104 cases (85%) were interviewed and available for analysis. The controls were selected randomly from population registers in Denmark, Italy, and Sweden, from electoral rolls in France and from municipality registers in Germany during the case recruitment period and were frequency- matched to the cases by gender, year of birth (5-year strata), and residence area. In countries where population controls were difficult to identify, colon and stomach cancer patients were regarded as appropriate alternatives to population controls, as no occupational exposure to organic solvents is suspected to play an important role in these cancers. Hospital-based cancer controls were selected randomly among incident colon cancer patients in Latvia and Spain, and among colon or stomach cancer patients in Portugal. The controls served as a common pool of controls for each of the seven groups of rare cancer cases included in the European study. The participation rate among male popula- tion controls was high in France (81%) and Italy (74%) but relatively low in Denmark, Germany, and Sweden (<60%). In countries using a hospital-based design, the participation rate among cancer controls was >95%. The overall partici- pation rate among controls was 67%. For the present study, we selected only male controls sampled in study areas from which ≥1 MBC patient was included. Data collection A structured questionnaire was first developed in English, and then translated into the language of each participating country. Back-translation to English was performed for quality control, and no major departure from the original version was observed. A trained interviewer administered this questionnaire face to face or by telephone and col- lected information on sociodemographic characteristics, previous medical conditions, lifestyle factors, anthropo- metric characteristics, alcohol and tobacco consumption, and detailed occupational activities in each job held for >6 months. For each occupational period, we recorded data on products and production processes, the year the job started and ended, job title, and working hours per week. The materials handled, chemical exposures, and occupations held by nearby workers were also recorded. The specific nature of the work was also addressed, such as work tasks, machines or products used, and duration of their use (hours per week). Specific questionnaires were also developed for 27 definite jobs or tasks, such as welding or painting. pp Each job held by a case or a control included in the European study was assigned the exposure indices reported in the JEM. In order to improve specificity, exposure to a given solvent was assigned only to jobs whose probability of exposure according to the JEM was >10%, while the jobs below this cut-off were consid- ered as non-exposed. A job-specific exposure score was then calculated as the result of the product of exposure probability, frequency and intensity, and duration of the job in years. An individual Cumulative Exposure Score (CES) was then calculated for each study subject as the sum of the job-specific exposure scores over his entire work history. Recruitment of cases and controls Finally, 1901 male controls (1395 population controls and 506 cancer controls) were available for the analyses. Occupational exposures to organic solvents were assessed using JEM developed at the Occupational Health Department of the French National Agency for Public Health (Santé Publique France) by experts in occupational hygiene (29, 30). Jobs were defined according to both the ISCO code and the Nomenclature d’Activité Française (NAF: Institut National de la Statis- tique et des Etudes Economiques, revision 1, 2003), subsequently converted to the European NACE code for the purpose of this study. Exposures to occupational hazards were assessed by job and calendar period to account for changes in exposures over time. In the JEM, exposure to chlorinated solvents [TCE, perchloroethylene (tetrachloroethylene), methylene chloride (dichloromethane), chloroform (trichloro- methane) and carbon tetrachloride], petroleum sol- vents [benzene, special boiling point spirits and other aliphatic petroleum-based solvents (SBP), gasoline, white spirits, and kerosene/diesel oil/fuel oils (KDF)] and to oxygenated solvents (alcohol, ketones/esters, diethyl ether, ethylene glycol and tetrahydrofuran) was determined for each job using semi-quantitative indica- tors for exposure probability (ie, proportion of exposed workers 0, 0–10, 10–20, … ≥90%), exposure frequency (eg, for petroleum solvents 1=<30%, 2=30–70%, 3= >70% of working hours) and exposure intensity [semi- quantitative exposure scores on the basis of literature review of occupational measurement data, for example, for benzene: 1=0.1–1 ppm, 2>1–5 ppm, 3=>5–15 ppm, and 4=>15 ppm]. Statistical analysis The distributions of baseline characteristics of study population according to cases and controls status were expressed as means and standard deviation (SD) for con- Scand J Work Environ Health – online first Results The main characteristics of cases (N=104) and controls (N=1901) are presented in table 1. Compared to con- trols, cases tended to be older and have a lower educa- tion level. The BMI and the mean number of jobs did not differ significantly. An elevated consumption of alcohol >60 g/day was associated with a 2.6 increased risk of MBC, as reported in a previous paper (5). The proportion of men ever exposed to organic sol- vent of any kind was higher among cases than among controls (table 2). As regards the most common expo- sures, 44% of the cases and 32% of the controls had been exposed to TCE, 30% and 20% respectively to white spirits, 21% and 17% to KDF, 21% and 9% to benzene. Each of the other exposures accounted for ≤10% of cases and controls. Each subject was typically exposed to more than one substance. For example, 144 subjects had been exposed to both benzene and TCE, a number that repre- Table 1. Comparison of cases and controls by selected character- istics (European study on male breast cancer) [BMI=body mass index; CI=confidence interval; SD=standard deviation] Cases (N=104) Controls (N=1901) OR a 95% CI N % N % Country Denmark 8 7.7 195 10.3 Sweden 7 6.7 140 7.4 Latvia 3 2.9 69 3.6 France 29 27.9 308 16.2 Germany 10 9.6 542 28.5 Italy 20 19.2 210 11 Spain 19 18.3 365 19.2 Portugal 8 7.7 72 3.8 Age (years) <40 6 5.8 218 11.5 40–44 6 5.8 193 10.1 45–49 10 9.6 190 10 50–54 14 13.4 202 10.6 55–59 16 15.4 264 13.9 60–64 20 19.2 329 17.3 ≥65 32 30.8 505 26.6 Alcohol intake (g/day) 0–30 43 41.4 1130 59.4 1.0 reference 30–60 31 29.8 503 26.5 1.3 0.8–2.2 >60 30 28.8 268 14.1 2.6 1.5–4.4 Education b Left school at ≤18 63 60.6 866 45.6 1.5 0.8–2.8 Professional training 25 24.0 515 27.1 1.5 0.8–2.9 University 16 14.4 516 27.1 1.0 reference BMI (kg/m²) <18.5 8 7.7 83 4.4 2.0 0.9–4.7 18.5–25 40 38.5 733 38.6 1.0 reference 25–30 40 38.5 883 46.4 0.8 0.5–1.3 >30 16 15.4 202 10.6 1.5 0.8–2.7 a Adjusted for age and country. b Four missing values. Table 2. Prevalence of occupational exposure to organic solvents for cases and controls (European study on male breast cancer) [KDF=kerosene/diesel oil/fuels oil; SBP=special-boiling-point spirits] Table 2. Exposure assessment As previously described, trained coders coded the jobs 3 Organic solvents and male breast cancer where included simultaneously to control for potential confounding between exposures; model C using a step- wise multiple logistic regression method that allowed dropping and adding solvents at each step to identify independent predictors of MBC. F probabilities were used as stepping criteria: 0.05 was the entry cut-off and 0.10 was the removal cut-off (31). tinuous variables or number (percentage) for dichotomous variables. The CES was categorized into “not exposed” plus two classes of exposed workers, according to the median among the exposed controls. Regarding ethylene glycol and tetrahydrofuran, subjects were simply defined as “never or ever exposed” because of the small number of exposed workers. We could not estimate associations between MBC and exposure to perchloroethylene, carbon tetrachloride, chloroform and diethyl ether because no case of breast cancer was exposed to these solvents. All analyses were performed using Statistical Analy- sis Systems (SAS) software, version 9.3 (SAS Institute, Inc, Cary, NC, USA). We conducted unconditional logistic regressions to estimate odds ratios (OR) and their 95% confidence intervals (95% CI) adjusting for age, country, alcohol intake in three categories (0–≥30; >30–≤60; >60 g/day), education in three categories (left school at age ≤18; professional training; university) and body mass index (BMI) in four categories (≤18.5, >18.5–≤25, >25–≤30 and >30 kg/m²) as these variables are potentially associ- ated with both breast cancer and solvent exposure. We used three different models: model A where each solvent was included separately; model B where all solvents Scand J Work Environ Health – online first a Adjusted for age and country. Results Prevalence of occupational exposure to organic solvents for cases and controls (European study on male breast cancer) [KDF=kerosene/diesel oil/fuels oil; SBP=special-boiling-point spirits] Solvents Cases (N=104) ever exposed Controls (N=1901) ever exposed N % N % Petroleum solvents Benzene 22 21.1 178 9.3 SBP 6 5.7 83 4.3 Gasoline 9 8.6 112 5.8 White spirits 31 29.8 376 19.7 KDF 22 21.1 317 16.6 Chlorinated solvents Trichloroethylene 46 44.2 613 32.2 Perchloroethylene 0 0.0 16 0.8 Methylene chloride 8 7.6 96 5.0 Chloroform 0 0.0 6 0.3 Carbon tetrachloride 0 0.0 1 0.0 Oxygenated solvents Alcohol 10 9.6 171 8.9 Ketones/esters 11 10.5 137 7.2 Diethyl ether 0 0.0 11 0.5 Ethylene glycol 5 4.8 25 1.3 Tetrahydrofuran 2 1.9 8 0.4 4 Scand J Work Environ Health – online first Laouali et al sented 72% of all those exposed to benzene (N=200) and 21% of all those exposed to TCE (N=659). were associated with similarly increased OR of 1.9, 1.8, and 1.7, respectively, using model A where each solvent was included separately. However no evidence of associa- tion was found in model B including all the solvents at a time. Exposure to SBP was not associated with MBC. Table 3 shows the adjusted OR for MBC associated with cumulative exposure scores of organic solvents obtained with three different models. i For chlorinated solvents, high exposure to TCE was associated with OR≈2 in all models, with a dose– response trend (P<0.01). The three models gave very similar estimates, including the one with stepwise vari- able selection. Among the petroleum solvents, significant two- to threefold increased OR were observed in all the models for the low benzene exposure levels compared to non- exposure. The narrower CI were observed in model C, which reports a 2.2 increase in risk. However the OR were smaller and did not reach statistical significance in the high exposure group. Low level exposure to methylene chloride was associ- ated with an OR of 2.4 in model A, but this decreased to 1.9 and was non-significant after adjustment for other solvents High exposures to gasoline, white spirits and KDF Table 3. Odds ratios (OR) for male breast cancer for exposure to organic solvents. [CES=cumulative exposure score; CI=confidence interval; KDF=kerosene/diesel oil/fuel oils; SBP=special-boiling-point spirits] Solvent CES (ppm. Adjusted for variables in model A and all solvents. Adjusted for variables in model B and regression model performed respectively with stepwise backward and stepwise forward variabl b Adjusted for variables in model A and all solvents. c Adjusted for variables in model B and regression model performed respectively with stepwise backward and stepwise forward variable selection. j g , y, , y , p b Adjusted for variables in model A and all solvents. j g , y, , y b Adjusted for variables in model A and all solvents. a Adjusted for age, country, education, body mass index, and alcohol consumption. b Adjusted for variables in model A and all solvents. Scand J Work Environ Health – online first Adjusted for age, country, education, body mass index, and alcohol consumption. Adjusted for variables in model A and all solvents. Scand J Work Environ Health – onlin exposure in model B indicating no strong association. Occupational exposure to oxygenated solvents was not associated with MBC except for ethylene glycol in the least-adjusted model based on five ever-exposed cases. In order to estimate the stability of the results, we repeated the analyses using 10-year lagtime between exposure and diagnosis or interview (table 4). Overall, we observed a shift towards null results. OR for benzene were lower and not statistically significant anymore, while the association between exposure to TCE and MBC was more stable. OR≈2 (not significant) were found in association with ethylene glycol exposure. The stepwise analysis selected TCE (statistically significant) and eth- ylene glycol, but not benzene. Results years) Cases (N=104) Controls (N=1901) Model A a Model B b Model C c OR 95% CI OR 95% CI OR 95% CI Benzene Not exposed 82 1723 1.0 reference 1.0 reference 1.0 reference >0–<0.87 13 89 2.6 1.3–5.1 3.1 1.1–8.9 2.2 1.1–4.3 ≥0.87 9 89 1.9 0.9–4.1 2.6 0.7–9.4 1.5 0.7–3.4 Gasoline Not exposed 95 1789 1.0 reference 1.0 reference Not selected >0–<0.16 4 56 1.3 0.4–3.8 1.2 0.4–3.6 ≥ 0.16 5 56 1.9 0.7–5.0 0.7 0.2–3.0 SBP Not exposed 98 1818 1.0 reference 1.0 reference Not selected >0–<1.10 3 42 1.1 0.3–3.8 0.5 0.1–1.8 ≥1.10 3 41 1.2 0.4–4.2 0.6 0.1–2.6 White spirits Not exposed 73 1525 1.0 reference 1.0 reference Not selected >0–<0.13 14 188 1.5 0.8–2.8 0.9 0.4–2.0 ≥0.13 17 188 1.8 1.0–3.1 0.8 0.3–2.1 KDF Not exposed 82 1584 1.0 reference 1.0 reference Not selected >0–<0.08 9 158 1.1 0.6–2.4 0.8 0.4–1.8 ≥0.08 13 159 1.7 0.9–3.3 0.9 0.4–2.1 Trichloroethylene Not exposed 58 1288 1.0 reference 1.0 reference 1.0 reference >0–<23.90 17 306 1.5 0.8–2.8 1.3 0.7–2.6 1.4 0.7–2.5 ≥23.90 29 307 2.2 1.3–3.7 2.1 1.2–4.0 1.9 1.1–3.3 Methylene chloride Not exposed 96 1805 1.0 reference 1.0 reference Not selected >0–<17.71 6 49 2.4 1.0–6.0 1.9 0.7–5.1 ≥17.71 2 47 0.7 0.2–2.9 0.5 0.1–2.6 Alcohol Not exposed 94 1730 1.0 reference 1.0 reference Not selected >0–<10.25 7 84 1.9 0.8–4.3 1.3 0.5–3.4 ≥10.25 3 87 0.6 0.2–2.1 0.6 0.2–2.7 Ketones/esters Not exposed 93 1764 1.0 reference 1.0 reference Not selected >0–<20.25 7 70 2.0 0.9–4.6 1.4 0.5–3.7 ≥20.25 4 67 1.0 0.3–2.8 1.1 0.3–4.5 Ethylene glycol Not exposed 99 1876 1.0 reference 1.0 reference Not selected Exposed 5 25 2.4 1.1–4.9 1.1 0.4–3.1 Tetrahydrofuran Not exposed 102 1893 1.0 reference 1.0 reference Not selected Exposed 2 8 0.7 0.2–3.0 0.4 0.1–2.6 a Adjusted for age, country, education, body mass index, and alcohol consumption. b Adjusted for variables in model A and all solvents. c Adjusted for variables in model B and regression model performed respectively with stepwise backward and stepwise forward variable selection. Table 3. Odds ratios (OR) for male breast cancer for exposure to organic solvents. [CES=cumulative exposure score; CI=confidence interval; KDF=kerosene/diesel oil/fuel oils; SBP=special-boiling-point spirits] 5 Scand J Work Environ Health – online first Organic solvents and male breast cancer Table 4. Odds ratios (OR) for male breast cancer according to occupational exposures to organic solvents with 10-year lag time period. Results [CES=cumulative exposure score; CI=confidence interval; KDF=kerosene/diesel oil/fuels oil; SBP=special-boiling-point spirits] [CES=cumulative exposure score; CI=confidence interval; KDF=kerosene/diesel oil/fuels oil; SBP=special-boiling-point spirits] Solvent CES (ppm.years) Cases (N=104) Controls (N=1901) Model A a Model B b Model C c OR 95% CI OR 95% CI OR 95% CI Benzene Not exposed 91 1762 1.0 reference 1.0 reference Not selected >0–<0.51 8 70 1.8 0.7–4.6 1.8 0.7–4.7 ≥0.51 5 69 1.7 0.6–5.0 1.6 0.5–4.9 Gasoline Not exposed 96 1805 1.0 reference 1.0 reference Not selected >0–<0.09 4 49 1.0 0.4–2.7 1.0 0.4–2.8 ≥ 0.09 4 47 1.1 0.3–3.2 0.8 0.2–2.8 SBP Not exposed 98 1823 1.0 reference 1.0 reference Not selected 0–1.10 3 39 1.2 0.4–4.1 0.9 0.2–2.1 ≥1.10 3 39 1.2 0.3–4.2 0.7 0.1–2.4 White spirits Not exposed 80 1559 1.0 reference 1.0 reference Not selected 0–0.05 11 172 1.2 0.6–2.3 1.1 0.6–2.2 ≥0.05 13 170 1.2 0.6–2.6 1.1 0.5–2.5 KDF Not exposed 86 1648 1.0 reference 1.0 reference Not selected 0–0.05 8 126 1.6 0.8–3.0 1.4 0.7–2.8 ≥0.05 10 127 1.3 0.6–2.8 1.0 0.5–2.4 Trichloroethylene Not exposed 64 1363 1.0 reference 1.0 reference 1.0 reference 0–15.30 15 269 1.2 0.6–2.2 1.0 0.5–1.9 1.2 0.6–2.1 ≥15.30 25 269 2.2 1.3–3.6 1.6 0.8–2.9 1.9 1.2–3.2 Methylene chloride Not exposed 99 1826 1.0 reference 1.0 reference Not selected 0–9.37 3 37 0.9 0.3–3.0 1.0 0.3–3.5 ≥9.37 2 38 1.0 0.4–2.5 1.1 0.4–3.2 Alcohol Not exposed 97 1763 1.0 reference 1.0 reference Not selected 0–3.48 4 69 0.5 0.2–1.4 0.5 0.2–1.5 ≥3.48 3 69 0.7 0.3–1.9 0.7 0.2–1.9 Ketones/esters Not exposed 95 1790 1.0 reference 1.0 reference Not selected 0–7.00 5 56 2.1 0.8–5.5 1.7 0.6–5.1 ≥7.00 4 55 1.9 0.7–5.4 1.4 0.4–4.8 Ethylene glycol Not exposed 101 1878 1.0 reference 1.0 reference 1.0 reference Exposed 3 23 2.3 0.9–5.6 2.1 0.7–6.3 1.8 0.8–3.9 Tetra-hydrofuran Not exposed 102 1895 1.0 reference 1.0 reference Not selected Exposed 2 6 0.5 0.1–2.4 0.6 0.1–3.0 a Adjusted for age, country, education, body mass index and alcohol consumption. b Adjusted for variables in model A and all solvents. c Adjusted for variables in model A and stepwise selection of solvents. Results were not substantially modified in sensitiv- ity analyses performed by excluding the hospital-based cancer controls in all countries (supplemental table S1, www.sjweh.fi/show_abstract.php?abstract_id=3717). exposure in model B indicating no strong association. Petroleum solvents The toxicity of benzene has been largely proven towards leukemia risk, and the IARC has classified benzene as a carcinogen (group 1) (12). It is generally recognized that aromatic hydrocarbons, such as benzene or methylene chloride, are mammary carcinogens in animals (7, 8, 32, 33). Benzene induces oxidative stress, is genotoxic and immunosuppressive. It causes genomic instabil- ity and induces apoptosis (12). In a study on female mice, benzene-induced mammary tumors exhibited a distinct pattern in the p53 and H-ras mutations com- pared to spontaneous tumors, suggesting that benzene induces specific genetic alterations (34). In addition, benzene and other aromatic hydrocarbons have exhib- ited endocrine-disrupting properties (35, 36) making the mammary gland more prone to tumor cell proliferation through hormonal mechanisms (37, 38). Apart from benzene, petroleum solvents in our analysis were complex mixtures of polycyclic aromatic hydrocarbons (PAH). We found elevated OR for white spirits, but the association disappeared when benzene was included in the model, suggesting that benzene content in white spirits explains this association. Previ- ous studies on MBC reported no association between MBC mortality and PAH exposure (23) or a doubling of MBC incidence in workers exposed to PAH or gasoline (24). Occupational exposures to aromatic, aliphatic and alicyclic hydrocarbon solvents were positively associ- ated with female breast cancer in some studies (16, 20). However, other studies gave inconsistent or null results (19, 42, 43). In another study, PAH DNA adducts, con- sidered as a body-burden measure of exposure to PAH of any source, were more frequent in female breast cancer patients than controls (44). In total, these studies are very sparse, are based on weak exposure characteriza- tion, and do not allow to conclude about breast cancer risk in relation to petroleum derivatives in general. g ( , ) In our study, exposure to benzene was associated with MBC, although the highest OR was found in the low-exposure group. Overall, this is consistent with the two-fold increased OR of MBC among motor vehicle mechanics and painters with known exposure to benzene previously reported in the same study population (25). According to the JEM, in addition to motor vehicle mechanics and painters, the most frequent occupations involving exposure to benzene included machinery fit- ters, shoe makers, printers, precision-instrument makers, tire makers and vulcanizers. Discussion In this European case–control study on MBC, one of the largest ever conducted, occupational exposure to organic solvents was assessed using a detailed JEM. High exposure to TCE was significantly associated with a doubling of the risk of MBC, a finding that remained 6 Scand J Work Environ Health – online first Laouali et al stable in different models and after accounting for a 10-year lag time period. A dose–response trend was also indicated. In addition, a possible role for benzene and ethylene glycol in the etiology of MBC was suggested. tatively. One study among shoe factory workers in Italy reported a two-fold increased standardized incidence ratio among women exposed to benzene >40 ppm years, but this results was based on only one case (41). In total, our study provides some support for the hypothesis that benzene may increase breast cancer risk, but this should be scrutinized in further studies. Petroleum solvents To explain the non-mono- tonic dose–response curve between benzene exposure and MBC in our study, it is possible that the participa- tion rate of the most highly exposed cases was reduced, eg, because of survival bias, thus leading to a smaller OR in the high exposure group. It is also possible that the risk of health outcome induced at low doses of ben- zene exposure does not increase linearly at higher doses, as is frequently observed in studies on health effects of endocrine disrupting chemicals (39). Several epidemio- logical studies have examined breast cancer in men or women in relation to benzene exposure. One study has evaluated the risk of MBC associated with exposure to benzene among US marines exposed to drinking water contaminated with solvents (40), but no association was found. However, the context and route of exposure were not the same as in our study and exposure levels were very low. As regards female breast cancer, population- based or occupational cohort studies reported positive associations with occupational exposure to benzene (17, 18, 41), but other did not (19, 42). However, most studies did not assess benzene exposure levels quanti- Scand J Work Environ Health – online first Oxygenated solvents Oxygenated solvents include a large variety of chemi- cals. Among alcoholic solvents, ethanol has well-known carcinogenic effects on many organs including the breast. These effects have been investigated intensively among alcohol drinkers in numerous studies. However exposure to alcoholic solvents from occupational origin through inhalation or cutaneous absorption may have different effects on health. Exposure to oxygenated solvents in general has been known to cause reprotoxic or neurologic effects, but their carcinogenic properties are not documented. It has been shown for example, that female workers exposed to ethylene glycol ethers had prolonged menstrual cycles and time-to-pregnancy com- pared to those who were not exposed (48, 49). Experi- mental studies have also demonstrated an increased pro- duction of progesterone in ovarian luteal cells exposed to ethylene glycol ethers (50, 51). It can be hypothesized that modifications of hormone synthesis in exposed women has an impact on breast cancer risk. i Regarding the quality of exposure assessment, we used a detailed JEM elaborated for assessing occupa- tional exposures in France as far back as the 1950s and based on the expertise of experienced industrial hygienists, extensive literature review and relevant exposure monitoring data at the work places. Using a JEM is considered to be a less accurate method than expert judgment based on the review of individual job histories (54). Moreover, the same JEM were used for all countries, while the jobs and the job tasks could differ to some extent between European countries. This could have led to errors in exposure estimates, however mis- classification errors are expected to be non-differential. Our findings suggest a possible role of occupational exposure to ethylene glycol in MBC. OR were ≈2 in the model without adjustment for other solvents and in the analyses (five exposed cases) and in the models accounting for a 10-year lag time (three exposed cases). Conversely, we did not observe any association with occupational exposure to alcohol (seven exposed cases), tetrahydrofuran (two exposed cases), and ketones/esters (nine exposed cases) in our study. The small number of cases in our data and the lack of known mechanistic pathway does not allow us to draw conclusions. Although this is the largest case–control study ever conducted to evaluate the association of MBC and occupational exposure to organic solvents, an impor- tant limitation is the relatively small number of cancer cases for certain exposures evaluated due to the rarity of the disease. Exposure to chlorinated solvents TCE is one of the most important chlorinated solvents, and the IARC recently upgraded it from group 2A (prob- ably carcinogenic to humans) to group 1 (carcinogenic to humans) based on epidemiological and animal evi- dence of a role in the risk of kidney cancer (13). TCE is metabolized to multiple mutagenic and carcinogenic metabolites that contribute to the carcinogenicity of the parent compound via genotoxic or non-genotoxic mechanisms (45). However, the biological mechanisms that could explain an excess breast cancer risk associated with TCE exposure are not known. We found that TCE exposure was associated with an increased risk of MBC. This association was stable in different models, robust in the analysis accounting for a 10-year lagtime and reinforced by a dose–response trend. The most frequent occupations exposed to TCE in our data were motor mechanics, machine fitters, plumb- ers and painters. Conversely, no consistent association with methylene chloride was detected in our study after adjustment for exposure to other solvents based on eight exposed cases, whereas no cases were exposed to the three remaining chlorinated solvents (perchloroethylene, chloroform and carbon tetrachloride). In the study of US marines exposed to contaminated drinking water, exposure to TCE did not increase the 7 Organic solvents and male breast cancer risk of MBC (40). However exposure to TCE in this study (median cumulative exposure 159 ppb months) was several orders of magnitude lower than in our data. Previous studies on female breast cancer, including a large population-based cohort in Finland (20) and a cohort of military women in the US (15) occupationally exposed to TCE, reported no evidence of an association with exposure to chlorinated solvents. No association of breast cancer with urinary biomarkers of exposure to TCE was found in a Scandinavian study, which included 260 female and 2 male cases (46). However, in an electron- ics workers cohort study in Taiwan, an increased risk of female breast cancer in association with a long duration of exposure to chlorinated solvents was reported (47). Overall, results of epidemiological studies do not allow for the drawing of firm conclusions, but our data add to the evidence that TCE may play a role in breast cancer. breast cancer (43). Study strengths and limitations As reported earlier (27), the number of eligible cases identified during the study period was close to the expected number, based on incidence data from cancer registries in the participating areas. The case participa- tion rate was 85%. The overall participation rate among controls was 67%, but there were large disparities across countries. Among countries that used population controls, the response rate was low in Northern Europe (Denmark, Germany, and Sweden). A differential participation of cases and controls, eg, according to social class, is thus possible, but its effects on our results were attenuated by adjustment on education level. Selection bias could occur in countries using hospital-based controls, mainly colon cancer patients. However, the results were not different when the analyses were conducted separately for subjects recruited with a population- and hospital-based design. Exposure to chlorinated solvents Of note, cohort studies of workers exposed to ethylene oxide, an intermediate chemical used in the production of ethylene glycol that is also used as a sterilizing agent for medical equipment, found that breast cancer incidence was increased in exposed women (52, 53). However, the data on breast cancer in relation to oxygenated solvents are sparse and require further investigations in groups of workers with well- characterized exposure. Scand J Work Environ Health – online first Oxygenated solvents This may lead to low statistical power for detecting significant associations or concluding on the lack thereof, particularly for the less-frequent exposures. However, our study permitted the detection of some associations between occupational factors and breast cancer that could have been difficult to observe in studies on women. To the best of our knowledge, no study has evaluated the risk of MBC specifically associated with exposure to oxygenated solvents to date. One case–control study in Australia evaluated occupational exposure to alcohol solvents in women and found no association with female Concluding remarks Concluding remarks 8 Laouali et al Laouali et al Ligue Nationale contre le Cancer, Fondation de France (Grant No. 955368), Institut National de la Santé et de la Recherche Médicale, Programme Environnement-Santé, Ministère de l’Environnement. Germany: Federal Min- istry for Education, Science, Reseach and Technology (BMBF), Grant No. 01-HP-684/8. Italy: MURST, Italian Association for Cancer Research, Compagnia San Paolo/ FIRMS. Portugal: Junta Nacional de Investigacao Cienti- fica e Tecnologica, Praxis XXI, No. 2/2.1/SAU/1178/95. Spain: Fondo de Investigacion de la Sanitarie, Minis- terio de Sanidad y Consumo, Unidad de Investigacion Clinico Epidemiologica, Hospital Dr. Peset. General- itet Valenciana; Departmento de Sanidad y Consumo, Gobierno Vasco; Fondo de Investigacion de la Sanitaria (FIS), Ministerio de Sanidad y Consumo, Ayuda a la Investigacion del Departamento de Salud del Gobierno de Navarra. Sweden: Swedish Council for Work Life Research, Research Foundation of the Department of Oncology in Umea, Swedish Society of Medicine, Lund University Hospital Research Foundation, Gunnar, Arvid and Elsebeth Nilsson Cancer Foundation. In conclusion, this study suggests that men occupation- ally exposed to TCE, and possibly benzene or ethylene glycol, are at increased risk of breast cancer, a finding that conceivably can be extrapolated to female breast cancer. Mechanisms that can explain these associa- tions are still to be elucidated. Further epidemiological studies should be conducted in populations with well- characterized exposure to organic solvents in order to confirm these findings. Acknowledgements Rare Cancer Study Group members. Project manage- ment group: Wolfgang Ahrens, Mikael Eriksson, Pascal Guenel, Henrik Kolstad, Linda Kaerlev, Jean-Michel Lutz, Elsebeth Lynge, Franco Merletti, Maria M. Morales Suarez-Varela, Jorn Olsen, and Svend Sabroe. Other members: Denmark: Herman Autrup, Lisbeth Norum Pedersen, Preben Johansen, Stein Poulsen, Peter Stubbe Teglbjaerg, and Mogens Vyberg. France: Antoine Buemi, Paule-Marie Carli, Gilles Chaplain, Jean-Pierre Dau- res, Jean Faivre, Joelle Fevotte, Pascale Grosclaude, Anne-Valérie Guizard, Michel Henry-Amar, Guy Launoy, Francois Menegoz, Nicole Raverdy, and Paul Schaffer. Germany: Cornelia Baumgardt-Elms, Sibylle Gotthardt, Ingeborg Jahn, Karl-Heinz Jockel, Hiltrud Merzenick, Andreas Stang, Christa Stegmaier, Antje Timmer, and Hartwig Ziegler. Italy Terri Ballard, Franco Bertoni, Giuseppe Gorini, Sandra Gostinicchi, Giovanna Masala, Enzo Merler, Lorenzo Richiardi, Lorenzo Simonato, and Paola Zambon. Latvia: Irena Rogovska, Galina Sharkova, and Aivars Stengrevics. Lithuania: Jolita Gibaviciene, Laimonas Jazukevicius, Juozas Kurtinaitis, and Roma Pociute. Portugal: Noemia Afonso, Altamiro Costa-Pereira, Sonia Doria, Carlos Lopes, Jose Manuel Lopes, Ana Miranda, and Cristina Santos. Spain: Daniel Almenar, Ines Aguinaga, Juan J. Aurrekoetxea, Con- cepcion Brun, Alicia Cordoba, Francisco Guillen, Rosa Guarch, Agustin Llopis, Rosa Llorente, Blanca Marın, Amparo Marquina, Miguel Angel Martınez, JM Martınez Penuela, Ana Puras, Ma Adela Sanz, Francisco Vega, and Ma Aurora Villanueva. Sweden: Lennart Hardell, Irene Larsson, Hakan Olson, Monica Sandstrom, and Gun Wingren. The United Kingdom: Janine Bell, Ian Cree, Tony Fletcher, and Alex JE Foss. References 10. Telang NT, Basu A, Modak MJ, Osborne MP, Cellular ras protooncogene expression in human mammary explant cultures. A potential marker for chemical carcinogenesis. Ann NY Acad Sci. 1990;586:230–7. https://doi. org/10.1111/j.1749-6632.1990.tb17811.x. 22. Peplonska B, Stewart P, Szeszenia-Dabrowska N, Lissowska, J, Brinton LA, Gromiec JP, et al. Occupational exposure to organic solvents and breast cancer in women. Occupational Environ Med 2010. 67(11): p. 722-729. https://doi. org/10.1136/oem.2009.046557 23. Cocco P, Figgs L, Dosemeci M, Hayes R, Linet MS, Hsing AW. Case-control study of occupational exposures and male breast cancer. Occup Environ Med. 1998; 55(9):599–604. https://doi. org/10.1136/oem.55.9.599. 11. Dunnick JK, Elwell MR, Huff J, Barrett CJ. Chemically induced mammary gland cancer in the National Toxicology Program’s carcinogenesis bioassay. Carcinogenesis. 1995;16(2):173–9. https://doi.org/10.1093/carcin/16.2.173. 24. Hansen J. Elevated risk for male breast cancer after occupational exposure to gasoline and vehicular combustion products. Am J Ind Med. 2000;37(4):349–52. https://doi. org/10.1002/(SICI)1097-0274(200004)37:4<349::AID- AJIM4>3.0.CO;2-L. 12. Loomis D, Guyton KZ, Grosse Y, El Ghissassi F, Bouvard V, Benbrahim-Tallaa L, et al. Carcinogenicity of benzene. Lancet Oncol. 2017;18(12):1574–5.https://doi.org/10.1016/S1470- 2045(17)30832-X. 13. International Agency for Research on Cancer (IARC). Trichloroehtylene, tetrachloroethylene and some other chlorinated agents. Lyon: IARC; 2014. IARC monographs on the evaluation of the carcinogenic risks of chemicals to humans, vol 106, 2014. 25. Villeneuve S, Cyr D, Lynge E, Orsi L, Sabroe S, Merletti F, et al. Occupation and occupational exposure to endocrine disrupting chemicals in male breast cancer: a case-control study in Europe. Occup Environ Med. 2010;67(12):837–44. https://doi.org/10.1136/oem.2009.052175. 14. Lamm SH, Engel A, Joshi KP, Byrd DM 3rd, Chen R, Chronic myelogenous leukemia and benzene exposure: a systematic review and meta-analysis of the case-control literature. Chem Biol Interact. 2009.182(2-3):93–7. https://doi.org/10.1016/j. cbi.2009.08.010. 26. Fevotte J, Dananche B, Delabre L, Ducamp S, Garras L, Houot M, et al. Matgene: a program to develop job- exposure matrices in the general population in France. Ann Occup Hyg. 2011;55(8):865–78. https://doi.org/10.1136/ oemed-2011-100382.256. 15. Blair A, Hartge P, Stewart PA, McAdams M, Lubin J, Mortality and cancer incidence of aircraft maintenance workers exposed to trichloroethylene and other organic solvents and chemicals: extended follow up. Occup Environ Med. 1998;55(3):161–71. https://doi.org/10.1136/oem.55.3.161. 27. Lynge E, Afonso N, Kaerlev L, Olsen J, Sabroe S, Ahrens W, et al. European multi-centre case-control study on risk factors for rare cancers of unknown aetiology. Eur J Cancer. 2005;41(4):601–12. https://doi.org/10.1016/j. ejca.2004.12.016. 16. Labreche F, Goldberg MS, Valois MF, Nadon L. Postmenopausal breast cancer and occupational exposures. Occup Environ Med. 2010.67(4):263–9. https://doi. org/10.1136/oem.2009.049817. 28. References 1. Torre LA, Bray F, Siegel RL, Ferlay J, Lortet-Tieulent J, Jemal A. Global cancer statistics, 2012. CA: Cancer J Clin. 2015.65(2):87–108. https://doi.org/10.3322/caac.21262. 2. Curado M, Edwards B, Shin H, Storm H, Ferlay J, Heanue M, et al. Cancer incidence in five continents. Lyon: International Agency for Research (IARC); Scientific publication No. 160, vol IX, 2007. 3. Ferzoco RM, Ruddy KJ, The Epidemiology of Male Breast Cancer. Curr Oncol Rep. 2016;18(1):1. https://doi. org/10.1007/s11912-015-0487-4. 4. Anderson WF, Jatoi I, Tse J, Rosenberg PS. Male breast cancer: a population-based comparison with female breast cancer. J Clin Oncol. 2010;28(2):232–9.https://doi.org/10.1200/ JCO.2009.23.8162. 5. Guenel P, Cyr D, Sabroe S, Lynge E, Merletti F, Ahrens W, et al. Alcohol drinking may increase risk of breast cancer in men: a European population-based case-control study. Cancer Causes Control. 2004;15(6):571–80. https://doi.org/10.1023/ B:CACO.0000036154.18162.43. 6. Ruddy KJ, Winer EP. Male breast cancer: risk factors, biology, diagnosis, treatment, and survivorship. Ann Oncol. 2013;24(6):1434–43. https://doi.org/10.1093/annonc/ mdt025. We acknowledge collaboration from the patients, con- trols, participating hospitals and data providers. The con- tribution of the individual cancer registries is gratefully acknowledged. The study was financially supported by the European Commission, DGXII, Grants No. BMH1 CT 931630 and ERB CIPD CT 940285, and national funding agencies. Denmark: The Strategic Environment Programme and the Danish Science Foundation. France: 7. Rodgers KM, Udesky JO, Rudel RA, Brody J. Environmental chemicals and breast cancer: An updated review of epidemiological literature informed by biological mechanisms. Environ Res. 2018;160:152–82. https://doi.org/10.1016/j. envres.2017.08.045. 7. Rodgers KM, Udesky JO, Rudel RA, Brody J. Environmental chemicals and breast cancer: An updated review of epidemiological literature informed by biological mechanisms. Environ Res. 2018;160:152–82. https://doi.org/10.1016/j. envres.2017.08.045. 9 9 Scand J Work Environ Health – online first Organic solvents and male breast cancer 20. Weiderpass E, Pukkala E, Kauppinen T, Mutanen P, Paakkulainen H, Vasama-Neuvonen K, et al. Breast cancer and occupational exposures in women in Finland. Am J Ind Med. 1999;36(1):48–53. https://doi.org/10.1002/(SICI)1097- 0274(199907)36:1<48::AID-AJIM7>3.0.CO;2-2. 8. Rudel RA, Attfield KR, Schifano JNB, Brody JG, Chemicals causing mammary gland tumors in animals signal new directions for epidemiology, chemicals testing, and risk assessment for breast cancer prevention. Cancer. 2007;109(12 Suppl):2635–66. https://doi.org/10.1002/cncr.22653. 9. Labreche FP, Goldberg MS, Exposure to organic solvents and breast cancer in women: a hypothesis. Am J Ind Med. 1997;32(1):1–14. https://doi.org/10.1002/(SICI)1097- 0274(199707)32:1<1::AID-AJIM1>3.0.CO;2-3. 21. Ekenga CC, Parks CG, Sandler DP. Chemical exposures in the workplace and breast cancer risk: A prospective cohort study. Int J Cancer. 2015;137(7):1765–74. https://doi.org/10.1002/ ijc.29545. References Kaerlev L, Teglbjaerg PS, Sabroe S, Kolstad HA, Ahrens W, Eriksson M, et al. Occupational risk factors for small bowel carcinoid tumor: a European population-based case-control study. J Occup Environ Med. 2002;44(6):516–22. https://doi. org/10.1097/00043764-200206000-00012. 17. Petralia SA, Chow WH, McLaughlin J, Jin F, Gao YT, Dosemeci M. Occupational risk factors for breast cancer among women in Shanghai. Am J Ind Med. 1998;34(5): 477–83. https://doi. org/10.1002/(SICI)1097-0274(199811)34:5<477::AID- AJIM8>3.0.CO;2-N. 29. Pilorget C, Dananche B, Luce D, Fevotte J. Éléments techniques sur l’exposition professionnelle aux carburants et solvants pétroliers. matrice emplois-expositions aux carburants et solvants pétroliers. [Technical data on occupational exposure to fuel oils and petroleum solvents. Job-exposure matrix for fuel oils and petroleum solvents]. Saint-Maurice; Institut de veille sanitaire; 2007. 18. Petralia S, Vena JE, Freudenheim JL, Dosemeci M, Michalek A, Goldberg MS, et al. Risk of premenopausal breast cancer in association with occupational exposure to polycyclic aromatic hydrocarbons and benzene. Scand J Work Environ Health. 1999;25(3):215–21. https://doi.org/10.5271/sjweh.426. 30. Dananche B, Fevotte J. Éléments techniques sur l’exposition professionnelle à cinq solvants chlorés. Matrices emplois- expositions à cinq solvants chlorés : trichloroéthylène, perchloroéthylène, chlorure de méthylène, tétrachlorure de carbone, chloroforme. [Technical data on occupational exposure to five chlorinated solvents. Job-exposure 19. Ray RM, Gao DL, Li W, Wernli KJ, Astrakianakis G, Seixas NS, et al., Occupational exposures and breast cancer among women textile workers in Shanghai. Epidemiology. 2007;18(3):383– 92. https://doi.org/10.1097/01.ede.0000259984.40934.ae. 10 Scand J Work Environ Health – online first Laouali et al matrices for five chlorinated solvents: trichloroethylene, perchloroethylene, methylene chloride, carbon tetrachloride, chloroform]. Saint-Maurice; Institut de veille sanitaire; 2009. 44. Gammon MD, Sagiv SK, Eng SM, Shantakumar S, Gaudet MM, Teitelbaum SL, et al. Polycyclic aromatic hydrocarbon- DNA adducts and breast cancer: a pooled analysis. Arch Environ Health. 2004;59(12):640–9. https://doi. org/10.1080/00039890409602948. 31. Zhang Z. Variable selection with stepwise and best subset approaches. Ann Transl Med. 2016;4(7):136. https://doi. org/10.21037/atm.2016.03.35. 45. Rusyn I, Chiu WA, Lash LH, Kromhout H, Hansen J, Guyton KZ. Trichloroethylene: Mechanistic, epidemiologic and other supporting evidence of carcinogenic hazard. Pharmacol Ther. 2014;141(1):55–68. https://doi.org/10.1016/j. pharmthera.2013.08.004. 32. Brody JG, Rudel RA. Environmental pollutants and breast cancer. Environ Health Persp. 2003;111(8):1007–19. https:// doi.org/10.1289/ehp.6310. 33. Bennett LM, Davis BJ. Identification of mammary carcinogens in rodent bioassays. Environ Mol Mutagen. 2002; 39(2- 3):150–7. https://doi.org/10.1002/em.10068. 46. Hansen J, Sallmen M, Selden AI, Anttila A, Pukkala E, Andersson K, et al. Risk of cancer among workers exposed to trichloroethylene: analysis of three Nordic cohort studies. References J Natl Cancer Inst. 2013;105(12):869–77. https://doi. org/10.1093/jnci/djt107. 34. Houle CD, Ton TV, Clayton N, Huff J, Hong HH, Sills RC. Frequent p53 and H-ras mutations in benzene- and ethylene oxide-induced mammary gland carcinomas from B6C3F1 mice. Toxicol Pathol. 2006;34(6):752–62. https://doi. org/10.1080/01926230600935912. 47. Sung TI, Chen PC, Jyuhn-Hsiarn Lee L, Lin YP, Hsieh GY, Wang JD. Increased standardized incidence ratio of breast cancer in female electronics workers. BMC Public Health. 2007;7:102. https://doi.org/10.1186/1471-2458-7-102. 35. Reutman SR, LeMasters GK, Knecht EA, Shukla R, Lockey JE, Burroughs GE, et al. Evidence of reproductive endocrine effects in women with occupational fuel and solvent exposures. Environ Health Persp. 2002;110(8):805–11. https:// doi.org/10.1289/ehp.02110805. 48. Hsieh GYY, Wang JDD, Cheng TJJ, Chen PCC. Prolonged menstrual cycles in female workers exposed to ethylene glycol ethers in the semiconductor manufacturing industry. Occup Environ Med. 2005;62(8):510–6. https://doi.org/10.1136/ oem.2004.016014. 36. Bolden AL, Kwiatkowski CF, Colborn T. New Look at BTEX: Are Ambient Levels a Problem? Environ Sci Technol. 2015;49(9). https://doi.org/10.1021/es505316f. 49. Chen PCC, Hsieh GYY, Wang JDD, Cheng TJJ. Prolonged time to pregnancy in female workers exposed to ethylene glycol ethers in semiconductor manufacturing. Epidem. (Cambridge, Mass.). 2002;13(2):191–6. https://doi. org/10.1097/00001648-200203000-00014. 37. Diamanti-Kandarakis E, Bourguignon J-PP, Giudice LC, Hauser R, Prins GS, Soto AM, et al. Endocrine-disrupting chemicals: an Endocrine Society scientific statement. Endoc Rev. 2009;30(4):293–342. https://doi.org/10.1210/er.2009-0002. 38. Fenton SE. Endocrine-disrupting compounds and mammary gland development: early exposure and later life consequences. Endocrin. 2006;147(6 Suppl):24.https://doi.org/10.1210/ en.2005-1131. 50. Davis BJ, Almekinder JL, Flagler N, Travlos G, Wilson R, Maronpot RR. Ovarian luteal cell toxicity of ethylene glycol monomethyl ether and methoxy acetic acid in vivo and in vitro. Toxic Appl Pharmac. 1997;142(2):328–37. https://doi. org/10.1006/taap.1996.8035 39. Vandenberg LN, Colborn T, Hayes TB, Heindel JJ, Jacobs DR, Lee D-H, et al. Hormones and endocrine-disrupting chemicals: low-dose effects and nonmonotonic dose responses. Endocr Rev. 2012;33(3):378–455. https://doi.org/10.1210/er.2011- 1050. 51. Almekinder JL, Lennard DE, Walmer DK, Davis BJ. Toxicity of methoxyacetic acid in cultured human luteal cells. Fund App Toxic: Off J Soc Toxic. 1997;38(2):191–4. https://doi. org/10.1006/faat.1997.2332. 52. Norman SA, Berlin JA, Soper KA, Middendorf BF, Stolley PD. Cancer incidence in a group of workers potentially exposed to ethylene oxide. Int J Epidemiol. 1995; 24(2): 276–84. https://doi.org/10.1093/ije/24.2.276. 40. Ruckart PZ, Bove FJ, Shanley E 3rd, Maslia M. Evaluation of contaminated drinking water and male breast cancer at Marine Corps Base Camp Lejeune, North Carolina: a case control study. Environ Health. 2015;14:74. https://doi.org/10.1186/ s12940-015-0061-4. 53. References Mikoczy Z, Tinnerberg H, Bjork J, Albin M. Cancer incidence and mortality in Swedish sterilant workers exposed to ethylene oxide: updated cohort study findings 1972-2006. Int J Environ Res Public Health. 2011;8(6):2009–19. https://doi. org/10.3390/ijerph8062009. 41. Costantini AS, Gorini G, Consonni D, Miligi L, Giovannetti L, Quinn M. Exposure to benzene and risk of breast cancer among shoe factory workers in Italy. Tumori. 2009;95(1):8–12. 42. Peplonska B, Stewart P, Szeszenia-Dabrowska N, Lissowska J, Brinton LA, Gromiec JP, et al. Occupational exposure to organic solvents and breast cancer in women. Occup Environ Med. 2010;67(11):722–9. https://doi.org/10.1136/ oem.2009.046557. 54. Siemiatycki J, Dewar R, Richardson L. Costs and statistical power associated with five methods of collecting occupation exposure information for population-based case-control studies. Am J Epidemiol. 1989:130(6):1236–46. https://doi. org/10.1093/oxfordjournals.aje.a115452. 43. Glass DC, Heyworth J, Thomson AK, Peters S, Saunders C, Fritschi L. Occupational exposure to solvents and risk of breast cancer. Am J Ind Med. 2015;58(9):915–22. https://doi. org/10.1002/ajim.22478. Received for publication: 27 September 2017 11 Scand J Work Environ Health – online first
https://openalex.org/W4225310686	https://www.researchsquare.com/article/rs-1617822/latest.pdf	English	null	Effect of oral nirmatrelvir on Long COVID symptoms: a case series	Research Square (Research Square)	2,022	cc-by	2,744	Effect of oral nirmatrelvir on Long COVID symptoms: a case series Michael J. Peluso (  michael.peluso@ucsf.edu ) University of California, San Francisco https://orcid.org/0000-0003-0585-6230 Khamal Anglin University of California, San Francisco Matthew S. Durstenfeld University of California, San Francisco Matthew S. Durstenfeld University of California, San Francisco Jeffrey N. Martin University of California, San Francisco Jeffrey N. Martin University of California, San Francisco J. Daniel Kelly University of California, San Francisco J. Daniel Kelly University of California, San Francisco Priscilla Y. Hsue University of California, San Francisco Priscilla Y. Hsue University of California, San Francisco Timothy J. Henrich University of California, San Francisco Timothy J. Henrich University of California, San Francisco Steven G. Deeks University of California, San Francisco Case Report Posted Date: May 5th, 2022 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/7 Abstract As the SARS-CoV-2 pandemic continues to evolve, efforts to understand variability in COVID-19 recovery, as well as the impact of factors including viral variants, vaccine status, and COVID-19 treatment on the development and persistence of Long COVID symptoms have intensified. We report three cases that demonstrate that variability in the timing of nirmatrelvir therapy may be associated with different outcomes and underscores the need for systematic study of antiviral therapy for this disease condition. Introduction Many individuals do not fully recover from acute SARS-CoV-2 infection (“Long COVID,” a type of post- acute sequelae of SARS-CoV-2 [PASC]). Efforts to prevent or reverse this potentially disabling syndrome are just now emerging. Here, we report three cases in which individuals were treated with nirmatrelvir/ritonavir at various stages following infection. We believe these cases may inform our understanding of the pathophysiology of post-acute symptoms and provide rationale for studying antiviral therapies in people with Long COVID. Cases Participants were sequential volunteers in the UCSF Long-term Impact of Infection with Novel Coronavirus (LIINC) study (NCT04362150), who enrolled because of a history of Long COVID symptoms and reported a history of nirmatrelvir/ritonavir use. The study was approved by the UCSF Institutional Review Board. Volunteers provided written informed consent prior to collection of clinical data and consented to the presentation of their cases. Case 1 A 48-year-old man with a past medical history of presumed Bechet’s disease on colchicine developed fever, worsening headache, and pharyngitis in Spring 2022. He had previously received 2 doses of the Moderna SARS-CoV-2 vaccine and 1 dose of the Pfizer-BioNTech vaccine, most recently 5 months prior. A rapid antigen test was positive, as was a confirmatory PCR test. He was prescribed a 5-day course of nirmatrelvir/ritonavir, which he initiated within 24 hours of symptom onset, and experienced rapid improvement in his systemic symptoms. However, approximately four days following completion of the 5-day course, he experienced rebound symptoms with recurrence of fever, fatigue, rhinorrhea, cough, chest pain, rash on his upper and lower extremities, and trouble concentrating (“brain fog”). During this period, he wore a personal fitness device which recorded certain physiologic measurements including heart rate, respiratory rate, and change from baseline body temperature (Fig. 1). Approximately 3 weeks following the positive test and despite prior antiviral therapy, he experienced worsening of his fatigue and associated chest soreness, palpitations, brain fog, and symptoms of post-exertional malaise, which have now persisted beyond 30 days following initial symptom onset. Page 2/7 Case 2 A 42-year-old man with no significant medical history developed rhinorrhea and pharyngitis in early 2022, followed by fatigue, myalgia, and a pruritic rash on his upper extremities and groin that persisted for approximately 10 days. He had previously received 3 doses of the Pfizer-BioNTech SARS-CoV-2 vaccine, most recently 2 months prior. A SARS-CoV-2 antigen test was positive when performed two days after symptom onset; multiple additional antigen tests were subsequently positive. On Day 11, an antigen test was negative and all of his symptoms resolved. Approximately two weeks later, he experienced new onset of severe myalgia and bone pain across his upper body, which he described as post-exertional soreness in the absence of exertion. There was an associated increase in fatigue, as well as heightened awareness of breathing (described as “lung soreness”). These symptoms persisted, and he contacted his primary care doctor for evaluation approximately 10 days later. At this visit, he noted that he was experiencing ongoing fatigue, and described being at 80% of his pre-COVID baseline health. Laboratory testing was notable only for vitamin D insufficiency (25 ng/mL); a chest X-ray showed no abnormalities. Approximately 7 weeks following initial symptom onset, his symptoms worsened. He experienced ongoing myalgia, severe fatigue, post-exertional malaise, and trouble with concentration (“brain fog”). These symptoms profoundly impacted his ability to perform his activities of daily living, and he felt substantially debilitated, reporting that he was at 40–50% of his pre-COVID baseline health and that he was spending the majority of the day resting. He began to seek care for Long COVID because he was concerned about the duration for which his symptoms had persisted. As his symptoms continued to worsen, he was re-exposed to SARS-CoV-2 when his spouse and children tested positive on antigen tests. A repeat antigen test was negative, but he experienced further worsening of his symptoms which his provider attributed to possible re-infection with SARS-CoV-2. In this context, he received a prescription for nirmatrelvir/ritonavir. Within days of re-exposure, the patient began to note an improvement in his persistent symptoms, while his family members continued to experience worsening symptoms. After symptomatic improvement for 1–2 days, he initiated a 5-day course of nirmatrelvir/ritonavir. During this period, he continued to experience improvement in his symptoms. While they have not resolved entirely, he reports that he is gradually approaching his baseline health. Discussion As the SARS-CoV-2 pandemic continues to evolve, efforts to understand variability in COVID-19 recovery, as well as the impact of factors including viral variants, vaccine status, and COVID-19 treatment on the development and persistence of Long COVID symptoms have intensified. This case series demonstrates that variability in the timing of antiviral therapy may be associated with different outcomes and underscores the need for systematic study of antiviral therapy for this disease condition during both the acute and convalescent stages. It has been suggested that the viral burden during acute infection may be an important determinant of Long COVID,1 and that early antiviral therapy might mitigate this risk. In Case 1, the individual took nirmatrelvir/ritonavir according to the recent Emergency Use Authorization (EUA) criteria2 and shortly thereafter experienced rebound symptoms, which were associated with physiologic changes measured using a fitness device. Symptomatic relapses of SARS-CoV-2 infection are just now starting to be reported.3 Although he was not re-tested during this period, it is possible that this coincided with viral rebound upon the completion of therapy. Concerningly, he subsequently experienced worsening post- infectious symptoms which now meet U.S. Centers for Disease Control (CDC) criteria for Long COVID.4 This suggests that although a short course of early antiviral therapy is adequate to prevent severe acute disease in high-risk patients,5 it may be insufficient to prevent the development of Long COVID, and those experiencing rebound symptoms could remain at risk. It has been suggested that the viral burden during acute infection may be an important determinant of Long COVID,1 and that early antiviral therapy might mitigate this risk. In Case 1, the individual took nirmatrelvir/ritonavir according to the recent Emergency Use Authorization (EUA) criteria2 and shortly thereafter experienced rebound symptoms, which were associated with physiologic changes measured using a fitness device. Symptomatic relapses of SARS-CoV-2 infection are just now starting to be reported.3 Although he was not re-tested during this period, it is possible that this coincided with viral rebound upon the completion of therapy. Concerningly, he subsequently experienced worsening post- infectious symptoms which now meet U.S. Centers for Disease Control (CDC) criteria for Long COVID.4 This suggests that although a short course of early antiviral therapy is adequate to prevent severe acute disease in high-risk patients,5 it may be insufficient to prevent the development of Long COVID, and those experiencing rebound symptoms could remain at risk. Case 3 A 43-year-old woman with no significant medical history developed cough and pharyngitis in Spring 2022. She had previously received 3 doses of the Pfizer-BioNTech SARS-CoV-2 vaccine, most recently 4 months prior. While a PCR test was initially negative, she and one of her children subsequently tested positive on an antigen test 5 days later. She did not initially receive antiviral therapy. Over the course of the subsequent 3 weeks, she began to experience worsening fatigue and malaise, with associated myalgia and trouble concentrating (“brain fog”); 3 weeks following initial symptom onset she was spending the majority of the day resting and was unable to easily complete her activities of daily living. She received a prescription for nirmatrelvir/ritonavir, which she began 25 days following initial symptom Page 3/7 Page 3/7 onset. One day following completion of therapy, she experienced improvement in her fatigue symptoms. While she has residual shortness of breath and myalgias, she has now been able to re-engage with usual activities of daily living. onset. One day following completion of therapy, she experienced improvement in her fatigue symptoms. While she has residual shortness of breath and myalgias, she has now been able to re-engage with usual activities of daily living. Discussion A related hypothesis is that SARS-CoV-2 may persist for weeks to months in some individuals, causing inflammation, local tissue damage and end-organ disease.6–8 If this turns out to be the case, antiviral therapy during the acute and post-acute stages may prevent or even reverse Long COVID. Although not approved under the EUA,2 there are reports emerging about individuals accessing oral antiviral therapy at later points in the disease course and the potential effects of these therapies.9 We present two cases in which individuals were able to access nirmatrelvir/ritonavir for clinical care in the setting of persistent COVID-19 symptoms. There were notable differences between these cases, including the timing of antiviral therapy from initial infection (> 60 days and < 30 days, respectively), as well as potential lineage differences based on the timing of infection (Winter 2021–2022 versus Spring 2022). While single anecdotes must be interpreted with caution, these cases emphasize the urgent need for carefully designed studies to assess the impact of antiviral therapy beyond the acute window. Confirmation of a benefit in the context of such studies would support the hypothesis that persistent viral activity, particularly in the tissues, could be one contributor to ongoing symptoms in Long COVID. Page 4/7 An additional factor in one of these cases was a clear re-exposure and potential re-infection event, after which the individual began to experience improvement in Long COVID symptoms which had until that Page 4/7 point been escalating. Although vaccines are likely to reduce the risk of developing Long COVID,10 there are numerous, but inconsistent, reports of the impact of SARS-CoV-2 vaccination on pre-existing Long COVID symptoms.11,12 A dysregulated immune response has been proposed as a potential mechanism underlying Long COVID pathophysiology.1,13,14 The fact that re-exposure to viral antigen may have led to symptomatic improvement is intriguing, suggesting the possibility that a dysregulated immune response, if present, could be recalibrated. Although this case series is limited by a lack of intensive physiologic and laboratory measurements throughout the disease course, we believe these clinical anecdotes are informative as investigators try to understand the pathophysiology that drives the development and persistence of Long COVID. They suggest that antiviral therapy and/or antigen re-exposure could potentially impact the complex interplay between viral replication and the host immune response that likely underlies this syndrome but raise concern that brief early antiviral therapy alone may be insufficient to prevent the development of Long COVID. Conflicts of Interest The authors declare no competing interests. Acknowledgements We are grateful to the participants. References 1. Su Y, Yuan D, Chen DG, et al. Multiple Early Factors Anticipate Post-Acute COVID-19 Sequelae. Cell [Internet] 2022 [cited 2022 Jan 26];0(0). Available from: https://www.cell.com/cell/fulltext/S0092- 8674(22)00072-1?dgcid=raven_jbs_aip_email 2. Paxlovid EUA [Internet]. [cited 2022 May 1];Available from: https://aspr.hhs.gov/COVID- 19/Therapeutics/Products/Paxlovid/Pages/emergency-use-authorization.aspx 2. Paxlovid EUA [Internet]. [cited 2022 May 1];Available from: https://aspr.hhs.gov/COVID- 19/Therapeutics/Products/Paxlovid/Pages/emergency-use-authorization.aspx 3. Gupta K, Strymish J, Stack G, Charness M. Rapid relapse of symptomatic SARS-CoV-2 infection following early suppression with nirmatrelvir/ritonavir [Internet]. Research Square. 2022;Available from: https://www.researchsquare.com/article/rs-1588371/v1 3. Gupta K, Strymish J, Stack G, Charness M. Rapid relapse of symptomatic SARS-CoV-2 infection following early suppression with nirmatrelvir/ritonavir [Internet]. Research Square. 2022;Available from: https://www.researchsquare.com/article/rs-1588371/v1 4. CDC. Post-COVID conditions [Internet]. Centers for Disease Control and Prevention. 2022 [cited 2022 Apr 30];Available from: https://www.cdc.gov/coronavirus/2019-ncov/long-term-effects/index.html 4. CDC. Post-COVID conditions [Internet]. Centers for Disease Control and Prevention. 2022 [cited 2022 Apr 30];Available from: https://www.cdc.gov/coronavirus/2019-ncov/long-term-effects/index.html 5. Hammond J, Leister-Tebbe H, Gardner A, et al. Oral Nirmatrelvir for High-Risk, Nonhospitalized Adults with Covid-19. N Engl J Med 2022;386(15):1397–408. 5. Hammond J, Leister-Tebbe H, Gardner A, et al. Oral Nirmatrelvir for High-Risk, Nonhospitalized Adults with Covid-19. N Engl J Med 2022;386(15):1397–408. 6. Gaebler C, Wang Z, Lorenzi JCC, et al. Evolution of antibody immunity to SARS-CoV-2. Nature [Internet] 2021;Available from: http://dx.doi.org/10.1038/s41586-021-03207-w 6. Gaebler C, Wang Z, Lorenzi JCC, et al. Evolution of antibody immunity to SARS-CoV-2. Nature [Internet] 2021;Available from: http://dx.doi.org/10.1038/s41586-021-03207-w Page 5/7 Page 5/7 7. Chertow D, Stein S, Ramelli S, et al. SARS-CoV-2 infection and persistence throughout the human body and brain [Internet]. Research Square. 2021;Available from: https://www.researchsquare.com/article/rs-1139035/v1 7. Chertow D, Stein S, Ramelli S, et al. SARS-CoV-2 infection and persistence throughout the human body and brain [Internet]. Research Square. 2021;Available from: https://www.researchsquare.com/article/rs-1139035/v1 8. Natarajan A, Zlitni S, Brooks EF, et al. Gastrointestinal symptoms and fecal shedding of SARS-CoV-2 RNA suggest prolonged gastrointestinal infection. Med [Internet] 2022 [cited 2022 Apr 13];0(0). Available from: https://www.cell.com/med/fulltext/S2666-6340(22)00167-2 8. Natarajan A, Zlitni S, Brooks EF, et al. Gastrointestinal symptoms and fecal shedding of SARS-CoV-2 RNA suggest prolonged gastrointestinal infection. Med [Internet] 2022 [cited 2022 Apr 13];0(0). Available from: https://www.cell.com/med/fulltext/S2666-6340(22)00167-2 9. Geng LN, Bonilla HF, Shafer RW, Miglis MG, Yang PC. Case Report of Breakthrough Long COVID and the Use of Nirmatrelvir-Ritonavir. 2022;Available from: https://www.researchsquare.com/article/rs- 1443341/v1 10. Antonelli M, Penfold RS, Merino J, et al. Risk factors and disease profile of post-vaccination SARS- CoV-2 infection in UK users of the COVID Symptom Study app: a prospective, community-based, nested, case-control study. References Lancet Infect Dis 2022;22(1):43–55. 10. Antonelli M, Penfold RS, Merino J, et al. Risk factors and disease profile of post-vaccination SARS- CoV-2 infection in UK users of the COVID Symptom Study app: a prospective, community-based, nested, case-control study. Lancet Infect Dis 2022;22(1):43–55. 11. Groff D, Sun A, Ssentongo AE, et al. Short-term and Long-term Rates of Postacute Sequelae of SARS- CoV-2 Infection: A Systematic Review. JAMA Netw Open 2021;4(10):e2128568. 11. Groff D, Sun A, Ssentongo AE, et al. Short-term and Long-term Rates of Postacute Sequelae of SARS- CoV-2 Infection: A Systematic Review. JAMA Netw Open 2021;4(10):e2128568. 12. Tran V-T, Perrodeau E, Saldanha J, Pane I, Ravaud P. Efficacy of COVID-19 vaccination on the symptoms of patients with long COVID: a target trial emulation using data from the ComPaRe e- cohort in France [Internet]. Research Square. 2022;Available from: https://www.researchsquare.com/article/rs-1350429/v1 12. Tran V-T, Perrodeau E, Saldanha J, Pane I, Ravaud P. Efficacy of COVID-19 vaccination on the symptoms of patients with long COVID: a target trial emulation using data from the ComPaRe e- cohort in France [Internet]. Research Square. 2022;Available from: https://www.researchsquare.com/article/rs-1350429/v1 13. Peluso MJ, Lu S, Tang AF, et al. Markers of Immune Activation and Inflammation in Individuals With Postacute Sequelae of Severe Acute Respiratory Syndrome Coronavirus 2 Infection. J Infect Dis [Internet] 2021;Available from: http://dx.doi.org/10.1093/infdis/jiab490 13. Peluso MJ, Lu S, Tang AF, et al. Markers of Immune Activation and Inflammation in Individuals With Postacute Sequelae of Severe Acute Respiratory Syndrome Coronavirus 2 Infection. J Infect Dis [Internet] 2021;Available from: http://dx.doi.org/10.1093/infdis/jiab490 14. Phetsouphanh C, Darley DR, Wilson DB, et al. Immunological dysfunction persists for 8 months following initial mild-to-moderate SARS-CoV-2 infection. Nat Immunol 2022;1–7. 14. Phetsouphanh C, Darley DR, Wilson DB, et al. Immunological dysfunction persists for 8 months following initial mild-to-moderate SARS-CoV-2 infection. Nat Immunol 2022;1–7. Figures Page 6/7 Figure 1 Physiologic measurements as recorded by a personal fitness device used by Patient 1, showing rebound Figure 1 Physiologic measurements as recorded by a personal fitness device used by Patient 1, showing rebound tachycardia, tachypnea, and elevated body temperature coinciding with completion of antiviral therapy. Physiologic measurements as recorded by a personal fitness device used by Patient 1, showing rebound tachycardia, tachypnea, and elevated body temperature coinciding with completion of antiviral therapy. Page 7/7
https://openalex.org/W2419445518	https://www.dora.lib4ri.ch/psi/islandora/object/psi%3A4214/datastream/PDF/Domnanich-2016-44Sc_for_labeling_of_DOTA--%28published_version%29.pdf	English	null	44Sc for labeling of DOTA- and NODAGA-functionalized peptides: preclinical in vitro and in vivo investigations	EJNMMI radiopharmacy and chemistry	2,016	cc-by	10,422	Abstract Background: Recently, 44Sc (T1/2 = 3.97 h, Eβ+ av = 632 keV, I = 94.3 %) has emerged as an attractive radiometal candidate for PET imaging using DOTA-functionalized biomolecules. The aim of this study was to investigate the potential of using NODAGA for the coordination of 44Sc. Two pairs of DOTA/NODAGA-derivatized peptides were investigated in vitro and in vivo and the results obtained with 44Sc compared with its 68Ga-labeled counterparts. DOTA-RGD and NODAGA-RGD, as well as DOTA-NOC and NODAGA-NOC, were labeled with 44Sc and 68Ga, respectively. The radiopeptides were investigated with regard to their stability in buffer solution and under metal challenge conditions using Fe3+ and Cu2+. Time-dependent biodistribution studies and PET/CT imaging were performed in U87MG and AR42J tumor-bearing mice. Background: Recently, 44Sc (T1/2 = 3.97 h, Eβ+ av = 632 keV, I = 94.3 %) has emerged as an attractive radiometal candidate for PET imaging using DOTA-functionalized biomolecules. The aim of this study was to investigate the potential of using NODAGA for the coordination of 44Sc. Two pairs of DOTA/NODAGA-derivatized peptides were investigated in vitro and in vivo and the results obtained with 44Sc compared with its 68Ga-labeled counterparts. DOTA-RGD and NODAGA-RGD, as well as DOTA-NOC and NODAGA-NOC, were labeled with 44Sc and 68Ga, respectively. The radiopeptides were investigated with regard to their stability in buffer solution and under metal challenge conditions using Fe3+ and Cu2+. Time-dependent biodistribution studies and PET/CT imaging were performed in U87MG and AR42J tumor-bearing mice. Results: Both RGD- and NOC-based peptides with a DOTA chelator were readily labeled with 44Sc and 68Ga, respectively, and remained stable over at least 4 half-lives of the corresponding radionuclide. In contrast, the labeling of NODAGA-functionalized peptides with 44Sc was more challenging and the resulting radiopeptides were clearly less stable than the DOTA-derivatized matches. 44Sc-NODAGA peptides were clearly more susceptible to metal challenge than 44Sc-DOTA peptides under the same conditions. Instability of 68Ga-labeled peptides was only observed if they were coordinated with a DOTA in the presence of excess Cu2+. Biodistribution data of the 44Sc-labeled peptides were largely comparable with the data obtained with the 68Ga-labeled counterparts. It was only in the liver tissue that the uptake of 68Ga-labeled DOTA compounds was markedly higher than for the 44Sc-labeled version and this was also visible on PET/CT images. RESEARCH Open Access 44Sc for labeling of DOTA- and NODAGA- functionalized peptides: preclinical in vitro and in vivo investigations Katharina A. Domnanich1,2, Cristina Müller3,4, Renata Farkas3, Raffaella M. Schmid3, Bernard Ponsard5, Roger Schibli3,4, Andreas Türler1,2 and Nicholas P. van der Meulen1,3* * Correspondence: nick.vandermeulen@psi.ch 1Laboratory of Radiochemistry, Paul Scherrer Institute, CH-5232 Villigen-PSI, Switzerland 3Center for Radiopharmaceutical Sciences ETH-PSI-USZ, Paul Scherrer Institute, 5232 Villigen-PSI, Switzerland Full list of author information is available at the end of the article * Correspondence: nick.vandermeulen@psi.ch 1Laboratory of Radiochemistry, Paul Scherrer Institute, CH-5232 Villigen-PSI, Switzerland 3Center for Radiopharmaceutical Sciences ETH-PSI-USZ, Paul Scherrer Institute, 5232 Villigen-PSI, Switzerland Full list of author information is available at the end of the article Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 DOI 10.1186/s41181-016-0013-5 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 DOI 10.1186/s41181-016-0013-5 EJNMMI Radiopharmacy and Chemistry © 2016 Domnanich 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. Background 44Sc is a novel radiometal which is attractive for positron emission tomography (PET) imaging, due to the emission of positrons with a high branching ratio (Eβ+ av = 632 keV, I = 94.3 %) (Rösch 2012; Müller et al. 2013). Its physical half-life of 3.97 h enables the acquisition of PET images several hours after injection of the 44Sc-radiopharmaceuti- cal and is of particular interest for application with biomolecules, providing slower kinetic profiles (Chakravarty et al. 2014). Importantly, application of 44Sc would allow a centralized production of radiopharmaceuticals, followed by transportation to more remote hospitals (van der Meulen et al. 2015). 44Sc is thought to be useful as a diagnostic match to therapeutic radionuclides with similar chemical properties, such as 90Y and 177Lu (Müller et al. 2013). Most interesting, however, would be to use 44Sc in combination with its therapeutic counterpart 47Sc, which provides excellent β−-decay properties for radionuclide therapy (Eβ− av = 162 keV, T1/2 = 3.35 d). The potential of 44Sc/47Sc as a theragnostic pair has been demonstrated recently in a preclinical pilot study with tumor-bearing mice (Müller et al. 2014). A crucial requirement for the application of radiopharmaceuticals is the formation of a thermodynamically stable and kinetically inert complex of the radiometal with a suitable chelator, which is then linked to the targeting agent (Majkowska-Pilip & Bilewicz 2011). The coordination of Sc(III) and Ga(III) has previously been investigated with several macrocyclic polyaminocarboxylic chelators, including 1,4,7,10-tetraazacyclo- dodecane-1,4,7,10-tetraacetic acid (DOTA) and 1,4,7-triazacyclononane-1,4,7-triacetic acid (NOTA) (Majkowska-Pilip & Bilewicz 2011; Huclier-Markai et al. 2011; Notni et al. 2012). The studies were performed with natSc(III) stock solutions containing 46Sc as tracer and the sole chelator without an attached biomolecule. As a result of these investigations, it was found that Sc(III) forms complexes with both chelators, DOTA and NOTA, how- ever, the stability of Sc-DOTA was superior to Sc-NOTA complexes (Majkowska-Pilip & Bilewicz 2011; Huclier-Markai et al. 2011). On the other hand, Ga(III) displays a reverse behavior, forming more stable complexes with NOTA than with DOTA (Majkowska-Pilip & Bilewicz 2011; Notni et al. 2012). The DOTA-chelator provides eight coordination sites, which are all coordinated by Sc(III), whereas NOTA can only form six coordinative bonds. Due to the higher denticity the Sc-DOTA complex is believed to be thermodynamically more stable than the Sc-NOTA complex (Huclier-Markai et al. 2011; Port et al. 2008). Abstract The 44Sc-labeled NODAGA-peptides showed a similar tissue distribution to those of the DOTA peptides without any obvious signs of in vivo instability. Conclusions: Although DOTA revealed to be the preferred chelator for stable coordination of 44Sc, the data presented in this work indicate the possibility of using NODAGA in combination with 44Sc. In view of a clinical study, thorough investigations will be necessary regarding the labeling conditions and storage solutions in order to guarantee sufficient stability of 44Sc-labeled NODAGA compounds. Keywords: 44Sc, PET, Imaging, Stability, DOTA-RGD, NODAGA-RGD, DOTA-NOC, NODAGA-NOC, 68Ga, AR42J, U87MG Keywords: 44Sc, PET, Imaging, Stability, DOTA-RGD, NODAGA-RGD, DOTA-NOC, NODAGA-NOC, 68Ga, AR42J, U87MG © 2016 Domnanich 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. Page 2 of 19 Page 2 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Background Due to the preference of Ga(III) for the coordination number six, all coordination sites of NOTA are used, while two sites of DOTA remain uncoordinated in a Ga-DOTA-complex (Majkowska-Pilip & Bilewicz 2011; Viola-Villegas & Doyle 2009). Despite the reduced stability of a Ga-DOTA complex compared to a Ga-NOTA and Ga-1,4,7-triazacyclononane,1-glutaric acid-4,7-acetic acid (NODAGA) complex, there are several examples of 68Ga-DOTA labeled peptides which showed promising in vivo proper- ties. The 68Ga-labeled somatostatin receptor analogues 68Ga-DOTA-TOC, 68Ga-DOTA- TATE and 68Ga-DOTA-NOC represent the most prominently applied radiopharmaceuti- cals in clinical studies for imaging of neuroendocrine tumors (Kwekkeboom et al. 2010). Recent PET imaging studies in patients using the αvβ3 integrin-targeting radiotracer 68Ga- NOTA-RGD indicated promising results (Kim et al. 2012; Choi et al. 2013; Yoon et al. 2014). These reports demonstrate that 68Ga is successfully used in clinics with both DOTA- and NOTA/NODAGA-derivatized targeting agents. Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 3 of 19 Page 3 of 19 The stable complexation of 44Sc with DOTA initiated a number of preclinical stud- ies with a range of DOTA-derivatized biomolecules, including bombesin analogues (Koumarianou et al. 2012), puromycin (Eigner et al. 2013), folate-conjugates (Müller et al. 2013) and dimeric cyclic RGD peptides (Hernandez et al. 2014). To date, all re- search with regard to the in vivo and in vitro behavior of 44Sc-labeled radiopharma- ceuticals was performed with DOTA chelators connected to the respective targeting agent. The only exception to our knowledge was a study in which an EGFR-targeted antibody was labeled with 44Sc using a CHX-A-DTPA (N-[(R)-2-amino-3-(para-isothio- cyanato-phenyl)propyl]trans-(S,S)-cyclohexane-1,2-diamine N,N,N’,N”N”-pentaacetic acid) chelator (Chakravarty et al. 2014). The question on whether or not a NOTA or NODAGA chelator would be suited for 44Sc-labeling has remained unclear thus far. Using these che- lators for coordination of 44Sc would be of interest in combination with radiopharmaceu- ticals where only the NOTA- or NODAGA-derivatized species are available for clinical studies, as is the case for NODAGA-RGD: a peptide which is currently employed in clin- ical trials when labeled with 68Ga. Since application of NODAGA-RGD at later time points after injection of the radioconjugate would be of interest for nuclear physicians, we set out to investigate whether 44Sc-labeling with NODAGA-derivatized biomolecules is possible and whether the in vitro and in vivo behavior of the radiolabeled peptides would be equal to the DOTA-derivatized matches. Methods 44CaCO3, 97.0 % enriched (Trace Sciences International, USA) and graphite powder, 99.9999 % (Alfa Aesar, Germany) were used for target preparation. The N,N,N’,N’-tetra- n-octyldiglycolamide, non-branched resin (DGA, particle size 50–100 μm, TrisKem International, France) was used for the separation of Sc(III) from Ca(II). The chemical separations were performed using MilliQ water (resistance >18 MΩ) and hydrochloric acid (HCl, 30 % Suprapur, Merck KGaA, Germany). The recycling of the target material was performed with oxalic acid dihydrate, (Trace SELECT, ≥99.9999 % metals basis, Fluka Analytical, Germany) and 25 % ammonia solution (Suprapur, Merck KGaA, Germany). 68Ga was obtained from a 68Ge/68Ga generator IGG100 (Eckert & Ziegler, Berlin, Germany). The generator was eluted in fractions and the fraction of eluate containing the highest quantity of 68Ga (approximately 200–250 MBq in 700 μL 0.1 M HCl) was directly used for radiolabeling purposes without purification. DOTA-RGD (DOTA-cyclo(RGDfK) acetate, Cat-N° 9863), NODAGA-RGD (NODAGA-RGD tri- fluoroacetate, Cat-N° 9805), DOTA-NOC (DOTA-NOC acetate, Cat-N° 9712) and NODAGA-NOC (NODAGA-NOC acetate, Cat-N° 9718) were obtained from ABX GmbH, advanced biochemical compounds, Germany. Copper chloride dihydrate was purchased from Merck Millipore, while iron chloride hexahydrate was obtained from Sigma-Aldrich GmbH. Phosphate buffered saline (PBS) pH 7.4 was prepared in-house (Additional file 1). Background The aim of this study was, therefore, to compare the in vitro and in vivo behavior of two pairs of peptides with a DOTA- and NODAGA-chelator (Fig. 1), respectively, after radiolabeling with 44Sc and 68Ga. Cyclic RGD peptides based on the Arg-Gly-Asp se- quence and NOC, a somatostatin analogue ([Tyr3,1-NaI3]octreotide), were chosen as targeting agents. The in vitro stability of 44Sc- and 68Ga-labeled DOTA/NODAGA- RGD and DOTA/NODAGA-NOC was examined in saline in the presence and absence of competing metal cations. The in vivo behavior was evaluated by the performance of biodistribution studies and in vivo PET imaging of tumor-bearing mice. Fig. 1 Chemical structures of DOTA-RGD (a), NODAGA-RGD (b), DOTA-NOC (c) and NODAGA-NOC (d) Fig. 1 Chemical structures of DOTA-RGD (a), NODAGA-RGD (b), DOTA-NOC (c) and NODAGA-NOC (d) Page 4 of 19 Page 4 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Production of 44Sc 44Sc was prepared by proton irradiation of enriched 44Ca targets at the Injector 2 cyclo- tron at PSI, as previously reported (van der Meulen et al. 2015). The irradiation of targets with 11 MeV proton beam energy, and a beam current of 50 μA, lasted for 90 min. A col- umn (1 mL cartridge fitted with 20 μm frit ISOLUTE SPE Accessories, UK) was filled with 50 – 70 mg DGA extraction chromatographic resin and a second column with 20– 25 mg of the same resin. A 20 μm frit was placed on top of the resin in each column. DGA columns were preconditioned with 3.0 M HCl. The first step of the separation was performed as previously reported (van der Meulen et al. 2015). In brief, the target was dis- solved in 2.5 mL 3.0 M HCl and loaded onto the first DGA column. The rinsing of the target container and the first DGA column with 3.0 M HCl ensured a complete transfer of the 44Sc radioactivity and complete removal of residual Ca(II), respectively. 44Sc was eluted from the first DGA column with 3.5 mL 0.1 M HCl. Subsequently, the solution was acidified with the addition of 3.3 mL 6.0 M HCl to yield a 3.0 M HCl solution, which was then passed through the second DGA column, to which the 44Sc activity was sorbed. The elution of 44Sc from the second column was performed with 700 μL 0.05 M HCl (pH 1.3) and was used directly for labeling reactions. The radionuclidic purity of the 44Sc- eluate was quantified by γ-spectrometry using an N-type high-purity germanium (HPGe) coaxial detector (EURISYS MESURES, France) and the Ortec InterWinner 7.1 software. The 44Ca-contained waste fraction from the first DGA column was collected and processed to recover the target material, as described previously (van der Meulen et al. 2015). Page 5 of 19 Page 5 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Radiolabeling of DOTA- and NODAGA-functionalized peptides The total 44Sc and 68Ga activity in the obtained eluate was quantitatively determined with a dose calibrator (ISOMED 2010, Nuclear–Medizintechnik Dresden GmbH, Germany) so that the activity required for radiolabeling could be withdrawn. Production of 44Sc Sodium acetate solution (0.5 M, pH 8) was added at a ratio of 1:1 to the 44Sc eluate (0.05 M HCl, pH 0.4–0.6) and at a ratio of 1:2 to the 68Ga generator eluate (0.1 M HCl, pH 1) to give a pH of 4–4.5. The corresponding peptides (DOTA-RGD, NODAGA-RGD, DOTA-NOC and NODAGA-NOC, in a 1 mM stock solution) were added to obtain a specific activity of up to 10 MBq/nmol and the reaction mixture incubated at 95 °C for 10 min. High-performance liquid chromatography (HPLC) with a C-18 reversed–phase column (XterraTM MS, C18, 5 μm, 150 × 4.6 mm; Waters) was used for quality control. The mobile phase consisted of MilliQ water containing 0.1 % trifluoracetic acid (A) and acetonitrile (B) with a gradient of 95 % A and 5 % B to 20 % A and 80 % B over a period of 15 min at a flow rate of 1.0 mL/min. In vitro stability of 44Sc- and 68Ga-labeled peptides 44Sc- and 68Ga-labeled peptides (radiochemical purity >95 %) were used for the investi- gation of stability in 0.9 % NaCl. An activity of 80–100 MBq of 44Sc or 68Ga labeling solution was diluted with 0.9 % NaCl solution to a volume of 800 μL and incubated for four half-lives of the corresponding radionuclide at 37 °C. Aliquots were taken from the prepared solutions at different time points over at least four half-lives, respectively, of 44Sc and 68Ga and analyzed by TLC (TLC was used, as the high metal concentrations used in this study may impair the long-term performance of HPLC columns). If initial experiments suggested instability of a com- pound, aliquots were retrieved more frequently. The TLC plates (Silica-gel 60, Merck) were developed using 0.1 M sodium citrate (pH 4.7) as mobile phase. The quantitative distribution of radioactivity was determined with an autoradiography system (Cyclone Plus, Perkin Elmer) and its associated software (Optiquant, version 5.0). Rf values of 0.2 were observed for 44Sc- and 68Ga-labeled DOTA-RGD and NODAGA-RGD, and of 0.1 for DOTA-NOC and NODAGA-NOC, respectively, whereas for unlabeled 44Sc and 68Ga a Rf value of 0.9 was calculated. The influence of high metal cation concentration on the stability of 44Sc and 68Ga labeled peptides was monitored in solutions containing 0.01 M Fe3+ or Cu2+. An aqueous metal cation solution (16 μL of a 0.5 M Fe3+/Cu2+ solution) was added to 80–100 MBq of 44Sc or 68Ga labeling solution and diluted with 0.9 % NaCl solution to a volume of 800 μL. The incubation conditions, time points of retrieving samples and analysis by TLC were kept the same, as described above. Biodistribution studies Biodistribution studies were performed with U87MG tumor and AR42J tumor-bearing mice 2 weeks after tumor cell inoculation. 44Sc- and 68Ga-labeled peptides (~5 MBq, ~1 nmol per mouse) were intravenously injected in a volume of 100–200 μL. Mice were sacrificed at 30 min, 2 h and 5 h after injection (p.i.) of the 44Sc-labeled peptides. Mice which were injected with 68Ga-labeled peptides were sacrificed at 30 min and 2 h p.i. Selected tissues and organs were collected, weighed, and counted for radioactivity using a γ-counter (Wallac Wizard 1480, Perkin Elmer). The results were listed as a percent- age of the injected activity per gram of tissue mass (% IA/g), using counts of a defined volume of the original injection solution counted at the same time. The data were analyzed for significance using a two-way ANOVA test (Graph Pad Prism 6 software, version 6.05). A p-value of < 0.05 was considered statistically significant. Tumor mouse models In vivo experiments were approved by the local veterinarian department and conducted in accordance with the Swiss law of animal protection. Female athymic nude mice (CD-1 nude), age 5–6 weeks, were obtained from Charles River Laboratories, Sulzfeld, Germany. U87MG cells and AR42J-cells were suspended in PBS (5 × 106 cells in 100 μL) and subcutaneously inoculated on each shoulder. Two to three weeks later, when the tumor reached a size of about 300–500 mm3, the mice were used for the in vivo studies. Preclinical PET imaging A bench-top preclinical PET scanner (G8, Sofie Biosciences, California, U.S.A. and Perkin Elmer, Massachusetts, U.S.A.) was employed for the PET scans of the tumor-bearing mice. The energy window was set to 150–650 keV. Mice were injected intravenously with 44Sc- or 68Ga-labeled peptides (~10 MBq, ~1 nmol per mouse) in a volume of 100– 200 μL. The PET scans were performed 3 h and 5 h after injection of the 44Sc-labeled pep- tides and 3 h after injection of the 68Ga-labeled peptides, using G8 acquisition software (version 2.0.0.10). All static PET scans lasted for 20 min. During the acquisition the mice were anesthetized by inhalation of a mixture of isoflurane and oxygen. The images were reconstructed with maximum-likelihood expectation maximization (MLEM). Gauss post- reconstruction filtering was performed using VivoQuant post-processing software (version 2.10, inviCRO Imaging Services and Software, Boston, U.S.A.). Cell culture U87MG cells (human glioblastoma cells; ACC® HTB-14TM) and AR42J cells (rat exocrine pancreatic tumor cells; ACC® CRL1492TM) were purchased from European Collection of Cell Cultures (ECACC, operated by Public Health England). U87MG cells were grown in MEM cell culture medium supplemented with 1 % non-essential amino acids (MEM NEAA solution 100×, Bioconcept), 1 mM sodium pyruvate (Bioconcept), 10 % fetal calf serum, L-glutamine and antibiotics. AR42J cells were grown in RPMI cell culture medium Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 6 of 19 supplemented with 10 % fetal calf serum, L-glutamine and antibiotics. Routine cell culture was performed twice a week using trypsin (Gibco by life technologies 0.25 % trypsin- EDTA) for detachment of the cells. Production and separation of 44Sc 44Sc was quantitatively sorbed on DGA resin in 3.0 M HCl solution, whereas Ca was not retained. Rinsing the resin with additional 4 mL 3.0 M HCl ensured the complete removal of Ca, after which 44Sc was eluted with 3.5 mL 0.1 M HCl. The 44Sc eluate was further acidified with the addition of 6.0 M HCl to yield a 3.0 M HCl solution. The Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 7 of 19 Page 7 of 19 Page 7 of 19 resultant solution was passed through a second, smaller DGA column at a flow rate of ~0.3 mL/min, retaining 97 % of the eluted 44Sc activity. 44Sc was eluted quantitatively (85 ± 2 %), at activities of ~2.0 GBq, with 700 μL 0.05 M HCl and was used directly for labeling experiments. The separation procedure was initially developed using trace activities of 46Sc, produced by the 45Sc(n,γ) nuclear reaction at the BR2 reactor at SCK.CEN, Mol, Belgium. The 44Sc activity separated from proton irradiated targets containing recycled 44CaCO3 was of the same quality and quantity as using the originally-purchased 44CaCO3 target material. Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides Radiolabeling with 44Sc was readily achieved with DOTA-compounds, but found to be more challenging for NODAGA-compounds, which did not allow reproducible labeling procedures at high specific activities. Radiolabeling with 68Ga was reproducibly achieved for both DOTA- and NODAGA-functionalized peptides, however. The radio- chemical yield of the 44Sc and 68Ga radiosyntheses at the specific activity of 10 MBq/ nmol was >95 %. Quantitative 68Ga-labeling of NODAGA-RGD and NODAGA-NOC was also possible at room temperature in less than 10 min. TLC and HPLC quality con- trol were in good agreement. HPLC analysis demonstrated a peak of free 44Sc and 68Ga at a retention time of 2.2 ± 0.1 min, while the retention times of the radiopeptides were between 6 and 10 min (Table 1). The stability of 44Sc- and 68Ga-labeled DOTA- and NODAGA-peptides was first in- vestigated in 0.9 % NaCl over a period of four half-lives of the corresponding nuclide by means of TLC. 44Sc- and 68Ga-DOTA-RGD and DOTA-NOC exhibited a high sta- bility. After four half-lives at 37 °C the amount of intact compound did not decrease below 98 %. 68Ga-labeled NODAGA-RGD and NODAGA-NOC remained stable, but the 44Sc-NODAGA peptides became more unstable over time. The amount of intact 44Sc-NODAGA-RGD dropped to 77 % and, in the case of 44Sc-NODAGA-NOC, a mere 37 % after more than 4 half-lives. The presence of different metal cations can cause displacement of the radio- nuclide from the chelator and its release into solution (Pruszynski et al, 2012). The stability of 44Sc- and 68Ga-labeled DOTA- and NODAGA-peptides was investigated in the presence of Fe3+ and Cu2+ over four half-lives of the corresponding radio- nuclide (Fig. 2). The addition of solutions containing these metal cations to result in final metal concentrations as high as 10 mM did not induce any transmetalation of 44Sc-labeled DOTA- and 68Ga-labeled NODAGA-compounds. Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides The integrity of 68Ga-DOTA-peptides was not impaired by the presence of Fe3+, however, the addition of 10 mM Cu2+ reduced the amount of intact 68Ga-DOTA-RGD to 10 % Table 1 HPLC retention times of radiolabeled peptides Radiopeptide 44Sc-/68Ga-DOTA-RGD 44Sc-/68Ga-NODAGA-RGD Retention time 6.8 ± 0.1 min 6.1 ± 0.1 min Radiopeptide 44Sc-/68Ga-DOTA-NOC 44Sc-/68Ga-NODAGA-NOC Retention time 9.3 ± 0.1 min 9.8 ± 0.1 min Table 1 HPLC retention times of radiolabeled peptides Radiopeptide 44Sc-/68Ga-DOTA-RGD 44Sc-/68Ga-NODAGA-RGD Retention time 6.8 ± 0.1 min 6.1 ± 0.1 min Radiopeptide 44Sc-/68Ga-DOTA-NOC 44Sc-/68Ga-NODAGA-NOC Retention time 9.3 ± 0.1 min 9.8 ± 0.1 min Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 8 of 19 Page 8 of 19 and 68Ga-DOTA-NOC to 50 %, respectively, after two half-lives. The presence of both metal cations further destabilized the already less stable 44Sc-labeled NODAGA-peptides. Fig. 2 Stability of 44Sc-DOTA-RGD (a), 44Sc-NODAGA-RGD (b), 68Ga-DOTA-RGD (c), 68Ga-NODAGA-RGD (d), 44Sc-DOTA-NOC (e), 44Sc-NODAGA-NOC (f), 68Ga-DOTA-NOC (g), 68Ga-NODAGA-NOC (h) in saline with and without the presence of 10 mM Cu2+ and Fe3+, respectively 2 Stability of 44Sc-DOTA-RGD (a), 44Sc-NODAGA -DOTA-NOC (e), 44Sc-NODAGA-NOC (f), 68Ga-DO 2+ 3+ and 68Ga-DOTA-NOC to 50 %, respectively, after two half-lives. The presence of both metal cations further destabilized the already less stable 44Sc-labeled NODAGA-peptides. Page 9 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 9 of 19 Biodistribution studies with 44Sc- and 68Ga-labeled peptides Biodistribution studies were performed with 44Sc- and 68Ga-labeled DOTA/NODAGA- RGD and DOTA/NODAGA-NOC in mice bearing U87MG and AR42J tumor xenografts, respectively (Additional file 1: Table S2-S5). Biodistribution studies with 44Sc- and 68Ga-labeled peptides Biodistribution studies were performed with 44Sc- and 68Ga-labeled DOTA/NODAGA- RGD and DOTA/NODAGA-NOC in mice bearing U87MG and AR42J tumor xenografts, respectively (Additional file 1: Table S2-S5). Time-dependent distribution studies of 44Sc-DOTA-RGD and 44Sc-NODAGA-RGD revealed a similar pattern for both compounds, resulting in a tumor uptake of 4.88 ± 0.67 % IA/g and 4.50 ± 0.77 % IA/g, respectively, at 0.5 h after injection (Fig. 3a). The wash-out of radioactivity from the tumor tissue was somewhat faster for the 44Sc- DOTA-RGD than for the 44Sc-NODAGA-RGD, but at 5 h after injection the values were almost the same (3.00 ± 0.38 % IA/g vs 3.01 ± 0.55 % IA/g). Clearance from the blood was fast for both 44Sc-DOTA-RGD and 44Sc-NODAGA-RGD. Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides This was also reflected by the high renal uptake shortly after injection (4.44 ± 0.51 % IA/g vs 3.89 ± 0.68 % IA/g, 0.5 h p.i.) which decreased with time, resulting in a retention of <2 % IA/g at 5 h p.i. Whereas the accumulation in non-targeted organs and tissues was generally comparable, the liver uptake of 44Sc-DOTA-RGD (5.16 ± 0.99 % IA/g, 0.5 h) was clearly higher than for 44Sc-NODAGA-RGD (1.49 ± 0.06 % IA/g, 0.5 h) at early time points. At 5 h after injection uptake in the liver was ~1 % IA/g for both radiopeptides (Fig. 3a). The tissue distribution pattern of 44Sc-DOTA-NOC and 44Sc-NODAGA-NOC was comparable with regard to the uptake in AR42J tumors (9.49 ± 0.76 % IA/g vs. 9.90 ± 0.66 % IA/g), kidneys (12.6 ± 3.36 % IA/g vs. 12.4 ± 1.41 % IA/g) and liver (1.49 ± 0.49 % IA/g vs. 1.22 ± 0.11 % IA/g) at 0.5 h after injection (Fig. 3b). The tissue distribution profiles at 2 h after injection of the NOC-based radiopeptides were also comparable. At 5 h after injection, retention of 44Sc-DOTA-NOC in tumors (5.56 ± 0.40 % IA/g) was lower than for 44Sc-NODAGA-NOC (10.8 ± 0.37 % IA/g), which indicates a faster wash-out of 44Sc-DOTA-NOC. Renal retention of 44Sc-DOTA-NOC decreased further over the period of investigation (6.54 ± 0.30 % IA/g, 5 h p.i.) while the accumulation of activity in the kidneys (13.0 ± 2.98 % IA/g) was increased at 5 h after injection of 44Sc-NODAGA-NOC (Fig. 3b). For selected organs and tissues, the uptake of the 44Sc-labeled peptides was compared with the uptake of the 68Ga-labeled peptides in mice at 2 h after injection (Fig. 4). The tis- sue distribution of DOTA-RGD was almost the same, independent of the radionuclide (44Sc vs 68Ga) used (Fig. 4a), although the tumor uptake was higher for 44Sc-DOTA-RGD (2.99 ± 0.16 % IA/g) than for 68Ga-DOTA-RGD (2.35 ± 0.27 % IA/g, p <0.05). The NODAGA-derivatized RGD-peptides revealed the same trend: the tumor uptake of 44Sc-NODAGA-RGD (4.05 ± 0.89 IA/g) was significantly higher (p <0.05) than the tumor uptake of 68Ga-NODAGA-RGD (3.13 ± 0.27 % IA/g). Undesired accumulation of 44Sc-NODAGA-RGD in the liver (1.34 ± 0.27 % IA/g, p <0.05) was significantly reduced compared to 68Ga-NODAGA-RGD (2.09 ± 0.08 % IA/g, Fig. 4b). Renal uptake of 44Sc- NODAGA-RGD was also slightly lower than for 68Ga-NODAGA-RGD. Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides The DOTA-RGD accumulated to a higher extent in the liver than the NODAGA-RGD, irrespective of which radionuclide was used for labeling (Fig. 4a/b). In all other organs of interest, the distribution profile was roughly the same, irrespective of the chelator used for coordin- ation of the radionuclide. Accumulation of 44Sc-DOTA-NOC in the tumor xenografts (8.83 ± 0.57 % IA/g) was significantly lower (p <0.05) when compared to the uptake of 68Ga-DOTA-NOC (12.2 ± 2.29 % IA/g) (Fig. 4c). This was also the case in the liver, where the uptake of 44Sc-DOTA- Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 10 of 19 Fig. 3 Biodistribution data obtained at different time points after injection of ~5 MBq (1 nmol) 44Sc-DOTA/NODAGA- RGD in U87MG tumor-bearing mice (a) and after injection of ~5 MBq (1 nmol) 44Sc-DOTA/NODAGA-NOC in AR42J tumor-bearing mice (b), respectively. Data bars represent the average ± SD of values obtained from n = 3 mice NOC (0.68 ± 0.07 % IA/g) was significantly (p <0.05) lower than for 68Ga-DOTA-NOC (5.52 ± 0.88 % IA/g). In all other organs and tissues the distribution of radioactivity was comparable among the radiopeptides, irrespective of whether they were labeled with 44Sc or 68Ga (Fig. 4c). The tissue distribution of 44Sc-NODAGA-NOC and 68Ga-NODAGA- NOC was also comparable (Fig. 4d). The only significant difference (p <0.05) were the kidneys, in which 44Sc-NODAGA-NOC was less retained (8.64 ± 1.62 % IA/g) than the 68Ga-NODAGA-NOC (11.5 ± 1.15 % IA/g). The liver uptake of 44Sc-NODAGA-NOC (0.88 ± 0.15 % IA/g) was lower than for 68Ga-NODAGA-NOC (1.68 ± 0.19 % IA/g) but the difference was not significant (Fig. 4c). Comparison of the DOTA-NOC and NODAGA-NOC revealed similar distribution profiles, irrespective of the radionuclide employed. As the only exception, it should be mentioned that 68Ga-DOTA-NOC showed a clearly higher retention in the liver than all other NOC-based radiopeptides (Fig. 4c/d). As a result of the tissue distribution data reported above, the tumor-to-background ratios were mostly similar between 44Sc-labeled DOTA-RGD and NODAGA-RGD as Fig. 3 Biodistribution data obtained at different time points after injection of ~5 MBq (1 nmol) 44Sc-DOTA/NODAGA- RGD in U87MG tumor-bearing mice (a) and after injection of ~5 MBq (1 nmol) 44Sc-DOTA/NODAGA-NOC in AR42J tumor-bearing mice (b), respectively. Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides Data bars represent the average ± SD of values obtained from n = 3 mice NOC (0.68 ± 0.07 % IA/g) was significantly (p <0.05) lower than for 68Ga-DOTA-NOC (5.52 ± 0.88 % IA/g). In all other organs and tissues the distribution of radioactivity was comparable among the radiopeptides, irrespective of whether they were labeled with 44Sc or 68Ga (Fig. 4c). The tissue distribution of 44Sc-NODAGA-NOC and 68Ga-NODAGA- NOC was also comparable (Fig. 4d). The only significant difference (p <0.05) were the kidneys, in which 44Sc-NODAGA-NOC was less retained (8.64 ± 1.62 % IA/g) than the 68Ga-NODAGA-NOC (11.5 ± 1.15 % IA/g). The liver uptake of 44Sc-NODAGA-NOC (0.88 ± 0.15 % IA/g) was lower than for 68Ga-NODAGA-NOC (1.68 ± 0.19 % IA/g) but the difference was not significant (Fig. 4c). Comparison of the DOTA-NOC and NODAGA-NOC revealed similar distribution profiles, irrespective of the radionuclide employed. As the only exception, it should be mentioned that 68Ga-DOTA-NOC showed a clearly higher retention in the liver than all other NOC-based radiopeptides (Fig. 4c/d). A lt f th ti di t ib ti d t t d b th t t b kg d NOC (0.68 ± 0.07 % IA/g) was significantly (p <0.05) lower than for 68Ga-DOTA-NOC (5.52 ± 0.88 % IA/g). In all other organs and tissues the distribution of radioactivity was comparable among the radiopeptides, irrespective of whether they were labeled with 44Sc or 68Ga (Fig. 4c). The tissue distribution of 44Sc-NODAGA-NOC and 68Ga-NODAGA- NOC was also comparable (Fig. 4d). The only significant difference (p <0.05) were the kidneys, in which 44Sc-NODAGA-NOC was less retained (8.64 ± 1.62 % IA/g) than the 68Ga-NODAGA-NOC (11.5 ± 1.15 % IA/g). The liver uptake of 44Sc-NODAGA-NOC (0.88 ± 0.15 % IA/g) was lower than for 68Ga-NODAGA-NOC (1.68 ± 0.19 % IA/g) but the difference was not significant (Fig. 4c). Comparison of the DOTA-NOC and NODAGA-NOC revealed similar distribution profiles, irrespective of the radionuclide employed. As the only exception, it should be mentioned that 68Ga-DOTA-NOC showed a clearly higher retention in the liver than all other NOC-based radiopeptides (Fig. 4c/d). As a result of the tissue distribution data reported above the tumor-to-background NOC (0.68 ± 0.07 % IA/g) was significantly (p <0.05) lower than for 68Ga-DOTA-NOC (5.52 ± 0.88 % IA/g). Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides In all other organs and tissues the distribution of radioactivity was comparable among the radiopeptides, irrespective of whether they were labeled with 44Sc or 68Ga (Fig. 4c). The tissue distribution of 44Sc-NODAGA-NOC and 68Ga-NODAGA- NOC was also comparable (Fig. 4d). The only significant difference (p <0.05) were the kidneys, in which 44Sc-NODAGA-NOC was less retained (8.64 ± 1.62 % IA/g) than the 68Ga-NODAGA-NOC (11.5 ± 1.15 % IA/g). The liver uptake of 44Sc-NODAGA-NOC (0.88 ± 0.15 % IA/g) was lower than for 68Ga-NODAGA-NOC (1.68 ± 0.19 % IA/g) but the difference was not significant (Fig. 4c). Comparison of the DOTA-NOC and NODAGA-NOC revealed similar distribution profiles, irrespective of the radionuclide employed. As the only exception, it should be mentioned that 68Ga-DOTA-NOC showed a clearly higher retention in the liver than all other NOC-based radiopeptides (Fig. 4c/d). As a result of the tissue distribution data reported above, the tumor-to-background ratios were mostly similar between 44Sc-labeled DOTA-RGD and NODAGA-RGD as As a result of the tissue distribution data reported above, the tumor-to-background ratios were mostly similar between 44Sc-labeled DOTA-RGD and NODAGA-RGD as Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 11 of 19 well as between 44Sc-labeled DOTA-NOC and NODAGA-NOC, respectively (Tables 2 and 3). When comparing variation of tumor-to-background ratios between the 44Sc- and 68Ga-labeled versions of each of the four peptides, the ratios appeared more pro- nounced for the NOC-based radiopeptides over their RGD-based counterparts (Tables 2 and 3). Fig. 4 Biodistribution data obtained 2 h after injection of ~5 MBq (~1 nmol) 44Sc- and 68Ga-labeled DOTA-RGD (a) and NODAGA-RGD (b) and ~5 MBq (~1 nmol) DOTA-NOC (c) and NODAGA-NOC (d). Data bars represent the average ± SD of values obtained from n = 3 mice (* significantly different uptake of the 44Sc-labeled peptide compared to the 68Ga-labeled peptide, p <0.05) Fig. 4 Biodistribution data obtained 2 h after injection of ~5 MBq (~1 nmol) 44Sc- and 68Ga-labeled DOTA-RGD (a) and NODAGA-RGD (b) and ~5 MBq (~1 nmol) DOTA-NOC (c) and NODAGA-NOC (d). Data bars represent the average ± SD of values obtained from n = 3 mice (* significantly different uptake of the 44Sc-labeled peptide compared to the 68Ga-labeled peptide, p <0.05) Fig. 4 Biodistribution data obtained 2 h after injection of ~5 MBq (~1 nmol) 44Sc- and 68Ga-labeled DOTA-RGD (a) and NODAGA-RGD (b) and ~5 MBq (~1 nmol) DOTA-NOC (c) and NODAGA-NOC (d). Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides Data bars represent the average ± SD of values obtained from n = 3 mice (* significantly different uptake of the 44Sc-labeled peptide compared to the 68Ga-labeled peptide, p <0.05) well as between 44Sc-labeled DOTA-NOC and NODAGA-NOC, respectively (Tables 2 and 3). When comparing variation of tumor-to-background ratios between the 44Sc- and 68Ga-labeled versions of each of the four peptides, the ratios appeared more pro- nounced for the NOC-based radiopeptides over their RGD-based counterparts (Tables 2 and 3). Preclinical PET imaging studies with 44Sc- and 68Ga-labeled peptides PET/CT experiments were performed with one or two mice 3 h after injection of 44Sc- and 68Ga-labeled DOTA-RGD, NODAGA-RGD, DOTA-NOC and NODAGA-NOC, respectively (Figs. 5 and 6). 44Sc-DOTA-RGD showed clearly less accumulation of radioactivity in the liver than 68Ga-DOTA-RGD (Fig. 5a/b). 44Sc-NODAGA-RGD was comparable to 68Ga-NODAGA-RGD but showed somewhat more background activity in the abdominal tract (Fig. 5c/d). The uptake pattern of 44Sc-DOTA-RGD was slightly more favorable over the distribution of 44Sc-NODAGA-RGD (Fig. 5a/c). Preclinical PET imaging studies with 44Sc- and 68Ga-labeled peptides PET/CT experiments were performed with one or two mice 3 h after injection of 44Sc- and 68Ga-labeled DOTA-RGD, NODAGA-RGD, DOTA-NOC and NODAGA-NOC, respectively (Figs. 5 and 6). 44Sc-DOTA-RGD showed clearly less accumulation of radioactivity in the liver than 68Ga-DOTA-RGD (Fig. 5a/b). 44Sc-NODAGA-RGD was comparable to 68Ga-NODAGA-RGD but showed somewhat more background activity in the abdominal tract (Fig. 5c/d). The uptake pattern of 44Sc-DOTA-RGD was slightly more favorable over the distribution of 44Sc-NODAGA-RGD (Fig. 5a/c). PET/CT scans obtained at 3 h after injection of 44Sc-DOTA-NOC revealed uptake of radioactivity only in the tumor xenografts and in the kidneys, while 68Ga-DOTA-NOC accumulated also to a significant extent in the liver (Fig. 6a/b). 44Sc-NODAGA-NOC and 68Ga-NODAGA-NOC accumulated solely in tumors and kidneys (Fig. 6c/d). The PET/CT scans obtained at 3 h after injection of 44Sc-DOTA-NOC revealed uptake of radioactivity only in the tumor xenografts and in the kidneys, while 68Ga-DOTA-NOC accumulated also to a significant extent in the liver (Fig. 6a/b). 44Sc-NODAGA-NOC and 68Ga-NODAGA-NOC accumulated solely in tumors and kidneys (Fig. 6c/d). The Page 12 of 19 Page 12 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Table 2 Tumor-to-background ratios at different time points after injection of 44Sc/68Ga-labeled DOTA-RGD and NODAGA-RGD 44Sc-DOTA-RGD 68Ga-DOTA-RGD U87MG 30 min p.i. 2 h p.i. 5 h p.i. 30 min p.i. 2 h p.i. Radiolabeling and stability of 44Sc- and 68Ga-labeled peptides Tumor-to-blood 4.82 ± 1.60 40.1 ± 15.6 47.2 ± 18.0 4.34 ± 0.57 28.0 ± 4.26 Tumor-to-liver 0.99 ± 0.14 0.61 ± 0.04 2.90 ± 0.38 0.55 ± 0.09 0.48 ± 0.07 Tumor-to-kidney 1.16 ± 0.29 1.77 ± 0.13 1.82 ± 0.21 0.88 ± 0.07 1.74 ± 0.19 44Sc-NODAGA-RGD 68Ga-NODAGA-RGD U87MG 30 min p.i. 2 h p.i. 5 h p.i. 30 min p.i. 2 h p.i. Tumor-to-blood 4.73 ± 1.24 34.1 ± 7.62 30.5 ± 9.93 4.17 ± 0.47 114 ± 35.0 Tumor-to-liver 3.02 ± 0.40 3.01 ± 0.13 2.82 ± 0.21 1.72 ± 0.16 1.50 ± 0.20 Tumor-to-kidney 1.18 ± 0.28 2.68 ± 0.14 2.58 ± 0.24 0.89 ± 0.03 1.49 ± 0.21 Table 2 Tumor-to-background ratios at different time points after injection of 44Sc/68Ga-labeled DOTA-RGD and NODAGA-RGD tumor uptake of 44Sc-NODAGA-NOC in the mouse, which was used for PET imaging, was reduced, resulting in lower tumor-to-kidney ratios compared to 68Ga-NODAGA- NOC. In this context, it has to be mentioned that 44Sc-NODAGA-NOC was prepared at a low specific activity for the PET scan (Fig. 6c) which implies that the injected molar amount of peptide was significantly increased compared to the peptide amount injected with 68Ga-NODAGA-NOC and, as a result, the binding sites in the tumor tissue may have been saturated. An overview of the PET/CT scans of all four peptides labeled with 44Sc at 5 h after injection showed largely the same tissue distribution as was found at 3 h after injection (Fig. 7). While RGD-based peptides accumulated in the U87MG tumor xenografts and showed residual activity in the intestinal tract, NOC-based peptides accumulated in AR42J tumors xenografts and showed significant retention of radioactivity in the kidneys. Discussion A number of preclinical studies demonstrated the potential of 44Sc as an alternative PET radiometal to the currently-used 68Ga (Müller et al. 2013; Koumarianou et al. 2012; Hernandez et al. 2014; Miederer et al. 2011). With this in mind, the possibility to extend its applications to peptides of clinical relevance is of great interest for medical physicians. Herein, we reported on the first, to our knowledge, preclinical study Table 3 Tumor-to-background ratios at different time points after injection of 44Sc/68Ga-labeled DOTA-NOC and NODAGA-NOC 44Sc-DOTA-NOC 68Ga-DOTA-NOC AR42J 30 min p.i. 2 h p.i. 5 h p.i. 30 min p.i. 2 h p.i. Tumor-to-blood 3.69 ± 1.52 58.2 ± 6.64 52.0 ± 4.38 5.22 ± 1.07 46.2 ± 2.28 Tumor-to-liver 6.87 ± 2.38 13.1 ± 2.16 15.7 ± 2.82 2.44 ± 0.36 2.22 ± 0.22 Tumor-to-kidney 0.79 ± 0.19 1.01 ± 0.06 0.85 ± 0.04 1.20 ± 0.21 1.41 ± 0.18 44Sc-NODAGA-NOC 68Ga-NODAGA-NOC AR42J 30 min p.i. 2 h p.i. 5 h p.i. 30 min p.i. 2 h p.i. Tumor-to-blood 5.23 ± 0.22 50.3 ± 16.8 78.6 ± 6.95 4.32 ± 0.58 76.0 ± 4.31 Tumor-to-liver 8.13 ± 0.72 11.9 ± 3.26 11.6 ± 1.51 4.36 ± 0.91 5.88 ± 0.68 Tumor-to-kidney 0.80 ± 0.10 1.22 ± 0.34 0.87 ± 0.24 0.58 ± 0.10 0.85 ± 0.08 Table 3 Tumor-to-background ratios at different time points after injection of 44Sc/68Ga-labeled DOTA-NOC and NODAGA-NOC Page 13 of 19 Page 13 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Fig. 5 PET/CT scans of U87MG tumor-bearing mice 3 h after injection of ~10 MBq (~1 nmol) 44Sc-DOTA- RGD (a), (~10 MBq/~1 nmol) 68Ga-DOTA-RGD (b), ~10 MBq (~1 nmol) 44Sc-NODAGA-RGD (c) and ~10 MBq (~1 nmol) 68Ga-NODAGA-RGD (d). During the PET (20 min) and the CT (1.5 min) scans the mice were anes- thetized with isoflurane/oxygen (Tu = U87MG tumor xenografts, Li = liver, Int = intestines, Bl = urinary bladder) Fig. 5 PET/CT scans of U87MG tumor-bearing mice 3 h after injection of ~10 MBq (~1 nmol) 44Sc-DOTA- RGD (a), (~10 MBq/~1 nmol) 68Ga-DOTA-RGD (b), ~10 MBq (~1 nmol) 44Sc-NODAGA-RGD (c) and ~10 MBq (~1 nmol) 68Ga-NODAGA-RGD (d). Discussion During the PET (20 min) and the CT (1.5 min) scans the mice were anes- thetized with isoflurane/oxygen (Tu = U87MG tumor xenografts, Li = liver, Int = intestines, Bl = urinary bladder) concerning the 44Sc-labeling of peptides comprising a NODAGA-chelator, as well as their in vitro and in vivo behavior. Several authors proposed the DOTA-chelator as the most suitable ligand for binding Sc(III), whereas for Ga(III) it is known that NODAGA com- plexes provide higher thermodynamic stability than the DOTA complex (Huclier-Markai et al. 2011; Notni et al. 2012). Currently, the 68Ga-DOTA functionalized somatostatin receptor analogues are among the most prominent radiopharmaceuticals for clinical PET imaging (Banerjee & Pomper 2013). Since NODAGA derivatized biomolecules have not been used for labeling with 44Sc to date, the question arose on whether or not sufficient stability can be achieved for in vivo application of 44Sc-NODAGA compounds. In order to perform the preclinical experiments effectively, it was necessary to obtain the 44Sc in a small solution volume, suitable for direct radiolabeling and subsequent in vivo application, without extensive dilution. Previously, we reported on the Page 14 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Fig. 6 PET/CT scans of AR42J tumor-bearing mice 3 h after injection of ~10 MBq (~1 nmol) 44Sc-DOTA-NOC (a), ~10 MBq (~1 nmol) 68Ga-DOTA-NOC (b), ~10 MBq (~3 nmol) 44Sc-NODAGA-NOC (c) and ~10 MBq (~1 nmol) 68Ga-NODAGA-NOC (d). During the PET (20 min) and the CT (1.5 min) scans the mice were anesthetized with isoflurane/oxygen (Tu = U87MG tumor xenografts, Li = liver, Int = intestines, Bl = urinary bladder) Fig. 6 PET/CT scans of AR42J tumor-bearing mice 3 h after injection of ~10 MBq (~1 nmol) 44Sc-DOTA-NOC (a), ~10 MBq (~1 nmol) 68Ga-DOTA-NOC (b), ~10 MBq (~3 nmol) 44Sc-NODAGA-NOC (c) and ~10 MBq (~1 nmol) 68Ga-NODAGA-NOC (d). During the PET (20 min) and the CT (1.5 min) scans the mice were anesthetized with isoflurane/oxygen (Tu = U87MG tumor xenografts, Li = liver, Int = intestines, Bl = urinary bladder) implementation of SCX resin to concentrate the 44Sc radioactivity in a small volume (van der Meulen et al. 2015). Although the use of this resin is already established for the concentration of the 68Ga eluate from the 68Ge generator, the high osmolarity of the eluate is not suitable for direct in vivo application. Discussion Herein, we report the use of DGA extraction chromatographic resin to effectively concentrate ~85 % of the 44Sc radioactivity in a small volume of 700 μL. The acidic solution containing the 44Sc was mixed with sodium acetate to obtain a pH of 4–4.5 for radiolabeling reactions. This procedure allowed in vivo application of the radiolabeled peptides without excessive dilution. Reproducible labeling with 44Sc at high specific activities (10 MBq/nmol) was achieved for peptides functionalized with a DOTA-chelator. The same was possible for 68Ga with both DOTA- and NODAGA-functionalized peptides. Radiolabeling of NODAGA- compounds with 44Sc, however, proved to be more challenging and was not achieved Page 15 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Fig. 7 PET/CT scans of U87MG tumor-bearing mice 5 h after injection of ~10 MBq (~1 nmol) 44Sc-DOTA-RGD (a) and ~10 MBq (~1 nmol) 44Sc-NODAGA-RGD (b). PET/CT scans of AR42J tumor-bearing mice 5 h after injection of ~10 MBq (~1 nmol) 44Sc-DOTA-NOC (c) and ~10 MBq (~3 nmol) 44Sc-NODAGA-NOC (d). During the PET (20 min) and the CT (1.5 min) scans the mice were anesthetized with isoflurane/oxygen (Tu = U87MG (a/b) or AR42J tumor xenografts (c/d), Ki = kidney, Int = intestines, Bl = urinary bladder) Fig. 7 PET/CT scans of U87MG tumor-bearing mice 5 h after injection of ~10 MBq (~1 nmol) 44Sc-DOTA-RGD (a) and ~10 MBq (~1 nmol) 44Sc-NODAGA-RGD (b). PET/CT scans of AR42J tumor-bearing mice 5 h after injection of ~10 MBq (~1 nmol) 44Sc-DOTA-NOC (c) and ~10 MBq (~3 nmol) 44Sc-NODAGA-NOC (d). During the PET (20 min) and the CT (1.5 min) scans the mice were anesthetized with isoflurane/oxygen (Tu = U87MG (a/b) or AR42J tumor xenografts (c/d), Ki = kidney, Int = intestines, Bl = urinary bladder) reproducibly at high specific activity. It may be a result of potentially interfering metal contaminations to which NOTA/NODAGA is more susceptible than DOTA, as previ- ously reported (Simecek et al. 2013). If NODAGA-functionalized peptides should be used with 44Sc for clinical studies, it will be important to determine the maximum concentra- tion of metal contaminants which would still allow high specific and reproducible labeling with 44Sc. A potential optimization of the labeling may also be accessible by thorough inves- tigation of different buffer systems and the use of microwave heating (Elander et al. 2000). Discussion Finally, even when the labeling was achieved successfully, the stability of 44Sc-labeled NODAGA-peptides was clearly inferior to the stability of 44Sc-labeled DOTA- compounds. It will be important, thus, to investigate the conditions which enhance the stability of 44Sc-NODAGA-peptides, potentially allowing an increased shelf-life which would be necessary in view of a clinical application. In the initial in vitro test, the stability of 44Sc- and 68Ga-labeled peptides was investi- gated in saline with and without addition of excess and Fe3+ and Cu2+, respectively, to de- termine the possibility of metal challenge. None of the conditions impaired the integrity of 44Sc-labeled DOTA-RGD and DOTA-NOC, even after four half-lives of incubation at 37 °C. The obtained results are in agreement with those of Pruszynski et al., who reported an unchanged stability of 44Sc-DOTA-TOC in the presence of metal cations (Pruszynski et al. 2012). The amount of intact 68Ga-DOTA-RGD and 68Ga-DOTA-NOC was only decreased after the addition of Cu2+ (0.01 M) which was comparable with the time- and Cu2+-concentration dependent transmetalation of 68Ga-DOTA-TATE (Oehlke et al. 2013). 44Sc-labeled NODAGA-peptides were significantly less stable, indicating an onset of release of the radionuclide from the chelator only one half-life after labeling. This was in clear contrast to the 68Ga-NODAGA-peptides, which were completely stable over the whole period of investigation. The NODAGA-chelator revealed less stable coordination of 44Sc compared to the DOTA under the experimental conditions in this work. Distribution coefficients determined in n-octanol and PBS pH 7.4 revealed logD values in the same range (−4.70 to −4.26) for all RGD-based peptides, irrespective of the chelator and radio- nuclide which was employed. The logD values obtained with NOC-based peptides were slightly higher (−1.68 to −2.54) for all four radiopeptides (44Sc/68Ga-DOTA/ Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 16 of 19 Page 16 of 19 NODAGA-NOC), indicating increased lipophilic properties compared to the RGD- peptides (Additional file 1: Table S1). Biodistribution studies were performed with tumor-bearing mice at different time points after administration of the 44Sc-labeled peptides. Application of 44Sc-DOTA/ NODAGA-RGD and 44Sc-DOTA/NODAGA-NOC, respectively, resulted in only small variations of the kinetics, independent of whether a DOTA or a NODAGA chelator was used (Fig. 3b). Discussion Comparing the tissue distribution profiles of 44Sc-labeled peptides with those of 68Ga-labeled peptides revealed that the differences between DOTA- and NODAGA-derivatized compounds were largely due to the different properties of these peptides, rather than the consequence of any kind of instability (Fig. 4). Indications of in vivo instability of the 44Sc-NODAGA-compounds were not apparent, with the excep- tion of the increasing kidney uptake from 3 to 5 h after injection of 44Sc-NODAGA-NOC (Fig. 3b). It remained unclear, however, whether this was due to release of 44Sc from the NODAGA chelator since injection of free 44Sc resulted in unspecific retention of radio- activity in the liver and intestinal tract, rather than in the kidneys (Additional file 1: Figure S1). Generally, the instability of 44Sc-NODAGA compounds in solution was time- dependent and, as a consequence, it appeared not to be an issue for imaging purposes at relatively short time points (<1 half-life of 44Sc) after injection. The tissue distribution pro- files with each peptide were similar, independent of whether it was labeled with 44Sc or 68Ga. One of the most conspicuous differences between 44Sc- and 68Ga-labeled peptides, however, was the increased liver uptake of 68Ga-DOTA-peptides, which was seen in PET images obtained with both DOTA-RGD and DOTA-NOC, respectively. Biodistribution studies confirmed these differences in liver uptake for DOTA-NOC (Fig. 4c), however, in the case of DOTA-RGD the liver uptake was relatively high for both 68Ga-DOTA-RGD and 44Sc-DOTA-RGD (Fig. 4a). This was in contrast to 44Sc- and 68Ga-labeled NODAGA-RGD peptides, which showed a clearly reduced liver uptake at 2 h p.i. in com- parison (Fig. 4b). Since uncoordinated 68Ga may also accumulate to a significant extent in the bones as it was shown in a separate experiment (Additional file 1: Figure S1), it is un- likely that the described results are a consequence of released 68Ga from the DOTA- chelator. Previously, it was reported that 68Ga-DOTA-RGD showed a higher blood pool activity than 68Ga-NODAGA-RGD (Knetsch et al. 2011; Decristoforo et al. 2008). As a result and in agreement with our studies, it was found that 68Ga-DOTA-RGD accumulates to a higher extent in the liver than 68Ga-NODAGA-RGD (Fig. 4a/b). Discussion In this context, it is also import- ant to note that the peptide structure of the DOTA-RGD comprises a phenylalanine, whereas in the case of the NODAGA-RGD the phenylalanine was replaced with a tyrosine, which may have an influence on the pharmacokinetics of these peptides (Fig. 1). Overall, 68Ga-NODAGA-RGD revealed a more favorable tissue distribution profile than 68Ga-DOTA-RGD, shown in previous studies as well as in the experiments presented in this work (Knetsch et al. 2011; Pohle et al. 2012). A similar trend, albeit less pronounced, was seen for the 44Sc-labeled RGD peptides, showing lower background activity of 44Sc-NODAGA-RGD compared to that of 44Sc-DOTA-RGD (Fig. 4a/b). The most striking difference between 44Sc- and 68Ga-labeled NOC-based peptides was the reduced liver uptake of 44Sc-DOTA-NOC as compared to 68Ga-DOTA-NOC, although 68Ga-DOTA-NOC accumulated to a higher extent in the tumor tissue than Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 17 of 19 Page 17 of 19 44Sc-DOTA-NOC (Fig. 4c). When looking at the tissue distribution profiles of the 44Sc- and 68Ga-NODAGA-NOC peptides, they were found to be largely comparable with slightly higher retention of 68Ga-NODAGA-NOC in the kidneys (Fig. 4d). These results were also largely comparable to previously published data obtained with 68Ga- NODAGA-TOC (Eisenwiener et al. 2002). Even though the 44Sc-NODAGA-NOC revealed to be the least stable in vitro, its tissue distribution profile was largely compar- able to 68Ga-labeled NODAGA-NOC indicating that the compound was stable in vivo. Overall, it was found that 44Sc-DOTA-peptides were of significantly higher stability than the corresponding 44Sc-NODAGA-peptides, as expected, based on stability constants previously reported (Huclier-Markai et al. 2011; Port et al. 2008). Our experience also revealed that metal impurities would clearly interfere more distinctly with the stability of 44Sc-NODAGA-compounds than in the case of 44Sc-DOTA-compounds. Finally, it is important to note that the 44Sc-labeling of a NODAGA-functionalized biomolecule ap- pears to be dependent on the overall chemical structure of the compound, as the 44Sc- NODAGA-RGD was found to be clearly more stable than the 44Sc-NODAGA-NOC. NODAGA-chelators are, thus, not excluded from use for labeling with 44Sc, but a thorough investigation of each case will be necessary in order to guarantee sufficient stability of the radiopharmaceutical. Ethical approval This article does not contain any studies with human participants performed by any of the authors. Discussion Based on our results, which show clear differences in kinetics between labeled DOTA- and NODAGA-functionalized peptides, it is likely that 44Sc would be the pre- ferred nuclide to be used with DOTA-functionalized biomolecules, in order to reflect the tissue distribution of 177Lu-labeled compounds more accurately, than if 68Ga was used for the same purpose. Conclusions In this work, it was demonstrated that 44Sc can be used for the labeling of biomolecules with both a DOTA and NODAGA chelator, although using a NODAGA-chelator proved to be more challenging. Other than with 68Ga, which shows clearly better results if coordinated with a NODAGA-chelator, 44Sc appears to be more stably complexed with DOTA. Based on these results, we conclude that even though 44Sc would be most favorably coordinated with a DOTA-chelator, coordination with a NODAGA-chelator is possible if the labeling conditions and storage buffers are validated. When using 44Sc and 47Sc for theragnostic application, however, DOTA is clearly the chelator of choice. Ethical approval All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. Authors’ contributions KD and RF performed the separation of 44Sc, did radiolabeling and stability experiments and drafted the manuscript. RMS performed in vitro and in vivo experiments and assisted in writing the manuscript. CM was responsible for the performance of the in vitro and in vivo studies and supervised these experiments. NvdM was responsible for the development of the production and separation process of 44Sc and supervised the whole study. NvdM and CM designed and wrote the final manuscript. RS and AT reviewed the manuscript. All authors read and approved the final manuscript. Competing interests The authors declare that they have no competing interests. Competing interests The authors declare that they have no competing interests. Acknowledgements g The authors thank Dr. Christiaan Vermeulen, Klaudia Siwowska, Susan Cohrs, David Bölsterli, Walter Hirzel, Alexander Sommerhalder, Muhamet Djelili and André Isenschmid for technical assistance. Funding information h h f d g The research was funded by the Swiss National Science Foundation (CR23I2_156852 and IZLIZ3_156800) and Paul Scherrer Institut (internal grant). References Banerjee SR, Pomper MG. Clinical applications of gallium-68. Appl Radiat Isot. 2013;76:2–13. doi:10.1016/j.apradiso.2013 01.039. Chakravarty R, Goel S, Valdovinos HF, Hernandez R, Hong H, Nickles RJ, et al. Matching the decay half-life with the biological half-life: ImmunoPET imaging with 44Sc-labeled cetuximab Fab fragment. Bioconjug Chem. 2014;25(12): 2197–204. doi:10.1021/bc500415x. Choi H, Phi JH, Paeng JC, Kim SK, Lee YS, Jeong JM, et al. Imaging of integrin alpha(V)beta(3) expression using 68Ga-RGD positron emission tomography in pediatric cerebral infarct. Mol Imaging. 2013;12(4):213–7. 68 111 Choi H, Phi JH, Paeng JC, Kim SK, Lee YS, Jeong JM, et al. Imaging of integrin alpha(V)beta(3) expression using 68Ga-RGD positron emission tomography in pediatric cerebral infarct. Mol Imaging. 2013;12(4):213–7. Decristoforo C, Hernandez Gonzalez I, Carlsen J, Rupprich M, Huisman M, Virgolini I, et al. 68Ga- and 111In-labelled DOTA-RGD peptides for imaging of avb3 integrin expression. Eur J Nucl Med Mol Imaging. 2008;35(8):1507–15. doi:10.1007/s00259-008-0757-6. Eigner S, Vera DR, Fellner M, Loktionova NS, Piel M, Lebeda O, et al. Imaging of protein synthesis: in vitro and in vivo evaluation of 44Sc-DOTA-puromycin. Mol Imaging Biol. 2013;15(1):79–86. doi:10.1007/s11307-012-0561-3. Eigner S, Vera DR, Fellner M, Loktionova NS, Piel M, Lebeda O, et al. Imaging of protein synthesis: in vitro and in vivo evaluation of 44Sc-DOTA-puromycin. Mol Imaging Biol. 2013;15(1):79–86. doi:10.1007/s11307-012-0561-3. Eisenwiener KP, Prata MI, Buschmann I, Zhang HW, Santos AC, Wenger S, et al. NODAGATOC, a new chelator-coupled somatostatin analogue labeled with [67/68Ga] and [111In] for SPECT, PET, and targeted therapeutic applications of somatostatin receptor (hsst2) expressing tumors. Bioconjug Chem. 2002;13(3):530–41. Eisenwiener KP, Prata MI, Buschmann I, Zhang HW, Santos AC, Wenger S, et al. NODAGATOC, a new chelator-coupled somatostatin analogue labeled with [67/68Ga] and [111In] for SPECT, PET, and targeted therapeutic applications of somatostatin receptor (hsst2) expressing tumors. Bioconjug Chem. 2002;13(3):530–41. Elander N, Jones JR, Lu SY, Stone-Elander S. Microwave-enhanced radiochemistry. Chem Soc Rev. 2000;29(4):239–49. doi:10.1039/a901713e. Hernandez R, Valdovinos HF, Yang Y, Chakravarty R, Hong H, Barnhart TE, et al. 44Sc: an attractive isotope for peptide- based PET imaging. Mol Pharm. 2014;11(8):2954–61. doi:10.1021/mp500343j. Huclier-Markai S, Sabatie A, Ribet S, Kubicek V, Paris M, Vidaud C, et al. Chemical and biological evaluation of scandium(III)-polyaminopolycarboxylate complexes as potential PET agents and radiopharmaceuticals. Radiochim A Huclier-Markai S, Sabatie A, Ribet S, Kubicek V, Paris M, Vidaud C, et al. Author details 1L b f R 1Laboratory of Radiochemistry, Paul Scherrer Institute, CH-5232 Villigen-PSI, Switzerland. 2Department of Chemistry and Biochemistry, University of Bern, 3012 Bern, Switzerland. 3Center for Radiopharmaceutical Sciences ETH-PSI-USZ, Paul Scherrer Institute, 5232 Villigen-PSI, Switzerland. 4Department of Chemistry and Applied Biosciences, ETH Zurich, 8093 Zurich, Switzerland. 5SCK.CEN, BR2 Reactor, 2400 Mol, Belgium. Received: 16 February 2016 Accepted: 21 April 2016 Received: 16 February 2016 Accepted: 21 April 2016 Additional file Additional file 1: Supplementary experimental data. (DOCX 777 kb) Page 18 of 19 Page 18 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Notni J, Pohle K, Wester HJ. Comparative gallium-68 labeling of TRAP-, NOTA-, and DOTA-peptides: practical consequences for the future of gallium-68-PET. EJNMMI Res. 2012;2(1):28. doi:10.1186/2191-219X-2-28. Oehlke E, Le VS, Lengkeek N, Pellegrini P, Jackson T, Greguric I, et al. Influence of metal ions on the 68Ga-labeling of DOTATATE. Appl Radiat Isot. 2013;82:232–8. doi:10.1016/j.apradiso.2013.08.010. Pohle K, Notni J, Bussemer J, Kessler H, Schwaiger M, Beer AJ. 68Ga-NODAGA-RGD is a suitable substitute for 18F-Galacto- RGD and can be produced with high specific activity in a cGMP/GRP compliant automated process. Nucl Med Biol. 2012;39(6):777–84. doi:10.1016/j.nucmedbio.2012.02.006. Port M, Idee JM, Medina C, Robic C, Sabatou M, Corot C. Efficiency, thermodynamic and kinetic stability of marketed gadolinium chelates and their possible clinical consequences: a critical review. Biometals. 2008;21(4):469–90. doi:10.1007/s10534-008-9135-x. Pruszynski M, Majkowska-Pilip A, Loktionova NS, Eppard E, Rösch F. Radiolabeling of DOTATOC with the long-lived positron emitter 44Sc. Appl Radiat Isot. 2012;70(6):974–9. doi:10.1016/j.apradiso.2012.03.005. Rösch F. Scandium-44: benefits of a long-lived PET radionuclide available from the 44Ti/44Sc generator system. Curr Radiopharm. 2012;5(3):187–201. doi:CRP-EPUB-20120529-10. Simecek J, Hermann P, Wester HJ, Notni J. How is 68Ga labeling of macrocyclic chelators influenced by metal ion contaminants in 68Ge/68Ga generator eluates? ChemMedChem. 2013;8(1):95–103. doi:10.1002/cmdc.201200471. van der Meulen NP, Bunka M, Domnanich KA, Müller C, Haller S, Vermeulen C, et al. Cyclotron production of 44Sc: From bench to bedside. Nucl Med Biol. 2015;42(9):745–51. doi:10.1016/j.nucmedbio.2015.05.005. Viola-Villegas N, Doyle R. The coordination chemistry of 1,4,7,10-tetraazacyclododecane-N, N’, N”, N‴-tetraacetic acid (H4DOTA): Structural overview and analyses on structure-stability relationships. Coordin Chem Rev. 2009;253:1906–25. Yoon HJ, Kang KW, Chun IK, Cho N, Im SA, Jeong S, et al. Correlation of breast cancer subtypes, based on estrogen receptor, progesterone receptor, and HER2, with functional imaging parameters from 68Ga-RGD PET/CT and 18F- FDG PET/CT. Eur J Nucl Med Mol Imaging. 2014;41(8):1534–43. doi:10.1007/s00259-014-2744-4. References Chemical and biological evaluation of di (III) l i l b l t l t ti l PET t d di h ti l R di hi Huclier-Markai S, Sabatie A, Ribet S, Kubicek V, Paris M, Vidaud C, et al. Chemical and biological evaluation of scandium(III)-polyaminopolycarboxylate complexes as potential PET agents and radiopharmaceuticals. Radiochim Acta. 2011;99:653–62. scandium(III)-polyaminopolycarboxylate complexes as potential PET agents and radiopharmaceuticals. Radiochim Acta. 2011;99:653–62. 68 Kim JH, Lee JS, Kang KW, Lee HY, Han SW, Kim TY, et al. Whole-body distribution and radiation dosimetry of 68Ga- NOTA-RGD, a positron emission tomography agent for angiogenesis imaging. Cancer Biother Radiopharm. 2012; 27(1):65–71. doi:10.1089/cbr.2011.1061. Knetsch PA, Petrik M, Griessinger CM, Rangger C, Fani M, Kesenheimer C, et al. [68Ga]NODAGA-RGD for imaging avb3 integrin expression. Eur J Nucl Med Mol Imaging. 2011;38(7):1303–12. doi:10.1007/s00259-011-1778-0. 44 Koumarianou E, Loktionova NS, Fellner M, Roesch F, Thews O, Pawlak D, et al. 44Sc-DOTA-BN[2–14]NH2 in comparison to 68Ga-DOTA-BN[2–14]NH2 in pre-clinical investigation. Is 44Sc a potential radionuclide for PET? Appl Radiat Isot. 2012;70(12):2669–76. doi:10.1016/j.apradiso.2012.08.004. Kwekkeboom DJ, Kam BL, van Essen M, Teunissen JJ, van Eijck CH, Valkema R, et al. Somatostatin-receptor-based imaging and therapy of gastroenteropancreatic neuroendocrine tumors. Endocr Relat Cancer. 2010;17(1):R53–73. doi:10.1677/ERC-09-0078. Majkowska-Pilip A, Bilewicz A. Macrocyclic complexes of scandium radionuclides as precursors for diagnostic and therapeutic radiopharmaceuticals. J Inorg Biochem. 2011;105(2):313–20. doi:10.1016/j.jinorgbio.2010.11.003. Miederer M, Loktionova NS, Bausbacher N, Buchholz H, Piel M, Schreckenberger M, et al. Small Animal PET-Imaging with 44Sc-DOTATOC. Eur J Nucl Med Mol Imaging. 2011;38:S158-S. Müller C, Bunka M, Reber J, Fischer C, Zhernosekov K, Türler A, et al. Promises of cyclotron-prod diagnostic match for trivalent b−-emitters: In vitro and in vivo study of a 44Sc-DOTA-folate 2013;54(12):2168–74. doi:10.2967/jnumed.113.123810. Müller C, Bunka M, Haller S, Köster U, Groehn V, Bernhardt P, et al. Promising prospects for 44Sc-/47Sc-based theragnostics: application of 47Sc for radionuclide tumor therapy in mice. J Nucl Med. 2014;55(10):1658–64. doi:10.2967/jnumed.114.141614. Müller C, Bunka M, Haller S, Köster U, Groehn V, Bernhardt P, et al. Promising prospects for 44Sc-/47Sc-based theragnostics: application of 47Sc for radionuclide tumor therapy in mice. J Nucl Med. 2014;55(10):1658–64. doi:10.2967/jnumed.114.141614. Page 19 of 19 Domnanich et al. EJNMMI Radiopharmacy and Chemistry (2016) 1:8 Page 19 of 19 Page 19 of 19 Notni J, Pohle K, Wester HJ. Comparative gallium-68 labeling of TRAP-, NOTA-, and DOTA-peptides: practical consequences for the future of gallium-68-PET. EJNMMI Res. 2012;2(1):28. doi:10.1186/2191-219X-2-28. References Oehlke E, Le VS, Lengkeek N, Pellegrini P, Jackson T, Greguric I, et al. Influence of metal ions on the 68Ga-labeling of DOTATATE. Appl Radiat Isot. 2013;82:232–8. doi:10.1016/j.apradiso.2013.08.010. Pohle K, Notni J, Bussemer J, Kessler H, Schwaiger M, Beer AJ. 68Ga-NODAGA-RGD is a suitable substitute for 18F-Galacto- RGD and can be produced with high specific activity in a cGMP/GRP compliant automated process. Nucl Med Biol. 2012;39(6):777–84. doi:10.1016/j.nucmedbio.2012.02.006. Port M, Idee JM, Medina C, Robic C, Sabatou M, Corot C. Efficiency, thermodynamic and kinetic stability of marketed gadolinium chelates and their possible clinical consequences: a critical review. Biometals. 2008;21(4):469–90. doi:10.1007/s10534-008-9135-x. Pruszynski M, Majkowska-Pilip A, Loktionova NS, Eppard E, Rösch F. Radiolabeling of DOTATOC with the long-lived positron emitter 44Sc. Appl Radiat Isot. 2012;70(6):974–9. doi:10.1016/j.apradiso.2012.03.005. Rösch F. Scandium-44: benefits of a long-lived PET radionuclide available from the 44Ti/44Sc generator system. Curr Radiopharm. 2012;5(3):187–201. doi:CRP-EPUB-20120529-10. Simecek J, Hermann P, Wester HJ, Notni J. How is 68Ga labeling of macrocyclic chelators influenced by metal ion contaminants in 68Ge/68Ga generator eluates? ChemMedChem. 2013;8(1):95–103. doi:10.1002/cmdc.201200471. van der Meulen NP, Bunka M, Domnanich KA, Müller C, Haller S, Vermeulen C, et al. Cyclotron production of 44Sc: From bench to bedside. Nucl Med Biol. 2015;42(9):745–51. doi:10.1016/j.nucmedbio.2015.05.005. Viola-Villegas N, Doyle R. The coordination chemistry of 1,4,7,10-tetraazacyclododecane-N, N’, N”, N‴-tetraacetic acid (H4DOTA): Structural overview and analyses on structure-stability relationships. Coordin Chem Rev. 2009;253:1906–25. Yoon HJ, Kang KW, Chun IK, Cho N, Im SA, Jeong S, et al. Correlation of breast cancer subtypes, based on estrogen receptor, progesterone receptor, and HER2, with functional imaging parameters from 68Ga-RGD PET/CT and 18F- FDG PET/CT. Eur J Nucl Med Mol Imaging. 2014;41(8):1534–43. doi:10.1007/s00259-014-2744-4. Notni J, Pohle K, Wester HJ. Comparative gallium-68 labeling of TRAP-, NOTA-, and DOTA-peptides: practical consequences for the future of gallium-68-PET. EJNMMI Res. 2012;2(1):28. doi:10.1186/2191-219X-2-28. 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
https://openalex.org/W3197283102	https://europepmc.org/articles/pmc8467967?pdf=render	English	null	One-Step Microwave-Assisted Synthesis of PtNiCo/rGO Electrocatalysts with High Electrochemical Performance for Direct Methanol Fuel Cells	Nanomaterials	2,021	cc-by	14,353	Citation: Shih, K.-Y.; Wei, J.-J.; Tsai, M.-C. One-Step Microwave-Assisted Synthesis of PtNiCo/rGO Electrocatalysts with High Electrochemical Performance for Direct Methanol Fuel Cells. Nanomaterials 2021, 11, 2206. https:// doi.org/10.3390/nano11092206 Citation: Shih, K.-Y.; Wei, J.-J.; Tsai, M.-C. One-Step Microwave-Assisted Synthesis of PtNiCo/rGO Electrocatalysts with High Electrochemical Performance for Direct Methanol Fuel Cells. Nanomaterials 2021, 11, 2206. https:// doi.org/10.3390/nano11092206 Keywords: direct methanol fuel cells; PtNiCo/rGO; graphene; microwave-assisted synthesis; nanocomposites Received: 29 July 2021 Accepted: 25 August 2021 Published: 27 August 2021 One-Step Microwave-Assisted Synthesis of PtNiCo/rGO Electrocatalysts with High Electrochemical Performance for Direct Methanol Fuel Cells Kun-Yauh Shih , Jia-Jun Wei and Ming-Chi Tsai Kun-Yauh Shih , Jia-Jun Wei and Ming-Chi Tsai Department of Applied Chemistry, National Pingtung University, Pingtung County 90003, Taiwan; sg974408@gmail.com (J.-J.W.); mikechi614@gmail.com (M.-C.T.) * Correspondence: sky@mail.nptu.edu.tw Abstract: Platinum (Pt) is widely used as an activator in direct methanol fuel cells (DMFCs). However, the development of Pt catalyst is hindered due to its high cost and CO poisoning. A multi-metallic catalyst is a promising catalyst for fuel cells. We develop a simple and rapid method to synthesize PtNiCo/rGO nanocomposites (NCs). The PtNiCo/rGO NCs catalyst was obtained by microwave- assisted synthesis of graphene oxide (GO) with Pt, Ni, and Co precursors in ethylene glycol (EG) solution after heating for 20 min. The Pt-Ni-Co nanoparticles showed a narrow particle size dis- tribution and were uniformly dispersed on the reduced graphene oxide without agglomeration. Compared with PtNiCo catalyst, PtNiCo/rGO NCs have superior electrocatalytic properties, in- cluding a large electrochemical active surface area (ECSA), the high catalytic activity of methanol, excellent anti-toxic properties, and high electrochemical stability. The ECSA can be up to 87.41 m2/g at a scan rate of 50 mV/s. They also have the lowest oxidation potential of CO. These excellent electrochemical performances are attributed to the uniform dispersion of PtNiCo nanoparticles, good conductivity, stability, and large speciﬁc surface area of the rGO carrier. The synthesized PtNiCo/rGO nanoparticles have an average size of 17.03 ± 1.93 nm. We also investigated the effect of catalyst material size on electrocatalytic performance, and the results indicate that PtNiCo/rGO NC catalysts can replace anode catalyst materials in fuel cell applications in the future. nanomaterials nanomaterials nanomaterials 1. Introduction Therefore, adding other non-precious transition metals such as Ni, Co, Ru, etc. to replace the expensive Pt can reduce the consumption of expensive Pt materials while maintaining excellent performance [18]. Rethinasabapathy et al. [19] synthesized ternary PtRuFe nanoparticles supported by N-doped graphene as efﬁcient methanol oxidation exhibiting higher ECSA. MOR activity is two to three times higher compared to other mono- and bimetallic catalysts. The addition of Fe signiﬁcantly reduces the amount of Pt used in fuel cells. Other high-quality ternary PtRhCu nanocrystals with highly dendritic nanostructures were synthesized. The high speciﬁc activity and mass activ- ity of the catalyst are due to the synergistic effect between Pt, Rh, and Cu elements and their highly dendritic nanostructure [20]. Sui et al. [21] prepared ternary Au@PdNi core–shell nanoparticles by a facile method. The results indicated that the electronic effects and the core–shell nanostructure played an important role in enhancing the catalytic activity and stability. They also enhanced the toxicity resistance of catalyst intermediates [22], improved the performance and durability of catalysts, and increased the overall energy conversion efﬁciency [23]. Pt1-x-yIrxNiy nanocrystals were synthesized by a one-step process at room temperature and showed excellent tolerance to poisoning and stability [24]. In addition, ternary PtIrCu nanocrystals exhibit high durability and toxicity tolerance due to their large surface area, composition, and strain effects [25]. Lee et al. [26] developed a carbon-loaded PtRuNi/C ternary electrocatalyst. Compared to Pt/C and PtRu/C catalysts, PtRuNi/C catalysts exhibit enhanced CO tolerance. PtNiCo ternary alloy nanoframe crystals exhibit excellent activity and durability as efﬁcient electrocatalysts for hydrogen evolution reaction (HER) [27]. The stable and highly efﬁcient ordered Pt2CoNi ternary alloy electrocatalyst has 5–6 times higher electrocatalytic ORR activity than commercial Pt/C catalysts [28]. Therefore, the ternary alloy Pt-Ni-Co nanoparticles have higher catalytic activities, better stability, and a CO anti-poisoning effect [29]. y p g [ ] Several methods to prepare electrocatalysis materials have been developed. Bhu- nia et al. [30] exhibited a simple one-pot and one-step solvothermal synthesis of PtAuNi nanoparticles as electrocatalysts with a diameter distribution of 3–7 nm by heating in an oven at 200 ◦C for 72 h. Sial et al. [31] used a typical hydrothermal method synthesis of trimetallic PtCoFe alloy nanosheets to obtain fuel cell catalysts with excellent electrocat- alytic activity and durability. Lee et al. [26] reported that a PtRuNi/C ternary metal-based electrocatalyst can be possibly used as a CO-tolerant anode catalyst for PEMFC. 1. Introduction Fuel cells have been considered green energy, renewable, and efﬁcient energy devices in recent years due to environmental and energy challenges [1]. Direct methanol fuel cells (DMFCs) have attracted signiﬁcant attention due to their high energy density and the abundance of liquid methanol [2,3]. The advantages of DMFCs are the simple structure of the system [4], low pollution [5], low operating temperature [6], and high energy conversion efﬁciency. They mainly convert methanol to produce electric energy through the catalyst of the electrode. Therefore, they can be used in small and portable electronic products such as laptops and mobile phones [7,8]. However, the electro-oxidation reaction of methanol is very complicated and slow, which is needed to improve the rate of electro-catalytic reaction. Moreover, the cost of DMFC systems is still very high, so the large-scale application of DMFCs is still quite limited [9]. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional afﬁl- iations. 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/). In DMFC, methanol is oxidized to produce H+, e−, and CO2 according to the reaction formula CH3OH + H2O →CO2 + 6H+ + 6e−. However, carbon–oxygen bonds in methanol are not easy to break, so the methanol electro-oxidation reaction (MOR) on the anode is crit- ical to the overall performance of DMFCs [10,11]. Pt-based alloys are widely used as highly reactive anode catalysts as an important component of DMFCs [12,13]. However, Pt-based alloys are easily hindered by the formation of toxic intermediates during electrochemical https://www.mdpi.com/journal/nanomaterials Nanomaterials 2021, 11, 2206. https://doi.org/10.3390/nano11092206 Nanomaterials 2021, 11, 2206 2 of 17 oxidation [12,14]. To avoid catalyst poisoning and oxygen reduction reaction (ORR), PtM alloys are proposed to have enhanced performance compared to pure platinum [13,15]. The electrocatalytic ability of Pt alloy increases because of the ligand effect and bifunctional effect [16]. It has been reported that Pt–Ni nanocrystals always have higher electrochemical activity, and Pt-Co nanocrystals are more stable than other Pt–M alloy materials attributed to the contribution of ﬁne-tuned electronic structure [17]. Compared with monometallic or bimetallic catalysts, the ternary metallic catalysts have better electrochemical performance and signiﬁcant catalytic activity. 1. Introduction In their study, a protective coating was used to prepare a product with a Pt-rich shell to prevent the Ni dissolution and sintering effect. Nugraha et al. [32] synthesized mesoporous AuCuNi alloy ﬁlms by electrodeposition from an electrolyte solution containing three metal pre- cursors with a micellar sacriﬁcial template at the ﬁxed applied potential. The mesoporous AuCuNi ﬁlms were synthesized for nonenzymatic glucose sensing with high sensitivity, se- lectivity, and low detection limit. Yang et al. [13] fabricated porous Pt–Pd nanoparticles by a reﬂux method. Hong et al. [33] developed a galvanic replacement method to obtain Pd–Pt with Pd nanocrystals with different shapes as sacriﬁcial templates. Wang, H. et al. [34] pre- pared a Pt nanocomposite by dopamine self-polymerization and a displacement reaction. Choi et al. [35] showed that extremely dispersed Pt and PtNi nanoparticles can be synthe- sized on supports by an impregnation process employing thiometallate precursors. Pd−Co nanowires with a jagged appearance were obtained via the template-conﬁned electrode- position ﬁrst and afterward excessive etching in phosphoric acid by Wang, C. et al. [14]. Wang, P. et al. [36] fabricated PtPdCu porous nanodendrites and nanocubes by using a surfactant assisted coreduction method with a solvent of water. However, the methods Nanomaterials 2021, 11, 2206 3 of 17 3 of 17 described above have several problems, including a comparatively low price–performance ratio, more elementary reaction steps, and inefﬁcient synthesis. Therefore, more economical and novel tactics need to be utilized in the synthesis of electrocatalysts to improve catalytic properties. In this study, we found a simple, efﬁcient, one-step synthesis, time-saving, environmentally friendly, non-metallic protective coating, and template-free method for the preparation of electrocatalysts. Microwave irradiation indicates ultra-high frequency electromagnetic waves with a certain wavelength. The range is from 1 m to 1 mm and the frequency range of 300 MHz–300 GHz [37]. The advantages of using microwaves compared to traditional heating methods are uniform heating, high speed [38], and energy efﬁciency. On the other hand, the heating material does not need to be in direct contact with the heat source, so the thermal resistance effect in the heat transfer process can be reduced. Microwave- assisted heating is a simple, effective, and energy-efﬁcient heating method that has been successfully applied in organic synthesis [39], functionalization of carbon nanotubes [40], and preparation of exfoliated graphite [41]. Pipus et al. 1. Introduction [42] found that the esteriﬁcation reaction of benzoic acid is a slow process and requires several days to reach equilibrium at 80 ◦C. However, microwave heating (140 ◦C, 7 atm) was able to increase the rate of the esteriﬁcation reaction in a short time. Yangá Lee et al. [43] showed that the nanoparticles obtained by microwave-assisted heating with the smallest particle size could be uniformly dispersed on the carbon carrier and had high electrocatalytic activity. In this study, a ternary metallic nanocatalyst was synthesized by the microwave- assisted method. In addition, the appropriate carbon support is also essential for the conduction of electrons and the dispersion of precious metal particles during the catalyst design process. Graphene is chosen as a carbon carrier due to its unique advantages such as large speciﬁc surface area, ﬂexible two-dimensional structure, high mechanical prop- erty [44], and good conductivity [17]. The loading of PtNiCo nanoparticles on graphene improves the utilization of the metal and its uniform dispersion on the graphene. It con- tributes to the accessibility of surface active sites and electron transfer kinetics [23]. The PtNiCo/rGO nanocomposite catalysts were prepared by GO with different ratios of Pt, Ni, Co precursors in an ethylene glycol solution microwave-assisted system at different temperatures for 20 min. In this process, a reductant is used to obtain better performance to eliminate oxygen-containing functional groups [17]. The PtNiCo particles were success- fully loaded on the supports and the structural characteristics of the prepared catalysts were evaluated by transmission electron microscopy (TEM), energy-dispersive X-ray spec- troscopy (EDX), X-ray diffraction (XRD), and Raman. The electrochemical measurements included cyclic voltammetry (CV) scanning, CO stripping, and chronoamperometry (CA). 2.3. Preparation of Samples 2.3. Preparation of Samples A 0.02 M K2PtCl6, 0.02 M NiCl2, 0.02 M Co(NO3)2 solution was mixed with 30 mL of ethylene glycol, 20 mg GO, and stirred for 30 min. The resulting homogeneous dark brown solution was adjusted to pH 10 with 0.2M KOH solution and transferred to 100 mL Teﬂon digestion vessels. The solution was heated at 200 ◦C for 20 min by a microwave system and then cooled to room temperature naturally. The samples were ﬁltered and dried in an oven at 70 ◦C. Nanocomposites at different temperatures were prepared using a similar process. PtNiCo/rGO was synthesized at 160, 180, 200, and 220 ◦C, labeled as PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220, respectively. PtNi2Co/rGO, PtNiCo/rGO, and PtNiCo2/rGO were synthesized in the mole ratios of Pt:Ni:Co 1:2:1, 1:1:1, and 1:1:2, respectively. A graph of the synthesis process is shown in Figure 1. A 0.02 M K2PtCl6, 0.02 M NiCl2, 0.02 M Co(NO3)2 solution was mixed with 30 mL of ethylene glycol, 20 mg GO, and stirred for 30 min. The resulting homogeneous dark brown solution was adjusted to pH 10 with 0.2M KOH solution and transferred to 100 mL Teflon digestion vessels. The solution was heated at 200 °C for 20 min by a microwave system and then cooled to room temperature naturally. The samples were filtered and dried in an oven at 70 °C. Nanocomposites at different temperatures were prepared using a similar process. PtNiCo/rGO was synthesized at 160, 180, 200, and 220 °C, labeled as PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220, respectively. PtNi2Co/rGO, PtNiCo/rGO, and PtNiCo2/rGO were synthesized in the mole ratios of Pt:Ni:Co 1:2:1, 1:1:1, and 1:1:2, respectively. A graph of the synthesis process is shown in Figure 1. Figure 1. Illustration of the formation of the PtNiCo/rGO. Figure 1. Illustration of the formation of the PtNiCo/rGO. Figure 1. Illustration of the formation of the PtNiCo/rGO. Figure 1. Illustration of the formation of the PtNiCo/rGO. 2.2. Fabrication of Graphene Oxide by a Modiﬁed Hummers Method 2.2. Fabrication of Graphene Oxide by a Modified Hummers Method GO was prepared by the modified Hummers method fro 2.2. Fabrication of Graphene Oxide by a Modiﬁed Hummers Method 2.2. Fabrication of Graphene Oxide by a Modified Hummers Method GO was prepared by the modified Hummers method fr GO was prepared by the modiﬁed Hummers method from graphite powder [45]. Seventy milliliters of H2SO4 was added to an ice bath and cooled to 5 ◦C. Graphite powder, NaNO3, and KMnO4 were added to the ﬂask and stirred well for 2 h. Then, 300 mL of deionized water was added and the color of the solution changed to yellowish brown, then 10 mL of 30% hydrogen peroxide was added to stop the reaction. After suction and ﬁltration, the sample was put into 500 mL of 5% hydrochloric acid to remove the metal ions, then washed repeatedly with DI water until neutral. The products were then dried overnight in an oven at 80 ◦C to obtain graphene oxide. GO was prepared by the modified Hummers method from graphite powder [45]. Seventy milliliters of H2SO4 was added to an ice bath and cooled to 5 °C. Graphite powder, NaNO3, and KMnO4 were added to the flask and stirred well for 2 h. Then, 300 mL of deionized water was added and the color of the solution changed to yellowish brown, then 10 mL of 30% hydrogen peroxide was added to stop the reaction. After suction and filtration, the sample was put into 500 mL of 5% hydrochloric acid to remove the metal ions, then washed repeatedly with DI water until neutral. The products were then dried overnight in an oven at 80 °C to obtain graphene oxide. 2.4. Preparation of Catalyst Ink 2.4. Preparation of Catalyst Ink The PtNiCo/rGO ink was prepared by adding 4 mg of sample powder to 0.8 mL of DI water and 0.2 mL of ethanol solution. The platinum working electrode was polished and used as a working electrode. It was coated with 15 μL of the above catalyst ink and 15 μL of Nafion ionomer, and dried at room temperature. After the electrodes were dried, electrochemical measurements could be carried out. The PtNiCo/rGO ink was prepared by adding 4 mg of sample powder to 0.8 mL of DI water and 0.2 mL of ethanol solution. The platinum working electrode was polished and used as a working electrode. It was coated with 15 µL of the above catalyst ink and 15 µL of Naﬁon ionomer, and dried at room temperature. After the electrodes were dried, electrochemical measurements could be carried out. 2.1. Materials Graphite powder (99.99%) and potassium hexachloroplatinate (K2PtCl6, 99.99%) were purchased from Alfa Aesar (Haverhill, MA, USA). Sodium nitrate (NaNO3, 99.5%), potas- sium permanganate (KMnO4, 99.3%), and nickel chloride hexahydrate (NiCl2·6H2O, 96%) were purchased from Hayashi Pure Chemical (Osaka, Japan) and cobalt(II) nitrate hex- ahydrate (Co(NO3)2·6H2O, 99%) was purchased from JT Baker Chemicals (Phillipsburg, NJ, USA). Hydrogen peroxide (H2O2, 30%) was purchased from Showa Chemical (Tokyo, Japan). Ethylene glycol (EG, 99.9%) was purchased from TEDIA (Fairﬁeld, OH, USA). Hydrochloric acid (HCl, >35%) and potassium hydroxide (KOH, 85%) were purchased from Union Chemical Works (Hsinchu, Taiwan). Liquid Naﬁon (5 wt%) was purchased from DuPont (Wilmington, DE, USA). Sulfuric acid (H2SO4, 97%), acetone ((CH3)2CO, 95%), and methanol (CH3OH, 99.5%) were purchased from Nihon Shiyaku Reagent (Kyoto, Japan). Millipore water (18 MΩ) was used for all electrochemistry measurements. All chemicals used in this experiment were analysis reagents (A.R.). 4 of 17 measu- Nanomaterials 2021, 11, 2206 2.6. Electrochemical Characterization The electrochemical measurements were carried out on a Bio-Logic CLB-500 electro- chemical Workstation (Knoxville, TN, USA) at room temperature. The electrochemical test was carried out with a three-electrode apparatus: a platinum working electrode (diameter: 2 mm), a Pt wire, and Ag/AgCl electrodes as the working, counter, and reference elec- trodes, respectively. The measured potentials are all compared with Ag/AgCl electrodes for convenience of comparison. Cyclic voltammetry (CV), CO stripping, and chronoam- perometry (CA) were employed to estimate the electrochemical activity and stability of catalysts. CV and CA techniques were employed to measure at room temperature and in a N2 saturated environment, while CO stripping is tested under saturated CO gas. The ECSA was calculated from the hydrogen desorption peak of the CV method in 0.5 M H2SO4 electrolyte, which was conducted by cycling the potential between −0.25 and 1.0 V, with a scan rate of 50 mV/s. CO stripping evaluates the ability of the catalyst to metabolize toxic substances, and its scanning potential is between −0.2 and 1.0 V, carried out in 0.5 M H2SO4 electrolyte. Electrochemical activity and catalytic ability of the catalysts for MOR were determined by CV and CA methods in 0.5 M H2SO4 + 1.0 M CH3OH electrolytic solution. 2.5. Structural Catalyst Characterization PtNiCo/rGO nanocomposite was synthesized by a microwave (Flexiwave T660, Mile- stone srl, Sorisole, Italy). The crystal structure of the samples was measured by X-ray diffraction (XRD, D8A25 eco, BRUKER Co. Ltd., Billerica, MA, USA) with CuKα X-ray radiation (λ = 1.5418 Å) at 40 kV and 25 mA. The morphology of the samples was observed by transmission electron microscopy (TEM, Hitachi H-7500, Tokyo, Japan) with an acceler- ating voltage of 80 kV. The elemental composition of the prepared nanocomposites was Nanomaterials 2021, 11, 2206 5 of 17 5 of 17 analyzed by energy dispersive X-ray analysis (EDS, INCA x-act) equipped with scanning electron microscopy (SEM, JEOL JSM-6390, Tokyo, Japan). Raman spectra were analyzed with a nitrogen-cooled CCD detector (Shamrock 750 spectrograph, Andor Technology Ltd., Belfast, Northern Ireland, UK). A randomly polarized 533 nm laser with an excitation power of 0.45 mW was used. 3.1.1. XRD Analysis The crystal structure of nanocomposites was analyzed by powder XRD. The diffraction peaks of these Pt-rich phases are fase center cubic (fcc) structures (Figure 2a) [17]. Note that the peaks become progressively broader after thermal treatment. The diffraction peaks are located at the corresponding angles of 39.7◦, 46.2◦, and 67.4◦, and the lattice constants are (111), (200), and (220) diffraction lattice planes, respectively. The diffraction peaks of Ni and Co are located at the top of the diffraction peak, corresponding to the angles of 44.5◦, 51.8◦, 76.4◦and 43.7◦, 51.0◦, 74.7◦, respectively. Based on the Joint Committee on Powder Diffraction Standards (JCPDS) card number #04-0802, it is determined that Pt metal represents Pt (111), Pt (200), Pt (220) crystallographic planes. In the PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 samples, we found a much broader peak at 2θ = 26◦, while the diffraction peak at 2θ = 11.9◦disappeared signiﬁcantly. This indicates that GO has been reduced to rGO (Figure 2b). The diffraction peak at 26◦ can be marked as rGO, with no additional peaks for phase separation structures such as pure Ni or Co. This indicates an excellent degree of alloying between Pt, Ni, and Co [46]. The XRD results showed that the diffraction peaks of Ni and Co in composites are not obvious and may be related to the small amount of Ni and Co in composites [47–49]. However, the presence of Ni and Co can also be explained by a slight shift of the Pt (111) peak to a higher angle in the XRD analysis. The peak position of Pt (111) shifts to higher 2θ values due to the introduction of smaller Ni and Co atoms, resulting in reduced lattice distances and a composite with Pt to form the PtNiCo ternary alloy [28,50,51]. The catalysts exhibited broader shoulder peaks, assigned to the characteristic peak of the rGO support [52]. Moreover, the higher diffraction shifts of Pt (111) of the synthesized nanocomposites at 200 ◦C indicate that the synthesis is relatively complete and PtNiCo ternary alloy is better. 6 of 17 Nanomaterials 2021, 11, 2206 pea synth ure 2. XRD patterns of (a) PtNiCo/rGO of different temperatures. (b) GO and rGO. Figure 2. XRD patterns of (a) PtNiCo/rGO of different temperatures. (b) GO and rGO. RD patterns of (a) PtNiCo/rGO of different temperatures. (b) GO and rGO. Figure 2. XRD patterns of (a) PtNiCo/rGO of different temperatures. 3.1.1. XRD Analysis (b) GO and rGO. phological Characterization 3.1.2. Morphological Characterization phological Characterization re 3a–c show the synthesized PtNiCo/rGO ternary alloys with different ratios. phology of the ternary alloy/reduced graphene oxide nanocomposite was by TEM. Moreover, the dispersion of the catalyst was observed to find the opti- o of the atomic catalyst. In these samples, the TEM diagrams display two-dimen- ages of PtNiCo/rGO 200 and PtNiCo2/rGO 200 with a characteristic size of about 93 nm and 22.28 ± 3.09 nm (Figure 4a,b). The ternary alloy PtNiCo/rGO 200 has ti l t i f l th 20 t Figure 3a–c show the synthesized PtNiCo/rGO ternary alloys with different ratios. The morphology of the ternary alloy/reduced graphene oxide nanocomposite was analyzed by TEM. Moreover, the dispersion of the catalyst was observed to ﬁnd the optimum ratio of the atomic catalyst. In these samples, the TEM diagrams display two-dimensional images of PtNiCo/rGO 200 and PtNiCo2/rGO 200 with a characteristic size of about 17.03 ± 1.93 nm and 22.28 ± 3.09 nm (Figure 4a,b). The ternary alloy PtNiCo/rGO 200 has an average particulate size of less than 20 nanometers. 7 of 17 7 f 18 7 of 18 Nanomaterials 2021, 11, 2206 l O i l O Figure 3. Morphology and structural characterization. TEM images of (a) PtNi2Co/rGO 200, (b) PtNiCo/rGO 200, (c) PtNiCo2/rGO 200, (d) PtNiCo/rGO 160, (e) PtNiCo/rGO 180, (f) PtNiCo/rGO 220. Figure 3. Morphology and structural characterization. TEM images of (a) PtNi2Co/rGO 200, (b) PtNiCo/rGO 200, (c) PtNiCo2/rGO 200, (d) PtNiCo/rGO 160, (e) PtNiCo/rGO 180, (f) PtNiCo/rGO 220. Figure 3. Morphology and structural characterization. TEM images of (a) PtNi2Co/rGO 200, (b) PtNiCo/rGO 200, (c) PtNiCo2/rGO 200, (d) PtNiCo/rGO 160, (e) PtNiCo/rGO 180, (f) PtNiCo/rGO 220. Figure 3. Morphology and structural characterization. TEM images of (a) PtNi2Co/rGO 200, (b) PtNiCo/rGO 200, (c) PtNiCo2/rGO 200, (d) PtNiCo/rGO 160, (e) PtNiCo/rGO 180, (f) PtNiCo/rGO 220. Figure 3. Morphology and structural characterization. TEM images of (a) PtNi2Co/rGO 200, (b) PtNiCo/rGO 200, (c) PtNiCo2/rGO 200, (d) PtNiCo/rGO 160, (e) PtNiCo/rGO 180, (f) PtNiCo/rGO 220. Figure 3. Morphology and structural characterization. TEM images of (a) PtNi2Co/rGO 200, (b) PtNiCo/rGO 200, (c) PtNiCo2/rGO 200, (d) PtNiCo/rGO 160, (e) PtNiCo/rGO 180, (f) PtNiCo/rGO 220. Figure 4. Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, (e) PtNiCo/rGO 220. Figure 4. 3.1.1. XRD Analysis The main reason for the small particle size is the short time and fast rate of nu- cleation at optimal temperature conditions [54,55]. Under other synthesis conditions, the average particle size was about 20 nm (Figure 4). The temperature-dependent agglomera- tion is the dominant mechanism of the particle size difference [56,57]. The results showed that the dispersed PtNiCo nanoparticles were easily grown on rGO and were active on electrochemical properties [58]. p p The composition of the synthesized nanocomposites was obtained by energy dis- persive X-ray analysis (EDS) equipped with SEM for evaluation. The EDX spectrum of PtNiCo/rGO 200 in Figure 5 shows that it consists mainly of three metal elements, Pt, Ni, and Co, in the sample. The peaks of Ni and Co can be clearly detected in the EDX image, and the atomic percentages of Pt, Ni, and Co are 2.73, 2.67, and 2.69%, respectively. The atomic composition of the composite is almost identical to that of the metal precur- sor solution. The EDX image demonstrates the presence of C elements in rGO, while O elements are mainly derived from residual oxygen-containing functional groups in rGO. Other elements can also be observed in the ﬁgure, mainly added from the material during the experiments. The appearance of Cu peaks originates from the copper gels used in the analysis [59]. The EDX images provide evidence for the presence of Pt, Ni, Co, rGO, and PtNiCo atomic ratios in the electrocatalyst. e ect oc e ica p ope ties [58]. The composition of the synthesized nanocomposites was obtained by energy disper- sive X-ray analysis (EDS) equipped with SEM for evaluation. The EDX spectrum of PtNiCo/rGO 200 in Figure 5 shows that it consists mainly of three metal elements, Pt, Ni, and Co, in the sample. The peaks of Ni and Co can be clearly detected in the EDX image, and the atomic percentages of Pt, Ni, and Co are 2.73, 2.67, and 2.69%, respectively. The atomic composition of the composite is almost identical to that of the metal precursor so- lution. The EDX image demonstrates the presence of C elements in rGO, while O elements are mainly derived from residual oxygen-containing functional groups in rGO. Other el- ements can also be observed in the figure, mainly added from the material during the experiments. The appearance of Cu peaks originates from the copper gels used in the analysis [59]. 3.1.1. XRD Analysis Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, (e) PtNiCo/rGO 220. Figure 4. Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, (e) PtNiCo/rGO 220. Figure 4. Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, Figure 4. Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, Figure 4. Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, Figure 4. Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, (e) PtNiCo/rGO 220. Figure 4. Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, (e) PtNiCo/rGO 220. Figure 4. Particle size distribution of (a) PtNiCo/rGO 200, (b) PtNiCo2/rGO 200, (c) PtNiCo/rGO 160, (d) PtNiCo/rGO 180, (e) PtNiCo/rGO 220. Furthermore, PtNiCo/rGO in Figure 3 shows a large number of PtNiCo nanoparticles surrounded by the rGO nanosheets. The metal precursor solutions were reduced to PtNiCo ternary alloy nanoparticles with ethylene glycol at different microwave tempera- tures, which were dispersed on the surface or embedded in the layered structure of rGO, as shown in Figure 3d f The 2D sheet like structure and slight folds can be observed in Furthermore, PtNiCo/rGO in Figure 3 shows a large number of PtNiCo nanoparticles surrounded by the rGO nanosheets. The metal precursor solutions were reduced to PtNiCo ternary alloy nanoparticles with ethylene glycol at different microwave tempera- tures, which were dispersed on the surface or embedded in the layered structure of rGO, as shown in Figure 3d f The 2D sheet like structure and slight folds can be observed in Furthermore, PtNiCo/rGO in Figure 3 shows a large number of PtNiCo nanoparticles surrounded by the rGO nanosheets. The metal precursor solutions were reduced to PtNiCo ternary alloy nanoparticles with ethylene glycol at different microwave temperatures, which were dispersed on the surface or embedded in the layered structure of rGO, as Nanomaterials 2021, 11, 2206 8 of 17 8 of 17 shown in Figure 3d–f. The 2D sheet-like structure and slight folds can be observed in TEM images. Figure 4 shows the particle size and distribution of various samples and various reaction conditions. 3.1.1. XRD Analysis The particles formed gradually at 160 ◦C and 180 ◦C, but the reaction showed that the metal ions did not fully composite into the ternary metallic nanoparticles. The particles have a mean radius of about 21 nm and a wide distribution width. The particle size decreases as the temperature rises to 200 ◦C. This is due to the larger number of seeds growing and the particles reacting more thoroughly. The metal ions can be converted to ternary metallic nanoparticles and exhibit an average radius of 17.03 ± 1.93 nm and a narrower distribution width. Larger particles are formed as the temperature increases (Figure 4a,b). At this time, the agglomeration mechanism leads to more clusters agglomerating at higher temperatures and forming nanoparticles [53]. The results show that the average particle size of the synthesized nanoparticles at 200 ◦C is the smallest, 17 nm, and has a relatively dispersed structure compared with other synthesis conditions. The main reason for the small particle size is the short time and fast rate of nucleation at optimal temperature conditions [54,55]. Under other synthesis conditions, the average particle size was about 20 nm (Figure 4). The temperature-dependent agglomera- tion is the dominant mechanism of the particle size difference [56,57]. The results showed that the dispersed PtNiCo nanoparticles were easily grown on rGO and were active on electrochemical properties [58]. TEM images. Figure 4 shows the particle size and distribution of various samples and various reaction conditions. The particles formed gradually at 160 °C and 180 °C, but the reaction showed that the metal ions did not fully composite into the ternary metallic na- noparticles. The particles have a mean radius of about 21 nm and a wide distribution width. The particle size decreases as the temperature rises to 200 °C. This is due to the larger number of seeds growing and the particles reacting more thoroughly. The metal ions can be converted to ternary metallic nanoparticles and exhibit an average radius of 17.03 ± 1.93 nm and a narrower distribution width. Larger particles are formed as the tem- perature increases (Figure 4a,b). At this time, the agglomeration mechanism leads to more clusters agglomerating at higher temperatures and forming nanoparticles [53]. The results show that the average particle size of the synthesized nanoparticles at 200 °C is the sma- llest, 17 nm, and has a relatively dispersed structure compared with other synthesis con- ditions. 3.1.1. XRD Analysis The EDX images provide evidence for the presence of Pt, Ni, Co, rGO, and PtNiCo atomic ratios in the electrocatalyst Figure 5. EDX spectrum of PtNiCo/rGO 200 composite. Figure 5. EDX spectrum of PtNiCo/rGO 200 composite. Figure 5. EDX spectrum of PtNiCo/rGO 200 composite. 3.1.3. Raman Spectrum Figure 6a shows the Raman spectra of the samples in the range of 1000–3000 cm−1 They all display obvious D bands and G bands originating from carbon sp2 domains and structural defects [60]. The G band is attributed to the E2g mode of C sp2 atoms and the D band arises due to the A1g symmetry [61]. The structural disorder of a graphitic structure could be estimated according to the ID/IG. It can be seen from Figure 6a,b that the D-band Figure 5. EDX spectrum of PtNiCo/rGO 200 composite. 3.1.3. Raman Spectrum Figure 6a shows the Raman spectra of the samples in the range of 1000–3000 cm−1. They all display obvious D bands and G bands originating from carbon sp2 domains and structural defects [60]. The G band is attributed to the E2g mode of C sp2 atoms and the D band arises due to the A1g symmetry [61]. The structural disorder of a graphitic structure Figure 5. EDX spectrum of PtNiCo/rGO 200 composite. Figure 5. EDX spectrum of PtNiCo/rGO 200 composite. 3.1.3. Raman Spectrum 3.1.3. Raman Spectrum 3.1.3. Raman Spectrum 3.1.3. Raman Spectrum Figure 6a shows the Raman spectra of the samples in the range of 1000–3000 cm−1. They all display obvious D bands and G bands originating from carbon sp2 domains and structural defects [60]. The G band is attributed to the E2g mode of C sp2 atoms and the D band arises due to the A1g symmetry [61]. The structural disorder of a graphitic structure ld b ti t d di t th I /I It b f Fi 6 b th t th D b d Figure 6a shows the Raman spectra of the samples in the range of 1000–3000 cm−1. They all display obvious D bands and G bands originating from carbon sp2 domains and structural defects [60]. The G band is attributed to the E2g mode of C sp2 atoms and the D band arises due to the A1g symmetry [61]. The structural disorder of a graphitic structure 9 of 17 Nanomaterials 2021, 11, 2206 could be estimated according to the ID/IG. It can be seen from Figure 6a,b that the D- band and G-band of GO, rGO, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 were 1339 cm−1 and 1586 cm−1, 1346 cm−1 and 1594 cm−1, 1356 cm−1 and 1590 cm−1, 1356 cm−1 and 1596 cm−1, 1358 cm−1 and 1594 cm−1, and 1363 cm−1 and 1590 cm−1, respectively. The ID/IG of GO, rGO, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 were 0.81, 1.07, 1.16, 1.15, 1.18, and 1.15, respectively. Afterwards, GO was reduced to rGO because large numbers of sp3 carbon were reduced to sp2 carbon, which increases the ID/IG value. In general, the D-band and G-band intensities were reduced after the PtNiCo nanoparticles were composited with reduced graphene oxide. This is because the exposred area of the rGO sheet to the excitation light in PtNiCo/rGO nanocomposite is reduced in Raman measurements. The broadening of the D-band and G-band of the nanocomposite is caused by the lattice strain between rGO and PtNiCo [62]. nd of GO, rGO, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and GO 220 were 1339 cm−1 and 1586 cm−1, 1346 cm−1 and 1594 cm−1, 1356 cm−1 and 1356 cm−1 and 1596 cm−1, 1358 cm−1 and 1594 cm−1, and 1363 cm−1 and 1590 cm−1, ly. The ID/IG of GO, rGO, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, Co/rGO 220 were 0.81, 1.07, 1.16, 1.15, 1.18, and 1.15, respectively. 3.1.3. Raman Spectrum 3.1.3. Raman Spectrum Afterwards, educed to rGO because large numbers of sp3 carbon were reduced to sp2 carbon, reases the ID/IG value. In general, the D-band and G-band intensities were re- er the PtNiCo nanoparticles were composited with reduced graphene oxide. This the exposred area of the rGO sheet to the excitation light in PtNiCo/rGO nano- e is reduced in Raman measurements. The broadening of the D-band and G-band ocomposite is caused by the lattice strain between rGO and PtNiCo [62]. re 6. Raman spectra of (a) PtNiCo/rGO at various temperatures. (b) GO and rGO. Figure 6. Raman spectra of (a) PtNiCo/rGO at various temperatures. (b) GO and rGO. aman spectra of (a) PtNiCo/rGO at various temperatures. (b) GO and rGO. Figure 6. Raman spectra of (a) PtNiCo/rGO at various temperatures. (b) GO and rGO. aman spectra of (a) PtNiCo/rGO at various temperatures. (b) GO and rGO. Figure 6. Raman spectra of (a) PtNiCo/rGO at various temperatures. (b) GO and rGO. Nanomaterials 2021, 11, 2206 10 of 17 10 of 17 The ID/IG value of PtNiCo/rGO is higher than that of GO, indicating that the double bond of GO is broken and composited with metal nanoparticles. The larger the ID/IG value, the more successfully the sample is composited [19]. In addition, the ID/IG ratio could be related to the existence of defects in graphene structure as a result of electronic interaction with PtNiCo metal nanoparticles, afﬁrming the reduction of the functional groups during microwave-assisted treatment. These defects introduced by GO act as the anchoring sites for the attachment of PtNiCo metal nanoparticles. According to the description in the literature, the increase in the intensity is related to the PtNiCo nanoparticles incorporated into the rGO as both peaks increased similarly in intensity [63]. A higher degree of graphitization is beneﬁcial to promote the overall conductivity of the ﬁnal product and enhance the electrochemical activity. In addition, the crystallinity of the 2D band (about 2700 cm−1) in PtNiCo/rGO 200 is higher, which is predicted to be more corrosion resistant in DMFC [61]. The wide 2D band shows the multilayer structure of rGO, afﬁrming the existence of graphene and mainly coming from a double resonance process that links phonons to the electronic band structure [64]. The 2D and D + G band peaks, near 2700 and 2900 cm−1, correspond to the combination mode induced by the disorder. the more successfully the sample is composited [19]. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l To obtain the ECSA of the catalysts, they were prepared as a slurry and coated on the Pt working electrode. In this study, CV was used to measure and analyze the characteristics of the catalysts. Figure 7 shows the CV curves of the different catalysts recorded in N2-purged sulfuric acid solution at a scan rate of 50 mV/s in the potential range of −0.2 V to 1.0 V. CV curves are divided into three parts to show the typical Pt-H under the potential deposition region, double-layer region, and Pt oxide region. Typical absorption and desorption of hydrogen occur in the low potential region. There is a signiﬁcant redox peak from −0.2~0.1 V, indicating the adsorption/desorption of hydrogen on Pt [65]. To obtain the ECSA of the catalysts, they were prepared as a slurry and coated on the Pt working electrode. In this study, CV was used to measure and analyze the characteris- tics of the catalysts. Figure 7 shows the CV curves of the different catalysts recorded in N2-purged sulfuric acid solution at a scan rate of 50 mV/s in the potential range of −0.2 V to 1.0 V. CV curves are divided into three parts to show the typical Pt-H under the poten- tial deposition region, double-layer region, and Pt oxide region. Typical absorption and desorption of hydrogen occur in the low potential region. There is a significant redox peak from −0.2~0.1 V, indicating the adsorption/desorption of hydrogen on Pt [65]. Figure 7. Electrocatalytic performance of PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 catalysts. Cyclic voltammograms obtained at room temperature in N2- purged 0.5 M H2SO4 aqueous solution at scan rate of 50 mV/s. Figure 7. Electrocatalytic performance of PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 catalysts. Cyclic voltammograms obtained at room temperature in N2-purged 0.5 M H2SO4 aqueous solution at scan rate of 50 mV/s. Figure 7. Electrocatalytic performance of PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 catalysts. Cyclic voltammograms obtained at room temperature in N2- purged 0.5 M H2SO4 aqueous solution at scan rate of 50 mV/s. Figure 7. Electrocatalytic performance of PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 catalysts. Cyclic voltammograms obtained at room temperature in N2-purged 0.5 M H2SO4 aqueous solution at scan rate of 50 mV/s. 3.1.3. Raman Spectrum 3.1.3. Raman Spectrum In addition, the ID/IG ratio could be related to the existence of defects in graphene structure as a result of electronic interaction with PtNiCo metal nanoparticles, affirming the reduction of the functional groups during microwave-assisted treatment. These defects introduced by GO act as the anchoring sites for the attachment of PtNiCo metal nanoparticles. According to the description in the lit- erature, the increase in the intensity is related to the PtNiCo nanoparticles incorporated into the rGO as both peaks increased similarly in intensity [63]. A higher degree of graph- itization is beneficial to promote the overall conductivity of the final product and enhance the electrochemical activity. In addition, the crystallinity of the 2D band (about 2700 cm−1) in PtNiCo/rGO 200 is higher, which is predicted to be more corrosion resistant in DMFC [61]. The wide 2D band shows the multilayer structure of rGO, affirming the existence of graphene and mainly coming from a double resonance process that links phonons to the electronic band structure [64]. The 2D and D + G band peaks, near 2700 and 2900 cm−1, correspond to the combination mode induced by the disorder. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l The corresponding ECSA was calculated by integrating the hydrogen desorption zones and all the CV curves were normalized. According to the following formula [66]: The corresponding ECSA was calculated by integrating the hydrogen desorption zones and all the CV curves were normalized. According to the following formula [66]: ECSA (cmଶ gିଵ) = charge (ܳୌ, μC cm ) 210 (μC cmିଶ) × electrode loading (g Pt cmିଶ) QH (μC cm−2) is the hydrogen desorption charge 210 (μC cm−2) is the ECSA cm2g−1 = charge QH, µC cm−2 210 (µC cm−2) × electrodeloading (g Ptcm−2) Nanomaterials 2021, 11, 2206 11 of 17 11 of 17 where QH (µC cm−2) is the hydrogen desorption charge, 210 (µC cm−2) is the charge required to oxidize the layer of hydrogen on Pt, and the electrode loading (g Pt cm−2) is the Pt working electrode loading. The individual ECSAs of these catalysts were measured as 18.19, 61.43, 82.65, 87.41, and 41.41 m2/g for PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220. The PtNiCo/rGO 200 ECSA is larger than the others because of the dispersed and well-anchored Pt nanoparticles on the surface of rGO nanocomposites. The nanocomposite of 200 ◦C has a good distribution of ﬁne PtNiCo nanoparticles, which is attributed to the largest ECSA. This result can be seen from the TEM image, which has the smallest particle size distribution under the 200 ◦C condition. Pt working electrode loading. The individual ECSAs of these catalysts were measured as 18.19, 61.43, 82.65, 87.41, and 41.41 m2/g for PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220. The PtNiCo/rGO 200 ECSA is larger than the oth- ers because of the dispersed and well-anchored Pt nanoparticles on the surface of rGO nanocomposites. The nanocomposite of 200 °C has a good distribution of fine PtNiCo na- noparticles, which is attributed to the largest ECSA. This result can be seen from the TEM image, which has the smallest particle size distribution under the 200 °C condition. h d ff l f h l f h To investigate the different ratio catalysts in terms of the catalytic activity of methanol oxidation, CV measurements were performed (Figure 8). In the forward scan, the positive scan anode peak is about 0.7 V compared to Ag/AgCl, which is caused by the oxidation of methanol. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l In the reverse scan, the oxidation peak appears at about 0.5 V, possibly owing to incompletely oxidized carbonic matters formed in the forward scan. The current density of CVs at ∼0.7 V on PtNiCo/rGO 200 is 196.82 mA cm−2, which is about 2.5 times larger than the PtNiCo (80.89 mA cm−2). To clearly illustrate the catalytic capacity, MOR performance is compared by normalizing peak currents to speciﬁc areas and qualities of Pt expressed as mass activity [67]. The mass activity of all catalysts is shown in Figure 9. It can be observed that PtNiCo/rGO 200 has the highest mass activity (102.96 mA mg−1) compared with other catalysts, indicating that the PtNiCo/rGO 200 catalysts had the highest catalytic activity for MOR. To investigate the different ratio catalysts in terms of the catalytic activity of metha- nol oxidation, CV measurements were performed (Figure 8). In the forward scan, the pos- itive scan anode peak is about 0.7 V compared to Ag/AgCl, which is caused by the oxida- tion of methanol. In the reverse scan, the oxidation peak appears at about 0.5 V, possibly owing to incompletely oxidized carbonic matters formed in the forward scan. The current density of CVs at ∼0.7 V on PtNiCo/rGO 200 is 196.82 mA cm−2, which is about 2.5 times larger than the PtNiCo (80.89 mA cm−2). To clearly illustrate the catalytic capacity, MOR performance is compared by normalizing peak currents to specific areas and qualities of Pt expressed as mass activity [67]. The mass activity of all catalysts is shown in Figure 9. It can be observed that PtNiCo/rGO 200 has the highest mass activity (102.96 mA mg−1) compared with other catalysts, indicating that the PtNiCo/rGO 200 catalysts had the high- est catalytic activity for MOR. Figure 8. CV curves of methanol oxidation on PtNi2Co/rGO, PtNiCo/rGO, PtNiCo, and PtNiCo2/rGO in N2-saturated 0.5 M H2SO4 + 1.0 M CH3OH solution, sweep rate 20 mV/s. Figure 8. CV curves of methanol oxidation on PtNi2Co/rGO, PtNiCo/rGO, PtNiCo, and PtNiCo2/rGO in N2-saturated 0.5 M H2SO4 + 1.0 M CH3OH solution, sweep rate 20 mV/s. Figure 8. CV curves of methanol oxidation on PtNi2Co/rGO, PtNiCo/rGO, PtNiCo, and PtNiCo2/rGO in N2-saturated 0.5 M H2SO4 + 1.0 M CH3OH solution, sweep rate 20 mV/s. Figure 8. CV curves of methanol oxidation on PtNi2Co/rGO, PtNiCo/rGO, PtNiCo, and PtNiCo2/rGO in N2-saturated 0.5 M H2SO4 + 1.0 M CH3OH solution, sweep rate 20 mV/s. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l The results show that PtNiCo/rGO 200 has the highest catalytic activity and mass activity in these synthesized catalysts. The high performance of PtNiCo/rGO is attributed to the electronic state of Pt, which is mainly because the crystal structure and electronic structure of Pt nanoparticles can be changed by introducing other metals, thus improving the binding energy between Pt and toxic species [9,68]. In addition, the TEM images of PtNiCo/rGO 200 can signiﬁcantly improve the catalytic activity due to the dispersion of small PtNiCo nanoparticles on rGO [59]. This is attributed to the fact that the presence of Ni and Co can mitigate Pt poisoning, resulting in the higher electrochemical activity of PtNiCo/rGO [69,70]. From the Raman spectrum, PtNiCo/rGO 200 has the highest ID/IG intensity, meaning that a greater degree of composite PtNiCo and rGO, which will also enhance the electrocatalytic activity [71]. 12 of 17 12 of 18 Nanomaterials 2021, 11, 2206 Nanomaterials 2021, 11, x FOR Figure 9. Histogram of mass activities of different catalysts for MOR. Figure 9. Histogram of mass activities of different catalysts for MOR. Figure 9. Histogram of mass activities of different catalysts for MOR. Figure 9. Histogram of mass activities of different catalysts for MOR. The results show that PtNiCo/ 3.2.2. CO Stripping Measurements activity in these synthesized catalysts. The high performance of PtNiCo/rGO is attributed to the electronic state of Pt, which is mainly because the crystal structure and electronic structure of Pt nanoparticles can be changed by introducing other metals, thus improving the binding energy between Pt and toxic species [9,68]. In addition, the TEM images of PtNiCo/rGO 200 can significantly improve the catalytic activity due to the dispersion of small PtNiCo nanoparticles on rGO [59]. This is attributed to the fact that the presence of Ni and Co can mitigate Pt poisoning, resulting in the higher electrochemical activity of PtNiCo/rGO [69,70]. From the Raman spectrum, PtNiCo/rGO 200 has the highest ID/IG in- tensity, meaning that a greater degree of composite PtNiCo and rGO, which will also en- hance the electrocatalytic activity [71]. 3.2.2. CO Stripping Measurements CO stripping experiments were carried out on the oxidative removal of CO. The CO gas is pumped to make the CO adhere to the catalyst at a lower potential, and then the CV CO stripping experiments were carried out on the oxidative removal of CO. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l The CO gas is pumped to make the CO adhere to the catalyst at a lower potential, and then the CV test is carried out. Figure 10 shows the voltammograms of CO oxidation for various samples at various peak potentials. In addition, the CO oxidation potentials (vs. Ag/AgCl) of PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 were 0.69 V, 0.64 V, 0.62 V, and 0.65 V vs. Ag/AgCl, respectively. These peaks disappear after the ﬁrst scan in the forward direction. The PtNiCo/rGO 200 expresses a more negative peak potential and onset potential, indicating that the afﬁnity between Pt and CO is weakened [72]. The results show that the introduction of Ni and Co elements can improve the oxidation ability of CO [73]. The electronic interaction between Pt, Ni, and Co could lead to the removal of CO poisoning substances and enhance the stability of the electrodes. Therefore, there are fewer intermediates adsorbed on the surface of PtNiCo/rGO 200, and the oxidation efﬁciency of COads is higher, which is beneﬁcial to increase catalytic active sites [73]. gas is pumped to make the CO adhere to the catalyst at a lower potential, and then the CV test is carried out. Figure 10 shows the voltammograms of CO oxidation for various sam- ples at various peak potentials. In addition, the CO oxidation potentials (vs. Ag/AgCl) of PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 were 0.69 V, 0.64 V, 0.62 V, and 0.65 V vs. Ag/AgCl, respectively. These peaks disappear after the first scan in the forward direction. The PtNiCo/rGO 200 expresses a more negative peak po- tential and onset potential, indicating that the affinity between Pt and CO is weakened [72]. The results show that the introduction of Ni and Co elements can improve the oxi- dation ability of CO [73]. The electronic interaction between Pt, Ni, and Co could lead to the removal of CO poisoning substances and enhance the stability of the electrodes. There- fore, there are fewer intermediates adsorbed on the surface of PtNiCo/rGO 200, and the oxidation efficiency of COads is higher, which is beneficial to increase catalytic active sites [73] The synergistic effect of CO electrooxidation can be clearly seen from the oxidation onset potential to more negative values. Due to the catalytic effect, PtNiCo/rGO 200 could activate CO at lower potentials than the other catalysts. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l This result also contributes to illustrating the higher activity of PtNiCo/rGO for the oxidation of methanol at 200 ◦C [74]. Table 1 shows the comparison of the ECSA, forward peak current density for methanol oxidation, and CO oxidation potential of PtNiCo/rGO 200 to those of various electrocata- lysts investigated in previous studies. The ECSA and methanol oxidation current density of PtNiCo/rGO 200 were higher than those of the other materials previously studied. These results indicated the enhanced Pt hydrogen absorption/desorption area and methanol electrocatalytic activity of PtNiCo/rGO 200. In addition, the CO oxidation potential of PtNiCo/rGO 200 was mostly lower than other materials, showing better CO anti-poisoning ability. gas is pumped to make the CO adhere to the catalyst at a lower potential, and then the CV test is carried out. Figure 10 shows the voltammograms of CO oxidation for various sam- ples at various peak potentials. In addition, the CO oxidation potentials (vs. Ag/AgCl) of PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 were 0.69 V, 0 64 V 0 62 V a d 0 65 V A /A Cl e e ti ely The e eak di a ea afte the fi t The synergistic effect of CO electrooxidation can be clearly seen from the oxidation onset potential to more negative values. Due to the catalytic effect, PtNiCo/rGO 200 could activate CO at lower potentials than the other catalysts. This result also contributes to illustrating the higher activity of PtNiCo/rGO for the oxidation of methanol at 200 ◦C [74]. , , g/ g , p y p pp scan in the forward direction. The PtNiCo/rGO 200 expresses a more negative peak po- tential and onset potential, indicating that the affinity between Pt and CO is weakened [72]. The results show that the introduction of Ni and Co elements can improve the oxi- dation ability of CO [73]. The electronic interaction between Pt, Ni, and Co could lead to the removal of CO poisoning substances and enhance the stability of the electrodes. There- fore, there are fewer intermediates adsorbed on the surface of PtNiCo/rGO 200, and the oxidation efficiency of COads is higher, which is beneficial to increase catalytic active sites [73] Table 1 shows the comparison of the ECSA, forward peak current density for methanol oxidation, and CO oxidation potential of PtNiCo/rGO 200 to those of various electrocata- lysts investigated in previous studies. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l The ECSA and methanol oxidation current density of PtNiCo/rGO 200 were higher than those of the other materials previously studied. These results indicated the enhanced Pt hydrogen absorption/desorption area and methanol electrocatalytic activity of PtNiCo/rGO 200. In addition, the CO oxidation potential of PtNiCo/rGO 200 was mostly lower than other materials, showing better CO anti-poisoning ability. 13 of 17 13 of 18 Nanomaterials 2021, 11, 2206 Nanomaterials 2021, 11, x FOR 13 of 17 13 of 18 Figure 10. CO stripping curves of (a) PtNiCo/rGO 160, (b) PtNiCo/rGO 180, (c) PtNiCo/rGO 200, and (d) PtNiCo/rGO 220 in CO−purged 0.5 M H2SO4 solution at scan rate of 50 mV/s. Figure 10. CO stripping curves of (a) PtNiCo/rGO 160, (b) PtNiCo/rGO 180, (c) PtNiCo/rGO 200, and (d) PtNiCo/rGO 220 in CO−purged 0.5 M H2SO4 solution at scan rate of 50 mV/s. Figure 10. CO stripping curves of (a) PtNiCo/rGO 160, (b) PtNiCo/rGO 180, (c) PtNiCo/rGO 200, and (d) PtNiCo/rGO 220 in CO−purged 0.5 M H2SO4 solution at scan rate of 50 mV/s. Figure 10. CO stripping curves of (a) PtNiCo/rGO 160, (b) PtNiCo/rGO 180, (c) PtNiCo/rGO 200, and (d) PtNiCo/rGO 220 in CO−purged 0.5 M H2SO4 solution at scan rate of 50 mV/s. The synergistic effect of CO electrooxidation can be clearly seen from the onset potential to more negative values Due to the catalytic effect PtNiCo/rGO Table 1. ECSA and current density of methanol oxidation of various electrocatalysts in previous works. The synergistic effect of CO electrooxidation can be clearly seen from onset potential to more negative values Due to the catalytic effect PtNiCo/r Table 1. ECSA and current density of methanol oxidation of various electrocatalysts in previous works. onset potential to more negative values. Due to the catalytic effect, PtNiCo/rGO 200 could activate CO at lower potentials than the other catalysts. This result also contributes to il- lustrating the higher activity of PtNiCo/rGO for the oxidation of methanol at 200 °C [74]. Table 1 shows the comparison of the ECSA, forward peak current density for meth- anol oxidation, and CO oxidation potential of PtNiCo/rGO 200 to those of various electro- catalysts investigated in previous studies. The ECSA and methanol oxidation current den- sity of PtNiCo/rGO 200 were higher than those of the other materials previously studied. These results indicated the enhanced Pt hydrogen absorption/desorption area and meth- anol electrocatalytic activity of PtNiCo/rGO 200. 4. Conclusions 4. Conclusions In summary, a ternary alloy structured was prepared with one-step microwave-as- sisted reduction for uniformly dispersed Pt-Ni-Co nanoparticles supported on rGO. The PtNiCo/rGO exhibited the highest electrochemical properties and the smallest average grain diameter (17 nm) at a microwave reaction temperature of 200 °C. The aggregation of PtNiCo is maximally suppressed by the rGO, attributed to the unique two-dimensional flexible microstructure of rGO. In addition, PtNiCo/rGO 200 had the highest ID/IG values, indicating an increase in reduction. The PtNiCo/rGO nanocomposite presents excellent electrocatalytic ability in terms of high electrocatalytic activity, high poison tolerance, en- hanced stability toward MOR compared with PtNiCo. Electrochemical measurements show that PtNiCo/rGO 200 nanoalloy displays functionality enhancement in both mass and catalytic activities over two times that of the pure PtNiCo catalyst. The unique dis- persion of rGO and the synergistic effect between Pt, Ni, and Co improve the catalytic performance of PtNiCo/rGO composites. To satisfy the challenges of rapid fabrication and low environmental impact, we obtained PtNiCo/rGO using a rapid synthesis method with In summary, a ternary alloy structured was prepared with one-step microwave- assisted reduction for uniformly dispersed Pt-Ni-Co nanoparticles supported on rGO. The PtNiCo/rGO exhibited the highest electrochemical properties and the smallest average grain diameter (17 nm) at a microwave reaction temperature of 200 ◦C. The aggregation of PtNiCo is maximally suppressed by the rGO, attributed to the unique two-dimensional ﬂexible microstructure of rGO. In addition, PtNiCo/rGO 200 had the highest ID/IG values, indicating an increase in reduction. The PtNiCo/rGO nanocomposite presents excellent electrocatalytic ability in terms of high electrocatalytic activity, high poison tolerance, en- hanced stability toward MOR compared with PtNiCo. Electrochemical measurements show that PtNiCo/rGO 200 nanoalloy displays functionality enhancement in both mass and catalytic activities over two times that of the pure PtNiCo catalyst. The unique dis- persion of rGO and the synergistic effect between Pt, Ni, and Co improve the catalytic performance of PtNiCo/rGO composites. To satisfy the challenges of rapid fabrication and low environmental impact, we obtained PtNiCo/rGO using a rapid synthesis method with a simple process and low-cost precursors. The PtNiCo/rGO electrocatalysts have the potential to be used as catalysts with high electrocatalytic activity, CO resistance, and stability in DMFC. It provides a fast and reduced energy consumption fabrication for designing other high-performance catalysts, which is a great prospect in the application of fuel cell catalyst materials for the future. Author Contributions: K.-Y.S. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l These results Nanomaterials 2021, 11, 2206 14 of 17 s due to eraction 14 of 17 s due to eraction conﬁrm that PtNiCo/rGO 200 shows great stability and higher intermediates of poison tolerance as a superior catalyst for MOR in DMFCs. on reduced graphene oxide also contributes to the catalytic activity [77]. These results confirm that PtNiCo/rGO 200 shows great stability and higher intermediates of poison tolerance as a superior catalyst for MOR in DMFCs. conﬁrm that PtNiCo/rGO 200 shows great stability and higher intermediates of poison tolerance as a superior catalyst for MOR in DMFCs. on reduced graphene oxide also contributes to the catalytic activity [77]. These results confirm that PtNiCo/rGO 200 shows great stability and higher intermediates of poison tolerance as a superior catalyst for MOR in DMFCs. Figure 11. Chronoamperometric curves of PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 in 0.5 M H2SO4 + 1.0 M CH3OH solution at constant potential of 0.70 V. 4 Conclusions Figure 11. Chronoamperometric curves of PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 in 0.5 M H2SO4 + 1.0 M CH3OH solution at constant potential of 0.70 V. 4 Conclusions Figure 11. Chronoamperometric curves of PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 in 0.5 M H2SO4 + 1.0 M CH3OH solution at constant potential of 0.70 V. Figure 11. Chronoamperometric curves of PtNiCo, PtNiCo/rGO 160, PtNiCo/rGO 180, PtNiCo/rGO 200, and PtNiCo/rGO 220 in 0.5 M H2SO4 + 1.0 M CH3OH solution at constant potential of 0.70 V. 3.2.1. CV and Mass Activity Analysis 3.2.1. CV and Mass Activity Analysis T bt i th ECSA f th t l In addition, the CO oxidation potential of PtNiCo/rGO 200 was mostly lower than other materials, showing better CO anti-poi- soning ability. y y p Electrocatalyst ECSA (m2/g) Current Density (mA cm−2) CO Stripping (V) References PtNiCo/rGO 200 87.41 196.82 0.62 vs. Ag/AgCl This work Pt/BG 58.8 ~1.7 ~0.8 vs. RHE [4] Pt-Ni/CNF 1:2 - ~2 ~0.7 vs. RHE [10] Pt-Pd (9:1) 31.59 0.67 - [13] Pt3Pd1-CeO2/C 30.33 ~4 ~1.0 vs. RHE [15] Hollow Pt-Ni-Co NDs 57.0 3.8 ~0.5 vs. SCE [17] PtCoFe 62.9 4.75 - [23] Au41Cu46Ni13 45.8 3.8 - [25] PtRuFe/rGO 56.4 1.33 - [45] Pd59Fe27Pt14 NMs - 4.36 - [50] nd current density of methanol oxidation 3.2.3. Chronoamperometric Study d current density of methanol oxidation of various electrocatalysts in previous works. 3.2.3. Chronoamperometric Study ECSA (m2/g) Current Density (mA cm−2) CO Stripping (V) References 87.41 196.82 0.62 vs. Ag/AgCl This work 58.8 ~1.7 ~0.8 vs. RHE [4] - ~2 ~0.7 vs. RHE [10] 31.59 0.67 - [13] 30.33 ~4 ~1.0 vs. RHE [15] 57.0 3.8 ~0.5 vs. SCE [17] 62.9 4.75 - [23] 45.8 3.8 - [25] 56.4 1.33 - [45] - 4.36 - [50] The catalytic stability of the catalyst for MOR in acidic media was further investigated by CA. Figure 11 shows the CA curve for the variation in current density with time. As the experiment progresses, the current decay of these catalysts slows down and gradually achieves quasi-steady state [75]. This is due to the formation of oxidation intermediate species such as CO, CH3OH, and CHO [76]. This causes the oxidation of methanol to produce adsorption on the Pt surface to hinder the active site. After testing at t = 600 s, the current density of PtNiCo/rGO 200 (65.92 mA/cm2) is still higher than that of PtNiCo (8.28 mA/cm2), PtNiCo/rGO 160 (45.86 mA/cm2), PtNiCo/rGO 180 (29.94 mA/cm2) and PtNiCo/rGO 220 (10.83 mA/cm2). The results show that PtNiCo/rGO 200 retains the highest steady current density and the highest initial current density when compared to the others. The catalysts supported on rGO show better stability than PtNiCo. This is due to the graphene sheets providing a number of oxygen groups to strengthen the interaction with Pt nanoparticles. As previously indicated, the existence of residual functional groups on reduced graphene oxide also contributes to the catalytic activity [77]. Conﬂicts of Interest: The authors declare no conﬂict of interest. References 1. Liu, M.; Yu, F.; Ma, C.; Xue, X.; Fu, H.; Yuan, H.; Yang, S.; Wang, G.; Guo, X.; Zhang, L. Effective Oxygen Reduction Reaction Performance of FeCo Alloys In Situ Anchored on Nitrogen-Doped Carbon by the Microwave-Assistant Carbon Bath Method and Subsequent Plasma Etching. Nanomaterials 2019, 9, 1284. [CrossRef] q g 2. Glüsen, A.; Dionigi, F.; Paciok, P.; Heggen, M.; Müller, M.; Gan, L.; Strasser, P.; Dunin-Borkowski, R.E.; Stolten, D. Dealloyed PtNi-Core–Shell Nanocatalysts Enable Signiﬁcant Lowering of Pt Electrode Content in Direct Methanol Fuel Cells. ACS Catal. 2019, 9, 3764–3772. [CrossRef] 3. Rezaee, S.; Shahrokhian, S.; Amini, M.K. Nanocomposite with Promoted Electrocatalytic Behavior Based on Bimetallic Pd–Ni Nanoparticles, Manganese Dioxide, and Reduced Graphene Oxide for Efﬁcient Electrooxidation of Ethanol. J. Phys. Chem. C 2018, 122, 9783–9794. [CrossRef] 4. Sun, Y.; Du, C.; An, M.; Du, L.; Tan, Q.; Liu, C.; Gao, Y.; Yin, G. Boron-doped graphene as promising sup with superior activity towards the methanol electrooxidation reaction. J. Power Sources 2015, 300, 245–2 with superior activity towards the methanol electrooxidation reaction. J. Power Sources 2015, 300, 245–253. [CrossRef] 5. Noor, T.; Pervaiz, S.; Iqbal, N.; Nasir, H.; Zaman, N.; Sharif, M.; Pervaiz, E. Nanocomposites of NiO/CuO Based MOF with rGO: oor, T.; Pervaiz, S.; Iqbal, N.; Nasir, H.; Zaman, N.; Sharif, M.; Pervaiz, E. Nanocomposites of NiO/CuO Base n Efﬁcient and Robust Electrocatalyst for Methanol Oxidation Reaction in DMFC. Nanomaterials 2020, 10, 16 6. Veizaga, N.S.; Rodriguez, V.I.; Bruno, M.; de Miguel, S.R. The Role of Surface Functionalities in PtGe and PtIn Catalysts for Direct Methanol Fuel Cells. Electrocatalysis 2018, 10, 125–133. [CrossRef] y 7. He, M.; Fei, G.; Zheng, Z.; Cheng, Z.; Wang, Z.; Xia, H. Pt Nanoparticle-Loaded Graphene Aerogel Microspheres with Excellent Methanol Electro-Oxidation Performance. Langmuir 2019, 35, 3694–3700. [CrossRef] 8. Menshchikov, V.; Alekseenko, A.; Guterman, V.; Nechitailov, A.; Glebova, N.; Tomasov, A.; Spiridonova, O.; Belenov, S.; Zelenina, N.; Safronenko, O. Effective Platinum-Copper Catalysts for Methanol Oxidation and Oxygen Reduction in Proton-Exchange Membrane Fuel Cell. Nanomaterials 2020, 10, 742. [CrossRef] mbrane Fuel Cell. Nanomaterials 2020, 10, 742. [CrossRe 9. Arán-Ais, R.M.; Dionigi, F.; Merzdorf, T.; Gocyla, M.; Heggen, M.; Dunin-Borkowski, R.E.; Gliech, M.; Solla-Gullón, J.; Herrero, E.; Feliu, J.M.; et al. Elemental Anisotropic Growth and Atomic-Scale Structure of Shape-Controlled Octahedral Pt-Ni-Co Alloy Nanocatalysts. Nano Lett. 2015, 15, 7473–7480. [CrossRef] y 10. Calderón, J.C.; Ráfales, M.R.; Nieto-Monge, M.J.; Pardo, J.I.; Moliner, R.; Lázaro, M.J. 4. Conclusions 4. Conclusions conceptualized the idea and designed the experiments; J.-J.W. per- formed the experiments; K.-Y.S., J.-J.W. and M.-C.T. analyzed the data; K.-Y.S. acquired funding; K.-Y.S., J.-J.W. and M.-C.T. explained and visualized the data and wrote the manuscript. All authors have read and agreed to the published version of the manuscript. Funding: This work was ﬁnancially supported by the National Pingtung University of Taiwan (Grant No. NPTU-110-002 and Grant No. NPTU-110-009). Data Availability Statement: All the data are available within the manuscript. Data Availability Statement: All the data are available within the manuscript. Conﬂicts of Interest: The authors declare no conﬂict of interest. 15 of 17 Nanomaterials 2021, 11, 2206 References Oxidation of CO and Methanol on Pd-Ni Catalysts Supported on Different Chemically-Treated Carbon Nanoﬁbers. Nanomaterials 2016, 6, 187. [CrossRef] 11. Rambabu, G.; Bhat, S.D.; Figueiredo, F.M.L. Carbon Nanocomposite Membrane Electrolytes for Direct Methanol Fuel Cells-A Concise Review. Nanomaterials 2019, 9, 1292. [CrossRef] [PubMed] 12. Higareda, A.; Kumar-Krishnan, S.; García-Ruiz, A.F.; Maya-Cornejo, J.; Lopez-Miranda, J.L.; Bahena, D.; Rosas, G.; Pérez, R.; Esparza, R. Synthesis of Au@Pt Core-Shell Nanoparticles as Efﬁcient Electrocatalyst for Methanol Electro-Oxidation. Nanomaterials 2019, 9, 1644. [CrossRef] [PubMed] [ ] [ ] 13. Yang, Y.; Cao, Y.; Yang, L.; Huang, Z.; Long, N.V. Synthesis of Pt-Pd Bimetallic Porous Nanostructures as Electrocatalysts for the Methanol Oxidation Reaction. Nanomaterials 2018, 8, 208. [CrossRef] [PubMed] 14. Wang, C.; Zheng, L.; Chang, R.; Du, L.; Zhu, C.; Geng, D.; Yang, D. Palladium-Cobalt Nanowires Decorated with Jagged Appearance for Efﬁcient Methanol Electro-oxidation. ACS Appl. Mater. Interfaces 2018, 10, 29965–29971. [CrossRef] [PubMed] 15. Yousaf, A.B.; Imran, M.; Uwitonze, N.; Zeb, A.; Zaidi, S.J.; Ansari, T.M.; Yasmeen, G.; Manzoor, S. Enhanced Electrocatalytic Performance of Pt3Pd1 Alloys Supported on CeO2/C for Methanol Oxidation and Oxygen Reduction Reactions. J. Phys. Chem. C 2017, 121, 2069–2079. [CrossRef] 16. Huang, J.; Liu, Z.; He, C.; Gan, L.M. Synthesis of PtRu Nanoparticles from the Hydrosilylation Reaction and Application as Catalyst for Direct Methanol Fuel Cell. J. Phys. Chem. B 2005, 109, 16644–16649. [CrossRef] 17. Sriphathoorat, R.; Wang, K.; Shen, P.K. Trimetallic Hollow Pt–Ni–Co Nanodendrites as Efﬁcient Anodic Electrocatalysts. ACS Appl. Energy Mater. 2019, 2, 961–965. [CrossRef] 18. Ghoshal, S.; Jia, Q.; Bates, M.K.; Li, J.; Xu, C.; Gath, K.; Yang, J.; Waldecker, J.; Che, H.; Liang, W.; et al. Tuning Nb-Pt Interactions to Facilitate Fuel Cell Electrocatalysis. ACS Catal. 2017, 7, 4936–4946. [CrossRef] 19. Rethinasabapathy, M.; Kang, S.M.; Haldorai, Y.; Jankiraman, M.; Jonna, N.; Choe, S.R.; Huh, Y.S.; Natesan, B. Ternary PtRuFe nanoparticles supported N-doped graphene as an efﬁcient bifunctional catalyst for methanol oxidation and oxygen reduction reactions. Int. J. Hydrogen Energy 2017, 42, 30738–30749. [CrossRef] J y g gy 20. Sun, B.; Huang, L.; Su, S.; Luo, S.; Meng, C.; Basit, H.M.A.; Xiao, J.; Bian, T.; Su, S. PtRhCu ternary alloy nanodendrites with enhanced catalytic activity and durability toward methanol electro-oxidation. Mater. Chem. Phys. 2020, 252, 123234. [CrossRef] 21. Sui, N.; Wang, T.; Bai, Q.; Yue, R.; Jiang, H.; Xiao, H.; Liu, M.; Wang, L.; Zhu, Z.; Yu, W.W. 23. Wang, C.; Zhang, Y.; Zhang, Y.; Xu, P.; Feng, C.; Chen, T.; Guo, T.; Yang, F.; Wang, Q.; Wang, J.; et al. Highly Ordered Hierarchical Pt and PtNi Nanowire Arrays for Enhanced Electrocatalytic Activity toward Methanol Oxidation. ACS Appl. Mater. Interfaces 2018, 10, 9444–9450. [CrossRef] References Thiometallate precursors for the synthesis of supported Pt and PtNi nanoparticle electrocatalysts: Size-focusing by S capping. Nanoscale 2020, 12, 10498–10504. [CrossRef] [ ] 36. Wang, P.; Zhang, Y.; Shi, R.; Wang, Z. Shape-Controlled Synthesis of Trimetallic PtPdCu Nanocrystals and Their Electrocatalytic Properties. ACS Appl. Energy Mater. 2019, 2, 2515–2523. [CrossRef] 37. Wang, Y.; Hou, Q.; Ju, M.; Li, W. New Developments in Material Preparation Using a Combination of Ionic Liquids and Microwave Irradiation. Nanomaterials 2019, 9, 647. [CrossRef] 38. Ramasahayam, S.K.; Nasini, U.B.; Bairi, V.; Shaikh, A.U.; Viswanathan, T. Microwave assisted synthesis and characterization of silicon and phosphorous co-doped carbon as an electrocatalyst for oxygen reduction reaction. RSC Adv. 2014, 4, 6306. [CrossRef] ll d h d h h d [ f] s co p osp o o s co ope c o s e ec oc ys o o yge e c o e c o , , 6306 [C oss e ] 39. Kappe, C.O. Controlled microwave heating in modern organic synthesis. Angew. Chem. Int. Ed. 2004, 43, 6250–6284. [CrossRef] 40. Brunetti, F.G.; Herrero, M.A.; Muñoz, J.d.M.; Dìaz-Ortiz, A.; Alfonsi, J.; Meneghetti, M.; Prato, M.; Vàzquez, E. Microwave-Induced Multiple Functionalization of Carbon Nanotubes. J. Am. Chem. Soc. 2008, 130, 8094–8100. [CrossRef] p p p y yg 39. Kappe, C.O. Controlled microwave heating in modern organic synthesis. Angew. Chem. Int. Ed. 2004, 4 pp g g y g 40. Brunetti, F.G.; Herrero, M.A.; Muñoz, J.d.M.; Dìaz-Ortiz, A.; Alfonsi, J.; Meneghetti, M.; Prato, M.; Vàzquez, E. Microwave-Induced Multiple Functionalization of Carbon Nanotubes. J. Am. Chem. Soc. 2008, 130, 8094–8100. [CrossRef] 41. Hou, S.; He, S.; Zhu, T.; Li, J.; Ma, L.; Du, H.; Shen, W.; Kang, F.; Huang, Z.H. E graphite and functional graphite sheets. J. Mater. 2021, 7, 136–145. [CrossRef] 41. Hou, S.; He, S.; Zhu, T.; Li, J.; Ma, L.; Du, H.; Shen, W.; Kang, F.; Huang, Z.H. Environment-friendly preparation of exfoliated graphite and functional graphite sheets. J. Mater. 2021, 7, 136–145. [CrossRef] 42. Pipus, G.; Plazl, I.; Koloini, T. Esteriﬁcation of benzoic acid in microwave tubular ﬂow reactor. Chem. E [CrossRef] 43. Liao, C.S.; Liao, C.T.; Tso, C.Y.; Shy, H.J. Microwave-polyol synthesis and electrocatalytic performance of Pt/graphene nanocom- posites. Mater. Chem. Phys. 2011, 130, 270–274. [CrossRef] p y 44. Ramirez-Barria, C.S.; Fernandes, D.M.; Freire, C.; Villaro-Abalos, E.; Guerrero-Ruiz, A.; Rodríguez-Ramos, I. References Hong, J.W.; Kang, S.W.; Choi, B.S.; Kim, D.; Lee, S.B.; Han, S.W. Controlled Synthesis of Pd Pt Alloy Hollow Nanostructures with with Controlled Compositions Using Block Copolymer Micelles as Templates. Small Methods 2018, 2, 1800283. [CrossRef] 33. Hong, J.W.; Kang, S.W.; Choi, B.S.; Kim, D.; Lee, S.B.; Han, S.W. Controlled Synthesis of Pd Pt Alloy Hollow Nanostructures with Enhanced Catalytic Activities for Oxygen Reduction ACS Nano 2012 6 2410–2419 [CrossRef] p g p y p [ ] 33. Hong, J.W.; Kang, S.W.; Choi, B.S.; Kim, D.; Lee, S.B.; Han, S.W. Controlled Synthesis of Pd Pt Alloy Hollow Nanostructures with Enhanced Catalytic Activities for Oxygen Reduction. ACS Nano 2012, 6, 2410–2419. [CrossRef] 33. Hong, J.W.; Kang, S.W.; Choi, B.S.; Kim, D.; Lee, S.B.; Han, S.W. Controlled Synthesis of Pd Pt Al Enhanced Catalytic Activities for Oxygen Reduction. ACS Nano 2012, 6, 2410–2419. [CrossRef] g, J ; g, ; , ; , ; , ; , y Enhanced Catalytic Activities for Oxygen Reduction. ACS Nano 2012, 6, 2410–2419. [CrossRef] 34. Wang, H.; Lu, S.; Xu, W.; Wu, B.; He, G.; Cui, S.; Zhang, Y. Synthesis of a Pt/reduced graphene oxide/polydopamine composite material for localized surface plasmon resonance and methanol electrocatalysis. New J. Chem. 2018, 42, 19458–19466. [CrossRef] 34. Wang, H.; Lu, S.; Xu, W.; Wu, B.; He, G.; Cui, S.; Zhang, Y. Synthesis of a Pt/reduced graphene oxide/polydopamine composite material for localized surface plasmon resonance and methanol electrocatalysis. New J. Chem. 2018, 42, 19458–19466. [CrossRef] 35. Choi, W.; Kim, J.M.; Hwang, C.K.; Choe, M.; Baek, S.; Ban, H.W.; Gu, D.H.; Jeong, H.; Chae, K.H.; Lim, Y.; et al. Thiometallate precursors for the synthesis of supported Pt and PtNi nanoparticle electrocatalysts: Size-focusing by S capping. Nanoscale 2020, 34. Wang, H.; Lu, S.; Xu, W.; Wu, B.; He, G.; Cui, S.; Zhang, Y. Synthesis of a Pt/reduced graphene oxide/polydopamine composite material for localized surface plasmon resonance and methanol electrocatalysis. New J. Chem. 2018, 42, 19458–19466. [CrossRef] 35. Choi, W.; Kim, J.M.; Hwang, C.K.; Choe, M.; Baek, S.; Ban, H.W.; Gu, D.H.; Jeong, H.; Chae, K.H.; Lim, Y.; et al. Thiometallate precursors for the synthesis of supported Pt and PtNi nanoparticle electrocatalysts: Size-focusing by S capping. Nanoscale 2020, 12, 10498–10504. [CrossRef] 35. Choi, W.; Kim, J.M.; Hwang, C.K.; Choe, M.; Baek, S.; Ban, H.W.; Gu, D.H.; Jeong, H.; Chae, K.H.; Lim, Y.; et al. References Preparation of CO-tolerant PtRuNi/C ternary electrocatalyst having a compos Eng. J. 2021, 414, 128792. [CrossRef] g J [ ] 27. Oh, A.; Sa, Y.J.; Hwang, H.; Baik, H.; Kim, J.; Kim, B.; Joo, S.H.; Lee, K. Rational design of Pt–Ni–Co ternary alloy nanoframe crystals as highly efﬁcient catalysts toward the alkaline hydrogen evolution reaction. Nanoscale 2016, 8, 16379–16386. [CrossRef] 27. Oh, A.; Sa, Y.J.; Hwang, H.; Baik, H.; Kim, J.; Kim, B.; Joo, S.H.; Lee, K. Rational design of Pt–Ni–Co ternary alloy nanoframe l hi hl fﬁi l d h lk li h d l i i N l 8 16379 16386 [C R f] 27. Oh, A.; Sa, Y.J.; Hwang, H.; Baik, H.; Kim, J.; Kim, B.; Joo, S.H.; Lee, K. Rational design of Pt–Ni–Co ternary alloy nanoframe crystals as highly efﬁcient catalysts toward the alkaline hydrogen evolution reaction. Nanoscale 2016, 8, 16379–16386. [CrossRef] 28. Lokanathan, M.; Patil, I.M.; Navaneethan, M.; Parey, V.; Thapa, R.; Kakade, B. Designing of stable and highly efﬁcient ordered y g y y y g 28. Lokanathan, M.; Patil, I.M.; Navaneethan, M.; Parey, V.; Thapa, R.; Kakade, B. Designing of stable and highly efﬁcient ordered Pt2CoNi ternary alloy electrocatalyst: The origin of dioxygen reduction activity. Nano Energy 2018, 43, 219–227. [CrossRef] 2 y y y g yg y gy , , [ ] 29. Min, J.; Jeffery, A.A.; Kim, Y.; Jung, N. Electrochemical Analysis for Demonstrating CO Tolerance of Catalysts in Polymer Electrolyte Membrane Fuel Cells. Nanomaterials 2019, 9, 1425. [CrossRef] 29. Min, J.; Jeffery, A.A.; Kim, Y.; Jung, N. Electrochemical Analysis for Demonstrating CO Tolera Electrolyte Membrane Fuel Cells. Nanomaterials 2019, 9, 1425. [CrossRef] 30. Bhunia, K.; Khilari, S.; Pradhan, D. Trimetallic PtAuNi Alloy Nanoparticles as an Efﬁcient Electrocatalyst for Methanol Electrooxi- dation Reaction. Dalton Trans. 2017, 46, 15558–15566. [CrossRef] 31. Sial, M.; Lin, H.; Zulﬁqar, M.; Ullah, S.; Ni, B.; Wang, X. Trimetallic PtCoFe Alloy Monolayer Sup Oxygen-Reduction and Ethanol-Oxidation Electrocatalysts. Small 2017, 13, 1700250. [CrossRef] yg y 32. Nugraha, A.S.; Malgras, V.; Kim, J.; Bo, J.; Li, C.; Iqbal, M.; Yamauchi, Y.; Asahi, T. Trimetallic Mesoporous AuCuNi Electrocatalysts with Controlled Compositions Using Block Copolymer Micelles as Templates. Small Methods 2018, 2, 1800283. [CrossRef] 32. Nugraha, A.S.; Malgras, V.; Kim, J.; Bo, J.; Li, C.; Iqbal, M.; Yamauchi, Y.; Asahi, T. Trimetallic Mesoporous AuCuNi Electrocatalysts with Controlled Compositions Using Block Copolymer Micelles as Templates. Small Methods 2018, 2, 1800283. [CrossRef] 33. References Facile synthesis of Ternary Au@PdNi core-shell nanoparticles with enhanced electrocatalytic performance for ethanol oxidation reaction. J. Alloys Compd. 2020, 817, 153335. [CrossRef] y g gy 20. Sun, B.; Huang, L.; Su, S.; Luo, S.; Meng, C.; Basit, H.M.A.; Xiao, J.; Bian, T.; Su, S. PtRhCu ternary alloy nanodendrites with enhanced catalytic activity and durability toward methanol electro-oxidation. Mater. Chem. Phys. 2020, 252, 123234. [CrossRef] 21. Sui, N.; Wang, T.; Bai, Q.; Yue, R.; Jiang, H.; Xiao, H.; Liu, M.; Wang, L.; Zhu, Z.; Yu, W.W. Facile synthesis of Ternary Au@PdNi y y y y , , [ ] 21. Sui, N.; Wang, T.; Bai, Q.; Yue, R.; Jiang, H.; Xiao, H.; Liu, M.; Wang, L.; Zhu, Z.; Yu, W.W. Facile synthesis of Ternary Au@PdNi core-shell nanoparticles with enhanced electrocatalytic performance for ethanol oxidation reaction. J. Alloys Compd. 2020, 817, 153335. [CrossRef] 22. Baronia, R.; Goel, J.; Kaswan, J.; Shukla, A.; Singhal, S.K.; Singh, S.P. PtCo/rGO nano-anode catalyst: Enhanced power density with reduced methanol crossover in direct methanol fuel cell. Mater. Renew. Sustain. Energy 2018, 7, 27. [CrossRef] 23. Wang, C.; Zhang, Y.; Zhang, Y.; Xu, P.; Feng, C.; Chen, T.; Guo, T.; Yang, F.; Wang, Q.; Wang, J.; et al. Highly Ordered Hierarchical Pt and PtNi Nanowire Arrays for Enhanced Electrocatalytic Activity toward Methanol Oxidation. ACS Appl. Mater. Interfaces 2018, 10, 9444–9450. [CrossRef] 16 of 17 16 of 17 Nanomaterials 2021, 11, 2206 24. Ahmad, Y.H.; Mohamed, A.T.; Youssef, K.M.; Kundu, S.; Mkhoyan, K.A.; Al-Qaradawi, S.Y. Rational synthesis of ternary PtIrNi nanocrystals with enhanced poisoning tolerance for electrochemical ethanol oxidation. Electrochem. Commun. 2019, 101, 61–67. [CrossRef] 25. Ahmad, Y.H.; El-Sayed, H.A.; Mohamed, A.T.; Aljaber, A.S.; Al-Qaradawi, S.Y. Rational one-pot syn nanocrystals as robust electrocatalyst for methanol oxidation reaction. Appl. Surf. Sci. 2020, 534, 14761 25. Ahmad, Y.H.; El-Sayed, H.A.; Mohamed, A.T.; Aljaber, A.S.; Al-Qaradawi, S.Y. Rational one-pot synthesis of ternary PtIrCu nanocrystals as robust electrocatalyst for methanol oxidation reaction. Appl. Surf. Sci. 2020, 534, 147617. [CrossRef] 26. Lee, H.; Park, S.; Kim, H. Preparation of CO-tolerant PtRuNi/C ternary electrocatalyst having a composition gradient shell. Chem. Eng. J. 2021, 414, 128792. [CrossRef] nanocrystals as robust electrocatalyst for methanol oxidation reaction. Appl. Surf. Sci. 2020, 534, 147617. [CrossRef] 26. Lee, H.; Park, S.; Kim, H. Preparation of CO-tolerant PtRuNi/C ternary electrocatalyst having a composition gradient shell. Chem. Eng. J. 2021, 414, 128792. [CrossRef] 26. Lee, H.; Park, S.; Kim, H. p p y 51. Lokanathan, M.; Patil, I.M.; Kakade, B. Trimetallic PtNiCo nanoﬂowers as efﬁcient electrocatalysts to reaction. Int. J. Hydrogen Energy 2018, 43, 8983–8990. [CrossRef] References [CrossRef] 60. Wang, Z.; Zhang, F.; Liu, Q.; Wang, Z. Electrochemical Synthesis of PtNiCo/3DGN Composite Catalyst and Its Performance for Methanol Electrooxidation. Int. J. Electrochem. Sci. 2017, 12, 10958–10968. [CrossRef] 61. Kasturi, S.; Torati, S.R.; Eom, Y.J.; Ahmad, S.; Lee, B.J.; Yu, J.S.; Kim, C. Real-time monitored photocatalytic activity and electrochemical performance of an rGO/Pt nanocomposite synthesized via a green approach. RSC Adv. 2020, 10, 13722. [CrossRef] 62. Mishra, A.; Singh, V.K.; Mohanty, T. Coexistence of interfacial stress and charge transfer in graphene oxide-based magnetic nanocomposites. J. Mater. Sci. 2017, 52, 7677–7687. [CrossRef] 63. Marinoiu, A.; Andrulevicius, M.; Tamuleviciene, A.; Tamulevicius, T.; Carcadea, E.; Raceanu, M.; Varlam, M. High performance catalytic system with enhanced durability in PEM fuel cell. Int. J. Hydrogen Energy 2020, 45, 10409–10422. [CrossRef] 64. Antony, R.P.; Preethi, L.K.; Gupta, B.; Mathews, T.; Dash, S.; Tyagi, A.K. Efﬁcient electrocatalytic p exfoliated reduced graphene oxide-Pt hybrid. Mater. Res. Bull. 2015, 70, 60–67. [CrossRef] 65. Woo, J.H.; Park, S.J.; Chung, S.; Kim, S. Effect of polydopamine-modiﬁed reduced graphene oxides on the catalytic activity of Pt nanoparticles catalysts for fuel cell electrodes. Carbon Lett. 2019, 29, 47–55. [CrossRef] p y 66. Ji, W.; Qi, W.; Tang, S.; Peng, H.; Li, S. Hydrothermal Synthesis of Ultrasmall Pt Nanoparticles as Highly Active Electrocatalysts for Methanol Oxidation. Nanomaterials 2015, 5, 2203–2211. [CrossRef] [PubMed] 66. Ji, W.; Qi, W.; Tang, S.; Peng, H.; Li, S. Hydrothermal Synthesis of Ultrasmall Pt Nanoparticles as High for Methanol Oxidation. Nanomaterials 2015, 5, 2203–2211. [CrossRef] [PubMed] 67. Luo, X.; Liu, C.; Wang, X.; Shao, Q.; Pi, Y.; Zhu, T.; Li, Y.; Huang, X. Spin Regulation on 2D Pd-Fe-Pt Nanomeshes Promotes Fuel Electrooxidations. Nano Lett. 2020, 20, 1967–1973. [CrossRef] 67. Luo, X.; Liu, C.; Wang, X.; Shao, Q.; Pi, Y.; Zhu, T.; Li, Y.; Huan Electrooxidations. Nano Lett. 2020, 20, 1967–1973. [CrossRef] 68. Liu, Y.; Wang, Y.; Zhang, J.; Shi, S.; Feng, P.; Wang, T. Observation of surface structural changes of Pt octahedron nanoparticles and its effect in electrocatalysis oxidation of methanol. Catal. Commun. 2009, 10, 1244–1247. [CrossRef] 69. Marinoiu, A.; Raceanu, M.; Carcadea, E.; Andrulevicius, M.; Tamuleviciene, A.; Tamulevicius, T.; Capris, C.; Varlam, M. Efﬁcient method to obtain Platinum-Cobalt supported on graphene oxide and electrocatalyst development. Int. J. Hydrogen Energy 2020, 45, 26226–26237. [CrossRef] 70. Baronia, R.; Goel, J.; Tiwari, S.; Singh, P.; Singh, D.; Singh, S.P.; Singhal, S.K. References ORR activity and stability of PtCr/C catalysts in a low tempera- ture/pressure PEM fuel cell: Effect of heat treatment temperature. Int. J. Hydrogen Energy 2018, 43, 5133–5144. [CrossRef] p p 53. Saloga, P.E.J.; Kästner, C.; Thünemann, A.F. High-Speed but N Nanoparticles. Langmuir 2018, 34, 147–153. [CrossRef] p p J y g gy 53. Saloga, P.E.J.; Kästner, C.; Thünemann, A.F. High-Speed but Not Magic: Microwave-Assisted Synthesis of Ultra-Small Silver Nanoparticles. Langmuir 2018, 34, 147–153. [CrossRef] 53. Saloga, P.E.J.; Kästner, C.; Thünemann, A.F. High Speed Nanoparticles. Langmuir 2018, 34, 147–153. [CrossRef] g , J ; , ; , g p g y Nanoparticles. Langmuir 2018, 34, 147–153. [CrossRef] p g 54. Nassr, A.B.A.A.; Sinev, I.; Pohl, M.M.; Grünert, W.; Bron, M. Rapid microwave-assisted polyol reduction for the preparation of highly active PtNi/CNT electrocatalysts for methanol oxidation. ACS Catal. 2014, 4, 2449–2462. [CrossRef] p g , A.B.A.A.; Sinev, I.; Pohl, M.M.; Grünert, W.; Bron, M. Rapid microwave-assisted polyol reduction for the 54. Nassr, A.B.A.A.; Sinev, I.; Pohl, M.M.; Grünert, W.; Bron, M. Rapid microwave-assisted polyol reduction for the preparation of highly active PtNi/CNT electrocatalysts for methanol oxidation ACS Catal 2014 4 2449 2462 [CrossRef] p p y p p CNT electrocatalysts for methanol oxidation. ACS Catal. 2014, 4, 2449–2462. [CrossRef] highly active PtNi/CNT electrocatalysts for methanol oxidation. ACS Catal. 2014, 4, 2449–2462. [CrossRef 55. McGinty, J.; Chong, M.W.S.; Manson, A.; Brown, C.J.; Nordon, A.; Sefcik, J. Effect of Process Conditions on Particle Size and Shape in Continuous Antisolvent Crystallisation of Lovastatin. Crystals 2020, 10, 925. [CrossRef] 56. Clark, M.D.; Kumar, S.K.; Owen, J.S.; Chan, E.M. Focusing Nanocrystal Size Distributions via Production Control. Nano Lett. 2011, 11, 1976–1980. [CrossRef] 57. Kozakova, Z.; Kuritka, I.; Kazantseva, N.E.; Babayan, V.; Pastorek, M.; Machovsky, M.; Bazant, P.; Saha, P. The formation mechanism of iron oxide nanoparticles within the microwave-assisted solvothermal synthesis and its correlation with the structural and magnetic properties. Dalton Trans. 2015, 44, 21099–21108. [CrossRef] g p p 58. Huang, X.; Zhao, Z.; Chen, Y.; Zhu, E.; Li, M.; Duan, X.; Huang, Y. A rational design of carbon-supported dispersive Pt-based octahedra as efﬁcient oxygen reduction reaction catalysts. Energy Environ. Sci. 2014, 7, 2957–2962. [CrossRef] 59. Sreenivasa Kumar, G.; Ramamanohar Reddy, N.; Sravani, B.; Subramanyam Sarma, L.; Veera Reddy, T.; Madhavi, V.; Adinarayana Reddy, S. Ultra-Range Bimetallic Pt–Pd Nanospheres Deposited on Reduced Graphene Sheet as Efﬁcient Electrocatalyst Towards Electrooxidation of Methanol. J. Clust. Sci. 2021, 32, 27–36. References Upgrading the Properties of Reduced Graphene Oxide and Nitrogen-Doped Reduced Graphene Oxide Produced by Thermal Reduction toward Efﬁcient ORR Electrocatalysts. Nanomaterials 2019, 9, 1761. y 45. Zaaba, N.I.; Foo, K.L.; Hashim, U.; Tan, S.J.; Liu, W.W.; Voon, C.H. Synthesis of Graphene Oxide using Modiﬁed Hummers Method: Solvent Inﬂuence. Procedia Manuf. 2017, 184, 469–477. [CrossRef] 46. Yang, D.; Gu, J.; Liu, X.; He, H.; Wang, M.; Wang, P.; Zhu, Y.; Fan, Q.; Huang, R. Monodispersed Pt3Ni Nanoparticles as a Highly Efﬁcient Electrocatalyst for PEMFCs. Catalysts 2019, 9, 588. [CrossRef] y y 47. He, F.G.; Yin, J.Y.; Sharma, G.; Kumar, A.; Stadler, F.J.; Du, B. Facile Fabrication of Hierarchical rGO/PANI@PtNi Nanocomposite via Microwave-Assisted Treatment for Non-Enzymatic Detection of Hydrogen Peroxide. Nanomaterials 2019, 9, 1109. [CrossRef] y 47. He, F.G.; Yin, J.Y.; Sharma, G.; Kumar, A.; Stadler, F.J.; Du, B. Facile Fabrication of Hierarchical rGO/PANI@PtNi Nanocomposite via Microwave-Assisted Treatment for Non-Enzymatic Detection of Hydrogen Peroxide. Nanomaterials 2019, 9, 1109. [CrossRef] 48. Gao, S.; Yang, X.; Liang, S.; Wang, Y.H.; Zang, H.Y.; Li, Y.G. One step synthesis of PtNi electrocatalyst for methanol oxidation. 48. Gao, S.; Yang, X.; Liang, S.; Wang, Y.H.; Zang, H.Y.; Li, Y.G. One step synthesis of PtNi electrocatalyst for methanol oxidation. Inorg. Chem. Commun. 2019, 106, 104–110. [CrossRef] 49. Hu, X.; Song, P.; Yang, X.; Wang, C.; Wang, J.; Tang, Y.; Zhang, J.; Mao, Z. One-step microwave-assisted synthesis of carbon- supported ternary Pt-Sn-Rh alloy nanoparticles for fuel cells. J. Taiwan Inst. Chem. Eng. 2020, 115, 272–278. [CrossRef] pp y y p g 50. Li, G.; Li, J.; Zhou, Z.; Li, C.; Cai, C.; Guo, B.; Priestley, R.D.; Han, L.; Liu, R. Silica-polydopamine templates for ultra-stable hollow Pt anchored N-doped carbon electrocatalysts. Dalton Trans. 2017, 46, 50. Li, G.; Li, J.; Zhou, Z.; Li, C.; Cai, C.; Guo, B.; Priestley, R.D.; Han, L.; Liu, R. Silica-polydopamine core-shell self-conﬁned templates for ultra-stable hollow Pt anchored N-doped carbon electrocatalysts. Dalton Trans. 2017, 46, 16419–16425. [CrossRef] 51. Lokanathan, M.; Patil, I.M.; Kakade, B. Trimetallic PtNiCo nanoﬂowers as efﬁcient electrocatalysts towards oxygen reduction reaction. Int. J. Hydrogen Energy 2018, 43, 8983–8990. [CrossRef] 51. Lokanathan, M.; Patil, I.M.; Kakade, B. Trimetallic PtNiCo nanoﬂowers as efﬁcient electrocatalysts towards oxygen reduction reaction. Int. J. Hydrogen Energy 2018, 43, 8983–8990. [CrossRef] 17 of 17 17 of 17 Nanomaterials 2021, 11, 2206 52. Kaewsai, D.; Yeamdee, S.; Supajaroon, S.; Hunsom, M. References Efﬁcient electro-oxidation of methanol using PtCo nanocatalysts supported reduced graphene oxide matrix as anode for DMFC. Int. J. Hydrogen Energy 2017, 42, 10238–10247. [CrossRef] [ ] 71. Wei, L.; Wu, Q.; Yang, Y.; Jiang, B.; Sun, G.; Feng, J.; Yu, F.; Kang, Y.; Dong, G. One-step synthesis of nitrogen-decorated CeO2/reduced graphene oxide nanocomposite and its electrocatalytic activity for triiodide/iodide reduction. Mater. Res. Lett. 2020, 35, 1461–1471. [CrossRef] [ ] 72. Wang, X.; Zhang, X.; He, X.; Ma, A.; Le, L.; Lin, S. Facile Electrodeposition of Flower-Like PMo12-Pt/rGO Composite with Enhanced Electrocatalytic Activity towards Methanol Oxidation. Catalysts 2015, 5, 1275–1288. [CrossRef] y y 73. Wang, Z.; Hu, S.; Ali, A.; Chen, H.; Shen, P.K. Facile One-Pot Synthesis of a PtRh Alloy Decorated on Ag Nanocubes as a Trimetallic Core–Shell Catalyst for Boosting Methanol Oxidation Reaction. ACS Appl. Energy Mater. 2021, 4, 1085–1092. [CrossRef] 73. Wang, Z.; Hu, S.; Ali, A.; Chen, H.; Shen, P.K. Facile One-Pot Synthesis of a PtRh Alloy Decorated on Ag Nanocubes as a Trimetallic Core–Shell Catalyst for Boosting Methanol Oxidation Reaction. ACS Appl. Energy Mater. 2021, 4, 1085–1092. [CrossRef] 74. Jia, Z.; Wang, Y.; Qi, T. Pd nanoparticles supported on Mg–Al–CO3 layered double hydroxide as an effective catalyst for methanol electro-oxidation. RSC Adv. 2015, 5, 62142. [CrossRef] y g pp gy 74. Jia, Z.; Wang, Y.; Qi, T. Pd nanoparticles supported on Mg–Al–CO3 layered double hydroxide as an effective catalyst for methanol electro-oxidation. RSC Adv. 2015, 5, 62142. [CrossRef] 75. Yang, Y.; Huang, H.; Yang, C.; He, H. Ultraﬁne Rh-Decorated 3D Porous Boron and Nitrogen Dual-Doped Graphene Architecture as an Efﬁcient Electrocatalyst for Methanol Oxidation Reaction. ACS Appl. Energy Mater. 2021, 4, 376–383. [CrossRef] y pp gy 76. Song, C.; Kim, S. Preparation and Electrochemical Characterization of Pt-Supported Flake-like Grap Reduced Graphene Oxide as Fuel Cell Catalysts. J. Electrochem. Soc. 2015, 162, F1181–F1190. [CrossRef 77. Kundu, P.; Nethravathi, C.; Deshpande, P.A.; Rajamathi, M.; Madras, G.; Ravishankar, N. Ultrafast Microwave-Assisted Route to Surfactant-Free Ultraﬁne Pt Nanoparticles on Graphene: Synergistic Co-reduction Mechanism and High Catalytic Activity. Chem. Mater. 2011, 23, 2772–2780. [CrossRef]
https://openalex.org/W2755963217	http://pdf.blucher.com.br/chemicalengineeringproceedings/cobeqic2017/336.pdf	Portuguese	null	PRODUÇÃO DE BIOETANOL COMBUSTÍVEL POR BATELADA ALIMENTADA UTILIZANDO SACCHAROMYCES CEREVISIAE SUPORTADO EM ESFERAS DE ALGINATO DE CÁLCIO REVESTIDAS COM QUITOSA.	Blucher Chemical Engineering Proceedings	2,017	cc-by	2,398	XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 1. INTRODUÇÃO O uso de células imobilizadas em processos de produção de bioetanol combustível vem sendo estudado para obtenção de ganho de produção (Lee et al., 2015; Duarte et al., 2013; Ghorbani et al., 2011; Nikolic et al., 2010; Wendhausen et al., 2000; Tyagi et al., 1992). Normalmente sistemas de fermentação que utilizam suportes celulares promovem maior tempo de processo e consequentemente maior produtividade, simplificam a separação e purificação dos produtos. Na literatura estão descritas técnicas de produção de bioetanol utilizando suportes e técnicas de imobilização utilizando alginato de sódio (Duarte et al., 2013; Călinescu et al., 2012; Najafpour et al., 2004), bem como o alginato de sódio é utilizado em processos industriais para produtos alimentícios e farmacêuticos (Najafpour et al., 2004; Kubota et al., 2000). Neste trabalho, nós reportamos o processo de produção de bioetanol combustível utilizando a levedura Saccharomyces cerevisiae imobilizada em esferas de alginato de cálcio recobertas com quitosana, utilizando ácido cítrico para solubilização quitosana e no recobrimento das esferas. O processo fermentativo desenvolvido conferiu alta resistência do suporte celular para ser utilizados em processos mais longos, forneceu alta produtividade de bioetanol devido a alta conversão da glicose, além de facilitar a separação do produto do meio fermentativo, o que demonstra ser possível a otimização do processo para maior escala. L. A. K. OURA1, T. S. BELLA DE JESUS1, J. R. NUNHEZ1, G. P. VALENÇA1, L. C. FARDELONE2, J. A. R. RODRIGUES2 e P. J. S. MORAN2 1 Universidade Estadual de Campinas, Faculdade Engenharia Química 2 Universidade Estadual de Campinas, Instituto de Química E-mail para contato: lucidio.fardelone@iqm.unicamp.br 1 Universidade Estadual de Campinas, Faculdade Engenharia Química 2 Universidade Estadual de Campinas, Instituto de Química E-mail para contato: lucidio.fardelone@iqm.unicamp.br RESUMO - Células imobilizadas de Saccharomyces cerevisiae em esferas de alginato de cálcio revestidas com quitosana foram utilizadas para o desenvolvimento de processo de produção de bioetanol combustível. Estes processos de bateladas alimentadas foram conduzidos por 96 horas, fornecendo rendimentos de 89-95%. PRODUÇÃO DE BIOETANOL COMBUSTÍVEL POR BATELADA ALIMENTADA UTILIZANDO SACCHAROMYCES CEREVISIAE SUPORTADO EM ESFERAS DE ALGINATO DE CÁLCIO REVESTIDAS COM QUITOSA. L. A. K. OURA1, T. S. BELLA DE JESUS1, J. R. NUNHEZ1, G. P. VALENÇA1, L. C. FARDELONE2, J. A. R. RODRIGUES2 e P. J. S. MORAN2 1 Universidade Estadual de Campinas, Faculdade Engenharia Química 2 Universidade Estadual de Campinas, Instituto de Química E-mail para contato: lucidio.fardelone@iqm.unicamp.br L. A. K. OURA1, T. S. BELLA DE JESUS1, J. R. NUNHEZ1, G. P. VALENÇA1, L. C. FARDELONE2, J. A. R. RODRIGUES2 e P. J. S. MORAN2 XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 2009). Os reagentes ácido cítrico, cloreto de cálcio, alginato de sódio (viscosidade média) e quitosana (peso molecular médio, grau de desacetilação 75-85%) foram adquiridos da Sigma-Aldrich Co, glicose da Synth, extrato de levedura, extrato de malte e peptona da BD Bioscience. Para a expansão celular da levedura S. cerevisae foi utilizado meio de cultura com composição de 3,0 g/L de extrato de malte, 3,0 g/L de extrato de levedura, 5,0 g/L peptona e 10 g/L glicose, ajustado o pH para 6,0, mantendo os erlenmeyers com reentrâncias (baffled flask), para melhor mistura, com 120 rpm (0,54 g) em shaker por 18h. Após o crescimento da massa celular, esta foi centrifugada, 1844 g por 20 minutos, sob medidas de assepsia, sendo que as células obtidas (30 g) foram utilizadas para imobilização em alginato de cálcio, seguido de revestimento com quitosana, a qual foi solubilizada com ácido cítrico (Duarte et al. 2013; Nagashima et al., 1984). A técnica utilizada para imobilização das células foi por aprisionamento em matriz porosa, através do uso de uma suspensão de alginato de sódio 3% e de S. cerevisiae, gotejada em solução de cloreto de cálcio, CaCl2 2%, formando esferas com diâmetro de aproximadamente 3-4 mm. As esferas foram mantidas em solução de CaCl2 2%, por 1 hora para que ocorresse uma melhor distribuição do cálcio no gel da matriz de suporte celular. Seguido de 30 minutos em solução de quitosana com ácido cítrico para o revestimento. Para que as esferas apresentassem uma melhor performance durante o processo fermentativo, e para que diminuísse a lixiviação do cálcio presente na rede polimérica das esferas, estas foram curadas de duas maneiras diferentes, após o recobrimento com quitosana solubilizada em ácido cítrico, uma mantendo-se por 1 hora as esferas em solução de CaCl2 2% e outra por 30 minutos. Em seguida, as esferas foram separadas por filtração, lavadas com água destilada e utilizadas nos processos fermentativos. As fermentações foram conduzidas de 4 maneiras diferentes, uma como controle utilizando as esferas de alginato de cálcio revestidas com quitosana, tal qual, uma utilizando o as esferas que passaram por cura por 30 minutos em CaCl2 2%, uma com a cura por 1 hora e a outra utilizando esferas de alginato de cálcio revestidas com quitosana (conforme o controle), mas com adição de 5 g de CaCl2 ao meio fermentativo. A concentração final de etanol e de glicose residual foram determinadas através de Cromatografia Líquida de Alta Eficiência (CLAE) utilizando o equipamento Agilent Technologies 1200 Series, com sistema de injeção automático, detector com índice de refração, acoplado a coluna Aminex HPX-87H, 300 x 7,8 mm, mantida à temperatura de 50 °C, utilizando como fase móvel solução 0,005 mol/L de H2SO4, sob fluxo de 0,6 mL/min. 2. METODOLOGIA Foi utilizada a cepa JAY270 de S. cerevisiae, fornecida pelo Instituto de Biologia da Unicamp (IB), que possui alta resistência à temperatura elevadas, variação de pH, estresse oxidativo, alta produção de massa celular e de bioetanol (Duarte et al. 2013; Argueso et al. 3. RESULTADOS E DISCUSSÃO Os processos fermentativos utilizando células imobilizadas de S. cerevisiae em alginato de cálcio revestidos com quitosana foram realizados nas mesmas condições utilizando 4,6 g de peso de células da levedura imobilizadas, 111 g/L de glicose inicial e adições a cada 24 horas de 142,0 g/L de glicose (concentrações determinadas por CLAE), pH 6, utilizando shaker a 120 rpm (0,54 g), 30 oC, por 96 h. XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 Na Figura 2 estão representadas imagens do início do processo e após 96 horas de produção de bioetanol combustível e é possível observar que poucas esferas estão rompidas, o que confere maior robustez ao processo. Figura 2 - a Esferas de alginato de cálcio como suporte para S. cerevisiae no início do processo fermentativo para produção de bioetanol combustível; b esferas de alginato de cálcio como suporte para S. cerevisiae após 96 h de processo fermentativo; c esferas de alginato de cálcio como suporte para S. cerevisiae curadas por 1 hora em CaCl2 2%, após 96 horas de processo; d esferas de alginato de cálcio como suporte para S. cerevisiae curadas por 30 minutos em CaCl2, após 96 horas de processo; e esferas de alginato de cálcio como suporte para S. cerevisiae com adição de 5 g de CaCl2 2% no processo fermentativo, após 96 horas de processo. a b c d e Normalmente quando se utiliza ácido acético no processo de recobrimento da quitosana nas esferas, há um residual do ácido acético, provocando uma maior fase lag, para o início da produção de bioetanol (Duarte et al., 2013), assim, a utilização do ácido cítrico em substituição ao ácido acético promoveu a diminuição desta fase de adaptação das células para produção de bioetanol, Figura 3. a b c d e Normalmente quando se utiliza ácido acético no processo de recobrimento da quitosana nas esferas, há um residual do ácido acético, provocando uma maior fase lag, para o início da produção de bioetanol (Duarte et al., 2013), assim, a utilização do ácido cítrico em substituição ao ácido acético promoveu a diminuição desta fase de adaptação das células para produção de bioetanol, Figura 3. Normalmente quando se utiliza ácido acético no processo de recobrimento da quitosana nas esferas, há um residual do ácido acético, provocando uma maior fase lag, para o início da produção de bioetanol (Duarte et al., 2013), assim, a utilização do ácido cítrico em substituição ao ácido acético promoveu a diminuição desta fase de adaptação das células para produção de bioetanol, Figura 3. Figura 3 - Processos em batelada de produção de bioetanol combustível por S. cerevisiae suportados em esferas de alginato de cálcio e alginato de cálcio revestido com quitosana. XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 Na Figura 1, estão demonstrados os perfis de produtividades de bioetanol ao longo de 96 horas nos processos de bateladas alimentadas e na Tabela 1, estão demonstrados os rendimentos, concentrações finais de bioetanol e de glicose residual. O rendimento final foi definido através do bioetanol produzido em relação ao máximo de conversão da fonte de carbono, glicose, em bioetanol. Sendo observado alta conversão da fonte de carbono (glicose) ao final dos processos fermentativos, 89-95% de rendimento. Figura 1 - Processos em batelada alimentada de produção de bioetanol combustível por S. cerevisiae suportados em esferas de alginato de cálcio revestido com quitosana. Figura 1 - Processos em batelada alimentada de produção de bioetanol combustível p cerevisiae suportados em esferas de alginato de cálcio revestido com quitosana. As inserções, 24 h, 48h, 72h são referentes as adições de glicose para que o processo mantivesse uma produção contínua de bioetanol combustível. As inserções, 24 h, 48h, 72h são referentes as adições de glicose para que o processo mantivesse uma produção contínua de bioetanol combustível. 1 – Resultados de produção de bioetanol combustível por S. cerevisiae suportados em esferas de alginato de cálcio revestido com quitosana. Tabela 1 – Resultados de produção de bioetanol combustível por S. cerevisiae suportados em esferas de alginato de cálcio revestido com quitosana. Células de S. cerevisiae Glicose residual (g/L) após 96 h Bioetanol (g/L) após 96 h Rendimento em bioetanol (%) Imobilizadas em alginato/ revestida com quitosana/ác. cítrico 0 152,4 93 Imobilizadas em alginato/ revestida com quitosana/ác. cítrico/ 30 min. em CaCl2 2% 0 150,9 89 Imobilizadas em alginato/ revestida com quitosana/ác. cítrico/ 1h em CaCl2 2% 0 155,2 95 Imobilizadas em alginato/ revestida com quitosana/ác. cítrico + 5g de CaCl2 0,6 148,9 91,2 5. AGRADECIMENTOS Os autores agradecem os suportes financeiros do projeto pela FAPESP (2016/12074- 7), CAPES e CNPq. 4. CONCLUSÕES O processo de produção de bioetanol combustível utilizando células de S. cerevisiae imobilizadas em alginato de cálcio, utilizando ácido cítrico no processo de imobilização e curadas com cloreto de cálcio, bem como, com adição de cloreto de cálcio no meio de cultura, promoveram maior robustez ao processo, isto é, forneceram maior resistência ao suporte utilizado, diminuindo a lixiviação do cálcio da matriz de imobilização, bem como maior produtividade de bioetanol. Este processo será ampliado e otimizado para biorreatores e os resultados serão reportados futuramente. XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 Figura 3 - Processos em batelada de produção de bioetanol combustível por S. cerevisiae suportados em esferas de alginato de cálcio e alginato de cálcio revestido com quitosana. Figura 3 - Processos em batelada de produção de bioetanol combustível por S. cerevisia suportados em esferas de alginato de cálcio e alginato de cálcio revestido com quitosan 6. REFERÊNCIAS ARGUESO J. L.; CARAZZOLLE, M. F.; MIECZKOWSKI, P. A.; DUARTE, F. M.; NETTO, O. V. C.; MISSAWA, S. K.; GALZERANI, F.; COSTA, G. G. L.; VIDAL, R. O.; NORONHA, M. F; DOMINSKA, M.; ANDRIETTA, M. G.S.; ANDRIETTA; S. R.; CUNHA, A. F.; GOMES, L. H.; TAVARES, F. C. A.; ALCARDE, A. R.; DIETRICH, F. S.; MCCUSKER, J. H; PEREIRA, G. A. G. Genome structure of a Saccharomyces cerevisiae strain widely used in bioethanol production. Gen. Res., v. 19, p. 2258-2270, 2009. CĂLINESCU,I., CHIPURICI, P., TRIFAN, A., BĂDOIU, C. Immobilization of Saccharomyces cerevisiae for the production of bioethanol. U.P.B. Sci. Bull., Series B 74, 33-40, 2012. DUARTE, J. C.; RODRIGUES, J. A. R.; MORAN, P. J. S.; VALENÇA, G. P.; NUNHEZ, J. R. Effect of immobilized cells in calcium alginate beads in alcoholic fermentation. ABM Express,, v. 31, p. 1-8, 2013. GHORBANI, F.; YOUNESI, H.; SARI, A. E.; NAJAFPOUR, G. Cane molasses fermentation for continuous ethanol production in an immobilized cells reactor by Saccharomyces cerevisiae. Ren. Energy, v. 36, p. 503-509, 2011. KUBOTA, N.; TATSUMOTO, N.; SANO, T.; TOYA, K. A. Simple preparation of half N- acetylated chitosan highly soluble in water and aqueous organic solvent. Carb. Res., p. 268-274, 2000. LEE, H.; KIM, S.; YOON, J.; KIM, K. H.; SEO, J.; PARK, Y. Evolutionary engineering of Saccharomyces cerevisiae for efficient conversion of red algal biosugars to bioethanol. Bioresour. Technol., v. 191, p. 445-451, 2015. NAGASHIMA, M.; AZUMA, M.; NOGUCID, S; INUZUKA, K; SAMEJIMA, H. Continuous fermentation using immobilized yeast cells. Biot. Bioeng., v. 26, p. 992-997, 1984. XII Congresso Brasileiro de Engenharia Química em Iniciação Científica UFSCar – São Carlos – SP 16 a 19 de Julho de 2017 NAJAFPOUR, G.; YOUNESI, H.; SYAHIDAH KU ISMAIL, K. Ethanol fermentation in an immobilized cell reactor using Saccharomyces cerevisiae. Biores. Technol., v. 92, p. 251-260, 2004. NIKOLIC, S.; MOJOVIC, L.; PEJIN, D.; RAKIN, M.; VUKASINOVIC, M. Production of bioethanol from corn meal hydrolizates by free and immobilized cells of Saccharomyces cerevisiae var. ellipsoideus. Biomass Bioenerg., v. 34, p. 1449-1456, 2010. TYAGI, R. D.; GUPTAB S. K.; CHAND, S. Process engineering studies on continuous ethanol production by immobilized S. cerevisiae. Proc. Bioch., v. 27, p. 23-32, 1992. WENDHAUSEN, R.; FREGONESI, A.; MORAN, P. J. S.; JOEKES, I.; RODRIGURES, J. A. R.; TONELLA, E.; ALTHOFF, K. Continuous Fermentation of Sugar Cane Syrup Using Immobilized Yeast Cells. J. Bio. Bioeng., v. 91, p. 48-52, 2000.
https://openalex.org/W3046576587	https://phsreda.com/e-articles/188/Action188-86003.pdf	Russian	null	Distance Learning in the Context of the COVID-19 Pandemic	null	2,020	cc-by	1,474	Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) Publishing house "Sreda" Патракова Галина Васильевна канд. филос. наук, канд. пед. наук, директор АНПОО «Сургутский институт экономики, управления и права» г. Сургут, Ханты-Мансийский автономный округ DOI 10.31483/r-86003 2 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) 2 https://phsreda.com О ДИСТАНЦИОННОЙ ФОРМЕ ОБУЧЕНИЯ В УСЛОВИЯХ ПАНДЕМИИ COVID-19 Аннотация: в статье говорится о том, что Сургутский институт эко- номики, управления и права не остался в стороне от обсуждения проблем, свя- занных с переходом на дистанционное обучение студентов в условиях панде- мии новой коронавирусной инфекции; актуализирована необходимость привле- чения внимания специалистов и общества к решению проблем в сфере удален- ного образования. Ключевые слова: дистанционное образование, обучение в условиях панде- мии новой коронавирусной инфекции, опрос участников образовательного про- цесса. Как известно, в Министерстве образования и науки Российской Федера- ции 15 марта 2020 г. были утверждены Рекомендации для предотвращения распространения вирусной инфекции COVID-19. Поэтому уже с понедельника 16 марта многие образовательные учреждения активно включились в процесс подготовки оптимальных условий для перехода на удаленное обучение. Ин- формация была принята к исполнению, хотя реакция среди преподавательской общественности, студентов и их родителей была неоднозначной. Как же отре- агировали участники образовательного процесса – студенты, их родители и преподаватели – на вынужденный переход к дистанционной форме обучения? Каким образом учебным заведениям страны удалось экстренно перейти на «удаленку»? Переходу на дистанционное обучение способствовало разрешение бес- платного доступа к различным онлайн-платформам страны. Например: 1 Издательский дом «Среда» Издательский дом «Среда» Издательский дом «Среда» ‒ SkyEng – сервис, позволяющий изучать языки, предоставил бесплатную возможность для всех желающих; ‒ SkyEng – сервис, позволяющий изучать языки, предоставил бесплатную возможность для всех желающих; ‒ Coursera – площадка, помогающая студентам заниматься через онлайн- курсы и просматривать видеолекции преподавателей из ведущих стран мира; ‒ «Открытое образование» – онлайн-курсы, где собраны видеолекции и за- дания от преподавателей ведущих вузов страны. Теперь сервис предлагает сту- дентам пройти бесплатную сертификацию и получить справку о пройденных курсах. Такой документ ценится при трудоустройстве и добавляет баллы при дальнейшем поступлении в университет, например, в магистратуру или аспи- рантуру; ‒ «Академия Хана», предоставляющая бесплатный доступ для студентов. Здесь собраны популярные лекции по всем направлениям. ‒ «Академия Хана», предоставляющая бесплатный доступ для студентов. Здесь собраны популярные лекции по всем направлениям. Имеются также платформы, которые предлагают учебные материалы по отдельным специальностям: ‒ «Нетология» – платформа, которая позволяет изучить маркетинг и ме- неджмент. Открыт бесплатный доступ к информации до конца карантина; ‒ «Pruffme» полезна тем, что преподаватели могут бесплатно проводить вебинары для студентов, а также загружать собственные тесты и задания, а за- тем проверять их. Телевизионный канал «Россия – Культура» также подготовил для студен- тов сервис «Академия» для занимательного самообразования. Сургутский институт экономики, управления и права не остался в стороне от обсуждения проблем, связанных с переходом на дистанционное образование в условиях пандемии новой коронавирусной инфекции. Весна 2019/2020-го учебного года стала для сотрудников, студентов и их родителей не слишком благоприятной, так как пришлось завершать учебный год нетрадиционным спо- собом. По инициативе администрации учебного заведения были проведены опросы мнения участников образовательного процесса о текущей ситуации. 2 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) Publishing house "Sreda" Примечателен тот факт, что по истечении пятой недели карантинных ме- роприятий отзывы были еще оптимистичными, поскольку сохранялась надежда в мае текущего года вернуться к традиционной форме обучения. Итак, на 4–5-й неделе карантина (середина апреля текущего года) 45% сту- дентов активно занимались самообразованием: осваивали учебный материал дистанционно, пользовались ресурсами Интернет, изучали языки, формировали новые навыки в спорте, кулинарии, творчестве, участвовали в дистанционных олимпиадах, онлайн-проектах, организованных институтом. 20% студентов за- нялись здоровьем, практикуя утренние зарядки, здоровое питание и участие в различных челенджах по здоровьесбережению и физической культуре. 15% студентов освоили декоративно-прикладное искусство, ремесло, занимая время на самоизоляции рисованием, вязанием, рукоделием и другими видами творче- ства. 10% студентов освоили фриланс и нашли себе способ заработка в дистан- ционном формате. Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) Издательский дом «Среда» 8% проявили себя в музыке, вокале и танцах при помощи мобильных приложений, а 2% студентов с удовольствием погрузились в про- смотры сериалов, популярных передач и шоу, посвящая этому свободное от учебных заданий время. У большинства студентов не возникало особых затруднений с обращением к интернет-ресурсам и выполнением заданий. Совсем немногие студенты испы- тывали растерянность перед техническими средствами обучения и задавали во- просы типа «Куда нажимать?», «Почему не загружается?» или «Как узнать ре- зультат?». Сложилось впечатление, что в основной своей массе студенты уже готовы к переходу на удаленное образование. Скорее всего, такая тенденция позволит сделать обучение более доступным для большинства людей. Как же чувствовали себя преподаватели? Как они реагировали на переход к дистанционным формам обучения в условиях пандемии? Многие отмечали быстроту перехода на удаленный доступ. Некоторые преподаватели считали, что дистанционный режим для некоторых предметов даже необходим. Также, вполне вероятно, по их мнению, что некоторый объем теоретического материа- ла резонно перевести на онлайн-формат и за пределами пандемии. Часть препо- 3 Издательский дом «Среда» давателей испытывали дискомфорт от нехватки эмоционального контакта с аудиторией. Некоторые преподаватели опасались того, что разрушение контак- та между студентом и наставником, который проходит в аудитории, впослед- ствии отрицательно скажется на исполнительской дисциплине обучающихся. По мнению преподавателей, с введением карантинных мер их профессио- нальная жизнь заметно усложнилась новыми задачами: организовать учебный процесс в непривычном формате, освоить образовательные платформы и новые интернет-ресурсы, усовершенствовать навыки работы с мобильными приложе- ниями и современными гаджетами, обеспечить передачу знаний студентам и обратную связь с ними. Результаты опроса свидетельствуют о том, что 65% преподавателей инсти- тута почти вдвое увеличили свое время на профессиональную деятельность. Это неудивительно, ведь за рабочий день каждый из них получал около 200 электронных писем от студентов, примерно столько же сообщений в мессен- джерах мобильных телефонов. Кроме этого, надо было проверить тесты и направить комментарии студентам по заданиям, указать на ошибки, ответить на вопросы, а также заполнить необходимую отчетную документацию по резуль- татам обучения. Лишь 21% опрошенных преподавателей смогли найти разум- ный баланс в профессиональных и домашних делах, а 14% преподавателей уда- лось выкроить время на спорт и хобби. Не остались в стороне и родители студентов. По результатам опроса они выразили просьбу снизить объемы домашних заданий студентам, чаще прово- дить онлайн-конференции по обучению и индивидуальному консультирова- нию, а также выкладывать на неделю вперед учебный график с полным разъяс- нением содержания учебной нагрузки, объемов домашних заданий и формы контроля знаний студентов. Однако в целом было отмечено, что дети стали бо- лее организованными, дисциплинированными и ответственными. ttps://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) 6 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) Издательский дом «Среда» 4 htt // h d Учебный год, так или иначе, завершен. Позади 3 месяца обучения в ди- станционной форме, и нами снова был организован опрос участников образова- тельного процесса. Результаты в целом таковы: 78% респондентов (в совокуп- 4 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) Publishing house "Sreda" ности 768 студентов, родителей и преподавателей) считают, что дистанционное обучение не может осуществляться в постоянном режиме, поскольку результат его гораздо ниже, чем при традиционной форме обучения. Они убеждены, что в личном контакте с преподавателем в аудитории материал усваивается лучше, а обсуждение его со сверстниками делает образовательный процесс более насы- щенным не только интеллектуально, но и эмоционально. На предложение дать оценку организации дистанционного обучения в течение прошедших 3 месяцев респонденты ответили следующим образом: 42% довольны уровнем организа- ции такого формата обучения, 26% совершенно недовольны, 32% затруднились ответить на этот вопрос. Несмотря на понимание того факта, что дистанционное обучение является вынужденной мерой в связи с распространением COVID-19, многие участники образовательного процесса остались недовольны переходом на дистанционный формат обучения и уже переживают за грядущий учебный год. Родителей обу- чающихся в первую очередь интересует, как студенты будут учиться с 1 сен- тября текущего года. Министерство просвещения сообщает, что на территории страны обучение будет проходить традиционно с 1 сентября 2020 года, если ситуация с пандеми- ей разрешится. По словам министра просвещения Сергея Кравцова, сфера обра- зования никогда не сможет отказаться от стандартного процесса обучения. Также ведомство никогда не сможет уйти от традиционных уроков и общения между сверстниками. Итак, дистанционное обучение позволит преподавателям и студентам не выходить на улицу и продолжить занятия в том же режиме, как и в учебном заведении. Администрация и преподаватели должны обеспечить строгий кон- троль посещаемости онлайн-курсов и выполнение студентами заданий учеб- ного плана. Образовательные программы никто не отменял, поэтому рассла- биться не удастся. Отрицательные аспекты дистанционного обучения имеют- ся, но все же самой главной положительной стороной удаленного преподава- 5 5 Издательский дом «Среда» ния в условиях пандемии является снижение риска заболевания коронавирус- ной инфекцией. ния в условиях пандемии является снижение риска заболевания коронавирус- ной инфекцией. Таким образом, реалии сегодняшнего дня свидетельствуют о том, что независимо от эпидемиологической ситуации в стране дистанционная форма обучения уверенно входит в образовательные процессы. В скором времени данный вид обучения будет зафиксирован и в правовом поле. Важно соблюсти баланс между традиционной и дистанционной формой обучения, не допуская снижения его результативности.
https://openalex.org/W2997951177	https://repub.eur.nl/pub/123465/Repub-123465-OA.pdf	English	null	Mycobacterium tuberculosis clinical isolates of the Beijing and East-African Indian lineage induce fundamentally different host responses in mice compared to H37Rv	Scientific reports	2,019	cc-by	9,996	www.nature.com/scientificreports www.nature.com/scientificreports Mycobacterium tuberculosis clinical isolates of the Beijing and East- African Indian lineage induce fundamentally different host responses in mice compared to H37Rv Bas C. Mourik 1, Jurriaan E. M. de Steenwinkel 1, Gerjo J. de Knegt1, Ruth Huizinga 2, Annelies Verbon3, Tom H. M. Ottenhoff 4, Dick van Soolingen5 & Pieter J. M. Leenen 2* Substantial differences exist in virulence among Mycobacterium tuberculosis strains in preclinical TB models. In this study we show how virulence affects host responses in mice during the first four weeks of infection with a mycobacterial strain belonging to the Beijing, East-African-Indian or Euro-American lineage. BALB/c mice were infected with clinical isolates of the Beijing-1585 strain or the East-African Indian (EAI)-1627 strain and host responses were compared to mice infected with the non-clinical H37Rv strain of the Euro-American lineage. We found that H37Rv induced a ‘classical’ T-cell influx with high IFN-γ levels, while Beijing-1585 and EAI-1627 induced an influx of B-cells into the lungs together with elevated pulmonary IL-4 protein levels. Myeloid cells in the lungs appeared functionally impaired upon infection with Beijing-1585 and EAI-1627 with reduced iNOS and IL-12 expression levels compared to H37Rv infection. This impairment might be related to significantly reduced expression in the bone marrow of IFN-γ, TNF-α and IFN-β in mice infected with Beijing-1585 and EAI-1627, which could be detected from the third day post infection onwards. Our findings suggest that increased virulence of two clinical isolates compared to H37Rv is associated with a fundamentally different systemic immune response, which already can be detected early during infection. Tuberculosis (TB) is a leading cause of death among infectious diseases worldwide and claimed more victims in 2017 than HIV and malaria combined1. While global efforts have resulted in a steady decline in TB-related deaths over the years, new threats present themselves in the form of drug resistance and the emergence of more virulent Mycobacterium tuberculosis (Mtb) genotypes1–3. Most TB cases are caused by modern lineage mycobacterial strains of the East Asian/Beijing lineage, the East-African Indian lineage and the Euro-American lineage2,4,5. Mycobacterial strains belonging to the Beijing genotype particularly have shown an aggressive global spread over the last century and have been associated with higher rates of treatment failure and disease relapse compared to other genotypes6–12. An important reason for the clinical impact of the Beijing genotypes, seems to be their increased capacity to acquire drug resistance13. A less well-defined characteristic concerns their hypervirulence14–16. Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 Mycobacterium tuberculosis clinical isolates of the Beijing and East- African Indian lineage induce fundamentally different host responses in mice compared to H37Rv Next, they enter a phase of chronic infection and become moribund between 22 and 38 weeks post infection26. In contrast, mice infected with Beijing-1585 or EAI-1627 reach peak infection at 14 dpi with histopathological signs of pneumonia comparable to H37Rv at 28 dpi and rapidly become moribund between three to five weeks post infection if left untreated26,27. In this study we aim to identify the differences in underlying host responses that might contribute to this marked difference in virulence. In the current study we evaluate the host response during acute infection against the virulent Beijing-1585 strain. This strain has previously demonstrated similar infection and mortality kinetics as other virulent Beijing strains18,24. Furthermore, Beijing-1585 was found associated with drug resistance and treatment failure18,25. We compare Beijing-1585 with the East-African/Indian (EAI)-1627 strain, that displays similar virulence as Beijing-1585 in our model18, and with the less virulent H37Rv strain belonging to the Euro-American lineage26. Previous studies in our BALB/c mouse TB model showed that mice infected with H37Rv reach maximal mycobacterial loads and start developing progressive pneumonia 28 days post infection (dpi). Next, they enter a phase of chronic infection and become moribund between 22 and 38 weeks post infection26. In contrast, mice infected with Beijing-1585 or EAI-1627 reach peak infection at 14 dpi with histopathological signs of pneumonia comparable to H37Rv at 28 dpi and rapidly become moribund between three to five weeks post infection if left untreated26,27. In this study we aim to identify the differences in underlying host responses that might contribute to this marked difference in virulence. Mycobacterium tuberculosis clinical isolates of the Beijing and East- African Indian lineage induce fundamentally different host responses in mice compared to H37Rv i yp In preclinical TB models, virulent Beijing strains cause higher mycobacterial loads, more lung damage and earlier mortality compared to strains from other lineages15,17,18. Mechanistic studies have suggested that Beijing 1Department Medical Microbiology & Infectious Diseases, Erasmus University Medical Center, Rotterdam, The Netherlands. 2Department of Immunology, Erasmus University Medical Center, Rotterdam, The Netherlands. 3Department of Internal Medicine, Erasmus University Medical Center, Rotterdam, The Netherlands. 4Department of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands. 5National Tuberculosis Reference Laboratory, National Institute of Public Health and the Environment (RIVM), Bilthoven, The Netherlands. email: p.leenen@erasmusmc.nl www.nature.com/scientificreports/ Figure 1. Mycobacterial loads in lungs and spleen. (A) Mycobacterial loads in the lungs after intratracheal infection with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars). After 14 days Beijing-1585- and EAI-1627-infected mice rapidly become moribund, therefore no later analyses for these strains are possible. (B) Mycobacterial loads in the spleen. Three mice were used for each group at 1 dpi and 6 mice for each group at the remaining time points. The numbers above the bars indicate the number of mice out of n = 6 in total with positive cultures. Inoculum sizes were 1.0.105 CFU for Beijing, 1.3.105 CFU for EAI and 1.8.105 CFU for H37Rv. p < 0.05 p < 0.01 p < 0.001 after Bonferroni correction. Figure 1. Mycobacterial loads in lungs and spleen. (A) Mycobacterial loads in the lungs after intratracheal infection with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars). After 14 days Beijing-1585- and EAI-1627-infected mice rapidly become moribund, therefore no later analyses for these strains are possible. (B) Mycobacterial loads in the spleen. Three mice were used for each group at 1 dpi and 6 mice for each group at the remaining time points. The numbers above the bars indicate the number of mice out of n = 6 in total with positive cultures. Inoculum sizes were 1.0.105 CFU for Beijing, 1.3.105 CFU for EAI and 1.8.105 CFU for H37Rv. p < 0.05 p < 0.01 p < 0.001 after Bonferroni correction. Figure 1. Mycobacterial loads in lungs and spleen. (A) Mycobacterial loads in the lungs after intratracheal infection with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars). After 14 days Beijing-1585- and EAI-1627-infected mice rapidly become moribund, therefore no later analyses for these strains are possible. (B) Mycobacterial loads in the spleen. Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 Results f i Infection with Beijing-1585 or EAI-1627 results in higher mycobacterial loads compared to H37Rv. After intratracheal infection of BALB/c mice with Mycobacterium tuberculosis (Mtb), we found no significant differences in mycobacterial load between Beijing-1585, EAI-1627 and H37Rv at 1 day post infection (dpi) and 3 dpi, indicating that all groups received a similar inoculum of mycobacteria (Fig. 1A). At 7 dpi, mice infected with Beijing-1585 had significantly higher mycobacterial loads and at 14 dpi, Beijing-1585 and EAI-1627 caused almost 2 log higher loads than H37Rv. Mycobacterial loads for H37Rv at peak infection (28 dpi) were still one log value lower than those observed for Beijing-1585 and EAI-1627 at 14 dpi. These findings are in agreement with our and others’ previous studies monitoring mycobacterial loads for Beijing strains and H37Rv15,28,29. p g y j g To assess whether the higher mycobacterial loads in the lungs caused by Beijing-1585 and EAI-1627 were associated with more rapid dissemination to other organs, we determined mycobacterial loads in the spleen (Fig. 1B). No significant differences in culture positivity rate were found between strains, but mycobacterial loads in the spleens of mice infected with Beijing-1585 were higher at 7 dpi and 14 dpi compared to other groups. Beijing-1585 and EAI-1627 induce lung influx of B-cells, while H37Rv induces T-cell influx. To explore whether the distinct in vivo mycobacterial growth profiles in our model correlated with differences in adaptive immune responses, we evaluated the numbers of B- and T-cells recruited to the lungs by the three different strains. Most notably, Beijing-1585 and EAI-1627 induced a strong influx of CD45R+ cells, identifying B-lymphocytes, at 14 dpi, which was not observed for H37Rv at either 14 dpi or 28 dpi (Fig. 2A). In contrast, H37Rv induced the recruitment of CD4+ and CD8+ T-cells at 28 dpi, which in turn was not observed upon infec- tion with the clinical strains (Fig. 2B,C). Despite the marked T-cell increase in the H37Rv group at 28 dpi, Foxp3+ regulatory T-cell percentages of total lung single cell suspension remained lower compared to Beijing-1585 and EAI-1627 at 14 dpi (Fig. 2D). The latter strains show a marked difference in FoxP3+ Treg at 14 dpi, where only mice infected with EAI-1627 show an elevated Treg level above background.h g g The associated cytokine protein levels in the lungs for each time point and genotype strain are shown in Fig. Mycobacterium tuberculosis clinical isolates of the Beijing and East- African Indian lineage induce fundamentally different host responses in mice compared to H37Rv Three mice were used for each group at 1 dpi and 6 mice for each group at the remaining time points. The numbers above the bars indicate the number of mice out of n = 6 in total with positive cultures. Inoculum sizes were 1.0.105 CFU for Beijing, 1.3.105 CFU for EAI and 1.8.105 CFU for H37Rv. p < 0.05 p < 0.01 p < 0.001 after Bonferroni correction. Figure 1. Mycobacterial loads in lungs and spleen. (A) Mycobacterial loads in the lungs after intratracheal infection with Beijing 1585 (black bars) EAI 1627 (open bars) or H37Rv (grey bars) After 14 days j g ( ) ( p ) (g y )t y eijing-1585- and EAI-1627-infected mice rapidly become moribund, therefore no later analyses for these trains are possible. (B) Mycobacterial loads in the spleen. Three mice were used for each group at 1 dpi and 6 mice for each group at the remaining time points. The numbers above the bars indicate the number of mice out f n = 6 in total with positive cultures. Inoculum sizes were 1.0.105 CFU for Beijing, 1.3.105 CFU for EAI and .8.105 CFU for H37Rv. p < 0.05 p < 0.01 p < 0.001 after Bonferroni correction. strains have enhanced capacity to inhibit protective immunity in the lungs through induction of higher levels of type-I interferons, leading to lower IL-12 and TNF-α levels and reduced T-cell activation19,20. Increased Beijing virulence also has been attributed to bacterial phenolic glycolipid (PGL), which suppresses the production of IL-12, IL-6 and TNF-α by host immune cells21,22. Lastly, Beijing strains may induce a stronger regulatory T-cell response compared to other strains, thereby down-regulating protective immunity17,23. p p y g g p y In the current study we evaluate the host response during acute infection against the virulent Beijing-1585 strain. This strain has previously demonstrated similar infection and mortality kinetics as other virulent Beijing strains18,24. Furthermore, Beijing-1585 was found associated with drug resistance and treatment failure18,25. We compare Beijing-1585 with the East-African/Indian (EAI)-1627 strain, that displays similar virulence as Beijing-1585 in our model18, and with the less virulent H37Rv strain belonging to the Euro-American lineage26. Previous studies in our BALB/c mouse TB model showed that mice infected with H37Rv reach maximal mycobacterial loads and start developing progressive pneumonia 28 days post infection (dpi). Results f i 3 n accordance with the increase in B-cells, Beijing-1585- and EAI-1627-infected mice showed elevated protein Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 www.nature.com/scientificreports/ p p Figure 2. Lymphoid cell populations in the lungs of mice infected with different Mtb strains. Lymphoid cells were determined in the lungs of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars) and compared to uninfected control mice (HC, striped bars). Gating strategies are shown in Fig. S5. (A) B-cells, identified based on scatterplot and high level CD45R expression, are significantly higher in number for Beijing-1585- and EAI-1627-infected mice at 14 dpi compared with H37Rv at 14 and 28 dpi. (B,C) Only H37Rv infection induces an increase in both CD4+ and CD8+ T-cells at 28 dpi. (D) Despite the increase in T-cells caused by H37Rv infection at 28 dpi, Foxp3+ regulatory T-cells are significantly lower compared to Beijing-1585- and EAI-1627-infected mice at 14 dpi. Gating strategies were similar as described previously51. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001 after Bonferroni correction. Data are shown as % of total lung single cell suspension. Figure 2. Lymphoid cell populations in the lungs of mice infected with different Mtb strains. Lymphoid cells were determined in the lungs of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars) and compared to uninfected control mice (HC, striped bars). Gating strategies are shown in Fig. S5. (A) B-cells, identified based on scatterplot and high level CD45R expression, are significantly higher in number for Beijing-1585- and EAI-1627-infected mice at 14 dpi compared with H37Rv at 14 and 28 dpi. (B,C) Only H37Rv infection induces an increase in both CD4+ and CD8+ T-cells at 28 dpi. (D) Despite the increase in T-cells caused by H37Rv infection at 28 dpi, Foxp3+ regulatory T-cells are significantly lower compared to Beijing-1585- and EAI-1627-infected mice at 14 dpi. Gating strategies were similar as described previously51. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001 after Bonferroni correction. Data are shown as % of total lung single cell suspension. Figure 2. Lymphoid cell populations in the lungs of mice infected with different Mtb strains. Lymphoid cell d d h l f f d h (bl k b ) ( b ) levels of IL-4 at 14 dpi. Results f i These were 4–5 fold higher than IL-4 levels observed for H37Rv at 14 or 28 dpi (Fig. 3A). Although Beijing-1585 and EAI-1627 also caused elevated protein levels of IFN-γ and IL-17a at 14 dpi, these remained 2-fold and 6-fold lower, respectively, compared to the H37Rv group at 28 dpi (Fig. 3B,C).hf p y p g p p g The TNF-α protein levels in the lungs closely correlated with strain-dependent differences in mycobacterial loads over time. At 7 dpi, TNF-α levels were significantly induced only in mice infected with Beijing-1585- and EAI-1627, which were almost 2-fold higher at 14 dpi compared to H37Rv at 28 dpi (Fig. 3D). This is in line with the role of TNF-α as general inflammation marker. In support of this, the inflammation marker IL-6 showed similar kinetics as TNF-α over time (Fig. S1). IL-10 and IL-23 levels were also measured in the lung homogenates but were below the limit of detection of our assay (data not shown). Quantitative PCR measurements of IFN-γ, IL-17a TNF-α, IL-6 in the lungs were performed with outcomes comparable to those at protein level as shown in Fig. 3 (Fig. S2). IL-10 expression levels were above the lower limit of detection but did not show strain-specific differences (Fig. S2). Beijing-1585 and EAI-1627 induce a qualitatively impaired myeloid response compared to H37Rv. The observed differences in lymphoid cell responses and cytokine levels raised the question whether CD11b+ myeloid cell influxes in the lungs might also vary between mice infected with different Mtb strains. Lung polymorphonuclear granulocyte (PMN) percentages were increased in the Beijing-1585 and EAI-1627 group compared to H37Rv at 7 dpi and 14 dpi (Fig. 4A), which was in line with the elevated mycobacterial loads and inflammation markers TNF-α and IL-6 at these time points. However, at 28 dpi the PMN frequency in the H37Rv group was comparable with that in the Beijing-1585 group and EAI-1627 group at 14 dpi.lli g p p j g g p g p p Inflammatory macrophage/dendritic cell (iM/DC; CD11b+Ly6CintCD11chigh) influx showed a similar profile as PMN (Fig. 4B). Monocyte-like cells (CD11b+Ly6ChighCD11clow) were present to a lesser extent than PMN and iM/DC, and were only higher in the EAI-1627 group at 14 dpi compared to the H37Rv group at 28 dpi (Fig. 4C). Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 www.nature.com/scientificreports/ Figure 3. Cytokine protein levels in the lungs of mice infected with different Mtb strains. Results f i Alveolar macrophages (AM) were reduced over time in all groups, associated with inflammatory cell influx, but most prominently at 14 dpi in the Beijing-1585 and EAI-1627 groups compared to the H37Rv group (Fig. 4D). We also evaluated lung eosinophils in each group since these cells are known IL-4 producers, but levels of these cells were not elevated in the Beijing-1585 and EAI-1627 groups compared to the H37Rv group at any time point evaluated (Fig. S3).h ( g ) The AM and iM/DC are important cellular sources of IL-12 in the lungs30, which is essential for initiation of T-cell responses31. Therefore, we measured the expression of IL-12p35 and IL-12p40. Most notably, Beijing-1585 caused the strongest down-regulation of IL-12p35 in the lungs compared to uninfected control mice and did not induce any IL-12p40 expression at all time points evaluated (Fig. 5A,B). y p p p g To investigate induction of putative bactericidal activity30,32 we measured the expression of inducible nitric oxide synthase (iNOS), typically inducible by IFN-γ and TNF-α, in both infiltrating iM/DC and the lung-resident AM at each time point. The AM from mice infected with the clinical strains essentially failed to up-regulate iNOS at any point during the course of infection (Fig. 5C). In contrast, iNOS expression was already significantly higher in AM from H37Rv-infected mice at 3 dpi and its expression continued to increase up to 14 dpi.hi p p p p The iM/DC in lungs of mice infected with Beijing-1585 or EAI-1627 also showed significantly lower iNOS expression compared to H37Rv-infected mice at all time points (Fig. 5D). At 14 dpi, when IFN-γ and TNF-α levels were high in the lungs of mice from the Beijing-1585 and EAI-1627 group (Fig. 3C), iNOS expression by iM/DC was increased accordingly. Nevertheless, iNOS expression by iM/DC in H37Rv-infected mice was sig- nificantly higher at 14 dpi despite lower levels of IFN-γ and TNF-α in the lungs compared to Beijing-1585- and EAI-1627-infected mice. iNOS expression by iM/DC in H37Rv-infected mice increased even further at 28 dpi. Infection with Beijing-1585 induces less inflammatory cytokines in bone marrow compared to H37Rv. The differential induction of iNOS in infiltrating iM/DC might be caused by distinct local inflamma- tory or inhibitory conditions. Alternatively, cells might be differently primed at an earlier developmental stage. Therefore, we determined cytokine mRNA expression in the bone marrow in the course of infection. Results f i Protein levels were determined in lung tissue homogenates of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars). (A) IL-4 levels are 4–5 fold higher for Beijing-1585 and EAI-1627 at 14 dpi compared to H37Rv at 14 dpi or 28 dpi. (B) IFN-γ levels are elevated for Beijing-1585 and EAI-1627 at 14 dpi, but are 2-fold lower compared to H37Rv at 28 dpi. (C) Beijing-1585 and EAI-1627 induced circa 7-fold lower levels of IL-17a at 14 dpi compared to H37Rv at 28 dpi. (D) TNF-α levels are circa 2-fold higher for Beijing-1585 and EAI-1627 at 14 dpi compared to H37Rv at 28 dpi. N = 5 mice per group per time point for Beijing-1585 and H37Rv and n = 4 mice for EAI, p < 0.05, p < 0.01, p < 0.001, **p < 0.0001 after Bonferroni correction. igure 3. Cytokine protein levels in the lungs of mice infected with different Mtb strains. Protein levels were Figure 3. Cytokine protein levels in the lungs of mice infected with different Mtb strains. Protein levels were determined in lung tissue homogenates of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars). (A) IL-4 levels are 4–5 fold higher for Beijing-1585 and EAI-1627 at 14 dpi compared to H37Rv at 14 dpi or 28 dpi. (B) IFN-γ levels are elevated for Beijing-1585 and EAI-1627 at 14 dpi, but are 2-fold lower compared to H37Rv at 28 dpi. (C) Beijing-1585 and EAI-1627 induced circa 7-fold lower levels of IL-17a at 14 dpi compared to H37Rv at 28 dpi. (D) TNF-α levels are circa 2-fold higher for Beijing-1585 and EAI-1627 at 14 dpi compared to H37Rv at 28 dpi. N = 5 mice per group per time point for Beijing-1585 and H37Rv and n = 4 mice for EAI, p < 0.05, p < 0.01, p < 0.001, **p < 0.0001 after Bonferroni correction. Alveolar macrophages (AM) were reduced over time in all groups, associated with inflammatory cell influx, but most prominently at 14 dpi in the Beijing-1585 and EAI-1627 groups compared to the H37Rv group (Fig. 4D). We also evaluated lung eosinophils in each group since these cells are known IL-4 producers, but levels of these cells were not elevated in the Beijing-1585 and EAI-1627 groups compared to the H37Rv group at any time point evaluated (Fig. S3). Results f i Similar to the lungs, IL-12p35 mRNA expression in the bone marrow was down-regulated most effectively by Beijing-1585 infection compared to uninfected mice at all time points evaluated (Fig. 6A). Interestingly, Beijing-1585 infection Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 www.nature.com/scientificreports/ Figure 4. Myeloid cell populations in the lungs of mice infected with different Mtb strains. CD11b+ cells were classified as PMN (CD11b+Ly6Ghigh), iM/DC (CD11b+Ly6CintCD11chigh), monocyte-like cells (Mo-like) (CD11b+Ly6ChighCD11clow) and alveolar macrophages (AM) (CD11bintCD11chighSiglec-F+) in the lungs of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars) compared to uninfected control mice (HC, striped bars). Gating strategies are shown in Fig. S5. (A) PMN cells showed a more rapid increase for Beijng-1585 and EAI-1627 compared to H37Rv. (B) iM/DC showed kinetics comparable to PMN for the different groups. (C) Monocyte-like cells are only higher in the EAI-1627 group at 14 dpi compared to the H37Rv group at 28 dpi. (D) Lung alveolar macrophages are lower in the Beijing-1585 and EAI-1627 group compared to the H37Rv group at 14 dpi. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001 after Bonferroni correction. Figure 4. Myeloid cell populations in the lungs of mice infected with different Mtb strains. CD11b+ cells i Figure 4. Myeloid cell populations in the lungs of mice infected with different Mtb strains. CD11b+ cells were classified as PMN (CD11b+Ly6Ghigh), iM/DC (CD11b+Ly6CintCD11chigh), monocyte-like cells (Mo-like) (CD11b+Ly6ChighCD11clow) and alveolar macrophages (AM) (CD11bintCD11chighSiglec-F+) in the lungs of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars) compared to uninfected control mice (HC, striped bars). Gating strategies are shown in Fig. S5. (A) PMN cells showed a more rapid increase for Beijng-1585 and EAI-1627 compared to H37Rv. (B) iM/DC showed kinetics comparable to PMN for the different groups. (C) Monocyte-like cells are only higher in the EAI-1627 group at 14 dpi compared to the H37Rv group at 28 dpi. (D) Lung alveolar macrophages are lower in the Beijing-1585 and EAI-1627 group compared to the H37Rv group at 14 dpi. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001 after Bonferroni correction. also showed a lack of induction, or even reduced expression of inflammatory cytokines IFN-γ, IL-17a and TNF-α compared to H37Rv as early as 3 dpi (Fig. 6B–D). Results f i IL-12 mRNA expression in total lung homogenate as determined by qPCR (A,B) and iNOS expression assessed by flow cytometry in (C) alveolar macrophages (AM) and (D) inflammatory macrophages and dendritic cells (iM/DC) in the lungs of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars) are compared to uninfected control mice (HC, striped bars). (A) IL-12p35 expression levels are lower in the Beijing-1585 group compared to the EAI-1627 and H37Rv group. (B) IL-12p40 expression is induced in EAI-1627- and H37Rv- infected mice, and reached its peak for both strains at 14 dpi with higher expression in the H37Rv group. Beijing-1585 does not induce any notable expression of either IL-12p35 or IL-12p40 at all time points evaluated. (C) Only AM in the lungs of H37Rv-infected mice at 3,7 and 14 dpi show iNOS expression levels higher than observed in uninfected control mice. (D) also iM/DC in H37Rv-infected mice show higher iNOS expression than iM/DC in the lungs of mice from the Beijing-1585 or EAI-1627 group at all time points evaluated. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001 after Bonferroni correction. For experiments depicted in this figure, iM/DC were defined as CD11bhighCD11chighMHC-II+ cells and AM were defined as CD11bintCD11c+F4/80+CD200R+ cells (gating strategies: Fig. S4A), based on a distinct panel of antibodies. Population frequencies through this gating were highly comparable to those in Fig. 4 (Fig. S4B). Figure 5. IL-12- and myeloid iNOS expression in the lungs of Mtb-infected mice. IL-12 mRNA expression in total lung homogenate as determined by qPCR (A,B) and iNOS expression assessed by flow cytometry in (C) alveolar macrophages (AM) and (D) inflammatory macrophages and dendritic cells (iM/DC) in the lungs of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars) are compared to uninfected control mice (HC, striped bars). (A) IL-12p35 expression levels are lower in the Beijing-1585 group compared to the EAI-1627 and H37Rv group. (B) IL-12p40 expression is induced in EAI-1627- and H37Rv- infected mice, and reached its peak for both strains at 14 dpi with higher expression in the H37Rv group. Beijing-1585 does not induce any notable expression of either IL-12p35 or IL-12p40 at all time points evaluated. (C) Only AM in the lungs of H37Rv-infected mice at 3,7 and 14 dpi show iNOS expression levels higher than observed in uninfected control mice. Results f i Especially for TNF-α, expression levels differed markedly between bone marrow cells from Beijing-1585- and H37Rv-infected mice over time with a decreased expression for Beijing-1585 at 14 dpi compared to 3 dpi, as opposed to a 34-fold increase for H37Rv. Measurement results for the EAI-1627 group consistently were intermediate between those for the Beijing-1585 and H37Rv groups. Induction of type 1 IFN signature genes in the lungs of infected mice essentially correlates with expression of IFN-β. Since mycobacterial virulence has been associated with increased lung induction of type 1 interferons, we tested the mRNA expression of IFN-α genes (subtypes 1, 2, 5, 6 and 7) in the lungs in a similar approach as originally described by Manca et al.20. We found only a limited expression of the tested IFN-α genes at 3 dpi for all strains, which decreased upon progressing infection and showed no significant inter-strain differences (Fig. 7A). f ( g ) IFN-α and IFN-β share the ability to bind to, and signal via the IFN-α/β receptor, therefore we evaluated IFN-β mRNA expression in the lungs. IFN-β expression in the H37Rv group was significantly higher than that in the EAI-1627 group at 3 dpi, and higher than that in the Beijing-1585 group, but without statistical significance (Fig. 7B). Type I interferons comprise several more subtypes than those for which we could test expression by qPCR and their expression is often transient. Therefore we decided to test the type 1 interferon response, represented by expression of Mx1, IFI44 and CCL2, which are known type 1 interferon-inducible genes33–35. Expression levels of such genes can be combined into a type 1 interferon signature, which provides an indication of type 1 inter- feron responsiveness in a tissue36. Our type 1 interferon signature showed different kinetics between Beijing-1585 and H37Rv, with EAI-1627 again showing intermediate results (Fig. 7C, see Fig. S5 for individual graphs). Most notably, at 7 dpi the Beijing-1585 group showed the strongest induction of type 1 interferon-inducible genes in Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 www.nature.com/scientificreports/ p p / Figure 5. IL-12- and myeloid iNOS expression in the lungs of Mtb-infected mice. Results f i (D) also iM/DC in H37Rv-infected mice show higher iNOS expression than iM/DC in the lungs of mice from the Beijing-1585 or EAI-1627 group at all time points evaluated. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001 after Bonferroni correction. For experiments depicted in this figure, iM/DC were defined as CD11bhighCD11chighMHC-II+ cells and AM were defined as CD11bintCD11c+F4/80+CD200R+ cells (gating strategies: Fig. S4A), based on a distinct panel of antibodies. Population frequencies through this gating were highly comparable to those in Fig. 4 (Fig. S4B). the lungs, while H37Rv-infected mice showed a higher peak induction at 14 dpi. The observed kinetics for the type 1 interferon-inducible genes at 7 dpi and 14 dpi closely matched the trends observed for IFN-β expression at these time points, and not IFN-α, except for the absence of a type 1 interferon signature in the lungs of mice infected with H37Rv at 3 dpi. This suggests that in this model IFN-β is more relevant for the induction of type 1 IFN-regulated genes during acute infection than the tested IFN-α subtypes.iff the lungs, while H37Rv-infected mice showed a higher peak induction at 14 dpi. The observed kinetics for the type 1 interferon-inducible genes at 7 dpi and 14 dpi closely matched the trends observed for IFN-β expression at these time points, and not IFN-α, except for the absence of a type 1 interferon signature in the lungs of mice infected with H37Rv at 3 dpi. This suggests that in this model IFN-β is more relevant for the induction of type 1 IFN-regulated genes during acute infection than the tested IFN-α subtypes.iff g g g yp To compare the findings on IFN-β in the lungs at 3 dpi with differential systemic effects observed in the bone marrow, we measured IFN-β mRNA expression in the bone marrow. This showed a significantly increased expres- sion of IFN-β in the H37Rv group compared to the Beijing-1585 group at 3 dpi (Fig. 7D). As such, the profile showed similarities to the expression of IFN-γ in bone marrow (Fig. 6C). Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 Discussion Cytokine mRNA expression levels in bone marrow of mice infected with different Mtb strains. Figure 6. Cytokine mRNA expression levels in bone marrow of mice infected with different Mtb strains. Expression levels of target cytokine mRNA are shown relative to Gapdh in the bone marrow of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars). (A) IL-12p35 expression at 3 dpi and 7 dpi was lower in the Beijing-1585 group compared to H37Rv. (B) IFN-γ expression was lower for Beijing-1585 at 3 dpi compared to H37Rv. (C) IL-17a levels were markedly lower in the Beijing-1585 group compared to the H37Rv group (mean of 0.26 vs. 26.7), but without statistical significance (p = 0.06 after Bonferroni correction) due to the high spread in the H37Rv group (5.3–71.8). (D) TNF-α levels were higher in the bone marrow of H37Rv-infected mice compared to Beijing-1585-infected mice at all time points evaluated. EAI-1627 consistently showed intermediate results between Beijing-1585 and H37Rv for all cytokines. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001, **p < 0.0001 after Bonferroni correction. Figure 6. Cytokine mRNA expression levels in bone marrow of mice infected with different Mtb strains. Expression levels of target cytokine mRNA are shown relative to Gapdh in the bone marrow of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars). (A) IL-12p35 expression at 3 dpi and 7 dpi was lower in the Beijing-1585 group compared to H37Rv. (B) IFN-γ expression was lower for Beijing-1585 at 3 dpi compared to H37Rv. (C) IL-17a levels were markedly lower in the Beijing-1585 group compared to the H37Rv group (mean of 0.26 vs. 26.7), but without statistical significance (p = 0.06 after Bonferroni correction) due to the high spread in the H37Rv group (5.3–71.8). (D) TNF-α levels were higher in the bone marrow of H37Rv-infected mice compared to Beijing-1585-infected mice at all time points evaluated. EAI-1627 consistently showed intermediate results between Beijing-1585 and H37Rv for all cytokines. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001, **p < 0.0001 after Bonferroni correction. as they reduced neutrophilia by limiting IL-17 responses40. We did find lower IL-17 protein levels in the lungs of Beijing-1585- and EAI-1627-infected mice compared to H37Rv at peak infection, but we observed a similar influx of PMN into the lungs. Discussion More recently it was shown that IL-17 and IL-17-producing innate lymphoid cells were protective during acute Mtb infection41–43. Our findings of higher IL-17 levels in the lungs of H37Rv-infected mice compared to Beijing-1585 and EAI-1627 are in accordance with this notion. Nevertheless, the exact role of B cells and potential involvement of IL-17 in this axis remains to be established.i p Beijing-1585 and EAI-1627 infection elicited significant IL-4 protein levels in the lungs, thus matching the observed increase in B-cells. The presence of IL-4 was previously shown to exert a pathogenic effect during TB by diverting the role of TNF-α from myeloid cell activator to tissue damage mediator44. This is in line with obser- vations by others that virulent Beijing strains induce higher IL-4 mRNA expression levels in the lungs of mice at 14 dpi compared to non-virulent Beijing strains24, and in vitro studies showing that Beijing-HN878 preferen- tially induced IL-4 expression in human peripheral blood mononuclear cells compared to the CDC1551 strain45. Together with our observations, this adds to the support for a host-detrimental role of IL-4 and B-cell responses during acute infection in which IL-4 stimulates macrophages to express a non-bactericidal, alternatively activated phenotype. However, differential exposure of myeloid cells to IL-4 may not be decisive in pathogenesis, since sim- ilar mortality, bacterial burden, histopathology and iNOS expression were observed after infection of wildtype or myeloid-specific IL-4Ralpha knockout mice with H37Rv or HN87846. The source of IL-4 during acute infection remains obscure as no notable T-cell responses were observed for Beijing-1585 and EAI-1627. Also eosinophil numbers, as another known source of IL-4 in the lungs, were not elevated in our study. Given the recently demon- strated protective role of IL-17- producing group 3 ILC in TB43, it might be interesting to study the role of IL-4 producing group 2 ILC during acute infection with Beijing or EAI strains. p g g p g j g Beijing-1585 induced lower lung IL-12p35 and IL-12p40 mRNA expression levels compared to H37Rv, which is in agreement with previous studies20,21,45,47. This might contribute to reduced T-cell responses given the role of IL-12 in this axis31,48. However, the differences in IL-12 expression between EAI-1627 and H37Rv were less pronounced and EAI-1627 infection still resulted in a B-cell response, making this a less likely explanation for the noticeable difference in lymphocyte response. Discussion We found that infection with the Mycobacterium tuberculosis Beijing-1585 strain or EAI-1627 strain is character- ized by an influx of B-cells in the lungs and higher pulmonary IL-4 protein levels compared to a T cell-dominated response to the less virulent H37Rv strain. Beijing- or EAI-strain infection is also associated with recruitment of myeloid cells that appear functionally impaired with low IL-12 and iNOS expression levels. In addition, especially Beijing-1585 infection is associated with a reduced expression of inflammatory cytokines in the bone marrow compared to H37Rv-infected mice, as early as from 3 dpi onwards, suggesting disrupted priming of developing myeloid cells in favor of the mycobacteria. y y Could the increased B-cell influx upon infection with clinical Mycobacterium tuberculosis strains contribute to the observed difference in virulence in our model? In chronic TB patients a protective role for B-cells and antibodies was recently demonstrated37,38. A study in non-human primates showed that B-cell depletion during acute infection resulted in lower levels of inflammation, but higher bacterial burdens39. We found that higher percentages of B-cells were associated with higher levels of inflammation, as expressed by inflammation markers TNF-α and IL-6, but also with higher bacterial burdens. In a mouse TB model of acute TB, B-cells were protective Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 www.nature.com/scientificreports/ Figure 6. Cytokine mRNA expression levels in bone marrow of mice infected with different Mtb strains. Expression levels of target cytokine mRNA are shown relative to Gapdh in the bone marrow of mice infected with Beijing-1585 (black bars), EAI-1627 (open bars) or H37Rv (grey bars). (A) IL-12p35 expression at 3 dpi and 7 dpi was lower in the Beijing-1585 group compared to H37Rv. (B) IFN-γ expression was lower for Beijing-1585 at 3 dpi compared to H37Rv. (C) IL-17a levels were markedly lower in the Beijing-1585 group compared to the H37Rv group (mean of 0.26 vs. 26.7), but without statistical significance (p = 0.06 after Bonferroni correction) due to the high spread in the H37Rv group (5.3–71.8). (D) TNF-α levels were higher in the bone marrow of H37Rv-infected mice compared to Beijing-1585-infected mice at all time points evaluated. EAI-1627 consistently showed intermediate results between Beijing-1585 and H37Rv for all cytokines. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001, **p < 0.0001 after Bonferroni correction. Figure 6. Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 Discussion iNOS-expression levels in AM and iM/DC were reduced both in mice infected with Beijing-1585 and EAI-1627 compared to H37Rv, consistent with previous studies15,24,49. Low iNOS expression was previously shown to be associated with low expression levels of IFN-γ and TNF-α mRNA32. Remarkably, we observed that lower iNOS expression by iM/DC was accompanied with higher IFN-γ and TNF-α Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 www.nature.com/scientificreports/ Figure 7. Expression of IFN-α, IFN-β and type 1 interferon-inducible genes in lung and bone marrow during infection with different Mtb strains. (A) Combined expression of IFN-α 1,2,5,6,7 mRNA in lung homogenate did not show significant differences between groups. No expression could be detected in uninfected mice (N/A). (B) Expression of IFN-β mRNA in lung homogenate was higher in the lungs of H37Rv-infected mice compared to EAI-1627-infected mice at 3 dpi. Other differences did not reach significance. (C) Combined expression in the lungs of type 1 interferon-inducible genes Mx1 (black), IFI44 (light grey) and CCL2 (dark grey), represented in a type 1 interferon signature, showed the highest level in the Beijing-1585 group at 7 dpi and in the H37Rv group at 14 dpi. For individual expression levels see Fig. S6. (D) Expression of IFN-β in the bone marrow is higher in the H37Rv group compared to Beijing-1585 at 3 dpi. N = 6 mice per group per time point, p < 0.05, p < 0.01, p < 0.001 after Bonferroni correction. Two-way, repeated measure ANOVA followed by Bonferroni correction was used to calculate significance for C. Figure 7. Expression of IFN-α, IFN-β and type 1 interferon-inducible genes in lung and bone marrow during infection with different Mtb strains. (A) Combined expression of IFN-α 1,2,5,6,7 mRNA in lung homogenate did not show significant differences between groups. No expression could be detected in uninfected mice (N/A). (B) Expression of IFN-β mRNA in lung homogenate was higher in the lungs of H37Rv-infected mice compared to EAI-1627-infected mice at 3 dpi. Other differences did not reach significance. (C) Combined expression in the lungs of type 1 interferon-inducible genes Mx1 (black), IFI44 (light grey) and CCL2 (dark grey), represented in a type 1 interferon signature, showed the highest level in the Beijing-1585 group at 7 dpi and in the H37Rv group at 14 dpi. For individual expression levels see Fig. S6. Discussion (D) Expression of IFN-β in the bone marrow is higher in the H37Rv group compared to Beijing-1585 at 3 dpi. N = 6 mice per group per time point, p < 0.05, p < 0.01, *p < 0.001 after Bonferroni correction. Two-way, repeated measure ANOVA followed by Bonferroni correction was used to calculate significance for C. protein levels in the lungs of Beijing-1585- and EAI-1627-infected mice compared to H37Rv at 14 dpi. This low level iNOS expression in AM and iM/DC in the lungs despite high local IFN-γ protein levels could be due to inhi- bition or prevention of iNOS induction at the site of infection. Alternatively, iNOS expression might be affected by differential priming of myeloid cells at an earlier stage of development.fl f In support of our hypothesis that differential priming might play a role we found that the expression of inflam- matory cytokines such as IFN-γ and TNF-α in the bone marrow differs significantly between mice infected with different bacterial strains. Interestingly, this occurs already in the early stage of infection before widespread bac- terial dissemination. While not conclusive, this suggests that differential bone marrow priming of myeloid cells might be an important shaping factor early during infection with Mycobacterium tuberculosis.fif Next to differential expression of IFN-γ in the bone marrow, we found significant differences in type 1 inter- feron expression and responses in both bone marrow and lung between Beijing-1585 and H37Rv. Previous studies have associated virulent Beijing-HN878 infection with elevated IFN-α mRNA expression in the lungs at 28 dpi in a low-dose BALB/c infection model19,20. We were unable to reproduce this preferential increase in IFN-α mRNA upon infection with the virulent Beijing strain in the lungs, which could be due to using a high-dose infection model and/or measurements at different time points. However, other parameters such as bacterial load kinetics and host responses in H37Rv infection in our model show strong similarities to low-dose (102 CFU) infection studies17,50 suggesting that strain-dependent virulence plays a dominant role in pathogenicity over bacterial load. Also, in our subsequent analysis of both generic type 1 interferon-inducible gene expression and pulmonary IFN-β expression we found generally lower expression levels after infection with Beijing-1585 and EAI-1627 compared to H37Rv, thus questioning the direct association of Mycobacterium tuberculosis virulence with type 1 interferon activity during acute infection. Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 Mycobacterial strains. Mycobacterial strains. Mice and infection. Female specific pathogen-free BALB/c mice aged 10–11 weeks and weighing 22–24 grams (Charles River, Les Oncins, France) were infected by intratracheal instillation under general anesthesia as described previously25. Inoculum sizes were confirmed by plating and were 1.0.105 colony forming units (CFU) for Beijing-1585, 1.3.105 CFU for EAI-1627 and 1.8.105 CFU for H37Rv. Mice infected with Beijing-1585 or EAI-1627 rapidly become moribund between 3–5 weeks18, therefore mycobacterial loads and other parame- ters for these two clinical strains were measured up to the peak of infection at 14 dpi, while measurements on H37Rv-infected animals were continued up to peak of infection at 28 dpi. Ethics statement. All protocols were approved by the Erasmus MC institutional animal ethics commit- tee (DEC number 117-12-13, EMC-number 3005) and adhered to the rules laid down in the Dutch Animal Experimentation Act and the EU Animal Directive 201/63/EU. All methods were performed in accordance with the relevant guidelines and regulations. The Beijing-1585 and EAI-1627 mycobacterial strains are part of the sample collection of the National Tuberculosis Reference Laboratory, National Institute of Public Health and the Environment (RIVM, Bilthoven, the Netherlands). Determination of mycobacterial load. Lungs and spleens were removed aseptically and homogenized in 2 mL PBS using the gentleMACS Octo Dissociator (Miltenyi Biotec BV, Leiden, the Netherlands) according to the manufacturer’s protocol. From each tissue homogenate 10-fold serial dilutions were made in PBS. Next, 200 µL aliquots were plated on 7H10 agar culture plates supplemented with 10% OADC. Plates were incubated for up to 42 days at 37 °C and 5% CO2 before colonies were counted. Flow cytometry. The flow cytometry protocol, fluorescent antibody panels, real-time quantitative PCR and cytokine assessments were essentially as described previously51. Briefly, lung single-cell suspensions were fixed for 30 min in fix/perm solution (Ebioscience, Vienna, AT) to eliminate live bacteria prior to flow cytometry analysis. Next, cells were washed and incubated for 10 min in permeabilization buffer (Ebioscience) and then incubated for 30 min with different mAb mixes as described Supplementary Table 1. After staining, the cells were washed again, resuspended in PBS/BSA/azide buffer and measured on a FACS Canto II flow cytometer (BD Biosciences, Breda, NL). Real-time quantitative PCR. RNA from mouse lung homogenate was purified and processed as described previously51. Primer sequences and manufacturers are listed in Supplementary Table 2. Cytokine protein levels. Discussion y g Taken together, we show in a mouse TB model that infection with highly virulent clinical Beijing- and EAI-strains is associated with influx of B-cells and elevated IL-4 levels in the lungs, while less virulent H37Rv www.nature.com/scientificreports/ bacteria induce T-cell influx and higher IFN-γ and IL-17a levels at peak infection. Induction of iNOS is hampered in lung-infiltrating myeloid cells, especially in Beijing-infected mice. A significantly reduced expression in the bone marrow of IFN-γ, TNF-α and IFN-β, early in infection with virulent clinical Mtb strains, might contribute to this poor responsiveness of myeloid cells recruited in the lungs. bacteria induce T-cell influx and higher IFN-γ and IL-17a levels at peak infection. Induction of iNOS is hampered in lung-infiltrating myeloid cells, especially in Beijing-infected mice. A significantly reduced expression in the bone marrow of IFN-γ, TNF-α and IFN-β, early in infection with virulent clinical Mtb strains, might contribute to this poor responsiveness of myeloid cells recruited in the lungs. Mycobacterial strains. To determine cytokine levels in the lungs, lung homogenate was placed in a low-adhesion tube (USA Scientific, Orlando, FL, USA) and centrifuged at 10.000 × g for 5 min. The supernatant was collected, placed in an Eppendorf tube with a 0.45 μm filter (Corning BV Life Sciences, Amsterdam, NL) and centrifuged again at 10.000 × g for 5 min to filter out all mycobacteria. IFN-γ, TNF-α, IL-4, IL-6, IL-17a, IL-23 and IL-10 concentrations were measured in the filtrate using a Luminex assay according to manufacturer’s instructions (Merck Millipore, Amsterdam, NL). Data analysis and statistics. Flow cytometry data were analyzed using Flowjo 7.6.5. Analyses were done and graphs were made using PRISM GraphPad 7. All data are expressed as mean ± SEM. Student’s t-test, followed by Bonferroni correction for multiple comparisons where applicable, was used to calculate significance, except for Fig. 7B. Here we used two-way, repeated measure ANOVA. P-values less than 0.05 were considered statistically significant. Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 Materials and Methods Mycobacterial strains. We used the Mycobacterium tuberculosis H37Rv strain (ATCC 27294) belonging to the Euro-American lineage and two strains isolated from patients in Vietnam in 2002, Beijing-1585, and EAI- 1627, as representatives for their respective lineage based on genotyping results27. References 1. World Health Organisation, Global tuberculosis report 2018, https://reliefweb.int/report/world/global-tuberculosis-report-2018 (2018). 2. Manson, A. L. et al. Genomic analysis of globally diverse Mycobacterium tuberculosis strains provides insights into the emergence and spread of multidrug resistance. Nat Genet 49, 395–402 (2017). 2. Manson, A. L. et al. Genomic analysis of globally diverse Mycobacterium tuberculosis strains provides insights into the emergence and spread of multidrug resistance. Nat Genet 49, 395–402 (2017). p g 3. Merker, M. et al. Evolutionary history and global spread of the Mycobacterium tuberculosis Beijing lineage. Nat Genet 47, 242–249 (2015). p g 3. Merker, M. et al. Evolutionary history and global spread of the Mycobacterium tuberculosis Beijing lineage. Nat Genet 47, 242–249 (2015). ( ) 4. Gagneux, S. & Small, P. M. Global phylogeography of Mycobacterium tuberculosis and implications for tuberculosis produc development. Lancet Infect Dis 7, 328–337 (2007). p f 5. Tientcheu, L. D. et al. Immunological consequences of strain variation within the Mycobacterium tuberculosis complex. European journal of immunology 47, 432–445 (2017). p f 5. Tientcheu, L. D. et al. Immunological consequences of strain variation within the Mycobacterium tuberculosis complex. European journal of immunology 47, 432–445 (2017). j f gy 6. Sun, Y. J., Lee, A. S., Wong, S. Y. & Paton, N. I. Association of Mycobacterium tuberculosis Beijing genotype with tuberculosis relapse in Singapore. Epidemiol Infect 134, 329–332 (2006).i j f gy 6. Sun, Y. J., Lee, A. S., Wong, S. Y. & Paton, N. I. Association of Mycobacterium tuberculosis Beijing genotype with tuberculosis relapse in Singapore. Epidemiol Infect 134, 329–332 (2006).i g p p f , ( ) 7. Huyen, M. N. et al. Tuberculosis relapse in Vietnam is significantly associated with Mycobacterium tuberculosis Beijing genotype infections. J Infect Dis 207, 1516–1524 (2013). Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 9 www.nature.com/scientificreports/ 8. Parwati, I. et al. Mycobacterium tuberculosis Beijing genotype is an independent risk factor for tuberculosis treatment failure in Indonesia. J Infect Dis 201, 553–557 (2010). f 9. Gurjav, U. et al. Transmission of multi-drug resistant tuberculosis in Mongolia is driven by Beijing strains of Mycobacterium tuberculosis resistant to all first-line drugs. Tuberculosis (Edinb) 101, 49–53 (2016).h i g 0. Hang, N. T., Maeda, S., Keicho, N., Thuong, P. H. & Endo, H. Sublineages of Mycobacterium tuberculosis Beijing genotype strain and unfavorable outcomes of anti-tuberculosis treatment. Tuberculosis (Edinb) 95, 336–342 (2015). 11. Burman, W. J. et al. References tuberculosis W/Beijing strains upregulate type I IF regulators of the Jak-Stat pathway. J Interferon Cytokine Res 25, 694–701 (2005). 21. Reed, M. B. et al. A glycolipid of hypervirulent tuberculosis strains that inhibits the innate immune response. Nature 431, 84–87 (2004). 22. Tsenova, L. et al. Virulence of selected Mycobacterium tuberculosis clinical isolates in the rabbit model of meningitis is depen on phenolic glycolipid produced by the bacilli. J Infect Dis 192, 98–106 (2005). p g y p p y f 3. Shang, S. et al. Increased Foxp3 expression in guinea pigs infected with W-Beijing strains of M. tuberculosis. Tuberculosis (Edinb 91, 378–385 (2011). 24. Aguilar, D. et al. Mycobacterium tuberculosis strains with the Beijing genotype demonstrate variability in virulence associated with transmission. Tuberculosis (Edinb) 90, 319–325 (2010). 5. Mourik, B. C. et al. Assessment of Bactericidal Drug Activity and Treatment Outcome in a Mouse Tuberculosis Model Using a Clinical Beijing Strain. Antimicrob Agents Chemother 61, e00696-17 (2017).ii j g g 6. De Steenwinkel, J. E. et al. Immunological parameters to define infection progression and therapy response in a well-defined tuberculosis model in mice. Int J Immunopathol Pharmacol 22, 723–734 (2009). 7. de Steenwinkel, J. E. et al. Drug susceptibility of Mycobacterium tuberculosis Beijing genotype and association with MDR TB. Emerg Infect Dis 18, 660–663 (2012).l 8. Krishnan, N. et al. Mycobacterium tuberculosis lineage influences innate immune response and virulence and is associated with distinct cell envelope lipid profiles. PLoS One 6, e23870 (2011).i p p pi ( ) 29. de Steenwinkel, J. E. et al. Course of murine tuberculosis and response to first-line therapy depends on route of infection and inoculum size. Int J Tuberc Lung Dis 15, 1478–1484 (2011). noculum size. Int J Tuberc Lung Dis 15, 1478–1484 (2011). g 0. Mayer-Barber, K. D. et al. Innate and adaptive interferons suppress IL-1alpha and IL-1beta production by distinct pulmonary myeloid subsets during Mycobacterium tuberculosis infection. Immunity 35, 1023–1034 (2011).i 31. Cooper, A. M. et al. Mice lacking bioactive IL-12 can generate protective, antigen-specific cellular responses to mycobact infection only if the IL-12 p40 subunit is present. J Immunol 168, 1322–1327 (2002).l 32. Mishra, B. B. et al. Nitric oxide controls the immunopathology of tuberculosis by inhibiting NLRP3 inflammasome-depen processing of IL-1beta. Nat Immunol 14, 52–60 (2013). p g ( ) 3. Antonelli, L. R. et al. References Intranasal Poly-IC treatment exacerbates tuberculosis in mice through the pulmonary recruitment of a pathogen-permissive monocyte/macrophage population. J Clin Invest 120, 1674–1682 (2010). 4. Rubio, D. et al. Crosstalk between the type 1 interferon and nuclear factor kappa B pathways confers resistance to a lethal viru infection. Cell Host Microbe 13, 701–710 (2013). 35. Greenberg, S. A. et al. Relationship between disease activity and type 1 interferon- and other cytokine-inducible gene expressi blood in dermatomyositis and polymyositis. Genes Immun 13, 207–213 (2012). y y y 36. Berry, M. P. et al. An interferon-inducible neutrophil-driven blood transcriptional signature in human tuberculosis. Nature 973–977 (2010). 37. Lu, L. L. et al. A functional role for antibodies in tuberculosis. Cell 167, 433–443 (2016). 37. Lu, L. L. et al. A functional role for antibodies in tuberculosis. Cell 167, 433–443 (2016). 38. Joosten, S. A. et al. Patients with Tuberculosis Have a Dysfunctional Circulating B-Cell Compartment, Which Normalizes follo Successful Treatment. PLoS Pathog 12, e1005687 (2016).f 39. Phuah, J. et al. Effects of B Cell Depletion on Early Mycobacterium tuberculosis Infection in Cynomolgus Macaques. Infect Immun 84, 1301–1311 (2016). ( ) 40. Kozakiewicz, L. et al. B cells regulate neutrophilia during Mycobacterium tuberculosis infection and BCG vaccination by modulating the interleukin-17 response. PLoS Pathog 9, e1003472 (2013). p g 1. Gopal, R. et al. Unexpected role for IL-17 in protective immunity against hypervirulent Mycobacterium tuberculosis HN878 infection. PLoS Pathog 10, e1004099 (2014). infection. PLoS Pathog 10, e1004099 (2014). 42. Domingo-Gonzalez, R. et al. Interleukin-17 limits hypoxia-inducible factor 1alpha and development of hypoxic granulomas during tuberculosis. JCI Insight 2, 92973 (2017). 43 A d i A l G 3 i l h id ll di l i i i i b l i N 570 528 532 (2019) g 3. Ardain, A. et al. Group 3 innate lymphoid cells mediate early protective immunity against tuberculosis. Nature 570, 528–532 (2019) 4 H d P d R & R k G A Th l f TNF l h i T ll di d i fl i d d h Th1/Th2 ki 44. Hernandez-Pando, R. & Rook, G. A. The role of TNF-alpha in T-cell-mediated inflammation depends on the Th1/Th2 cytokine balance. Immunology 82, 591–595 (1994).fi 45. Manca, C. et al. Differential monocyte activation underlies strain-specific Mycobacterium tuberculosis pathogenesis. Infect Immun 72, 5511–5514 (2004). 6. Guler, R. et al. Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 References Relapse associated with active disease caused by Beijing strain of Mycobacterium tuberculosis. Emerg Infec 15, 1061–1067 (2009). 12. Casali, N. et al. Evolution and transmission of drug-resistant tuberculosis in a Russian population. Nat Genet 46, 279–286 (20 g p p 13. Parwati, I., van Crevel, R. & van Soolingen, D. Possible underlying mechanisms for successful emergence of the Mycobacte tuberculosis Beijing genotype strains. Lancet Infect Dis 10, 103–111 (2010). j g g y f 14. Domenech, P. et al. Unique Regulation of the DosR Regulon in the Beijing Lineage of Mycobacterium tuberculosis. J Bacteriol 199, e00696-16 (2017).ff 5. Lopez, B. et al. A marked difference in pathogenesis and immune response induced by different Mycobacterium tuberculosis genotypes. Clin Exp Immunol 133, 30–37 (2003).l g yp p 6. van Laarhoven, A. et al. Low induction of proinflammatory cytokines parallels evolutionary success of modern strains within the Mycobacterium tuberculosis Beijing genotype. Infect Immun 81, 3750–3756 (2013).h y j g g yp f 17. Ordway, D. et al. The hypervirulent Mycobacterium tuberculosis strain HN878 induces a potent TH1 response followed by rapid down-regulation. J Immunol 179, 522–531 (2007).fi g 18. de Steenwinkel, J. E. et al. Consequences of noncompliance for therapy efficacy and emergence of resistance in murine tuberculosis caused by the Beijing genotype of Mycobacterium tuberculosis. Antimicrob Agents Chemother 56, 4937–4944 (2012).h y j g g yp y g 19. Manca, C. et al. Virulence of a Mycobacterium tuberculosis clinical isolate in mice is determined by failure to induce T i it d i i t d ith i d ti f IFN l h /b t P N tl A d S i USA 98 5752 5757 (2001) y j g g yp y g 9. Manca, C. et al. Virulence of a Mycobacterium tuberculosis clinical isolate in mice is determined by failure to induce Th1 type immunity and is associated with induction of IFN-alpha/beta. Proc Natl Acad Sci USA 98, 5752–5757 (2001). 19. Manca, C. et al. Virulence of a Mycobacterium tuberculosis clinical isolate in mice is determined by failure to induce Th1 immunity and is associated with induction of IFN-alpha/beta. Proc Natl Acad Sci USA 98, 5752–5757 (2001). 20. Manca, C. et al. Hypervirulent M. tuberculosis W/Beijing strains upregulate type I IFNs and increase expression of negative regulators of the Jak-Stat pathway. J Interferon Cytokine Res 25, 694–701 (2005). 20. Manca, C. et al. Hypervirulent M. Acknowledgements g The authors thank from the Erasmus University Medical Center: S. van den Berg and M. T. ten Kate from the Department of Medical Microbiology and Infectious Diseases, M. A. W. Smits, A. van Oudenaren and L. Hogenkamp from the Department of Immunology for their technical support, and J. Hagoort from the Department of Communication for his critical reading of the manuscript. Research for this manuscript was (in part) performed within the framework of the Erasmus post-graduate school Molecular Medicine. This work was supported by the Royal Dutch Academy of Sciences, the Dr. Hendrik Muller Vaderlandsch Fonds and the Erasmus post-graduate school Molecular Medicine. References IL-4Ralpha-dependent alternative activation of macrophages is not decisive for Mycobacterium tuberculosis pathology and bacterial burden in mice. PLoS One 10, e0121070 (2015).f gy 47. Reyes-Martinez, J. E. et al. Differential activation of dendritic cells by Mycobacterium tuberculosis Beijing genotype. Immunological investigations 43, 436–446 (2014).t 48. Khader, S. A. et al. Interleukin 12p40 is required for dendritic cell migration and T cell priming after Mycobacterium tuberculosis infection. The Journal of experimental medicine 203, 1805–1815 (2006).i h f p 9. Koo, M. S., Subbian, S. & Kaplan, G. Strain specific transcriptional response in Mycobacterium tuberculosis infected macrophages Cell Commun Signal 10, 2 (2012). Scientific Reports \| (2019) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6 10 www.nature.com/scientificreports/ 0. Manca, C. et al. Mycobacterium tuberculosis CDC1551 induces a more vigorous host response in vivo and in vitro, but is not more virulent than other clinical isolates. J Immunol 162, 6740–6746 (1999). 1. Mourik, B. C. et al. Immunotherapy added to antibiotic treatment reduces relapse of disease in a mouse model of tuberculosis. Am J Respir Cell Mol Biol 56, 233–241 (2017). Competing interestsh Competing interestsh p g The authors declare no competing interests. Author contributions B.M. contributed to experiments, data analysis and interpretation and drafting the manuscript. J.S. contributed to the study design, analysis and interpretation of mycobacterial data, the initial drafts of the manuscript and substantial revisions of later versions. G.K. contributed to the mouse experiments and acquisition and analysis of mycobacterial parameters. R.H. contributed to the acquisition and interpretation of the gene expression experiments. A.V. contributed to the interpretation of the data, study design and substantial revisions to the manuscript. T.O. contributed to the analysis and interpretation of the flow cytometry, protein level and gene- expression data. D.S. provided the mycobacterial strains and contributed to the experimental design, analysis of the mycobacterial data and revisions to the manuscript. P.L. contributed to study design, analysis and interpretation of immunological data, the initial drafts of the manuscript and substantial revisions of later versions. All authors approved the final version of the manuscript and have agreed to be personally accountable for their respective contributions and ensure that questions related to the accuracy or integrity of any part of the work, even ones in which the author was not personally involved, are appropriately investigated, resolved, and the resolution documented in the literature. Additional information Supplementary information is available for this paper at https://doi.org/10.1038/s41598-019-56300-6. Supplementary information is available for this paper at https://doi.org/10.10 Correspondence and requests for materials should be addressed to P.J.M.L. Reprints and permissions information is available at www.nature.com/reprints. 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 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) 9:19922 \| https://doi.org/10.1038/s41598-019-56300-6
https://openalex.org/W3083360348	https://bovine-ojs-tamu.tdl.org/AABP/article/download/7384/7249	English	null	Neonatal Calf Management	American Association of Bovine Practitioners Conference	1,983	cc-by	7,166	Weak Calf Syndrome (WCS) A disease characterized by late abortions, stillbirths, weak calves at birth, 1and surviving calves which become chronic "runts" was reported in Montana, Idaho, and Oregon in the I 960's. The condition has more recently been reported in Nebraska and other parts of the mid west. The most common sign was weak calves at birth or in the first few days of life. Most calves affected would die. Plant teratogens which cause lethal anomalies exist in certain localities, but these also are not important on a national scale. No etiological agent has been determined to date, and the name 'weak Calf Syndrome' has been adopted to encompass a single or perhaps a multitude of entities. 23 Cow - Calf Split Session I Dr. Duane Miksch, presiding 0 "'O (D 0 "'O (D ~ ~ (") (D 00 00 0.. ...... 00 ,-+- '"i ~ ~ ...... 0 p Causes Congenital anomalies have always been a small source of perinatal and neonatal fatalities. Occasionally the incidence of congenital anomalies can be alarmingly high in an individual herd. This occurs when ( 1) a teratogenic agent invades the embryo or fetus at a susceptible stage of development or (2) when the gene frequency for a genetic anomaly is very 1 high with1n the herd. A veterinary practitioner should seek diagnostic assistance before declaring an anomaly is of genetic etiology. Herds having a high incidence of NI should stop using the anaplasmosis vaccine, and calves should receive colostrum from non-vaccinated cows. The vaccine is very effective in preventing anaplasmosis, so the stockman and veterinarian should judge its value on the basis of total economic return. Current recommendations are to vaccinate no more frequently than every two years and the incidence of NI seems to be decreasing. Viral agents such as BVD and bluetongue may cause congenital anomalies if a susceptible dam is infected in the second trimester. These are most commonly exhibited as internal hydrocephalus, cerebellar hypoplasia, and other neurological anomalies. Some calves will be born dead, and others may live but have varying degrees of ataxia. It is important to remember that such conditions are sporadic unless many dams are susceptible and infected at the correct stage of gestation. Losses due to BYD can be prevented by vaccination of all replacements with BYD vaccine. Losses due to bluetongue infections have not been fully assessed, but some investigators feel bluetongue is an important cause of calf losses in endemic areas. 6 There is currently no vaccine for bluetongue. Neonatal Calf Management Lawrence E. Rice, D. V. M., M. S. Department of Medicine and Surgery College of Veterinary Medicine Oklahoma State University Stillwater, Oklahoma 74078 condition was identified to be a disease of passive immunity. Calves born to cows previously vaccinated with anaplasmosis vaccine would develop a hemolytic crisis following nursing. These calves would be normal at birth, stand, and nurse but become weak, collapse, and often die within 24 hours after nursing. Perinatal and neonatal calf mortality account for the second greatest loss to the beef cattle industry; the first being open or late calving cows. Large surveys have shown perinatal deaths to range from 4 to 7 percent and neonatal deaths to range from 2 to 6 percent. 18 27 Certain herds and areas have mortality rates much higher in epizootic situations. The purpose of this presentation is to discuss some of the causes of such losses and management practices to reduce the losses. Anaplasmosis vaccine is made from a homogenate of red blood cells from cows acutely infected with anaplasmosis. Antigenic determinants on the red blood cells may be foreign to some cows, so antibodies are produced in response to these RBC antigens. The antigenic determinants are genetic in nature and are linked to those that determine blood types. If the offspring inherits the same blood type from the sire that stimulated antibodies in the dam, NI may occur after colostrum is ingested. The maternal antibodies from colostrum react with the antigenic determinants of the calf s erythrocytes causing hemolysis, anemia, and often death. 24 Neonatal Isoerythrolysis (NI) Calves that survive more than a few days show atrophy of fat and atrophy or involution of the thymus. 5. Calves that survive more than a few days show atrophy of fat and atrophy or involution of the thymus. While no etiological agent has been isolated and identified, the following agents or conditions have been incriminated: ( l) Hypbthermia, 1 (2) Low prepartum protein for dam, 1 (3) BVD, 23 (4) Bovine adenovirus ,4 (5) Mycoplasma, 23 (6) Hemophilus somnus, 28 and (7) Bluetongue virus. 6 The most important protection for the neonate remains proper colostrum intake. Hypogammaglobulinemia may result from: Bull reported in Idaho that cows on very low prepartum protein levels had calves with a higher incidence of weak calf syndrome. He determined that each 0.1 pound of crude protein consumed below two pounds per day resulted in a one percent increase in weak calf syndrome in 12 Idaho herds. The condition was exacerbated by cold weather and did not seem to occur if either prepartum protein intake was adequate or if calves were not born during cold weather. I. Low colostrum intake due either to low volume or inadequate nursing. 2. Poor quality colostrum. 3. Late nursing. 4. Perinatal stress ( catch-all). The signs of WCS have been experimentally produced by intrauterine inoculation with Hemophilus somnus, but H. somnus has not been isolated from affected calves frequently enough to be identified as the etiological agent. Bovine adenovirus has also experimentally caused the signs and lesions. A calf should consume 6%-10% of its body weight as soon as possible after birth. First, the rate of absorption of immunoglobulins from the gut declines from maximum at birth to no absorption at approximately 24 hours postpartum (Table l). Second, there is a lag of approximate- ly 5 hours between sucking and 50% maximum antibody levels in the calfs circulation. 12 This means that a calf that nurses at 2 hours of age is still hypogammaglobulinemic at 6- 7 hours of age and susceptible to invasion by pathogens. Treatment is generally directed towards supportive therapy. It is most important that calves be kept warm and out of drafts. Most have to be fed colostrum with a stomach tube or esophageal feeder. Blood transfusions or plasma from cows that previously produced a weak calf have been reported to be helpful. TABLE 1. Effect of Time of Colostrum Feeding on lg Absorption. Neonatal Isoerythrolysis (NI) TABLE 1. Effect of Time of Colostrum Feeding on lg Absorption. Time of Feeding (Hrs. PP.) 6 12 24 36 48 Plasma lg Levels mg/ml 52.7 37.5 9.2 5.4 4.8 % Absorption 66 47 12 7 6 Weak calf syndrome ·may be a 'catch-all' for perinatal deaths in some areas of the United States. The symptoms and proposed etiological agents are varied, but there is little doubt th~t a specific condition exists in some areas. Besides the identified protein problem in Idaho, there is probably an infectious disease contacted in utero because it is seen most commonly in offspring from first-calf-heifers or new additions to the herd. The herds affected in Nebraska were in excellent nutritional status and an infectious agent was suspecte?. Prepartum nutrition levels have been shown to affect cow colostrum volume and calf serum lg levels following colostrum ingestion. Cows wintered on free choice silage produced 3.5 pints of colostrum at first milking, while cows wintered on available pasture produced 1.2 pints of colostrum. There was a marked difference in total colostral lg levels, and 70% of the calves from the pastured cows were hypogammaglobulinemic after nursing 14 (Table 2). Neonatal Isoerythrolysis (NI) Symptoms reported include: In the late 1960's a fatal hemolytic disease of newborn calves was observed with increasing frequency. This I. Stillbirths. 2. Weak calves at birth, depressed, unable or unwilling to APRIL, 1984 137 nurse, often dying in a few hours after birth. nurse, often dying in a few hours after birth. nurse, often dying in a few hours after birth. inutero against £. coli by i~jection of the antigens into the amnionic fluid. These calves withstood the challenge of inoculation with virulent organisms before nursing while controls succumbed to colibacillosis. 2 3. Diarrhea at birth. 4. Polyarthritis with heat and swelling of carpal and tarsal joints. 4. Polyarthritis with heat and swelling of carpal and tarsal joints. Calves are normally born agammaglo bulinemic. lmmunocompetence is probably the poorest from birth to 2 weeks of age than at any other time. The fetal corticosteroids that are responsible for the initiation of parturition remain at high levels for several days following birth. These high levels of corticosteroids suppress immunogenesis during the first two weeks of life. This immunosuppression is mediated through reduced thymus activity and lymphocyte numbers which results in interference with cellular and humoral immunity. 16 At seven to ten days of age the corticosteroid levels return to normal, and calves have been shown to be able to develop immunity against colibacillosis after this time. Calves that are agammaglobulinemic during this time are likely to die of colibacillosis. On the other hand, properly managed calves that received adequate colostrum soon after birth are likely to survive this critical period and develop immunity through natural exposure. 5. Affected calves are very susceptible to any stress at birth such as hypothermia and secondary infections. They are refractory to treatment and may become chronic 'runts.' 5. Affected calves are very susceptible to any stress at birth such as hypothermia and secondary infections. They are refractory to treatment and may become chronic 'runts.' The most consistent necropsy finds are: 1. Edema and hemorrhage of extremities, especially of the carpus and tarsus. 1. Edema and hemorrhage of extremities, especially of the carpus and tarsus. 2. Polyarthritis. 2. Polyarthritis. 2. Polyarthritis. 3. Hyperemia of small intestine. 4. Occasionally meningitis. 0 "'O (D ~ ~ (") (D 00 00 0.. ...... 00 ,-+- '"i ~ ~ ...... 0 p 5. Bovine Fetal and Neonatal Immunity The bovine fetus is immunocompetent during the last trimester, and calves have been experimentally immunized THE BOVINE PROCEEDINGS-No. 16 138 TABLE 4. Calf Heat Stress - lg Absorption and Calf Mortality. Temperature Humidity Index* 81.2 82.4 84.5 No. Calves 36 36 36 Calf 1 2 9 Mortality (2.7%) (5.6%) (25% ) 2 Hr. serum Cortisol (mg/ml) 42 42 56 lg Cone. 2 days (mg/ml) 25.5 22.0 18.6** * Comfort Maximum = 73 - P < 0.01 TABLE 2. Effect of Prepartum Nutrition on Cow Colostrum Levels and Calf Serum lg Levels TABLE 2. Effect of Prepartum Nutrition on Cow Colostrum Levels and Calf Serum lg Levels TABLE 2. Effect of Prepartum Nutrition on Cow Colostrum Levels and Calf Serum lg Levels First Milking Prepartum Vol. Calves Cows Body Weight Colostrum Total lg Hypo lg Silage 1,043 3.5 pts. 108 gm 27% Pasture 761 1.2 pts.** 39 gm* 70% * - P< 0.01 ** - P< 0.0005 Regardless of nutritional levels all hypogammaglobuline- mic calves were from cows with the lowest levels of lg in the colostrum. Poor nutrition reduced total volume of colostrum in the pastured cows. While the cortisol levels were higher in the extremely stressed calves, the role of cortisol in lg absorption is not clear in this study. The cortisol level may only be in response to the heat stress. Other studies have indicated that high cortisol levels in calves at birth did not interfere with lg absorption. 10 Also exogenous corticosteroids used for induced parturition apparently do not hinder absorption, but the early calving may reduce the volume of colostrum. 9 A similar study was conducted in Wyoming where cows were fed 50% energy requirements starting 100 days before expected calving. Thirty days before calving half of the cows were fed 117% energy requirements while other remained on the low level of nutrition. Calves born to nutritionally deprived cows were much more susceptible to diarrhea and mortality was markedly increased 3 (Table 3). Calf immuno- globulin levels were not reported, but prepartum nutrition had marked effects on calf survival and weaning weight. Dystocia remains the number one cause of perinatal calf mortality. Young in Australia reported that dystocia was responsible for 70% of all deaths at or near birth, and this amounted to 45% of all calf deaths including abortion and postnatal deaths. Bovine Fetal and Neonatal Immunity 29 Patterson and Bellows reported similar figures of 60% and 38% 18 In all reports most of the dystocia cases are first-calf-heifers. TABLE 3. Effect of Restricting Prepartum Energy Intake and Elevating it 30 Days Prior to Calving. g p gy g it 30 Days Prior to Calving. Continuous Low* Energy Level Cow prepartum Wt. Change -145 Calf Birth Weight 59 Calf Survival to Weaning 71 Calf Scours - % Affected 52 % Mortality 19 Low-High** -22 76 90 33 0 * 50% of energy requirement for last 100 days prepartum * * 50% of energy requirement for days -100 to -30 then switched to 117% for last 30 days prepartum. Stillbirths are not the only result of dystocia, but many calves are injured that later die or are permanently injured. Such injuries reported have been slipped femoral capitulum epiphysis, 7 femoral nerve stretching and paralysis, ruptured quadriceps muscles , meningeal hemorrhage and congestion. 8 Most of the above lesions are due to excessive forced traction where a caesarean section would have resulted in a live calf. * 50% of energy requirement for last 100 days prepartum * * 50% of energy requirement for days -100 to -30 then switched to 117% for last 30 days prepartum. Many factors have been attributed to causing dystocia. Stockmen have traditionally held to the premise that calving difficulty can be controlled by limiting prepartum nutrition so calf birth weight is reduced. Research and experience have conclusively shown that only extremes in body conditions cause high dystocia rates. That is, thin and small heifers have more difficulty because of weakness and small size in spite of somewhat reduced birth weights. Conversely, very fat heifers 1,have more difficulty because internal pelvic fat decreases the pelvic area of the dam. A general rule of thumb is to have heifers at 85% of their mature weight at calving (including calf weight) to minimize dystocia and maximize subsequent breeding efficiency. Perinatal Stress In spite of adequate colostrum consumption in the early hours of life approxirnately 20-30% of neonatal calves remain hypogammaglobulinemic. 16 The reason for this is not known but probably accounts for some of the consistent low levels of colibacillosis in the best managed herds: A generalized category of perinatal stress may explain some of the variation of immunoglobulin levels. These include: l. Environmental stress. Calf birth weight has consistently been shown to be the most important factor associated with dystocia. 22 Birth weight has moderate to high heritability, and the sire effects on dystocia and perinatal mortality are quite variable. This means dystocia and mortality rates can be reduced by select- ing bulls whose progeny cause few difficult births. 20 27 This discussion will not dwell further on causes or prevention of 2. Dystocia. 3. Poor 'mothering-up'. Arizona calves subjected to extreme heat stress at birth had higher cortisol levels, lower immunoglobulin levels, and higher mortality rates 25 (Table 4). All calves were subjected to heat above the comfort range, but those under extreme stress had lower lg levels and higher mortality rates. APRIL, 1984 139 TABLE 7. Cortisol and lgG, lgM, lgA Serum Levels in Dystocial Calves. TABLE 7. Cortisol and lgG, lgM, lgA Serum Levels in Dystocial Calves. dystocia but the effects of dystocia on neonatal life. The bovine fetal genotype has recently been shown to have marked effects on the fetal and maternal endocrine systems. This 'control' of the hormone levels influences parturition, milk production, and postpartum fertility. One report showed the effect of progeny from five sites on maternal urinary estrogens and stillbirths17 (Table 5). The bovine fetal genotype has recently been shown to have marked effects on the fetal and maternal endocrine systems. Hours Postpartum Dystocial Eutoclal Cortisol 0 162 mg/ml* 211 mg/ml lgM 24 2.4 mg/ml 2.1 mg/ml lgG 24 12.3 mg/ml 12.2 mg/ml lgA 24 1.9 mg/ml 1.8 mg/ml * -P < 0.01 a ed e ects o t e eta a d ate a e doc e syste s. This 'control' of the hormone levels influences parturition, milk production, and postpartum fertility. One report showed the effect of progeny from five sites on maternal urinary estrogens and stillbirths17 (Table 5). TABLE 5. Sire Effects on Maternal Hormone Relationships to Dystocia and Perinatal Deaths. Maternal Stillbirths Sire No. Perinatal Stress E2 (ub) in Heifers 1150* 3.0% 2 1543 5.5% 3 1297 9.1% 4 1098* 20.1% 5 1057* 33.0% * - < 0.05 TABLE 5. Sire Effects on Maternal Hormone Relationships to Dystocia and Perinatal Deaths. 24 hours, and had they not been force fed perhaps there would have been a difference in circulating antibodies at 24 hours. Recent studies reported by Doornbos in Montana and Putnam in Oklahoma, indicate that the length of Stage II of normal parturition is shorter than classically described in textbooks. Length of Stage II is reported to range from 0.5-4 hours with averages from 1-3 hours depending on the age of the dam. Most texts do not recommend intervention until Stage II has been in progress 2-3 hours without completion. With the exception of sire No. 1 there was a significant relationship between low maternal estrogens and higher dystocia and stillbirth rates. Such a relationship is possible since fetal cortisol initiates the events leading to a normal parturition. This includes elevated maternal estrogens prior to parturition. Both studies defined Stage II as clinically beginning with abdominal press and appearance of fetal membranes (water bag) at the vulva. In Montana the average length of Stage II in unassisted births of first-calf-heifers was approximately 1 hour and in adult cows approximately 0.5 hour. 5 p O'Brien and Stott showed a difference between prepartum maternal hormone levels in eutocial and dystocial parturitions 15 (Table 6). Putnam reported similar results with first-calf heifers. In this study, heifers that did not deliver after two hours of Stage II labor were to be assisted. Heifers that calved in less than two hours after the appearance of membranes did so in 55 minutes. Heifers that calved in more than two hours did so in 162 minutes 21 (Table 8). Of the 15 heifers that required more than two hours of labor, only four calved unassisted and these were obviously going to calve unassisted at the end of the two hour period. TABLE 6. Prepartum Maternal Hormones Related to Dystocia. Cortisol Estradiol 17 Progesterone * - P < 0.01 Eutocla Dystocia Higher Lower Higher Lower* (15-20 days prepartum) Lower Higher* TABLE 6. Prepartum Maternal Hormones Related to Dystocia. Cortisol Estradiol 17 Progesterone * - P < 0.01 Eutocla Dystocia Higher Lower Higher Lower* (15-20 days prepartum) Lower Higher* TABLE 6. Prepartum Maternal Hormones Related to Dystocia. TABLE 8. Perinatal Stress Length of Stage II For Heifers Calving Less Than 2 Hours and Those Calving Greater Than or Equal to 2 Hours. < 2 Hours (Mins) No. Heifers - 31 > 2 Hours (Mins) Average Length - 55.48 15 162.33 P < .001 < 2 Hours (Mins) No. Heifers - 31 > 2 Hours (Mins) Average Length - 55.48 15 162.33 P < .001 Other works by Stott also showed that dystocial calves consistently had lower cortisol levels at birth than eutocial calves25 (Table 7). The question is: Were the cortisol levels low because of the dystocia, or did the dystocia occur because of the low cortisol levels? Birth weight did not seem to vary between the groups. This allows one to theorize that fetal endocrine action (low cortisol) may not have initiated the normal chain of events of increased estrogens, prostaglandin f2a production, decreased progesterone, release of oxytocin, and perhaps relaxin with cervical dilation and parturition. Future research may answer some of these questions. Pelvic measurements were also taken at parturition. While there was no difference in dystocia rates between heifers with below average pelvic areas and those with above average areas, there was a significant difference in the length of Stage II. There was also a difference in the number of calves that needed assistance in nursing, dystocia or not. Immunoglobulin levels were not determined and there were no neonatal diseases, but what might have happened had there been a higher dystocia rate or if the sluggish calves had not been assisted after two hours of not nursing? Dystocial calves clinically appear to be slower at standing and nursing and more susceptible to infectious diseases. However, Stott found no difference in immunoglobulin levels when the calves were fed pooled colostrum four hours postpartum. The dystocial calves were sluggish for the first g Another question: If eutocias are apparently completed with 30-60 minutes why recommend waiting 2-3 hours after THE BOVINE PROCEEDINGS-No. 16 140 TABLE 9. Pelvic Area and Length of Stage II. Number Heifers Length (Minutes) No. of calves not nursing in 2 hour postpartum Large Pelvic Area 23 68 3 Small Pelvic Area 23 113 7 TABLE 9. Pelvic Area and Length of Stage II. The value of hyperimmunizing pregnant cows with Colostridium perfringens Type C toxoid to provide calves with passive immunity to hemorrhagic enteritis has long been known. Colostrum Management 1. The use of a refractometer to measure serum protein is quick and serum protein levels below 5 mg% can mean that the calf is hypogammaglobulinemic. However, it is the least accurate method. Recent studies have shown that force feeding one gallon of colostrum (10% of body weight) within six hours of birth reduced the incidence of colibacillosis. The object of forced feeding is to increase passive immunity on a herd basis. This may be practical in dairies but not in beef cattle. The rancher must observe his herd closely at calving for problems other than dystocia. 2. Single radial immunodiffusion and serum electrophore- sis are both very accurate but require laboratory facilities and are relatively expensive. 3. The zinc sulfate turbidity test is easy and inexpensive and is accurate enough to be of prognostic value. 1. He must see that the calves are 'mothered-up'. Studies have shown that when the dam licks and cares for the newborn calf, immunoglobulin absorption is significantly increased when colostrum is consumed. 25 Calves separated from the dams at birth had reduced antibody levels even when fed colostrum at the same time as the 'mothered-up' calves. Ranchers and veterinarians have observed that the neglected calf is less vigorous and always needs more attention to 'get going'. g p g 4. I prefer the sodium sulfite precipitation test because it is simple, inexpensive, and also accurate enough for a clinic prognosis. lmmunoglobulins can be precipitated from serum with concentrations of N a2SO3 ranging from 14 to 18%. 19 Lower concentrations fail to precipitate any protein, and concentrations above 18% will precipitate non-lg proteins. Concentrations of 14, 16, and 18% Na2SO3 may be prepared using 14, 16, and 18 gm ofanhydrous Na2SO3 in a total volume of 100 ml distilled water. Serum (0.1 ml) is added to 1.9 ml of each Na2SO3 solution. Mix samples and allow to stand for 30 minutes at room temperature. Observe samples for precipitation. Precipitation may vary but score positive for any flaking or precipitate. An estimation of serum lg levels are shown in Table 10. This test does not differen- tiate between immunoglobulins but is over 90% accurate for total immunoglobulins. Calves with serum lg levels above 50 mg/ ml will precipitate all 3 concentrations and may be considered to have adequate lg levels. Calves with levels below 50 mg / ml are considered hypogammaglo bulinemic. 4. Colostrum Management I prefer the sodium sulfite precipitation test because it is simple, inexpensive, and also accurate enough for a clinic prognosis. lmmunoglobulins can be precipitated from serum with concentrations of N a2SO3 ranging from 14 to 18%. 19 Lower concentrations fail to precipitate any protein, and concentrations above 18% will precipitate non-lg proteins. Concentrations of 14, 16, and 18% Na2SO3 may be prepared using 14, 16, and 18 gm ofanhydrous Na2SO3 in a total volume of 100 ml distilled water. Serum (0.1 ml) is added to 1.9 ml of each Na2SO3 solution. Mix samples and allow to stand for 30 minutes at room temperature. Observe samples for precipitation. Precipitation may vary but score positive for any flaking or precipitate. An estimation of serum lg levels are shown in Table 10. This test does not differen- tiate between immunoglobulins but is over 90% accurate for total immunoglobulins. Calves with serum lg levels above 50 mg/ ml will precipitate all 3 concentrations and may be considered to have adequate lg levels. Calves with levels below 50 mg / ml are considered hypogammaglo bulinemic. 2. He must observe the cow and calf to assure that nursing has occurred within six hours of birth. If the calf has not nursed, it must be assisted or forced fed 2-4 quarts of FIRST MILKING COLOSTRUM. Many beef cows do not give that much colostrum and there is no way of knowing how much colostrum a beef cow has or how much the calf nursed. Only problem calves can be supplemented, but losses will be reduced by feeding these calves. 2. He must observe the cow and calf to assure that nursing has occurred within six hours of birth. If the calf has not nursed, it must be assisted or forced fed 2-4 quarts of FIRST MILKING COLOSTRUM. Many beef cows do not give that much colostrum and there is no way of knowing how much colostrum a beef cow has or how much the calf nursed. Only problem calves can be supplemented, but losses will be reduced by feeding these calves. Even though many beef calves may be hypogammaglobu- linemic, neonatal diseases may be decreased by reduced exposure to organisms. This is achieved by having clean calving grounds and dispersion of pairs into cow-calf pastures. The management goal should be to maximize passive immunity and minimize exposure within the limits of the environment and type of cattle. Evaluation of Passive Immunity There are several diagnostic aids available for the evalua- tion of immunoglobulin levels of the newborn. Perinatal Stress The effectiveness of similar techniques to prevent other enteric conditions is not well documented. Commercial and autogenous Salmonella sp bacterins have been used with mixed results. Cows experimentally vaccinated twice in the last 30 days of gestation with£. coli antigens produced calves that withstood oral challenge following colostrum feeding. 14 Seventy-seven percent of the calves from vaccinated cows were protected from diarrhea. All calves from control cows suffered severe diarrhea, severe dehydration, and hypothermia. Calves not treated with electrolytes died. Commercial E. coli bacterins now available for pregnant cows might be useful in improving the quality of lg in colostrum. Regardless of the quality of lg in the colostrum, passive immunity will not be acquired unless adequate volumes are consumed within six hours of birth. 0 "'O (D ~ ~ ('") (D 00 00 0.. ...... 00 ,-+- '"i ~ ~ ...... 0 p labor has started to intervene and assist the birth? Doornbos found that intervention upon appearance of fetal membranes did no harm to cow or calf. Also subsequent cycling rates and pregnancy rates were significantly higher in early assisted heifers. Obviously more work needs to be done with these procedures but from these data it would appear that earlier assistance could improve fertility of the dam and viability of the calf. These observations may warrant recommending intervention within one hour after the appearance of membranes at the vulva, provided the cervix is completely dilated. 21 Neonatal Nutrition The nursing calf that is receiving enough milk needs no interference by man because it is getting what Mother Nature intended. A popular belief is that cows fed high energy feeds or on green grass give too much milk thus causing scours. Actually very few beef calves get too much milk. Calves nursing good milking cows not only gain better, but are less likely to contact infectioµs diseases. Calves that have marginal or deficient milk consumption are more susceptible to infectious diarrhea and pneumonia. 3 These calves are weak, emaciated, and more difficult to treat because they rapidly dehydrate. Early Weaning Rations Used in 1980 (OSU) Ration Ingredient Starter Ration II Ration Ill % % % Rolled Corn 64.0 56.5 50.1 Soybean meal 20.0 17.0 12.0 Cottonseed hulls 10.0 20.0 33.0 Cane molasses 5.0 5.0 3.0 Dicalcium phosphate .5 Limestone 0.5 0.5 0.5 Potassium chloride 0.5 0.5 Salt 0.5 0.5 0.5 Vit A (30,000 IU gm) 1 lb/ton lb/ton ½ lb/ton Early Weaning Rations Used in 1980 (OSU) Early Weaning Rations Used in 1980 (OSU) Economically there is litt_le that can be done for the beef calf that is receiving marginal amounts of milk. However, the calf that is being starved because its mother does not milk well, has mastitis, sore teats, etc. needs attention. Ranchers should be alert to notice these problems and supplement such calves or find foster mothers. When foster mothers are not available, starving calves can be raised on milk replacer and starter grain just as with dairy calves. Calves were fed the starter for two weeks, Ration I I was fed for six more weeks and Ration I I I was fed until approximately seven months of age. The early weaned calves were 88 pounds heavier (435 lbs. vs 347 lbs.) at seven months when the nursing calves were weaned. While early weaning will not be a routine management practice in beef herds, it can be a feasible salvage practice and should not be overlooked. Whole milk is the most desirable food for all calves. Economically this can not be fed to orphan calves so milk replacers are used. Calves should be fed two quarts twice a day (approximately 8-10 percent of body weight). They should also receive free choice clean water. This will meet hydration requirement and if the milk replacer is of high quality the energy, protein, and .. 0 "'O (D ~ ~ (") (D 00 00 0.. ...... 00 ,-t,, '"i ~ ~ ...... 0 p Calves should be offered a highly palatable calf starter. This should be offered free choice in a clean container that the calves cannot soil. After the calves are 6-8 weeks of age they should be consuming approximately I½ to 2 pounds of starter and can be weaned from the milk replacer. Limitation of milk replacer to one gallon per day encourages feeding on the starter. The neonate bovine is basically monogastric but with this stimulation the rumen soon develops and the calf can be weaned at two months of age. The administration of whole blood to hypogammaglobu- linemic calves may be helpful. Whole blood is suggested because of the difficulty of separating bovine plasma from citrated blood. One to two liters of whole blood can be administered subcutaneously or intraperitoneally with no adverse effect on the calf. Blood can also be administered intravenously but should not exceed 500 to 1000 ml. Intra- venous b'lood is also more time consuming and if the IV use is not necessary the other routes are just as satisfactory. The fiber content of this feed should be less than 10 percent and the protein content should be 18 percent with no urea. The protein supplement would ideally be soybean meal. A void cottonseed meal for the calf starter. Processing of cottonseed meal releases gossypol which is toxic to the preruminant calf. Vitamins 'A' and 'D', Calcium , Phosphorus, and Trace Minerals should be added to balance the calf starter. A study at Oklahoma State University on weaning beef calves at 6-8 weeks of age used the following ration: 13 Colostrum Management This is a simple prognostic test which can be run in the field to determine efficiency of colostrum intake and to plan course of therapy for affected calves. APRIL, 1984 141 TABLE 10. Estimation of Serum lg Content From Sodium-Sulfite Precipitation. Concentration of lg (mg/ml) Na2803 Concentration 14% 16% 18% <5 + 5-15 + + >15 + + + TABLE 10. Estimation of Serum lg Content From Sodium-Sulfite Precipitation. TABLE 10. Estimation of Serum lg Content From Sodium-Sulfite Precipitation. TABLE 10. Estimation of Serum lg Content From Sodium-Sulfite Precipitation. above three factors. Calves less than 4-6 week of age cannot utilize plant proteins. Many inferior milk replacers contain plant protein to make the product less expensive. Unknowing farmers will often buy these products with resulting starvation of the calves. I have seen starving conditions reversed in approxi- mately two weeks by simply changing milk replacers. Concentration of lg (mg/ml) Concentration of lg (mg/ml) management to his clients. 33: 11 66, (1964). 28. Waldham, D.G ., Hall, R.F., Meinershagen, W.A. , Card, C.S., and Frank, F.W.: Haemophilus Somnus Infection in the Cow as a Poss ible Contributing Factor to Weak Calf Syndrome: Isolation and Animal Inoculation Studies. Am. J . Vet. Res. 35: 140 I, ( 1974). 29. Young, J .S., and Balir, J .M.: Perinatal Calf Losses in a Beef Herd. Aust. Vet. J . 50:338, ( 1974). The following procedures should increase perinatal and neonatal survival: I. Have heifers big at calving 2. Breed heifers to sires selected for 'easy calving'. 3. Provide proper prepartum nutrition. 4. Hyperimmunize dams when indicated. 5. Provide adequate and clean calving facilities. 6. Provide proper surveillance and early assistance at calving. 7. Insure adequate and timely colostrum consumption. Neonatal Nutrition 26:211, (1976). 21. Putnam, M .R., Rice, L.E., Wettemann, R.P., Lusby, K.S .: Planipart (Clenbuterol) for the Postponement of Parturition and Alleviation of Dystocia in Cattle. Proc. Soc. Therio. ( 1982). 22. Price, T.D., and Wiltbank, J .N. : Predicting Dystocia in Heifers. Therio. 9:22 1, (1978). 23. Stauber, E.H .: Weak Calf Syndrome: A Continuing Enigma. J .A.V.M.A. 165:223, (1976). 24. Stormont, C. : Neonata l lsoerythrolysis in Domestic Animals: A Comparative Review. Cornell Vet. 25. Stott, G.H.: lmmunoglobulin Absorption in Calf Neonates with Special Consideration of Stress. J . Dairy Sci. (In Press). 26. Tryphonas, L., Hamilton, G.F., and Rhodes, C.S. : Perinatal Femoral Nerve Degeneration and Neurogenic Atrophy of Quadriceps Femoris Muscle in Calves. J .A.V.M.A. 164:801, (1974). 27. Van .Dieten, S.W.J .: Calf Mortality at Birth. Veetelt. en Zwivel Berichten. 7:20-29. In An. Brd. Abstr. 33: 11 66, (1964). 28. Waldham, D.G ., Hall, R.F., Meinershagen, W.A. , Card, C.S., and Frank, F.W.: Haemophilus Somnus Infection in the Cow as a Poss ible Contributing Factor to Weak Calf Syndrome: Isolation and Animal Inoculation Studies. Am. J . Vet. Res. 35: 140 I, ( 1974). 29. Young, J .S., and Balir, J .M.: Perinatal Calf Losses in a Beef Herd. Aust. Vet. J . 50:338, ( 1974). management to his clients. management to his clients. Yield and lmmunoglobulin Concentration in Beef Cows, Br. Vet. J . 133:120, (1977). 12. Logan, E.F., McMurray, D.H ., O'Neill, D.G ., McParland, P.J. , and Mc Rory F.J.: Absorption of Colostral lmmunoglo- bulins by the Neonatal Calf. Br. Vet. J. 134:258, (1978). 13. Lusby, K.S., and Parra, A.A.: Effect of Early Weaning on Calf Performance and on Reproduction in Mature Cows. An. Sci. Res. Rep. M P-108, Ok. St. U. p 64, (1981). 14. Myers, LL., Newman, F.S., Wilson, R.A., and Catlin, J.E.: Passive Immunization of Calves Against Experimentally Induced Enteric Colibacillosis by Vaccination of Dams. Am. J. Vet. Res. 34:29, (1973). 15. O'Brien, T., and Stott, G.H.: Prepartum Serum Hormone Concentrations Related to Dystocia in Holstein Heifers. J. Dairy Sci. 60:249, ( 1977). 16. Osburn, 8.1., Stabenfeldt, G.H ., Adams, A.A., Trees, C., and Sawyer, M: Perinatal Immunity in Calves. J.A. V. M.A. 164:295, ( 1974). 17. Osinga, A.: Endocrine Aspects of Bovine Dystocia with Special Reference to Estrogens. Therio. I 0: 149, ( 1978). 18. Patterson, D.J ., Bellows, R.A., Burfening, P.J ., Short, R.E. , and Carr, J.B.: Incidence and Causes of Neonatal and Postnatal Mortality in Range Cattle. {Abst. J . An. Sci., 1979, Vol. 49 (Suppl. I) p 325) Proc. 71 st Annual Mt. Am. Soc. An. Sci. 19. Pfeiffer, H.E., and McGuire, T.C.: A Sodium-Sulfite Precipitation Test for Assessment of Colostrial lmmunoglobulin Transfer to Calves, J .A. V. M.A. 170:809, ( 1977). 20. Philipsson, J.: Studies on Calving Difficulty, Stillbirth, and Associated factors in Swedish Cattle Breeds: Ill Genetic Parameters. Acta Ag. Scand . 26:211, (1976). 21. Putnam, M .R., Rice, L.E., Wettemann, R.P., Lusby, K.S .: Planipart (Clenbuterol) for the Postponement of Parturition and Alleviation of Dystocia in Cattle. Proc. Soc. Therio. ( 1982). 22. Price, T.D., and Wiltbank, J .N. : Predicting Dystocia in Heifers. Therio. 9:22 1, (1978). 23. Stauber, E.H .: Weak Calf Syndrome: A Continuing Enigma. J .A.V.M.A. 165:223, (1976). 24. Stormont, C. : Neonata l lsoerythrolysis in Domestic Animals: A Comparative Review. Cornell Vet. 25. Stott, G.H.: lmmunoglobulin Absorption in Calf Neonates with Special Consideration of Stress. J . Dairy Sci. (In Press). 26. Tryphonas, L., Hamilton, G.F., and Rhodes, C.S. : Perinatal Femoral Nerve Degeneration and Neurogenic Atrophy of Quadriceps Femoris Muscle in Calves. J .A.V.M.A. 164:801, (1974). 27. Van .Dieten, S.W.J .: Calf Mortality at Birth. Veetelt. en Zwivel Berichten. 7:20-29. In An. Brd. Abstr. Neonatal Nutrition mineral intake will be adequate. High quality milk replacers will have the following basic properties on a dry matter basis: The futility of treating hypogammaglobulinemic calves is well documented. The frustration that owners and veterina- rians have when faced with an outbreak of a severe perinatal or neonatal disease has been discussed in many poolrooms and professional meetings. These frustrations can be reduced through preventive medicine. The veterinarian should give timely instructions on preventive medicine - 20-22 percent protein from milk source, with no plant proteins. - 20-22 percent protein from milk source, with no plant proteins. - I 5-20 percent fat. - I 5-20 percent fat. - less than 0.5% (preferably 0.3%) fiber. Other additives such as vitamins and antibiotics will be present, but the real value (and cost) is determined by the THE BOVINE PROCEEDINGS-No. 16 142 Yield and lmmunoglobulin Concentration in Beef Cows, Br. Vet. J . 133:120, (1977). 12. Logan, E.F., McMurray, D.H ., O'Neill, D.G ., McParland, P.J. , and Mc Rory F.J.: Absorption of Colostral lmmunoglo- bulins by the Neonatal Calf. Br. Vet. J. 134:258, (1978). 13. Lusby, K.S., and Parra, A.A.: Effect of Early Weaning on Calf Performance and on Reproduction in Mature Cows. An. Sci. Res. Rep. M P-108, Ok. St. U. p 64, (1981). 14. Myers, LL., Newman, F.S., Wilson, R.A., and Catlin, J.E.: Passive Immunization of Calves Against Experimentally Induced Enteric Colibacillosis by Vaccination of Dams. Am. J. Vet. Res. 34:29, (1973). 15. O'Brien, T., and Stott, G.H.: Prepartum Serum Hormone Concentrations Related to Dystocia in Holstein Heifers. J. Dairy Sci. 60:249, ( 1977). 16. Osburn, 8.1., Stabenfeldt, G.H ., Adams, A.A., Trees, C., and Sawyer, M: Perinatal Immunity in Calves. J.A. V. M.A. 164:295, ( 1974). 17. Osinga, A.: Endocrine Aspects of Bovine Dystocia with Special Reference to Estrogens. Therio. I 0: 149, ( 1978). 18. Patterson, D.J ., Bellows, R.A., Burfening, P.J ., Short, R.E. , and Carr, J.B.: Incidence and Causes of Neonatal and Postnatal Mortality in Range Cattle. {Abst. J . An. Sci., 1979, Vol. 49 (Suppl. I) p 325) Proc. 71 st Annual Mt. Am. Soc. An. Sci. 19. Pfeiffer, H.E., and McGuire, T.C.: A Sodium-Sulfite Precipitation Test for Assessment of Colostrial lmmunoglobulin Transfer to Calves, J .A. V. M.A. 170:809, ( 1977). 20. Philipsson, J.: Studies on Calving Difficulty, Stillbirth, and Associated factors in Swedish Cattle Breeds: Ill Genetic Parameters. Acta Ag. Scand . Bibliography I. Bull, R.C., Loucks, R.R. , Edmiston, F.L., Hawkins; J .N., and Stauber, E.H.: Nutrition and Weak Calf Syndrome in Beef Cattle. Current Info. Series No. 246. U. of Idaho, Coll. of Ag., Sept. (1974). 2. Conner, G.H., Richardson, M., Carter, G.R., and Wamukoya, J.P.O.: Immune Responses of the Bovine Fetus. J . Dairy Sci., 60:289, (1977). 3. Corah, LR., Dunn, T.G., and Kaltenbach, C.C.: Influence of Prepartum Nutrition on Reproductive Performance of Beef Females and Their Progeny. J . An. Sci. 41:819, (1975). 4. Cutlip, R.C., and McClurkin, A.W.: Lesions and Patho- genesis of Diseases in Young Calves Experimentally Induced by Bovine Adenovirus Type 5 Isolated from a Calf with Weak Calf Syndrome. Am. J. Vet. Res. 36:1095, (1975). 5. Doornbos, D.E., Bellows, R.A. , and Burfening, P.J.: Effects of Obstetrical Assistance on Postpartum Reproduction in Beef Females (In Press). 6. Frey, M .: Personal Communication, (1983). 7. Hamilton, G.F., Turner, A.S., Ferguson, J .G., and Pharr, J. W.: Slipped Capital Femoral Epiphysis in Calves. J.A .V.M .A., 172:1318, (1978). 8. Haughey , K .G.: Meningeal Haemorrhage and Congestion Associated with the Perinatal Mortality of Beef Calves. Aust. Vet. J., 51 :22, (1975). 9. Hoerlein, A .B., and Jones, D.L.: Bovine lmmunoglobulins Following Induced Parturition. J .A.V.M.A. 170:325, (1977). 10. Johnston, N.E., and Oxender, W.D. : Effect of Altered Serum Glucorticoid Concentrations on the Ability of the Newborn Calf to Absorb Colostral Immunoglobulin. Am. J. Vet. Res. 40:32, ( 1979). 11. Logan, E. F.: The Influence of Husbandry on Colostrum I. Bull, R.C., Loucks, R.R. , Edmiston, F.L., Hawkins; J .N., and Stauber, E.H.: Nutrition and Weak Calf Syndrome in Beef Cattle. Current Info. Series No. 246. U. of Idaho, Coll. of Ag., Sept. (1974). 2. Conner, G.H., I. Bull, R.C., Loucks, R.R. , Edmiston, F.L., Hawkins; J .N., and Stauber, E.H.: Nutrition and Weak Calf Syndrome in Beef Cattle. Current Info. Series No. 246. U. of Idaho, Coll. of Ag., Sept. (1974). 2. Conner, G.H., Richardson, M., Carter, G.R., and Wamukoya, J.P.O.: Immune Responses of the Bovine Fetus. J . Dairy Sci., 60:289, (1977). 3. Corah, LR., Dunn, T.G., and Kaltenbach, C.C.: Influence of Prepartum Nutrition on Reproductive Performance of Beef Females and Their Progeny. J . An. Sci. 41:819, (1975). 4. Bibliography Cutlip, R.C., and McClurkin, A.W.: Lesions and Patho- genesis of Diseases in Young Calves Experimentally Induced by Bovine Adenovirus Type 5 Isolated from a Calf with Weak Calf Syndrome. Am. J. Vet. Res. 36:1095, (1975). 5. Doornbos, D.E., Bellows, R.A. , and Burfening, P.J.: Effects of Obstetrical Assistance on Postpartum Reproduction in Beef Females (In Press). 6. Frey, M .: Personal Communication, (1983). 7. Hamilton, G.F., Turner, A.S., Ferguson, J .G., and Pharr, J. W.: Slipped Capital Femoral Epiphysis in Calves. J.A .V.M .A., 172:1318, (1978). 8. Haughey , K .G.: Meningeal Haemorrhage and Congestion Associated with the Perinatal Mortality of Beef Calves. Aust. Vet. J., 51 :22, (1975). 9. Hoerlein, A .B., and Jones, D.L.: Bovine lmmunoglobulins Following Induced Parturition. J .A.V.M.A. 170:325, (1977). 10. Johnston, N.E., and Oxender, W.D. : Effect of Altered Serum Glucorticoid Concentrations on the Ability of the Newborn Calf to Absorb Colostral Immunoglobulin Am J Vet Res Richardson, M., Carter, G.R., and Wamukoya, J.P.O.: Immune Responses of the Bovine Fetus. J . Dairy Sci., 60:289, (1977). 3. Corah, LR., Dunn, APRIL, 1984 143
https://openalex.org/W3049339665	https://bmccancer.biomedcentral.com/track/pdf/10.1186/s12885-020-07278-2	English	null	Prognostic value of S1PR1 and its correlation with immune infiltrates in breast and lung cancers	BMC cancer	2,020	cc-by	9,414	Open Access Prognostic value of S1PR1 and its correlation with immune infiltrates in breast and lung cancers Limei Zhong1, Linling Xie2, Zhiyong Yang1, Lijuan Li1, Shaohua Song1, Donglin Cao1* and Yufeng Liu2,3* Abstract 16 Airport Road, Baiyun District, Guangzhou 510407, China Full list of author information is available at the end of the article Abstract Background: Sphingosine-1-phosphate receptor (S1PR1) is involved in vascular development, a key process in tumorigenesis. This study aimed to evaluate its roles in tumor development and prognosis. Methods: S1PR1 expression levels were analyzed using TIMER and Oncomine database, and the prognostic significance of S1PR1 was assessed using PrognoScan and Kaplan-Meier plotter databases. The relationship between S1PR1 and tumor-infiltrated immune cells was analyzed using TIMER. Results: S1PR1 expression was remarkably lower in breast and lung cancer tissues than in the corresponding normal tissues. Lower expression was related to poor overall survival and disease-free survival in breast invasive carcinoma (BRCA), lung adenocarcinoma (LUAD), and lung squamous cell carcinoma (LUSC). A functional network analysis confirmed the function of S1PR1 in regulating vasculogenesis. In addition, S1PR1 levels were significantly negative with regard to the tumor purity of BRCA (r = −0.508, P = 1.76e-66), LUAD (r = −0.353, P = 6.05e-16), and LUSC (r = −0.402, P = −5.20e-20). Furthermore, S1PR1 levels were significantly positive with regard to infiltrating CD8+ (r = 0.38, P = 5.91e-35) and CD4+ T cells (r = 0.335, P = 1.03e-26), macrophages (r = 0.219, P = 3.67e-12), neutrophils (r = 0.168, P = 2.03e-7), and dendritic cells (DCs) (r = 0.208, P = 9.14e-11) in BRCA; S1PR1 levels were significantly positive with regard to CD8+ T cells (r = 0.308, P = 3.61e-12), macrophages (r = 0.376, P = 1.01e-17), neutrophils (r = 0.246, P = 4.15e-8), and DCs (r = 0.207, P = 4.16e-6) in LUAD; and positive with regard to B cells (r = 0.356, P = 1.57e-15), CD8+ (r = 0.459, P = 3.83e-26) and CD4+ T cells (r = 0.338, P = 3.98e-14), macrophages (r = 0.566, P = 2.61e-45), neutrophils (r = 0.453, P = 1.79e-25), and DCs (r = 0.56, P = 2.12e-40) in LUSC. Conclusions: S1PR1 levels are positively correlated with multiple immune markers in breast and lung cancer. These observed correlations between S1PR1 and the prognosis and immune cell infiltration provide a foundation for further research on its immunomodulatory role in cancer. Keywords: S1PR1, Breast cancer, Lung cancer, Tumor-infiltrating, Prognosis biomarker p y 1Department of Laboratory Medicine, Guangdong Second Provincial General Hospital, No. 466 Xingang Middle Road, Haizhu District, Guangzhou 510317, Guangdong Province, China 2The First Affiliated Hospital, Guangzhou University of Chinese Medicine, No. Zhong et al. BMC Cancer (2020) 20:766 https://doi.org/10.1186/s12885-020-07278-2 Zhong et al. BMC Cancer (2020) 20:766 https://doi.org/10.1186/s12885-020-07278-2 Oncomine database analysis Oncomine database analysis The Oncomine database (https://www.oncomine.org/re- source/login.html) was used to evaluate the expression level of S1PR1 in various types of cancers [21]. The thresholds were a P-value of 0.0001, fold change of 2.0 and data type was mRNA. PrognoScan database analysis The PrognoScan database (www.prognoscan.org/) was used to test S1PR1 expression and survival in various types of cancers [22]. The threshold was an adjusted Cox P-value of < 0.05. © 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. Zhong et al. BMC Cancer (2020) 20:766 Zhong et al. BMC Cancer (2020) 20:766 Page 2 of 17 Page 2 of 17 Page 2 of 17 Methods g Sphingosine-1-phosphate (S1P), produced by sphingo- sine kinase (Sphk), is a biologically active signaling lipid [1]. S1P regulates vascular development and function, including vascular maturation [2, 3]. S1P re- ceptor (S1PR1) is a biologically active sphingolipid metabolite that mediates S1P activity and promotes cell proliferation and survival [4, 5]. S1PR1 is widely expressed in vascular endothelial cells and is required for embryonic vascular development and maturation [6]. Estrogen (the growth-stimulating hormone in breast cancer cells) was shown to stimulate endothe- lial cell growth via S1PR1 [7, 8]. In the tumor micro- environment, S1P exhibits multiple functions: (a) it increases the survival of macrophages; (b) it serves as the “come-and-get-me” signal of dead cells, attracting and enhancing macrophage migration by combining with S1PR1; (c) it stimulates the polarization of TAM/M2 macrophages by activating S1PR1/2/4 [9– 11]. Accumulating evidence demonstrated that tumor progression requires new blood vessel growth, which is achieved by producing angiogenic factors that can activate vascular endothelial cells [12]. Tumor cells release angiogenic stimuli, such as vascular endothelial growth factor (VEGF)-a, which leads to angiogenesis and tumor growth [13]. Studies have shown that S1PR1 inhibits VEGF signaling by pro- moting the interaction between VE-cadherin and VEGFR2, thereby inhibiting VEGF-induced vascular sprouting [14, 15]. C-BioPortal database analysis c-BioPortal (http://cbioportal.org) contains multidimen- sional cancer genomics data sets [23]. S1PR1 mutations and copy number variation (CNV) in breast and lung cancers were analyzed using c-BioPortal. The OncoPrint tab was used to obtain an overview of the genetic alter- ations for each sample. Gene correlation analysis using GEPIA Gene correlation analysis using GEPIA GEPIA (http://gepia.cancer-pku.cn/index.html) was used to confirm the genes with significantly correlated expression levels in TIMER [26]. The Spearman method was used to determine the correlation coeffi- cients. The tumor tissue datasets were used for analysis. Herein, we used Oncomine, Kaplan-Meier plotter, PrognoScan, UALCAN and GEPIA datasets to analyze S1PR1 expression and its relationship with the prognosis of cancer patients. Furthermore, we studied the correlation between S1PR1 and tumor- infiltrated immune cells in the tumor microenviron- ment using TIMER. Our results shed light on the important role of S1PR1 in breast and lung cancer, and determined that it is closely related to tumor immunity. Kaplan-Meier plotter Kaplan-Meier plotter Kaplan-Meier Plotter (https://kmplot.com/) was ap- plied to assess the prognostic value of S1PR1. Grouped according to the median expression of S1PR1 (high vs low expression), all patients were ana- lyzed for overall survival (OS) and progression-free survival (PFS), and Kaplan-Meier was used to draw a survival chart [24]. However, the role of S1PR1 in tumorigenesis and its prognostic value are unclear. A preclinical study on human breast cancer cells found that S1PR1 anti- body can enhance the cytotoxic and anti-proliferative effect of carboplatin on MDA-MB-231 and SK-BR-3 (HER2 subtype) cells, respectively [16]. Lei et al. found that S1PR1 signaling has tumor-suppressive effects and survival benefits in breast cancer [17]. Therefore, it is necessary to clarify the role of S1PR1 in tumor development and progression. Transcrip- tome analysis can be used to predict important is- sues, such as the intrinsic subtype of the primary tumor, tumor grade, drug reactivity, and recurrence risk [18–20]. Immune infiltrates analysis using the TIMER TIMER 2.0 (https://cistrome.shinyapps.io/timer/) was used to analyze immune infiltrates across different types of cancer [25]. Especially, the expression of S1PR1 in dif- ferent cancer types, and the correlation between the ex- pression of S1PR1 and the abundance of immune invasion was determined. In addition, the correlation be- tween S1PR1 expression and tumor infiltrating immune cell gene markers was also explored through related modules. UALCAN database analysis UALCAN (http://ualcan. The Oncomine database was used to analyze S1PR1 mRNA levels in tumor tissues and normal tissues of various cancer types. S1PR1 expression was lower in most tumor tissues, including sarcoma, bladder, brain, central nervous system, breast, colorectal, leukemia, lung, myeloma, and ovarian cancer tissues, than in normal tissues (Fig. 1a). The mRNA-seq data from TCGA were analyzed using TIMER to verify these findings. Data from TCGA shown that the differential expression of S1PR1 between the tumor and adjacent normal tissues is shown in Fig. 1b. Compared with adjacent normal tissues, S1PR1 expression was signifi- cantly reduced in bladder urothelial carcinoma (BLCA), BRCA, cholangiocarcinoma (CHOL), colon adenocarcinoma (COAD), esophageal carcinoma (ESCA), head and neck squamous cell carcinoma (HNSC), kidney chromophobe (KICH), kidney renal papillary cell carcinoma (KIRP), liver hepatocellular carcinoma (LIHC), LUAD, LUSC, prostate adenocar- cinoma (PRAD), rectum adenocarcinoma (READ), skin cutaneous melanoma (SKCM), stomach UALCAN (http://ualcan.path.uab.edu) included the Cancer Genome Atlas (TCGA) level RNA sequences. Clinical data from 31 cancer types were used to analyze the relative expression of genes in tumor and normal samples according to tumor stage, tumor grade or other clinicopathological characteris- tics [28]. S1PR1 mRNA expression level analysis Gene expression data of breast invasive carcinoma (BRCA), lung adenocarcinoma (LUAD), and lung squa- mous cell carcinoma (LUSC) in TCGA were downloaded in UCSC Xena (https://xenabrowser.net). S1PR1 mRNA expression level was compared between cancerous and normal tissue using Mann-Whitney test with P < 0.05 setting as a cut-off. LinkedOmics database analysis LinkedOmics database analysis The LinkedOmics database (http://www.linkedomics. org/login.php) was used to analyze S1PR1 co- expression based on Pearson’s correlation Zhong et al. BMC Cancer (2020) 20:766 Zhong et al. BMC Cancer (2020) 20:766 Page 3 of 17 Page 3 of 17 Page 3 of 17 coefficients. The results were visually evaluated using volcano plots and heat maps. The function module of LinkedOmics was used to analyze gene ontology (GO) biological processes (BP) and Kyoto Encyclopedia of Genes and Genomes (KEGG) path- ways by a gene set enrichment analysis (GSEA). The rank criterion was FDR < 0.05 and 500 simulations were performed [27]. Meier plots and PrognoScan database are displayed with HR and P or Cox P-values from a log-rank test. Spearman correlation analysis was used to evaluate the correlation of gene expression in TIMER and Lin- kedOmics databases. P < 0.05 was considered statisti- cally significant. Prognostic evaluation of S1PR1 in cancers og ost c e a uat o o S ca ce s We investigated whether S1PR1 expression is related to prognosis. The effect of S1PR1 expression on sur- vival was evaluated by PrognoScan. Two probes (204642_at and 239401_at) matching S1PR1 were de- tected. Notably, in three breast cancer cohorts (GSE1456-GPL96, GSE7378, and GSE12276), low S1PR1 expression was significantly associated with a poorer prognosis breast cancer (Fig. 2a–f). We used the Kaplan-Meier plotter database to further examine the prognostic value of S1PR1 in breast cancer. Poor prognosis based on recurrence-free survival (RFS) in breast cancer was significantly correlated with low S1PR1 expression (HR = 0.67, P = 7.1e-13), but a sig- nificant correlation was not observed for overall sur- vival (OS) (HR = 0.86, P = 0.17) and post-progression survival PPS (HR = 1.03, P = 0.82) (Fig. 2g–i). Its de- termined that the low expression of S1PR1 is an inde- pendent risk factor for poor prognosis of breast cancer. In addition, low S1PR1 expression was also related to poor prognosis in two cohorts of patients with lung cancer (GSE31210 and GSE8894), as deter- mined using two probes (204642_at and 239401_at) (Fig. 2j–l). Kaplan-Meier plotter database also showed that low expression of S1PR1 was an independent risk factor for poor prognosis of lung cancer (overall sur- vival, HR = 0.7, P = 6.9e-08; recurrence-free survival, HR = 0.71, P = 0.00035), but not related to post- progression survival in lung cancer (HR = 0.82, P = 0.14) (Fig. 2m–o). Statistical analysis Gene expression data in the Oncomine database was analyzed using p-value, fold change, and mRNA data type. The survival curves were generated via Kaplan- Fig. 1 S1PR1 expression levels in different types of human cancers. a Differences in S1PR1 between cancer tissues and normal tissues based on data in the Oncomine database. (P = 1E-04, Fold change = 2, Data type = mRNA) (b) Human S1PR1 expression levels in different tumor types from TCGA database were determined using TIMER 2.0. P < 0.05, P < 0.01, *P < 0.001 Fig. 1 S1PR1 expression levels in different types of human cancers. a Differences in S1PR1 between cancer tissues and normal tissues based on data in the Oncomine database. (P = 1E-04, Fold change = 2, Data type = mRNA) (b) Human S1PR1 expression levels in different tumor types from TCGA database were determined using TIMER 2.0. P < 0.05, P < 0.01, P < 0.001 Fig. 1 S1PR1 expression levels in different types of human cancers. a Differences in S1PR1 between cancer tissues and normal tissues based on data in the Oncomine database. (P = 1E-04, Fold change = 2, Data type = mRNA) (b) Human S1PR1 expression levels in different tumor types from TCGA database were determined using TIMER 2.0. P < 0.05, P < 0.01, P < 0.001 Zhong et al. BMC Cancer (2020) 20:766 Zhong et al. BMC Cancer (2020) 20:766 Page 4 of 17 Page 4 of 17 Correlations between clinical characteristics and S1PR1 expression in breast cancer and lung cancer Correlations between clinical characteristics and S1PR1 expression in breast cancer and lung cancer adenocarcinoma (STAD), and uterine corpus endo- metrial carcinoma (UCEC). However, S1PR1 expres- sion was significantly higher in kidney renal clear cell carcinoma (KIRC) and thyroid carcinoma (THCA) than in adjacent normal tissues (Fig. 1b). These data showed that alterations in S1PR1 expression depend on the tumor type, suggesting that this gene exerts diverse functions in various tumors. adenocarcinoma (STAD), and uterine corpus endo- metrial carcinoma (UCEC). However, S1PR1 expres- sion was significantly higher in kidney renal clear cell carcinoma (KIRC) and thyroid carcinoma (THCA) than in adjacent normal tissues (Fig. 1b). These data showed that alterations in S1PR1 expression depend on the tumor type, suggesting that this gene exerts diverse functions in various tumors. p g We used the Kaplan-Meier plotter to study the rela- tionship between S1PR1 expression and clinical char- acteristics in patients with breast cancer and lung cancer. Low expression of S1PR1 was associated with worse overall survival (OS) in male and female pa- tients with lung adenocarcinoma (P < 0.05) (Table 1). In particular, low S1PR1 mRNA expression was corre- lated with worse OS in stage 1 (P = 9.20E-13) and early-stage (AJCC stage M) (P = 0.013) lung cancer (Table 1). Low S1PR1 mRNA expression was related to poor OS in patients with (P = 0.023) or without (P = 0.00075) a smoking history (Table 1). In addition, low S1PR1 mRNA expression was related to worse OS in patients who did not receive chemotherapy or radiotherapy. These findings strongly suggest that low S1PR1 mRNA expression is correlated with poor OS in lung cancer (Table 1). In BRCA, low S1PR1 mRNA expression was related to poor OS in ER-positive or HER2-negative patients and in the luminal androgen receptor subtype (Table 2). Taken together, high ex- pression of S1PR1 could be considered a good prog- nostic indictor for breast and lung cancers depending on the clinical characteristics. Decreased expression of S1PR1 in breast cancer and lung cancer patients Survival curves based on OS, DSS, and DFS in three breast cancer cohorts [GSE1456-GPL96 (n = 159), GSE7378 (n = 54) and GSE12276 (n = 204)]. g–i Survival curves for breast cancers based on mRNA-seq data from TCGA of Kaplan–Meier plotter databases. j–l Kaplan– Meier survival curves comparing high and low expression of S1PR1 in lung cancers using PrognoScan. Survival curves based on RFS in two three lung cancer cohorts [GSE31210 (n = 204) and GSE8894 (n = 138)]. m–o Survival curves for lung cancers based on mRNA-seq data from TCGA of Kaplan–Meier plotter databases. OS = Overall survival; RFS = Relapse-Free Survival; PPS = Post-progression survival; DSS = Disease-specific survival; DFS = Disease-free survival Relationship between immune and S1PR1 expression in breast cancer and lung cancer cancer. Figure 4a–c shows genes with significantly positive (dark red dots) and negative (dark green dots) correlations with S1PR1 (false discovery rate, FDR < 0.01). The top 50 positively and negatively re- lated genes are shown in a heat map in Fig. 4d–f. A Gene Ontology (GO)-based gene set enrichment ana- lysis (GSEA) showed that genes that are co-expressed with S1PR1 are enriched for vasculogenesis and the purinergic receptor signaling pathway, while genes re- lated to mitochondria and RNA transcript processing were inhibited in breast cancer (Fig. 4g). Similarly, GO annotation results showed that genes co- expressed with S1PR1 are primarily associated with vasculogenesis, the purinergic receptor signaling path- way, and the phospholipase C-activating G protein coupled receptor signaling pathway, while tRNA metabolic process, RNA modification, and RNA tran- script processing were inhibited in lung cancer (Fig. 4h–i). A KEGG pathway analysis showed enrich- ment for hematopoietic cell lineage, Staphylococcus aureus infection, and renin secretion pathways in both breast cancer and lung cancer. Spliceosome, DNA replication, and proteasome pathways were inhibited in both tumor types (Fig. 4j-l). These results suggest that S1PR1 contributes to various processes in tumor development at least partially through regu- late vasculogenesis. g Tumor infiltrating lymphocytes (TIL) are lympho- cytes that leave the blood circulation and migrate to the vicinity of the tumor. The amount of TIL in the tumor is an important indicator to predict the prog- nosis of cancer patients and the response to im- munotherapy [29, 30]. Tumor purity is a key factor in analyses of immune infiltration by genomic ap- proaches [31]. Decreased expression of S1PR1 in breast cancer and lung cancer patients p We further analyzed the expression of S1PR1 in breast and lung cancers. Gene expression data of breast invasive carcinoma (BRCA), lung adenocarcin- oma (LUAD) and lung squamous cell carcinoma (LUSC) in TCGA were downloaded and S1PR1 mRNA expression level was compared between tumor and normal tissue. As shown in Fig. 3a, the expres- sion of S1PR1 was significantly decreased in tumor tissues of BRCA, LUAD and LUSC (Fig. 3a). In com- parison with normal control tissues, breast cancer and lung cancer tissues presented lower expression of S1PR1, which was also observed by GEPIA analysis (Fig. 3b). Furthermore, we analyzed TCGA data using the UALCAN database. Compared to normal tissues, S1PR1 mRNA expression was significantly decreased in primary tumors and tumor stages (stage 1, stage 2, stage 3, and stage 4) of BRCA, LUAD, and LUSC (Fig. 3c–e). Taken together, these data confirmed the down-regulation of S1PR1 expression in breast cancer and lung cancer patients. Furthermore, we found that low S1PR1 expression was associated with a poor prognosis in patients with soft tissue, blood, and brain cancers (Fig. S1a– c). In contrast, low S1PR1 expression was an inde- pendent risk factor for a good prognosis in gastric cancer (Fig. S1d–g). These results confirmed the prognostic value of S1PR1 in specific types of can- cer; both high and low S1PR1 expression was associ- ated with prognosis depending on the type of cancer. Based on the consistent results for the asso- ciations between S1PR1 expression and survival in breast and lung cancers, we focused on the precise effects of S1PR1 in these two cancer types, as well as the underlying mechanisms. Regulators of S1PR1 in breast cancer and lung cancer We used the LinkedOmics function module to detect the S1PR1 regulatory network to further understand the biological role of S1PR1 in breast cancer and lung Zhong et al. BMC Cancer (2020) 20:766 Page 5 of 17 Fig. 2 (See legend on next page.) Fig. 2 (See legend on next page.) Fig. 2 (See legend on next page.) Zhong et al. BMC Cancer (2020) 20:766 Page 6 of 17 (See figure on previous page.) Fig. 2 Prognostic value of S1PR1 in cancers. a–f Kaplan-Meier survival curves comparing high and low expression of S1PR1 in breast cancers using PrognoScan. Decreased expression of S1PR1 in breast cancer and lung cancer patients Therefore, we use TIMER to investi- gate whether the expression of S1PR1 in breast cancer and lung cancer is related to immune infiltra- tion. We found a significant negative correlation be- tween the S1PR1 expression level and tumor purity in both breast cancer and lung cancer (Fig. 6a–f, Left). S1PR1 was a determinant of immune infiltra- tion in BRCA (tumor purity; r = −0.508, P = 1.76e- 66), including subtypes of BRCA (BRCA-Basal: r = − 0.5411, P = 1.28e-06; BRCA-Her2: r = −0.505, P = 4.44e-06 and BRCA-Luminal: r = −0.557, P = 9.15e- 46). S1PR1 was related to immune infiltration in lung cancer, including LUAD (tumor purity; r = − 0.353, P = 6.05e-16) and LUSC (tumor purity; r = − 0.402, P = 5.20e-20). ) Furthermore, the relationship between S1PR1 and specific immune infiltrates in breast cancer and lung cancer were analyzed. The S1PR1 expression level was significantly positively correlated with levels of infiltrating CD8+ T cells (r = 0.38, P = 5.97e-35), CD4+ T cells (r = 0.335, P = 1.03e-26), macrophages (r = 0.219, P = 3.67e-12), neutrophils (r = 0.168 P = 2.03e-07), and DCs (r = 0.208, P = 9.14e-11) in BRCA (Fig. 6a). In BRCA-Basal, there were slight positive correlations between S1PR1 expression levels and levels of infiltrating CD8+ T cells (r = 0.279, P = 1.76e-03) and CD4+ T cells (r = 0.237, P = 8.52e-03). Similarly, there were positive correlations with infil- trating levels of CD8+ T cells (r = 0.546, P = 1.13e- 05), CD4+ T cells (r = 0.529, P = 2.00e-05), neutro- phils (r = 0.342, P = 8.57e-03), and DCs (r = 0.488, P = 1.35e-04) in BRCA-Her2. S1PR1 expression levels were positively correlated with levels of infiltrating CD8+ T cells (r = 0.147, P = 3.43e-21), CD4+ T cells (r = 0.316, P = 6.26e-14), macrophages (r = 0.151, P = Genomic alterations in S1PR1 in breast cancer and lung cancer cBioPortal database was used to determine the types and frequencies of S1PR1 alterations in BRCA, LUAD, and LUSC. S1PR1 was altered in 4% of patients with BRCA. These alterations included mRNA missense mutations, amplifications, and deletions (Fig. 5a). S1PR1 was altered in 6% of patients with LUAD and 2.3% of patients with LUSC, including mRNA mis- sense mutations, truncating mutations, amplifications, and deletions (Fig. 5a). Moreover, S1PR1 CNV was associated with OS in LUAD but not with OS or DFS in BRCA and LUSC (Fig. 5b–d). These results suggest that mutations in S1PR1 are associated with prognosis in LUAD. Zhong et al. Genomic alterations in S1PR1 in breast cancer and lung cancer BMC Cancer (2020) 20:766 Page 7 of 17 infiltrating CD8+ T cells (r = 0.308, P = 3.61e-12), Table 1 Correlation between S1PR1 mRNA expression and prognosis in lung cancer with respect to clinicopathological factors Clinicopathological characteristics Overall survival N Hazard ratio P-value Sex Female 715 0.72 (0.57–0.91) 0.0064 Male 1100 0.72 (0.61–0.84) 4.90E-05 Histology Adenocarcinoma 720 0.57 (0.45–0.73) 5.90E-06 Squamous cell carcinoma 524 0.85 (0.67–1.07) 0.1677 Stage 1 577 0.35 (0.26–0.47) 9.20E-13 2 244 0.74 (0.51–1.07) 1.13E-01 3 70 1.03 (0.6–1.77) 9.20E-01 4 4 NA NA Grade I 201 1.19 (0.83–1.71) 0.34 II 310 0.83 (0.6–1.13) 0.23 III 77 0.61 (0.32–1.19) 0.15 AJCC stage T 1 237 1.01 (0.76–1.34) 0.9527 2 389 0.77 (0.62–0.96) 0.019 3 81 1.47 (0.89–2.43) 0.13 4 46 0.98 (0.52–1.85 0.95 AJCC stage N 0 781 0.85 (0.68–1.04) 0.12 1 56 1.78 (0.89–3.57) 0.098 2 111 1.27 (0.84–1.9) 0.2515 AJCC stage M 0 681 0.77 (0.62–0.95) 0.013 1 10 NA NA Smoking history Exclude those never smoked 820 0.79 (0.64–0.94) 0.023 Only those never smoked 105 0.37 (0.21–0.68) 0.00075 Chemotherapy No 310 0.71 (0.51–1) 0.046 Yes 176 1.11 (0.74–1.67) 0.62 Radiotherapy No 271 0.69 (0.48–0.99) 0.042 Yes 70 1.04 (0.61–1.78) 0.8745 Bold values indicate P < 0.05; NA: none Table 2 Correlations between S1PR1 mRNA expression and clinical prognosis in breast cancer with respect to clinicopathological factors Clinicopathological characteristics Overall N Hazard ratio P-value ER status ER positive 2061 0.79 (0.67–0.94) 0.0057 ER negative 801 0.95 (0.7–1.18) 0.62 PR status PR positive 589 0.91 (0.64–1.29) 0.6024 PR negative 549 1.02 (0.76–1.36) 0.9124 HER2 status HER2 positive 252 1.13 (0.73–1.75) 0.5743 HER2 negative 800 0.75 (0.57–0.96) 0.0247 Intrinsic subtype Basal 241 1.23 (0.75–2.01) 0.41 Luminal A 611 0.75 (0.52–1.06) 0.1 Luminal B 433 0.97 (0.67–1.41) 0.88 HER2+ 147 0.67 (0.35–1.28) 0.2235 Lymph node status Lymph node positive 313 0.94 (0.64–1.38) 0.75 Lymph node negative 594 1.07 (0.73–1.55) 0.74 Grade 1 345 0.68 (0.4–1.15) 0.1461 2 901 0.94 (0.74–1.2) 0.63 3 903 0.93 (0.75–1.16) 0.5257 TP53 status Mutated 188 1.17 (0.73–1.88) 0.52 Wild type 273 0.81 (0.42–1.54) 0.52 Pietenpol subtype Basal-like 1 58 1.69 (0.55–5.17) 0.35 Basal-like 2 38 0.96 (0.28–3.34) 0.95 Immunomodulatory 100 1.67 (0.65–4.32) 0.28 Mesenchymal 73 0.79 (0.36–1.73) 0.56 Mesenchymal stem -like 19 NA NA Luminal androgen receptor 203 0.46 (0.3–0.71) 0.0002 Systemically untreated patients Include 1402 0.86 (0.69–1.07) 0.17 Exclude 3951 0.67 (0.6–0.75) 7.1E-13 Bold values indicate P < 0.05; NA: none Table 2 Correlations between S1PR1 mRNA expression and clinical prognosis in breast cancer with respect to clinicopathological factors infiltrating CD8+ T cells (r = 0.308, P = 3.61e-12), macrophages (r = 0.376, P = 1.01e-17), neutrophils (r = 0.246, P = 4.15e-08), and DCs (r = 0.207, P = 4.16e-06) in LUAD. Genomic alterations in S1PR1 in breast cancer and lung cancer In addition, there were positive correlations with levels of infiltrating B cells (r = 4.14e-04), neutrophils (r = 0.147, P = 6.67e-04), and DCs (r = 0.213, P = 6.44e-07) in BRCA-Luminal tu- mors (Fig. 6a). We also found that S1PR1 expression levels were positively correlated with levels of 4.14e-04), neutrophils (r = 0.147, P = 6.67e-04), and DCs (r = 0.213, P = 6.44e-07) in BRCA-Luminal tu- mors (Fig. 6a). We also found that S1PR1 expression levels were positively correlated with levels of Zhong et al. BMC Cancer (2020) 20:766 Page 8 of 17 Fig. 3 Decreased expression of S1PR1 in breast and lung cancer patients (a) Gene expression data of breast invasive carcinoma (BRCA), lung adenocarcinoma (LUAD), and lung squamous cell carcinoma (LUSC) in TCGA were downloaded in UCSC Xena. S1PR1 mRNA expression level was compared between cancerous and normal tissue using Mann-Whitney test with P < 0.05 setting as cut-off. b The expression of S1PR1 in BRCA, LUAD, and LUSC were analysis using GEPIA. T: tumor, N: normal tissue, NUM = number. c–e S1PR1 mRNA expression level was expressed as box plots using the UALCAN database. mRNA expression of S1PR1 in normal control and BRCA, LUAD, and LUSC tumors: (Left) primary tumors, (Right) individual cancer stage. P < 0.05, P < 0.01, P < 0.001 Fig. 3 Decreased expression of S1PR1 in breast and lung cancer patients (a) Gene expression data of breast invasive carcinoma (BRCA), lung adenocarcinoma (LUAD), and lung squamous cell carcinoma (LUSC) in TCGA were downloaded in UCSC Xena. S1PR1 mRNA expression level was compared between cancerous and normal tissue using Mann-Whitney test with P < 0.05 setting as cut-off. b The expression of S1PR1 in BRCA, LUAD, and LUSC were analysis using GEPIA. T: tumor, N: normal tissue, NUM = number. c–e S1PR1 mRNA expression level was expressed as box plots using the UALCAN database. mRNA expression of S1PR1 in normal control and BRCA, LUAD, and LUSC tumors: (Left) primary tumors, (Right) individual cancer stage. P < 0.05, P < 0.01, P < 0.001 Zhong et al. BMC Cancer (2020) 20:766 Page 9 of 17 Fig. 4 (See legend on next page.) Fig. 4 (See legend on next page.) Fig. 4 (See legend on next page.) Fig. 4 (See legend on next page.) Zhong et al. BMC Cancer (2020) 20:766 Page 10 of 17 (See figure on previous page.) Fig. Discussion We systematically analyzed the expression levels of S1PR1 and the prognostic value in different types of cancers. Compared with levels in normal tissues, S1PR1 expression was significantly lower in BLCA, BRCA, CHOL, COAD, ESCA, HNSC, KICH, KIRP, LIHC, LUAD, LUSC, PRAD, READ, SKCM, STAD, and UCEC and was significantly higher in KIRC and THCA. Accordingly, S1PR1 expression patterns de- pend on the type of cancer. Prognostic data from Kaplan-Meier plotter showed that low levels of S1PR1 are significantly related to poor prognosis in breast cancer and lung cancer. The down-regulation of S1PR1 was associated with worse prognosis in breast cancer and lung cancer and was significantly related to clinical characteristics, such as gender, population, smoking status, and stage. These results suggested that S1PR1 is a prognostic biomarker in breast cancer and lung cancer. However, some literatures have reported the oncogenic role of S1PR1 in breast cancer. Lee H demonstrated that Stat3-induced S1PR1 expression, as well as S1P/ S1PR1 pathway, is important for persistent Stat3 acti- vation in cancer cells and the tumor microenviron- ment and for malignant progression [32]. This may be one of the molecular mechanisms by which S1PR1 mediates such a complex biological response. We considered that the main reason for this inconsistency is that our study analyzed the expression of S1PR1 at the overall level. We further verified the significant reduction of S1PR1 expression in breast cancer and lung cancer patients through TCGA analysis. Another study has also claimed a survival function benefit of S1P/S1PR signaling in BRCA patients, which might explain the obstacle to relative antagonist therapy in clinics [17]. A recent study determined that attenu- ated endothelial S1PR1 function led to increased tumor growth and metastasis, whereas S1PR1 overex- pression led to smaller tumors, and strategies to en- hance S1PR1 function in the tumor vasculature may potentiate the efficacy of cytotoxic and targeted anti- cancer therapies [33]. These studies support our find- ings that high expression of S1PR1 is beneficial for tumor survival. Genomic alterations in S1PR1 in breast cancer and lung cancer 4 S1PR1 co-expression genes in breast and lung cancer. a–c The S1PR1 highly correlated genes identified by Pearson test in BRCA (a), LUAD (b), and LUSC (c). d–f The heat map shows that in BRCA (d), LUAD (e), and LUSC (f), the first 50 genes are positively (red) and negatively (blue) correlated with S1PR1. g–i Significantly enriched GO annotations of S1PR1 in BRCA (g), LUAD (h), and LUSC (i). j–l Significantly enriched KEGG pathways of S1PR1 in BRCA (j), LUAD (k), and LUSC (l) 0.358, P = 1.27e-15), CD8+ T cells (r = 0.459, P = 3.83e-26), CD4+ T cells (r = 0.338, P = 3.98e-14), macrophages (r = 0.586, P = 2.61e-45), neutrophils (r = 0.453, P = 1.79e-25), and DCs (r = 0.56, P = 2.12e- 40) in LUSC. These results strongly suggest that S1PR1 plays a special role in the immune infiltration of breast and lung cancers, and has a particularly strong effect on T cells, macrophages, neutrophils and DCs. These observed correlations between S1PR1 and various types of immune cells in breast and lung cancers indicated that S1PR1 may have high prognostic value. Correlations between S1PR1 expression and immune markers Correlations between S1PR1 expression and immune markers We further evaluated the correlations between S1PR1 and markers of various immune cells in breast cancer and lung cancer using TIMER (Table 3) and GEPIA databases (Table S1). The correlations between S1PR1 expression and immune marker genes for different immune cell populations, includ- ing CD8+ T cells, T cells (general), B cells, mono- cytes, TAMs, M1, and M2 macrophages, neutrophils, NK cells, DCs, and various functional T cells, such as Th1 cells, Th2 cells, Tfh cells, Th17 cells, and Tregs, as well as exhausted T cells were analyzed by TIMER. After adjusting for tumor purity, S1PR1 ex- pression levels were significantly positively correlated with marker sets for various immune cells, except for NK cells, Th17, and T cell exhaustion in BRCA (Table 3 and Fig. 7). However, S1PR1 expression levels were highly positively correlated with most im- mune marker sets and both T cell populations and exhausted T cells in LUAD and LUSC (Table 3 and Fig. 7). We further analyzed the correlation between S1PR1 expression and the markers using the GEPIA database, including data for BRCA, LUAD, and LUSC. The results for correlations between S1PR1 and markers of immune infiltrating cells were similar to those of the TIMER analysis (Table S1). This fur- ther confirms that S1PR1 is significantly related to immune infiltrating cells in lung and breast cancer, suggesting that high levels of S1PR1 could induce immune activity in the lung and breast cancer microenvironment. The tumor microenvironment refers to non-cancer cells in and around tumors; infiltrated of immune Zhong et al. BMC Cancer (2020) 20:766 Page 11 of 17 cells in the tumor microenvironment plays a vital function in the occurrence and development of tu microenvironment is an independent predictor of can- cer patient survival and lymph node metastasis [29 Fig. 5 S1PR1 genomic alterations in breast and lung cancer. a OncoPrint of S1PR1 alterations in BRCA, LUAD, and LUSC. Different types of genetic alterations highlighted in different colors. b–d The relationship between genetic alterations and S1PR1 (OS/DFS) in BRCA (b), LUAD (c), and LUSC (d). Logrank test was used in analysis of OS/DFS Fig. 5 S1PR1 genomic alterations in breast and lung cancer. a OncoP alterations highlighted in different colors b d The relationship betw Fig. 5 S1PR1 genomic alterations in breast and lung cancer. a OncoPrint of S1PR1 alterations in BRCA, LUAD, and LUSC. Correlations between S1PR1 expression and immune markers Different types of genetic alterations highlighted in different colors. b–d The relationship between genetic alterations and S1PR1 (OS/DFS) in BRCA (b), LUAD (c), and LUSC (d). Logrank test was used in analysis of OS/DFS microenvironment is an independent predictor of can- cer patient survival and lymph node metastasis [29, 30]. Studies have shown that S1PR1 can affect the cells in the tumor microenvironment plays a vital function in the occurrence and development of tu- mors [34, 35]. Lymphocyte infiltration in the tumor Zhong et al. BMC Cancer (2020) 20:766 Page 12 of 17 Fig. 6 Correlations between S1PR1 expression and immune infiltration levels in breast and lung cancer. a S1PR1 expression was significantly negatively related to tumor purity and significantly positively correlated with infiltrating levels of CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells in BRCA (n = 1093). b S1PR1 expression was significantly negatively related to tumor purity and was significant positively correlated with infiltrating levels of CD8+ T cells, CD4+ T cells, and dendritic cells in BRCA-Basal (n = 139). c S1PR1 expression was i ifi l i l l d i d i ifi l i i l l d i h i fil i l l f CD8 T ll CD4 T ll Fig. 6 Correlations between S1PR1 expression and immune infiltration levels in breast and lung cancer. a S1PR1 expression was significantly f f f Fig. 6 Correlations between S1PR1 expression and immune infiltration levels in breast and lung cancer. a S1PR1 expression was significantly negatively related to tumor purity and significantly positively correlated with infiltrating levels of CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells in BRCA (n = 1093). b S1PR1 expression was significantly negatively related to tumor purity and was significant positively correlated with infiltrating levels of CD8+ T cells, CD4+ T cells, and dendritic cells in BRCA-Basal (n = 139). c S1PR1 expression was significantly negatively related to tumor purity and was significantly positively correlated with infiltrating levels of CD8+ T cells, CD4+ T cells, neutrophils, and dendritic cells in BRCA-Her2 (n = 67). d S1PR1 expression was significantly negatively related to tumor purity and was significantly positively correlated with infiltrating levels of CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells in BRCA- Luminal (n = 611). e S1PR1 expression was significantly negatively related to tumor purity and was significantly positively correlated with infiltrating levels of CD8+ T cells, macrophages, neutrophils, and dendritic cells in LUAD (n = 457). f S1PR1 expression was significantly negatively related to tumor purity and was significant positively correlated with infiltrating levels of B cells, CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells in LUSC (n = 457). Spearman’s correlation coefficients were used for analyses. P < 0.01; P < 0.001; * P < 0.0001 CD8+ T, CD4+ T, neutrophils, macrophages, and DCs in BRCA. The correlation between S1PR1 expression and immune cell marker genes suggests that S1PR1 regulates lung cancer tumor immunity through mul- tiple immune cell populations. These results indicate that high levels of S1PR1 could increase the cytotox- icity of the immune system and immune activation in proliferation and differentiation of lymphocytes in the tumor microenvironment [36]. The evaluation of im- mune cell infiltration in breast and lung cancers using the TIMER database revealed strong negative correla- tions between S1PR1 and tumor purity in BRCA, LUAD, and LUSC. Furthermore, the S1PR1 expres- sion level was positively correlated with levels of Zhong et al. BMC Cancer (2020) 20:766 Page 14 of 17 Table 3 Correlations between S1PR1 and related genes and markers of immune cells, as evaluated using TIMER (Continued) BRCA LUAD LUSC STAT5A 0.165 1.81E-07 * 0.248 2.27E-08 * 0.413 4.22E-21 * IL13 0.048 1.27E-01 0.071 1.15E-01 0.199 1.20E-05 * Tfh BCL6 0.174 3.52E-08 *** 0.119 8.01E-03 * 0.004 9.24E-01 IL21 0.001 9.77E-01 0.054 2.34E-01 0.207 4.92E-06 * Th17 STAT3 0.043 1.75E-01 0.188 2.65E-05 * 0.158 6.09E-04 IL17A −0.053 9.29E-02 0.033 4.62E-01 −0.038 4.09E-01 Treg FOXP3 0.027 3.94E-01 0.058 1.98E-01 0.393 4.15E-19 * CCR8 0.014 6.71E-01 0.157 4.61E-04 0.464 7.27E-27 * STAT5B 0.283 8.58E-20 * 0.505 4.67E-12 * 0.138 2.47E-03 * TGFb (TGFB1) 0.321 3.21E-25 * 0.198 9.43E-06 * 0.064 1.64E-01 T cell exhaustion PD-1 (PDCD1) 0.112 4.12E-04 0.051 2.56E-01 0.361 3.80E-16 * CTLA4 0.018 5.75E-01 0.081 7.27E-02 * 0.404 3.88E-20 * LAG3 −0.109 6.00E-04 −0.035 4.39E-01 0.212 3.11E-06 * TIM-3 (HAVCR2) 0.039 2.19E-01 0.213 1.78E-06 * 0.589 8.44E-46 * GZMB 0.056 7.82E-02 0.024 5.99E-01 0.267 3.33E-09 * human malignancies, including breast cancer [38]. Angiogenesis mimicry leads to worse prognosis, in- creased tumor metastasis, low 5-year overall sur- vival, and increased mortality [39]. This shows that S1PR1 defects promote the occurrence of VM, and the knockout of S1PR1 in breast cancer cells in- creases the number of VMs. More importantly, tumor cells with low S1PR1 expression receive nu- trition through VM, and accelerate tumor growth in animal models [40]. Recent research has shown that S1PR1 signaling is an important vascular factor af- fecting tumor progression, metastasis, and responses to chemotherapy and immunotherapy [33]. Strat- egies to enhance S1PR1 function in the tumor vas- culature may enhance the cytotoxic killing effect and chemotherapy effect of targeted anti-cancer therapy. BRCA, LUAD and LUSC by increasing the infiltration of CTLs, CD4 + T cells, and DCs. On the contrary, low expression of S1PR1 could lead to reduced infil- trated effector cells in the tumor microenvironment. As shown in recently reports, endothelial loss of S1PR1 led to a reduction in CD45+ cells, macro- phages, and DCs, which influences tumor growth and metastasis [33]. In addition, S1P is involved in enhan- cing endocytosis and migration of mature dendritic cells through S1PR3, an event that may increase the immune response to cancer cells. BMC Cancer (2020) 20:766 Page 13 of 17 Table 3 Correlations between S1PR1 and related genes and markers of immune cells, as evaluated using TIMER BRCA LUAD LUSC Description Gene markers Purity Purity Purity varX cor p cor p cor p CD8+ T cell CD8A 0.267 1.26E-17 * 0.166 2.19E-04 0.411 6.51E-21 * CD8B 0.176 2.42E-08 * 0.108 1.66E-02 0.378 1.22E-17 * T cell (general) CD3D 0.217 4.71E-12 * 0.112 1.28E-02 0.411 7.60E-21 * CD3E 0.276 7.15E-19 * 0.226 8.85E-07 * 0.459 2.82E-26 * CD2 0.202 3.20E-10 * 0.159 4.00E-04 0.438 7.99E-24 * B cell CD19 0.156 7.38E-07 * 0.181 5.37E-05 * 0.324 3.78E-13 * CD79A 0.177 1.98E-08 * 0.172 1.21E-04 0.325 3.29E-13 * Monocyte CD86 0.044 1.28E-01 0.228 2.97E-07 * 0.588 1.27E-45 * CD115 (CSF1R) 0.202 1.29E-10 * 0.264 3.10E-08 * 0.64 2.67E-56 * TAM CCL2 0.111 4.68E-04 0.093 3.86E-02 0.44 5.89E-24 * CD68 0.023 4.63E-01 0.289 5.86E-11 * 0.494 1.18E-30 * IL10 0.055 8.35E-02 0.27 1.10E-09 * 0.534 1.49E-36 * M1 Macrophage INOS (NOS2) 0.257 1.76E-16 * 0.374 7.93E-18 * 0.079 8.64E-02 IRF5 0.016 6.18E-01 −0.042 3.55E-01 −0.036 4.31E-01 COX2 (PTGS2) 0.338 4.90E-28 * 0.095 3.58E-02 0.214 2.37E-06 * M2 Macrophage CD163 0.056 7.72E-02 0.331 4.36E-14 * 0.645 1.52E-57 * VSIG4 0.08 1.14E-02 0.271 9.75E-10 * 0.625 4.77E-53 * MS4A4A 0.23 1.96E-13 * 0.365 5.39E-17 * 0.628 9.28E-54 * Neutrophils CD66b (CEACAM8) 0.04 2.03E-01 0.25 1.95E-08 * 0.212 2.99E-06 * CD11b (ITGAM) 0.007 8.24E-01 0.199 8.16E-06 * 0.491 2.66E-30 * CCR7 0.316 1.55E-24 * 0.321 2.57E-13 * 0.514 1.70E-33 * Natural killer cell KIR2DL1 0.011 7.27E-01 0.216 1.30E-06 * 0.146 1.36E-03 * KIR2DL3 0.051 1.10E-01 0.148 9.96E-04 0.233 2.63E-07 * KIR2DL4 −0.027 3.95E-01 −0.03 5.06E-01 0.152 8.45E-04 ** KIR3DL1 0.095 2.63E-03 * 0.174 1.04E-04 0.295 4.85E-11 * KIR3DL2 0.068 3.19E-02 0.077 8.79E-02 0.217 1.68E-06 *** KIR3DL3 −0.005 8.75E-01 0.025 5.81E-01 0.044 3.43E-01 KIR2DS4 0.035 2.68E-01 0.119 8.34E-03 * 0.221 1.05E-06 * Dendritic cell HLA-DPB1 0.237 3.89E-14 * 0.261 4.13E-09 * 0.621 3.86E-52 HLA-DQB1 0.073 2.11E-02 0.089 4.79E-02 0.4 8.84E-20 * HLA-DRA 0.156 7.17E-07 * 0.219 8.69E-07 * 0.603 1.29E-48 HLA-DPA1 0.21 2.26E-11 * 0.225 4.53E-07 * 0.622 1.87E-52 BDCA-1(CD1C) 0.461 1.76E-53 * 0.271 1.00E-09 * 0.438 8.69E-24 * BDCA-4(NRP1) 0.484 1.58E-59 * 0.174 1.07E-04 0.473 6.69E-28 *** CD11c (ITGAX) 0.087 6.21E-03 * 0.135 2.69E-03 * 0.445 1.58E-24 * Th1 T-bet (TBX21) 0.227 4.72E-13 * 0.182 4.81E-05 * 0.403 5.17E-20 * STAT4 0.277 5.92E-19 * 0.131 3.66E-03 * 0.504 4.73E-32 * STAT1 0.116 2.61E-04 −0.046 3.10E-01 0.177 1.03E-04 IFN-g (IFNG) 0.009 7.84E-01 * −0.076 9.13E-02 0.108 1.85E-02 TNF-a (TNF) 0.193 8.08E+ 10 * −0.076 9.30E-02 0.069 1.34E-01 Th2 GATA3 0.078 1.43E-02 0.047 3.01E-01 0.232 3.00E-07 * STAT6 0.225 6.69E-13 * 0.138 2.20E-03 * 0.022 6.25E-01 Zhong et al. Our findings are consistent with such reports, and these discoveries imply that S1PR1 plays an important role in recruit- ing and governing immune infiltration in BRCA, LUAD and LUSC. To further elucidate the molecular mechanisms underlying the role of S1PR1 in breast and lung cancers, we used GSEA to identify pathways that are enriched in genes co-expressed with S1PR1. We found that S1PR1 was significantly associated with vasculogenesis, the purinergic receptor signaling pathway, and metabolism of nucleic acids in tumor conditions. This conclusion is consistent with previ- ous research reports that showed that S1PR1 regu- lates vasculogenesis [7]. Recent studies have provided potential explanations for the associations between S1PR1 expression, immune infiltration, and poor prognosis. Angiogenesis mimicry (VM) system is a blood vessel-like network in which tumor cells are co-expressed with endothelial cells and tumor markers [37]. VM is closely related to a variety of A limitation of our study was the lack of in vitro and animal experiments to confirm the role of S1PR1 in the growth and progression of breast cancer and lung cancer and its relationship with the infiltration of immune cells in the tumor microenvironment. Therefore, further research is needed to verify the role of S1PR1 in breast cancer and lung cancer using these models. Funding YFL was supported by National Natural Science Foundation of China (No. 81700512), and Natural Science Foundation of Guangdong Province of China (No. 2016A030310252). 16. Xiao S, Yang J. Preclinical study of the antitumor effect of sphingosine-1- phosphate receptor 1 antibody (S1PR1-antibody) against human breast cancer cells. Investig New Drugs. 2019;37(1):57–64. 16. Xiao S, Yang J. Preclinical study of the antitumor effect of sphingosine-1- phosphate receptor 1 antibody (S1PR1-antibody) against human breast cancer cells. Investig New Drugs. 2019;37(1):57–64. 17. Lei FJ, Cheng BH, Liao PY, et al. Survival benefit of sphingosin-1-phosphate and receptors expressions in breast cancer patients. Cancer Med. 2018;7(8): 3743–54. 17. Lei FJ, Cheng BH, Liao PY, et al. Survival benefit of sphingosin-1-phosphate and receptors expressions in breast cancer patients. Cancer Med. 2018;7(8): 3743–54. Abbreviations 6. Proia RL, Hla T. Emerging biology of sphingosine-1-phosphate: its role in pathogenesis and therapy. J Clin Invest. 2015;125(4):1379–87. S1P: Sphingosine-1-phosphate; S1PR1: Sphingosine-1-phosphate receptor; CNV: Copy number variation; GO: Gene Ontology; BP: Biological processes; KEGG: Kyoto Encyclopedia of Genes and Genomes; GSEA: Gene set enrichment analysis; TCGA: The Cancer Genome Atlas; TIL: Tumor infiltrating lymphocytes; BLCA: Bladder urothelial carcinoma; BRCA : Breast invasive carcinoma; CHOL: Cholangiocarcinoma; COAD: Colon adenocarcinoma; ESCA: Esophageal carcinoma; HNSC: Head and neck squamous cell carcinoma; KICH: Kidney chromophobe; KIRP: Kidney renal papillary cell carcinoma; LIHC: Liver hepatocellular carcinoma; LUAD: Lung adenocarcinoma; LUSC: Lung squamous cell carcinoma; PRAD: Prostate adenocarcinoma; READ: Rectum adenocarcinoma; SKCM: Skin cutaneous melanoma; STAD: Stomach adenocarcinoma; UCEC: Uterine corpus endometrial carcinoma; KIRC: Kidney renal clear cell carcinoma; THCA: Thyroid carcinoma; OS: Overall survival; RFS: Relapse-Free Survival; PPS: Post-Progression survival; DSS: Disease-specific survival; DFS: Disease-free survival 7. Sukocheva O, Wadham C, Gamble J, Xia P. Sphingosine-1-phosphate receptor 1 transmits estrogens' effects in endothelial cells. Steroids. 2015; 104:237–45. 8. Sukocheva OA. Expansion of Sphingosine kinase and Sphingosine-1- phosphate receptor function in Normal and Cancer cells: from membrane restructuring to mediation of estrogen signaling and stem cell programming. Int J Mol Sci. 2018;19(2):420. 9. Rodriguez YI, Campos LE, Castro MG, Aladhami A, Oskeritzian CA, Alvarez SE Sphingosine-1 phosphate: a new modulator of immune plasticity in the tumor microenvironment. Front Oncol. 2016;6:218. 10. Nakajima M, Nagahashi M, Rashid OM, Takabe K, Wakai T. The role of sphingosine-1-phosphate in the tumor microenvironment and its clinical implications. Tumour Biol. 2017;39(4):1010428317699133. PPS: Post-Progression survival; DSS: Disease-specific survival; DFS: Disease-free survival 11. Sukocheva OA, Furuya H, Ng ML, Friedemann M, Menschikowski M, Tarasov VV, et al. Sphingosine kinase and sphingosine-1-phosphate receptor signaling pathway in inflammatory gastrointestinal disease and cancers: a novel therapeutic target. Pharmacol Ther. 2020;207:107464. 11. Sukocheva OA, Furuya H, Ng ML, Friedemann M, Menschikowski M, Tarasov VV, et al. Sphingosine kinase and sphingosine-1-phosphate receptor signaling pathway in inflammatory gastrointestinal disease and cancers: a novel therapeutic target. Pharmacol Ther. 2020;207:107464. Acknowledgements We thank Xiaotao Jiang for his help in data analysis. 12. Folkman J. Angiogenesis: an organizing principle for drug discovery? Nat Rev Drug Discov. 2007;6(4):273–86. 12. Folkman J. Angiogenesis: an organizing principle for drug discovery? Nat Rev Drug Discov. 2007;6(4):273–86. Conclusions In conclusion, decreased S1PR1 expression was re- lated to poor prognosis together with reduction of ef- fect immune cell infiltration in breast and lung cancers. In addition, the down-regulation of S1PR1 Zhong et al. BMC Cancer (2020) 20:766 Page 15 of 17 Fig. 7 Correlations between S1PR1 expression and immune markers. Correlations between S1PR1 expression with markers of immune cells CD8+ T cell, T cell (general), B cells, monocytes, TAM, M1 macrophages, M2 macrophages, neutrophils, natural killer cells, dendritic cells, Th1, Th2, Tfh, Th17, Treg, and T cell exhaustion in BRCA, LUAD, and LUSC using TIMER 2.0 ations between S1PR1 expression with markers of immune cells CD8+ phages, neutrophils, natural killer cells, dendritic cells, Th1, Th2, Tfh, Fig. T cel Th17 Fig. 7 Correlations between S1PR1 expression and immune markers. Correlations between S1PR1 expression with markers of immune cells CD8+ T cell, T cell (general), B cells, monocytes, TAM, M1 macrophages, M2 macrophages, neutrophils, natural killer cells, dendritic cells, Th1, Th2, Tfh, Th17, Treg, and T cell exhaustion in BRCA, LUAD, and LUSC using TIMER 2.0 Page 16 of 17 Page 16 of 17 Page 16 of 17 Zhong et al. BMC Cancer (2020) 20:766 Zhong et al. BMC Cancer (2020) 20:766 may have profound effects on vasculogenic mimicry in tumor microenvironment. Accordingly, S1PR1 pre- sumably plays a critical part in immune infiltration and acts as a prognostic marker in breast cancer and lung cancer. may have profound effects on vasculogenic mimicry in tumor microenvironment. Accordingly, S1PR1 pre- sumably plays a critical part in immune infiltration and acts as a prognostic marker in breast cancer and lung cancer. References 1. Sun K, Zhang Y, D'Alessandro A, Nemkov T, Song A, Wu H, et al. Sphingosine- 1-phosphate promotes erythrocyte glycolysis and oxygen release for adaptation to high-altitude hypoxia. Nat Commun. 2016;15(7):12086. 1. Sun K, Zhang Y, D'Alessandro A, Nemkov T, Song A, Wu H, et al. Sphingosine- 1-phosphate promotes erythrocyte glycolysis and oxygen release for adaptation to high-altitude hypoxia. Nat Commun. 2016;15(7):12086. 1. Sun K, Zhang Y, D'Alessandro A, Nemkov T, Song A, Wu H, et al. Sphingosine- 1-phosphate promotes erythrocyte glycolysis and oxygen release for adaptation to high-altitude hypoxia. Nat Commun. 2016;15(7):12086. Additional file 1: Figure S1. Prognostic potential of S1PR1 in different cancers. (a–c) Kaplan-Meier survival curves comparing the high and low expression of S1PR1 in soft tissue cancer (a), blood cancer (b), and brain cancer (c) in the PrognoScan. (d–f) Survival curves of gastric cancer ana- lyzed with mRNA-seq data of TCGA of Kaplan-Meier plotter databases. OS = Overall survival; RFS = Relapse-Free Survival; PPS = Post-progression survival. 2. Asano Y, Stawski L, Hant F, Highland K, Silver R, Szalai G, et al. Endothelial Fli1 deficiency impairs vascular homeostasis: a role in scleroderma vasculopathy. Am J Pathol. 2010;176(4):1983–98. 3. Allende ML, Yamashita T, Proia RL. G-protein-coupled receptor S1P1 acts within endothelial cells to regulate vascular maturation. Blood. 2003;102(10): 3665–7. OS = Overall survival; RFS = Relapse-Free Survival; PPS = Post-progression survival. Additional file 2: Table S1. Correlations between S1PR1 and related genes and markers, as evaluated using GEPIA. 4. Spiegel S, Milstien S. The outs and the ins of sphingosine-1-phosphate in immunity. Nat Rev Immunol. 2011;11(6):403–15. y 5. Spiegel S, Milstien S. Sphingosine 1-phosphate, a key cell signaling molecule. J Biol Chem. 2002;277(29):25851–4. 6. Proia RL, Hla T. Emerging biology of sphingosine-1-phosphate: its role in pathogenesis and therapy. J Clin Invest. 2015;125(4):1379–87. y 5. Spiegel S, Milstien S. Sphingosine 1-phosphate, a key cell signaling molecule. J Biol Chem. 2002;277(29):25851–4. Author details 1 1Department of Laboratory Medicine, Guangdong Second Provincial General Hospital, No. 466 Xingang Middle Road, Haizhu District, Guangzhou 510317, Guangdong Province, China. 2The First Affiliated Hospital, Guangzhou University of Chinese Medicine, No. 16 Airport Road, Baiyun District, Guangzhou 510407, China. 3Lingnan Medical Research Center, Guangzhou University of Chinese Medicine, Guangzhou 510407, China. Supplementary information S l i f i y Supplementary information accompanies this paper at https://doi.org/10. 1186/s12885-020-07278-2. Supplementary information accompanies this paper at https://doi.org/10. 1186/s12885-020-07278-2. Received: 8 April 2020 Accepted: 9 August 2020 Received: 8 April 2020 Accepted: 9 August 2020 Authors’ contributions 13. Ferrara N. Pathways mediating VEGF-independent tumor angio Cytokine Growth Factor Rev. 2010;21(1):21–6. 13. Ferrara N. Pathways mediating VEGF-independ Cytokine Growth Factor Rev. 2010;21(1):21–6. LMZ, DLC, and YFL designed the study. LMZ, LLX and ZYY collected and analyzed the data. LMZ and YFL organized the manuscript. LLX, LJL and SHS reviewed the papers and revised the manuscript. All the authors have read and approved the final manuscript. All authors contributed toward data analysis, drafting, and revising the paper, and agree to be accountable for all aspects of the work. 14. Jung B, Obinata H, Galvani S, Mendelson K, Ding B, Skoura A, et al. Flow- regulated endothelial S1P receptor-1 signaling sustains vascular development. Dev Cell. 2012;23(3):600–10. 14. Jung B, Obinata H, Galvani S, Mendelson K, Ding B, Skoura A, et al. Flow- regulated endothelial S1P receptor-1 signaling sustains vascular development. Dev Cell. 2012;23(3):600–10. 15. Gaengel K, Niaudet C, Hagikura K, Hagikura K, Laviña B, Muhl L, et al. The sphingosine-1-phosphate receptor S1PR1 restricts sprouting angiogenesis by regulating the interplay between VE-cadherin and VEGFR2. Dev Cell. 2012;23(3):587–99. 15. Gaengel K, Niaudet C, Hagikura K, Hagikura K, Laviña B, Muhl L, et al. The sphingosine-1-phosphate receptor S1PR1 restricts sprouting angiogenesis by regulating the interplay between VE-cadherin and VEGFR2. Dev Cell. 2012;23(3):587–99. Consent for publication 20. Tang H, Huang X, Wang J, Yang L, Kong Y, Gao G, et al. circKIF4A acts as a prognostic factor and mediator to regulate the progression of triple- negative breast cancer. Mol Cancer. 2019;18(1):23. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. 18. Glinsky GV, Glinskii AB, Stephenson AJ, Hoffman RM, Gerald WL. Gene expression profiling predicts clinical outcome of prostate cancer. J Clin Invest. 2004;113(6):913–23. Ethics approval and consent to participate Not applicable. 19. Chen B, Tang H, Chen X, Zhang G, Wang Y, Xie X, et al. Transcriptomic analyses identify key differentially expressed genes and clinical outcomes between triple-negative and non-triple-negative breast cancer. Cancer Manag Res. 2018;11:179–90 Published 2018 Dec 21. Competing interests Competing interests The authors declare that they have no competing interests. The authors declare that they have no competing interests. Page 17 of 17 Zhong et al. BMC Cancer (2020) 20:766 Zhong et al. BMC Cancer (2020) 20:766 21. Rhodes DR, Kalyana-Sundaram S, Mahavisno V, Varambally R, Yu J, Briggs BB, et al. Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene expression profiles. Neoplasia. 2007;9(2):166–80. 22. Lánczky A, Nagy Á, Bottai G, Munkácsy G, Szabó A, Santarpia L, et al. miRpower: a web-tool to validate survival-associated miRNAs utilizing expression data from 2178 breast cancer patients. Breast Cancer Res Treat. 2016;160(3):439–46. 23. Gao J, Aksoy BA, Dogrusoz U, Dresdner G, Gross B, Sumer SO, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6(269):pl1. 24. Szász AM, Lánczky A, Nagy Á, Förster S, Hark K, Green JE, et al. Cross- validation of survival associated biomarkers in gastric cancer using transcriptomic data of 1,065 patients. Oncotarget. 2016;7(31):49322–33. 25. Li T, Fan J, Wang B, Traugh N, Chen Q, Liu JS, et al. TIMER: a web server for comprehensive analysis of tumor-infiltrating immune cells. Cancer Res. 2017; 77(21):e108–10. 26. Tang Z, Li C, Kang B, Gao G, Li C, Zhang Z. GEPIA: a web server for cancer and normal gene expression profiling and interactive analyses. Nucleic Acids Res. 2017;45(W1):W98–W102. 27. Vasaikar SV, Straub P, Wang J, Zhang B. LinkedOmics: analyzing multi-omics data within and across 32 cancer types. Nucleic Acids Res. 2018;46(D1): D956–63. 28. Chandrashekar DS, Bashel B, Balasubramanya SAH, Creighton CJ, Ponce- Rodriguez I, Chakravarthi BVSK, et al. UALCAN: a portal for facilitating tumor subgroup gene expression and survival analyses. Neoplasia. 2017;19(8):649–58. 29. Ohtani H. Focus on TILs: prognostic significance of tumor infiltrating lymphocytes in human colorectal cancer. Cancer Immun. 2007;7:4. 30. Azimi F, Scolyer RA, Rumcheva P, Moncrieff M, Murali R, McCarthy SW, et al. Tumor-infiltrating lymphocyte grade is an independent predictor of sentinel lymph node status and survival in patients with cutaneous melanoma. J Clin Oncol. 2012;30(21):2678–83. 31. Yoshihara K, Shahmoradgoli M, Martínez E, Vegesna R, Kim H, Torres-Garcia W, et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun. 2013;4:2612. 32. Lee H, Deng J, Kujawski M, Yang C, Liu Y, Herrmann A, et al. STAT3-induced S1PR1 expression is crucial for persistent STAT3 activation in tumors. Nat Med. 2010;16(12):1421–8. 33. Competing interests Cartier A, Leigh T, Liu CH, Hla T. Endothelial sphingosine 1-phosphate receptors promote vascular normalization and antitumor therapy. Proc Natl Acad Sci U S A. 2020;117(6):3157–66. 34. Zhu J, Petit PF, Van den Eynde BJ. Apoptosis of tumor-infiltrating T lymphocytes: a new immune checkpoint mechanism. Cancer Immunol Immunother. 2019;68(5):835–47. 35. Sharma P, Hu-Lieskovan S, Wargo JA, Ribas A. Primary, adaptive, and acquired resistance to Cancer immunotherapy. Cell. 2017;168(4):707–23. 36. Aoki M, Aoki H, Ramanathan R, Hait NC, Takabe K. Corrigendum to "Sphingosine-1-phosphate signaling in immune cells and inflammation: roles and therapeutic potential". Mediat Inflamm. 2016;2016:2856829. 37. Valdivia A, Mingo G, Aldana V, Pinto MP, Ramirez M, Retamal C, et al. Fact or fiction, it is time for a verdict on Vasculogenic mimicry? Front Oncol. 2019;9:680. 38. El Hallani S, Boisselier B, Peglion F, Rousseau A, Colin C, Idbaih A, et al. A new alternative mechanism in glioblastoma vascularization: tubular vasculogenic mimicry. Brain. 2010;133(Pt 4):973–82. https://doi.org/10.1093/ brain/awq044. 39. Cao Z, Bao M, Miele L, Sarkar FH, Wang Z, Zhou Q. Tumour vasculogenic mimicry is associated with poor prognosis of human cancer patients: a systemic review and meta-analysis. Eur J Cancer. 2013;49(18):3914–23. 40. Liu S, Ni C, Zhang D, Sun H, Dong X, Che N, et al. S1PR1 regulates the switch of two angiogenic modes by VE-cadherin phosphorylation in breast cancer. Cell Death Dis. 2019;10(3):200. 40. Liu S, Ni C, Zhang D, Sun H, Dong X, Che N, et al. S1PR1 regulates the switch of two angiogenic modes by VE-cadherin phosphorylation in breast cancer. Cell Death Dis. 2019;10(3):200. 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/W2789992655	https://link.springer.com/content/pdf/10.1007%2Fs00018-018-2773-4.pdf	English	null	Detailed analysis of the plasma extracellular vesicle proteome after separation from lipoproteins	Cellular and molecular life sciences	2,018	cc-by	11,060	Abstract The isolation of extracellular vesicles (EVs) from blood is of great importance to understand the biological role of circulat- ing EVs and to develop EVs as biomarkers of disease. Due to the concurrent presence of lipoprotein particles, however, blood is one of the most difficult body fluids to isolate EVs from. The aim of this study was to develop a robust method to isolate and characterise EVs from blood with minimal contamination by plasma proteins and lipoprotein particles. Plasma and serum were collected from healthy subjects, and EVs were isolated by size-exclusion chromatography (SEC), with most particles being present in fractions 8–12, while the bulk of the plasma proteins was present in fractions 11–28. Vesicle markers peaked in fractions 7–11; however, the same fractions also contained lipoprotein particles. The purity of EVs was improved by combining a density cushion with SEC to further separate lipoprotein particles from the vesicles, which reduced the contamination of lipoprotein particles by 100-fold. Using this novel isolation procedure, a total of 1187 proteins were identified in plasma EVs by mass spectrometry, of which several proteins are known as EV-associated proteins but have hitherto not been identified in the previous proteomic studies of plasma EVs. This study shows that SEC alone is unable to completely separate plasma EVs from lipoprotein particles. However, combining SEC with a density cushion significantly improved the separation of EVs from lipoproteins and allowed for a detailed analysis of the proteome of plasma EVs, thus making blood a viable source for EV biomarker discovery. Keywords Exosomes · Extracellular vesicles · Lipoproteins · Plasma · Serum · Size-exclusion chromatography · Density cushion · Mass spectrometry · Proteomics Detailed analysis of the plasma extracellular vesicle proteome after separation from lipoproteins Nasibeh Karimi1,2 · Aleksander Cvjetkovic1 · Su Chul Jang1,3 · Rossella Crescitelli1 · Mohammad Ali Hosseinpour Feizi2 · Rienk Nieuwland4 · Jan Lötvall1 · Cecilia Lässer1 Received: 2 November 2017 / Revised: 3 January 2018 / Accepted: 1 February 2018 / Published online: 13 February 2018 © The Author(s) 2018. This article is an open access publication * Cecilia Lässer cecilia.lasser@gu.se Keywords Exosomes · Extracellular vesicles · Lipoproteins · Plasma · Serum · Size-exclusion chromatography · Density cushion · Mass spectrometry · Proteomics Cellular and Molecular Life Sciences (2018) 75:2873–2886 https://doi.org/10.1007/s00018-018-2773-4 Cellular and Molecular Life Sciences (2018) 75:2873–2886 https://doi.org/10.1007/s00018-018-2773-4 Cellular and Molecular Life Sciences ORIGINAL ARTICLE ORIGINAL ARTICLE Introduction A blood sample is minimally invasive and is one of the most commonly used samples for diagnostic purposes [1]. Dis- eased cells, such as tumour cells, and injured or stressed tissues release molecules into the bloodstream, and these molecules can be used to monitor the status of different tis- sues and organs without obtaining an invasive biopsy, and blood samples are, therefore, often referred to as a “liquid biopsy” [2]. Cells and molecules that can be analysed in a liquid biopsy include circulating tumour cells, cell-free DNA and RNA, soluble proteins, and extracellular vesicles (EVs) [2]. Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00018-018-2773-4) contains supplementary material, which is available to authorized users. * Cecilia Lässer cecilia.lasser@gu.se 1 Krefting Research Centre, Institute of Medicine at Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden 1 Krefting Research Centre, Institute of Medicine at Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden 2 Department of Animal Biology, Faculty of Natural Sciences, University of Tabriz, Tabriz, Iran 3 Codiak BioSciences, Cambridge, MA 02139, USA 4 Laboratory of Experimental Clinical Chemistry, Department of Clinical Chemistry, and Vesicle Observation Centre, Academic Medical Centre of the University of Amsterdam, Amsterdam, The Netherlands 2 Department of Animal Biology, Faculty of Natural Sciences, University of Tabriz, Tabriz, Iran EVs are small (40–800 nm) membrane-enclosed ves- icles that are released by all cells into the extracellular space [3, 4], and they contain RNA, lipids, proteins, and DNA that can be shuttled to other cells to influence the recipient cell’s phenotype [4–7]. Furthermore, patients (0121 2874 N. Karimi et al. that EVs isolated from plasma are contaminated by HDL [12] and LDL [13]. suffering from diseases such as cancer have higher concen- trations of circulating EVs, and these EVs can carry dis- ease-specific molecules [3, 8–10]. Isolation of EVs from plasma and serum is, therefore, of great importance in the use of EVs as biomarkers for diseases such as cancer [3]. However, human plasma and serum contain a vast array of particles, including EVs, but also a dominating pool of lipid particles such as chylomicrons and multiple types of lipoprotein particles and plasma proteins, making blood one of the most difficult body fluids to isolate EVs from. Introduction Thus, a major hurdle in the characterisation of circulating EVs is that EVs are difficult to separate from lipoproteins and chylomicrons, not only because of these molecules’ abundance, but also because they resemble EVs in their physical features, including size and density. The aim of this study was, therefore, to develop a robust procedure for isolating EVs from human blood with minimal contamina- tion of lipoprotein particles to enable the proteome of highly purified plasma EVs to be determined. fil Chylomicrons are produced after ingestion of fat-con- taining meals, and they transport lipids and cholesterol to the liver via the peripheral blood plasma. The size of chylomicrons varies with the amount of ingested fat, ranging from 75 to 1200 nm in diameter [11]. The liver transforms the fat in chylomicrons into very low-density lipoproteins (VLDL, 30–80 nm), which in turn can be converted into smaller types of lipoproteins (5–35 nm). These types include intermediate-density lipoprotein (IDL), low-density lipoprotein (LDL), and high-density lipoprotein (HDL), and all of these transport triglycerides and cholesterol to and from the peripheral tissues. Lipo- protein particles are very abundant in the circulation, and estimates have suggested that there are 20- to 100-fold more lipoproteins than EVs in isolates from plasma [12, 13], with chylomicrons further increasing in numbers after food intake [13]. In addition, it is suggested that > 70% of the particles isolated from plasma are non-EVs [14] and Fig. 1 Schematic overview of the experimental workflow. Blood was collected in the morning from overnight-fasting healthy subjects, and plasma and serum were isolated. Several approaches were used to isolate EVs from plasma and serum and to separate them from lipoprotein particles and plasma proteins. HDL high-density lipoprotein, IDC iodixanol density cushion, RT room temperature, SEC size exclusion chromatography, UCF ultracentrifugation PageBlue protein staining An equal volume (40 µL) from each of the collected SEC fractions was boiled in reducing buffer and subjected to electrophoresis on precast Mini-Protean TGX 4–20% gra- dient gels (Bio-Rad Laboratory). Gels were incubated with PageBlue Protein Staining Solution (Thermo Scientific, Waltham, MA, USA) for 2 h with gentle agitation at RT and then washed three times in dH2O before being detected with a VersaDoc 4000 MP imaging system (Bio-Rad Laborato- ries, Hercules, CA, USA). Isolation of EVs by size‑exclusion chromatography The particle concentrations in the SEC fractions were meas- ured with a ZetaView® PMX 110 instrument according to the manufacturer’s instruction (Particle Metrix, Meerbusch, Germany). Aliquots from the SEC fractions were diluted 10- to 1000-fold in PBS. All samples were measured in duplicate and using the same instrument settings. The chamber tem- perature was automatically measured and integrated into the calculation, and the sensitivity of the camera was set to 80 and the shutter was set to 100. Data were analysed using the ZetaView® analysis software version 8.2.30.1 with a mini- mum size of 5, a maximum size of 5000, and a minimum brightness of 20. EVs were isolated with an in-house made size-exclusion chromatography (SEC) column as previously described [15]. Briefly, Sepharose CL-2B (GE Healthcare, Uppsala, Sweden) was packed in a Telos SPE column (Kinesis, Cam- bridgeshire, UK) to a final volume of 10 mL and equili- brated with PBS. Fresh plasma or serum samples (1 mL) were applied to the premade column, and up to 30 fractions of 0.5 mL were collected with PBS as the elution buffer. Blood collection and processing Peripheral blood was collected from healthy donors after overnight-fasting. Briefly, for plasma, the blood was col- lected into K2E EDTA tubes, while for serum, the blood was collected into clot activator tubes. For plasma, processing was carried out directly, while, for serum, the sample was left at room temperature (RT) for 30 min to allow clotting. Plasma and serum were isolated by centrifugation at 1880×g for 10 min at RT. The plasma and serum were transferred to new tubes and centrifuged at 2500×g for 10 min at RT to minimise contamination by platelets. EVs were isolated directly by several different methods (Fig. 1). For some of Fig. 1 Schematic overview of the experimental workflow. Blood was collected in the morning from overnight-fasting healthy subjects, and plasma and serum were isolated. Several approaches were used to isolate EVs from plasma and serum and to separate them from lipoprotein particles and plasma proteins. HDL high-density lipoprotein, IDC iodixanol density cushion, RT room temperature, SEC size exclusion chromatography, UCF ultracentrifugation Fig. 1 Schematic overview of the experimental workflow. Blood was collected in the morning from overnight-fasting healthy subjects, and plasma and serum were isolated. Several approaches were used to isolate EVs from plasma and serum and to separate them from lipoprotein particles 1 3 Detailed analysis of the plasma extracellular vesicle proteome after separation from… 2875 loaded onto an SEC column, and fractions were collected as described above. the validating experiments, frozen plasma was used. Sam- ples were collected with the approval of the Regional Ethical Approval Committee in Gothenburg, Sweden (no. 593-08). Protein concentration measurement An OptiPrep cushion was used for the removal of lipopro- tein before loading the samples onto an SEC column. In brief, 6 mL of plasma was layered on top of a 2 mL 50%, 2 mL 30%, and 2 mL 10% OptiPrep cushion. The cushion and sample were centrifuged at 178,000×gavg (SW 41 Ti rotor, k-factor 143.9, Beckman Coulter, Brea, CA, USA) for 2 h at 4 °C. A visible band between the 10 and 30% layers was collected (high-density band) as well as a band that was floating on top of the plasma (low-density band) (Fig. 1). Collected bands were loaded onto separate SEC columns. If the collected bands were less than 1 mL in volume, PBS was added to a final volume of 1 mL prior to loading. For some experiments, the same bands from two cushions were mixed to a final volume of 1 mL and loaded to one common SEC column. The SEC columns and vesicle isolation procedure were as described above. The protein concentrations in the SEC fractions were meas- ured with a BCA assay according to the manufacturer’s instruction (Pierce™ BCA Protein Assay Kit, Thermo Sci- entific). An aliquot from each fraction was mixed with 1% SDS Tris–HCl and sonicated three times for 5 min with 10 s intermediate-speed vortexing in-between sonication prior to analysis. Mass spectrometry Based on the protein concentration in fraction 10 (the frac- tion with the most particles), a volume equivalent to 500 ng proteins was used for ELISA. The same volume as for frac- tion 10 was then used for all other fractions, which were diluted in 1 mL PBS. Samples (100 µL per well) were added to a black-walled 96-well plate (Thermo Scientific) and incubated overnight at 4 °C. After incubation, the plate was washed three times with PBS and blocked with PBS with 1% bovine serum albumin (BSA) for 1 h at RT. The PBS-1% BSA was discarded, and the primary antibodies were added (1:200 dilution) and the plate and samples were incubated for 2 h at RT. Primary antibodies included CD9, CD63, and CD81 (all from Santa Cruz Biotechnology). The plates were then washed three times with PBS-1% BSA and then incubated with donkey anti-rabbit IgG HRP-linked F(ab′)2 fragment (1:2000 dilution) or sheep anti-mouse IgG HRP- linked F(ab′)2 fragment (1:2000 dilution) (both from GE Healthcare) for 1 h at RT. The plates and samples were then washed four times, and the BM Chemiluminescence ELISA substrate (Roche, Basel, Switzerland) was used according to the manufacturer’s protocols to measure the chemilumines- cence on a Varioskan™ LUX multimode microplate reader (Thermo Fisher Scientific). Proteomic analyses were performed at The Proteomics Core Facility at the Sahlgrenska Academy, University of Gothen- burg. The EV samples (30 µg) were lysed by the addition of sodium dodecyl sulphate (SDS) to a final concentration of 2% SDS and 50 mM triethylammonium bicarbonate (TEAB). Samples were digested with trypsin using the filter- aided sample preparation method [16]. Briefly, samples were reduced with 100 mM dithiothreitol at 60 °C for 30 min, transferred to 30 kDa MWCO Pall Nanosep centrifugation filters (Sigma-Aldrich), and washed several times with 8 M urea, and once with digestion buffer prior to alkylation with 10 mM methyl methanethiosulfonate in digestion buffer for 30 min. Digestion was performed by addition of trypsin (0.3 µg, Pierce MS-grade trypsin, Thermo Fisher Scientific) in 50 mM TEAB and 1% sodium deoxycholate (SDC) buffer at 37 °C overnight. An additional portion of enzyme was added and incubated for another 2 h. Peptides were collected by centrifugation and SDC was removed by acidification with 10% trifluoroacetic acid. Electron microscopy Formvar/carbon-coated copper grids (Ted Pella, Inc., Red- ding, CA, USA) were glow discharged before the samples were loaded. The grids and samples were incubated for 15 min, fixed sequentially in 2% paraformaldehyde and 2.5% glutaraldehyde, and contrasted in 2% uranyl acetate. The preparations were examined using an LEO 912AB Omega electron microscope (Carl Zeiss NTS, Jena, Germany). Mass spectrometry l Samples were desalted using PepClean C18 spin columns (Thermo Fisher Scientific) according to the manufacturer’s guidelines prior to analysis on a Q Exactive mass spectrom- eter (Thermo Fisher Scientific) interfaced with an Easy nLC 1200 liquid chromatography system. Peptides were sepa- rated using an in-house constructed C18 analytical column (300 mm × 0.075 mm I.D., 3 μm, Dr. Maisch, Germany) using a gradient from 6 to 27% acetonitrile in 0.1% formic acid over 45 min followed by an increase to 80% acetonitrile in 0.1% formic acid for 5 min at a flow rate of 3 nL/min. Precursor ion mass spectra were acquired at 70 K resolu- tion, and MS/MS analysis was performed in a data-depend- ent mode where the 10 most intense precursor ions were selected for fragmentation using HCD at a collision energy of 27. Charge states 2–6 were selected for fragmentation, and dynamic exclusion was set to 30 s. Western blot A combination of ultracentrifugation, iodixanol density cushion, and SEC was used for isolation of larger numbers of EVs of higher purity. Briefly, 40–80 mL of plasma pooled from several individuals was diluted in PBS and centrifuged at 16,500×gavg (Type 70 Ti rotor, k-factor 950.6, Beckman Coulter) for 20 min to pellet larger EVs such as microvesi- cles. The supernatant was subjected to ultracentrifugation at 118,000×gavg (Type 70 Ti rotor, k-factor 133.7, Beckman Coulter) for 2.5 h to pellet smaller EVs such as exosomes. Both EV pellets were re-suspended in PBS and mixed into one sample with a final volume of 6 mL that was loaded onto an iodixanol cushion (as described above). Again, the fraction between the 10 and 30% layer was collected and Forty microliters of each SEC fraction were loaded and sep- arated on Mini-Protean TGX precast 4–20% gels (Bio-Rad Laboratory), and proteins were blotted onto PVDF mem- branes using a Trans-Blot Turbo Transfer system (Bio-Rad Laboratory). Membranes were blocked with 5% non-fat dry milk in TBS containing 0.01% Tween-20 (TBST). Mem- branes were incubated with the following primary antibod- ies: CD81 (1:500 dilution; clone H-121, sc-9158, Santa Cruz Biotechnology, Santa Cruz, CA), TSG-101 (1:500 dilution; clone 4A10, ab83, Abcam, Cambridge, UK), flo- tillin-1 (1:1000 dilution; clone H-104, sc-25506, Santa Cruz 1 2876 N. Karimi et al. 6000 Pico chip on an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). Biotechnology), and Apo-A (1:1000 dilution; clone FL-267, sc-30089, Santa Cruz Biotechnology), diluted in TBST over- night at 4 °C. Membranes were washed three times before being incubated with the following secondary antibodies diluted in TBST; donkey anti-rabbit IgG HRP-linked F(ab′)2 fragment (1:10,000 dilution; NA9340V), and sheep anti- mouse IgG HRP-linked F(ab′)2 fragment (1:10,000 dilutions; NA9310V) (both from GE Healthcare, Buckinghamshire, UK). Blots were visualised with SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Scientific) and a VersaDoc 4000 MP imaging system (Bio-Rad Laboratory) with Quantity One software. Total RNA isolation RNA was isolated either from 300 µL of the high-density or low-density band from the density cushion or from the pel- lets of pooled SEC fractions (F1–6, F7–12, F13–18, F19–24, and F25–30). The pooled fractions were ultracentrifuged at 115,000×gavg for 1 h (TLA-100.3 rotor, k-factor 52.8, Beck- man Coulter). Total RNA was extracted using the miRCURY RNA Isolation Kit—Cell and Plant (Exiqon, Vedbaek, Den- mark) according to the manufacturer’s instructions. For the density bands, 700 µL lysis buffer was added to 300 µL sam- ple, and for the pooled SEC fractions, 350 µL lysis buffer was added to the pellets. One microliter of isolated RNA was examined by capillary electrophoresis using an Agilent RNA Data analysis was performed using the Proteome Dis- coverer version 1.4 software (Thermo Fisher Scientific) and the Human SWISSPROT Database version Jan 2017 1 3 3 2877 Detailed analysis of the plasma extracellular vesicle proteome after separation from… (Swiss Institute of Bioinformatics, Switzerland). Mascot 2.3 (Matrix Science) was used as the search engine with a precursor mass tolerance of 5 ppm and a fragment mass tolerance of 200 mmu. Tryptic peptides were accepted with one missed cleavage, and methionine oxidation and cysteine alkylation were set as variable modifications. The detected peptide threshold in the software was a 1% False Discovery Rate by searching against a reversed database, and identi- fied proteins were grouped by shared sequences to minimise redundancy. by electron microscopy (Fig. 2d, e). It is, however, impor- tant to note that when the degree of contamination is evalu- ated by measuring lipoproteins such as Apo-A, Apo-B, and Apo-E, proteins such as Apo-E have been reported to be present on exosomes, specifically those released from pig- ment cells [18], and EVs isolated from plasma can be cov- ered with LDL [13], thus making it difficult to discriminate between lipoprotein particle contaminants and EV-associ- ated lipoproteins. We continued by analysing fractions 8 and 9 with mass spectrometry; however, only 88 proteins were identified (data not shown). Most of these proteins were plasma pro- teins, such as Apo-B, which is the primary apolipoprotein of chylomicrons (Apo B-48) and LDL and VLDL (Apo B-100). Other apolipoproteins known to be associated with chylomi- crons and other lipoprotein particles were also observed. However, few EV-specific proteins were identified, with only 9 proteins identified from the top 100 human EV proteins listed on EVpedia [17]. Size‑exclusion chromatography fails to separate extracellular vesicles from lipoproteins One millilitre of plasma or serum was loaded onto an in- house-made Sepharose-based SEC column (Fig. 1) as previ- ous described [15], and 28 fractions were collected. Nano- particle tracking analysis (NTA) was used to determine the concentration of particles in each fraction. Because fractions 8–12 contained the highest concentrations of particles (Sup- plementary Figure 1), we limited the subsequent analysis to fractions 5–16. While fractions 8–12 contained the highest concentration of particles in both plasma and serum, peaking in fraction 10, the bulk of the proteins eluted in later frac- tions (Fig. 2a and Supplementary Figure 2), confirming the previous results [15]. Bioinformatics The proteins identified with LC–MS/MS were analysed using the Database for Annotation, Visualization and Inte- grated Discovery (DAVID; http://david.abcc.ncifcrf.gov/) to identify cellular components associated with or enriched in the vesicle proteome. Venny (http://bioinfogp.cnb.csic.es/ tools/venny/index.html) was used to compare lists of pro- teins to find common and unique molecules. Information from the exosome database EVpedia (http://www.evped ia.info/) [17] was accessed in April 2017. Total RNA isolation Together, these results show that although fractions 8–12 contained particles with the size of EVs (Fig. 2a) and contained the highest concentrations of EV markers (Fig. 2b, c), electron microscopy and Western blot revealed that these fractions also contained lipoproteins and plasma proteins (Fig. 2c–e). Böing et al. used the same SEC column as we have used here, and they showed that the purest vesicles were found in fraction 9 [15]. When frac- tions 10, 11, and 12 were included, the recovery increased as these fractions also contained vesicles. However, as these fractions also contained more contaminants, this led to an overall decrease in purity. We observed a similar trend of decreased purity beyond fraction 10, but when we analysed fractions 8 and 9 with mass spectrometry, it was also clear that mainly lipoprotein and plasma proteins and relatively few vesicle proteins, could be identified. Sequential density gradient and size‑exclusion chromatography separate EVs from lipoproteins Examples of EV-like structures (cup-shaped) are indicated by black arrows, and examples of lipopro- tein particle-like structures (white structures) are indicated by white arrows. e Enlargements from fraction 8–10 from the plasma sample showed in d. Scale bars are 200 nm in d and 100 nm in e ◂ Western blot showed that flotillin-1 was present in SEC fractions 7–9 of the low-density band and in fractions 7–14 of the high-density band (Fig. 4d). Interestingly, TSG-101 was mainly detected in SEC fractions 10–14 from both the low-density and high-density bands, suggesting the pres- ence of a subpopulation of EVs positive for TSG-101 but negative for or containing lower amounts of flotillin-1 and eluting in later fractions than the flotillin-1-positive vesi- cles (Fig. 4d). It was surprising to us that the SEC fractions from the lipoprotein-enriched band (low-density band) also contained detectable levels of established vesicle markers. This indicates that particles or vesicles positive for classi- cal EV markers are also present at densities < 1.025 g/cm3. This is puzzling as to our knowledge, there is little evidence suggesting that EVs can be found at this density. However, two studies have found EVs at these densities, one where prominin-1—containing particles were isolated from human epithelial colorectal adenocarcinoma cells at a density of 1.032–1.068 g/cm3 [20], and another study where vesicles were isolated from bone marrow-derived mesenchymal stem cells at a density of < 1.06 g/cm3 [21]. removes the lipoprotein particles with similar density to EVs but that differ in size (Fig. 3). f Six millilitres of plasma were loaded on top of an iodix- anol density cushion (Fig. 1) and centrifuged. The density cushion was carefully designed to allow the vesicles to float at approximately 1.06–1.16 g/cm3, while most lipoprotein particles have a lower density (Fig. 3) and, therefore, will float on top of the cushion. Following centrifugation, two bands were visible, one band containing material floating above the plasma at approximately < 1.025 g/cm3, here called the “low-density band”, which was enriched in lipo- protein particles, and a “high-density band” containing material floating at approximately 1.06–1.16 g/cm3, which was expected to be enriched in EVs (Fig. 4a). Next, the low- density and high-density bands were loaded onto individual SEC columns, and the concentrations of particles and pro- teins were measured in the eluted fractions. Sequential density gradient and size‑exclusion chromatography separate EVs from lipoproteins Because the concentration of chylomicrons increases in the blood after a meal, blood was collected from individu- als after overnight fast. Despite this precaution, SEC alone was still unable to separate EVs from contaminating factors (Fig. 2c–e). Several lipoproteins have a diameter of less than 35 nm and would, therefore, be eluted in later fractions than EVs. On the other hand, chylomicrons and VLDL are over- lapping in diameter with EVs, and co-isolation is expected during size-based separation by SEC. Because chylomicrons and VLDL differ in density from EVs, < 0.93, < 1.06, and > 1.10 g/cm3, respectively [11, 19], we decided to combine SEC with a density-based separation to further separate EVs from lipoprotein particles and chylomicrons. Density separa- tion was thus expected to remove lipoprotein particles that are similar in size to EVs but differ in density, whereas SEC We used an in-house developed ELISA system to deter- mine the presence of the common EV markers CD9, CD63, and CD81 in the SEC fractions. All three markers were the most prominent in the fractions containing the highest con- centrations of particles (Fig. 2b, fraction 8–12). Flotillin-1, also a marker of EVs, was most abundant in fractions 8 and 9 of both serum and plasma as determined by Western blot (Fig. 2c). However, Western blot also revealed that apolipo- protein A1 (Apo-A1), a marker for HDL and chylomicrons, was detectable in fractions 7–12, showing that lipoprotein particles were co-isolated with EVs (Fig. 2c). The presence of lipoprotein particles and plasma proteins was confirmed 1 3 1 3 N. Karim B A C D E 2878 N. Karimi et al. B A C A B B A C D C D D E E 1 3 3 Detailed analysis of the plasma extracellular vesicle proteome after separation from… 2879 in size with EVs (Fig. 3). Thus, our results demonstrate that it is important to combine separation steps, each based on its own physical separation principle, when working with plasma, and to characterise the obtained isolates with several methods to determine the composition of the isolates. Fig. 2 Evaluation of EVs isolated with size-exclusion chromatogra- phy (SEC). One millilitre of plasma or serum was loaded onto 10 mL Sepharose CL-2B columns, and up to 30 fractions of 500 µL were collected from each column. Sequential density gradient and size‑exclusion chromatography separate EVs from lipoproteins a Concentrations of particles and pro- teins in the SEC fractions were determined with nanoparticle track- ing analysis (NTA; ZetaView®, blue) and BCA (green), respectively. Data are presented as the percentage of the total amount of particles or proteins in fractions 5–16. N = 4–6, and the results are presented as the average ± SEM. b ELISA was used to determine the expres- sion of CD9, CD63, and CD81 on the vesicles in the SEC fractions. Data are presented as the percentage of the total expression for each protein in fractions 5–16. N = 3–5, and the results are presented as the average ± SEM. c Presence of the vesicle marker flotillin-1 and the HDL marker Apo-A1 was determined in fractions 7–12 (40 µL/ fraction) with Western blot. d, e Fifteen microliters (1–20 µg protein) from fractions 8–11 were loaded onto grids, negative stained, and evaluated with electron microscopy. Examples of EV-like structures (cup-shaped) are indicated by black arrows, and examples of lipopro- tein particle-like structures (white structures) are indicated by white arrows. e Enlargements from fraction 8–10 from the plasma sample showed in d. Scale bars are 200 nm in d and 100 nm in e ◂ Fig. 2 Evaluation of EVs isolated with size-exclusion chromatogra- phy (SEC). One millilitre of plasma or serum was loaded onto 10 mL Sepharose CL-2B columns, and up to 30 fractions of 500 µL were collected from each column. a Concentrations of particles and pro- teins in the SEC fractions were determined with nanoparticle track- ing analysis (NTA; ZetaView®, blue) and BCA (green), respectively. Data are presented as the percentage of the total amount of particles or proteins in fractions 5–16. N = 4–6, and the results are presented as the average ± SEM. b ELISA was used to determine the expres- sion of CD9, CD63, and CD81 on the vesicles in the SEC fractions. Data are presented as the percentage of the total expression for each protein in fractions 5–16. N = 3–5, and the results are presented as the average ± SEM. c Presence of the vesicle marker flotillin-1 and the HDL marker Apo-A1 was determined in fractions 7–12 (40 µL/ fraction) with Western blot. d, e Fifteen microliters (1–20 µg protein) from fractions 8–11 were loaded onto grids, negative stained, and evaluated with electron microscopy. Sequential density gradient and size‑exclusion chromatography separate EVs from lipoproteins The majority of particles were recovered in fractions 7–14 for the low-den- sity band, whereas the high-density band peaked at fraction 8 (Fig. 4b). When the absolute numbers of particles were compared between the same fractions of the low-density and high-density bands from the same sample, fractions 8–14 contained 30- to 100-fold more particles in the low-density band, suggesting that the chylomicrons and lipoprotein par- ticles are substantially more abundant then EVs in plasma (Fig. 4c), supporting the previous observations [12–14]. This emphasises the difficulty that the EV field faces when work- ing with highly complex samples such as plasma, where EVs are a tiny minority among other particles with similar physi- cal features. This also highlights the limitations of NTA. Although NTA efficiently measures the particle abundance in SEC fractions, NTA fails to distinguish EVs from non-EV components. This becomes a problem especially when work- ing with plasma and serum as lipoprotein particles overlap In this study, we used Sepharose CL-2B with a pore size of approximately 75 nm [22], which causes the EVs below this size to be eluted in later fractions together with the small lipoprotein particles and soluble plasma proteins. When evaluating the yield of vesicles in the different fractions, it is important to remember that NTA cannot detect smaller vesicles or particles that might be present in the analysed fractions nor in fractions 16, and later, because NTA has a lower limit of detection of about 70 nm for EVs, although this detection limit depends on the refractive index of the measured particles [23]. Thus, NTA cannot detect HDL and LDL particles that are eluted in the later fractions. Our find- ing that flotillin-1 and TSG-101 are also detectable in the fractions containing smaller particles stresses the need for a way to separate these potentially interesting EVs from small lipoproteins and plasma proteins, as well as the need for more sensitive instrumentation capable of detecting particles at such a minute size range. Apo-A1 was mainly detected in the SEC fractions of the low-density band, suggesting that by implementing the density cushion before SEC, the lipoprotein particles can be efficiently separated from the plasma EVs (Fig. 4d). The higher particle concentration in the low-density band compared to the high-density band was further supported by electron microscopy (Fig. 4e). Sequential density gradient and size‑exclusion chromatography separate EVs from lipoproteins Furthermore, electron microscopy revealed that the low-density band contained more lipoprotein-like particles, while the high-density band 1 3 1 3 N. Karimi et al. 2880 Fig. 3 Schematic overview of the size and density of lipoproteins and EVs. Sev- eral of the lipoproteins such as high-density lipoproteins (HDL), low-density lipoproteins (LDL), intermediate-density lipoproteins (IDL), very low- density lipoproteins (VLDL), and chylomicrons overlap with extracellular vesicles (EVs) in terms of size or density. With an iodixanol density cushion of 10%/30%/50%, the 10% layer will create a density cutoff at approximate 1.06 g/cm3 (indicated by the orange dashed line). With Sepharose CL-2B SEC columns, the size cutoff is approximately 75 nm (indicated by the blue dashed line). The picture is modified from [11] therefore, reasonable to conclude that the combination of flotation and density gradient is essential to remove con- taminating lipid protein particles and plasma proteins for the analysis of plasma EV protein content. In addition, with this approach, the starting volume of plasma could be increased. contained more EV-like structures (Fig. 4e). Thus, the SEC fractions from the high-density band contained less contami- nation, and a purer EV isolate was generated when flotation and SEC were combined. A commercial column, Exo-spin (CellGS), has previ- ously been used to successfully isolate EVs from cell cul- ture media from a prostate cancer cell line; however, when plasma was used, the fractions containing EVs overlapped with the fractions containing apolipoproteins [14]. Thus, this column also has a problem with separating EVs from lipo- protein particles, highlighting again that a combination of density cushion and SEC is probably essential when plasma or serum is used, although this problem might not be as great when cell culture media are used as a starting material, espe- cially if the cells are cultured in a serum-free environment. The starting amount of plasma can be further increased with sequential ultracentrifugation, density gradient, and size‑exclusion chromatography The concentrations of particles and proteins in the SEC fractions were determined by nanoparticle track- ng analysis (NTA; ZetaView®, blue) and BCA (green), respectively. Data are presented as the percentage of the total amount of particles or proteins in fractions 5–16. N = 3, and the results are presented as the average ± SEM. c Total number of particles was determined by ZetaView® in fractions 5–16 from the high-density and low-density bands isolated from the same plasma samples, and the fold change was calculated. N = 3, and the results are presented as the aver- age ± SEM. LD low-density, HD high-density. d Presence of the vesicle markers flotillin-1 and TSG-101 as well as the HDL marker Apo-A1 was determined by Western blot of fractions 7–14 (40 µL/ fraction) isolated from both the high-density and low-density band. e Fifteen microliters (1–6 µg protein) from fractions 8–10 from the high-density and low-density bands were loaded onto grids, nega- tive stained, and evaluated with electron microscopy. Scale bars are 200 nm A A B C B C C B C D D E E Fig. 4 Evaluation of EVs isolated with the combination of den- sity cushion and size-exclusion chromatography (IDC + SEC). Six millilitres of plasma were loaded on top of a density cushion (50%/30%/10% iodixanol), and visible bands after ultracentrifuga- tion were further loaded onto 10 mL Sepharose CL-2B columns, and up to 30 fractions of 500 µL each were collected. a After centrifug- ing of the plasma sample on top of the cushion, two bands were vis- ible. One band contained material floating above 1.025 g/cm3 (low- density band), and the second band contained material floating at approximately 1.06–1.16 g/cm3 (high-density band). b Low-density and high-density bands were loaded onto individual SEC columns, and fractions were collected. The concentrations of particles and proteins in the SEC fractions were determined by nanoparticle track- ing analysis (NTA; ZetaView®, blue) and BCA (green), respectively. Data are presented as the percentage of the total amount of particles or proteins in fractions 5–16. N = 3, and the results are presented as the average ± SEM. c Total number of particles was determined by ZetaView® in fractions 5–16 from the high-density and low-density bands isolated from the same plasma samples, and the fold change was calculated. N = 3, and the results are presented as the aver- age ± SEM. LD low-density, HD high-density. The starting amount of plasma can be further increased with sequential ultracentrifugation, density gradient, and size‑exclusion chromatography d Presence of the vesicle markers flotillin-1 and TSG-101 as well as the HDL marker Apo-A1 was determined by Western blot of fractions 7–14 (40 µL/ fraction) isolated from both the high-density and low-density band. e Fifteen microliters (1–6 µg protein) from fractions 8–10 from the high-density and low-density bands were loaded onto grids, nega- tive stained, and evaluated with electron microscopy. Scale bars are 200 nm “Membrane” (Fig. 5d), supporting the conclusion that EVs of high purity had, indeed, been isolated. present in fractions 7–10 when the volume of starting mate- rial was increased (Fig. 5c compared to Fig. 4e). Fractions 8 and 9 were combined and analysed with mass spectrometry, and in total, 1187 proteins were identified (Supplementary Table 1). A total of 85 proteins were identified from the top 100 human EV proteins listed on EVpedia [17] as well as several common EV proteins such as Rab proteins, annex- ins, tetraspanins, heat shock proteins, and ESCRT proteins (Table 1). The identified proteins were analysed with GO Term Finder to identify enriched cellular components com- pared to the genome frequency, and the top associated terms were “Extracellular exosome”, “Blood microparticle”, and The starting amount of plasma can be further increased with sequential ultracentrifugation, density gradient, and size‑exclusion chromatography To increase the concentration of vesicles, ultracentrifugation was added to the procedure (Fig. 1), thereby offering the advantage that we could start with any volume of plasma. Plasma was centrifuged at 16,500×g and 120,000×g, and both pellets were dissolved in PBS, mixed, and loaded on top of a density cushion. The high-density band was subse- quently loaded onto an SEC column and fractions were col- lected. In accordance with the previous observations, most particles eluted in fractions 8 and 9 (Fig. 5a). Western blot once more showed that the flotillin-1 signal was strongest in fractions 8 and 9; however, flotillin-1 was now also detect- able in fractions 10–14 (Fig. 5b). Furthermore, Western blot showed low levels of Apo-A1 in fractions 7–14 (Fig. 5b). Electron microscopy confirmed that more vesicles were Due to the small number of vesicles in the high-density band, bands from two density cushions were combined and loaded onto a single SEC column, resulting in a total of 12 mL plasma as starting material, and fractions 8 and 9 were analysed with mass spectrometry. In total, 634 and 608 proteins were identified in fractions 8 and 9, respectively (data not shown), with approximately 90% overlap, indicat- ing that vesicles in fractions 8 and 9 are similar in their protein cargo. In total, 86 proteins were identified from the top 100 human EV proteins listed at EVpedia [17]. It is, 3 Detailed analysis of the plasma extracellular vesicle proteome after separation from… 2881 t i f ti 7 10 h th l f t ti t “M b ” (Fi 5d) ti th l i th t EV A B C D E Fig. 4 Evaluation of EVs isolated with the combination of den- sity cushion and size-exclusion chromatography (IDC + SEC). Six millilitres of plasma were loaded on top of a density cushion 50%/30%/10% iodixanol), and visible bands after ultracentrifuga- ion were further loaded onto 10 mL Sepharose CL-2B columns, and up to 30 fractions of 500 µL each were collected. a After centrifug- ng of the plasma sample on top of the cushion, two bands were vis- ble. One band contained material floating above 1.025 g/cm3 (low- density band), and the second band contained material floating at approximately 1.06–1.16 g/cm3 (high-density band). b Low-density and high-density bands were loaded onto individual SEC columns, and fractions were collected. Protein analysis of plasma‑derived extracellular vesicles The pellets from 16,500×g and 118,000×g spins were re-suspended in PBS, mixed, loaded on top of a density cushion (50%/30%/10% iodixanol), and centrifuged. The band between 30 and 10% was subsequently loaded onto a 10 mL Sepharose CL-2B column, and up to 30 fractions of 500 µL each were collected. a Concentrations of particles and pro- teins in the SEC fractions were determined with nanoparticle track- ing analysis (NTA; ZetaView®, blue) and BCA (green), respectively. Data are presented as the percentage of the total amount of particles or proteins in fractions 5–16. N = 1. b Presence of the vesicle marker flotillin-1 and the HDL marker Apo-A1 was determined with Western blot in fractions 7–14 (40 µL/fraction). c Five micrograms of protein (11–19 µL) from SEC fractions 7–10 were loaded onto grids, nega- tive stained, and evaluated with electron microscopy. Scale bars are 500 nm. d LC–MS/MS was performed on the EVs isolated from frac- tions 8 and 9 and pooled. In total, 1187 proteins were identified and were analysed with DAVID Bioinformatics Resources 6.8 (https:// david.ncifcrf.gov/). The ten most associated cellular compartments (based on p value) are listed in the graph. e The 1187 identified pro- teins were compared to previously published proteomes of plasma EVs [29–31, 33] or proteins in fractions 5–16. N = 1. b Presence of the vesicle marker flotillin-1 and the HDL marker Apo-A1 was determined with Western blot in fractions 7–14 (40 µL/fraction). c Five micrograms of protein (11–19 µL) from SEC fractions 7–10 were loaded onto grids, nega- tive stained, and evaluated with electron microscopy. Scale bars are 500 nm. d LC–MS/MS was performed on the EVs isolated from frac- tions 8 and 9 and pooled. In total, 1187 proteins were identified and were analysed with DAVID Bioinformatics Resources 6.8 (https:// david.ncifcrf.gov/). The ten most associated cellular compartments (based on p value) are listed in the graph. e The 1187 identified pro- teins were compared to previously published proteomes of plasma EVs [29–31, 33] for hematopoietic stem cells, CD11c and CD123 for den- dritic cells, CD19 and CD20 for B cells, and CD3, CD4, and CD8 for T cells. Importantly, we did detect MHC class I but not MHC class II, the latter being expressed only on antigen presenting cells. Protein analysis of plasma‑derived extracellular vesicles In an attempt to determine the cellular origin of the isolated vesicles, the presence of markers for several cells was evalu- ated in the proteomic data set. In the plasma EVs, the fol- lowing cell markers were detected with mass spectrometry: CD235a (glycophorin-A) for erythrocytes; CD41, CD61, and CD62p for platelets; and CD56 for NK cells. However, 1 3 1 3 2882 N. Karimi et al. rons glia cells and skeletal muscle cells can also express for hematopoietic stem cells CD11c and CD123 for den A B C D E 5 Evaluation of EVs isolated with the combination of ultracen- gation, density cushion, and size-exclusion chromatography F + IDC + SEC). To be able to increase the starting volume lasma, two centrifugation steps were added. The pellets from 00×g and 118,000×g spins were re-suspended in PBS, mixed, ed on top of a density cushion (50%/30%/10% iodixanol), and rifuged. The band between 30 and 10% was subsequently loaded a 10 mL Sepharose CL-2B column, and up to 30 fractions of µL each were collected. a Concentrations of particles and pro- s in the SEC fractions were determined with nanoparticle track- analysis (NTA; ZetaView®, blue) and BCA (green), respectively. a are presented as the percentage of the total amount of particles or proteins in fractions 5–16. N = 1. b Presence of the vesicle marke flotillin-1 and the HDL marker Apo-A1 was determined with Wester blot in fractions 7–14 (40 µL/fraction). c Five micrograms of protei (11–19 µL) from SEC fractions 7–10 were loaded onto grids, nega tive stained, and evaluated with electron microscopy. Scale bars ar 500 nm. d LC–MS/MS was performed on the EVs isolated from frac tions 8 and 9 and pooled. In total, 1187 proteins were identified an were analysed with DAVID Bioinformatics Resources 6.8 (https:/ david.ncifcrf.gov/). The ten most associated cellular compartment (based on p value) are listed in the graph. e The 1187 identified pro teins were compared to previously published proteomes of plasm EVs [29–31, 33] B A B A C C C C D E E D Fig. 5 Evaluation of EVs isolated with the combination of ultracen- trifugation, density cushion, and size-exclusion chromatography (UCF + IDC + SEC). To be able to increase the starting volume of plasma, two centrifugation steps were added. Protein analysis of plasma‑derived extracellular vesicles used SEC to isolate plasma EVs, but in this study, only 21 proteins could be detected by mass spectrometry, and these proteins were mainly soluble plasma proteins, despite removal of more than > 97% of the proteins [33]. Thus, the authors used a protein array platform, SOMAscan®, instead of mass spectrometry and could then detect approximately 1000 proteins [33]. Because SOMAscan® is a multiplex aptamer-based protein array, only aptamer-binding proteins were detected. Therefore, several exosomal markers such as CD9, CD81, and ezrin were not found, as they were not among the 1300 proteins that the aptamer-based assay had been designed for. Thus, although the authors increased the identified proteins from 21 to over a 1000 using a multiplex protein assay, this technique still has limitations. Interestingly, the membrane proteins CD55, CD59, and CD47 were identified in plasma EVs. CD55 and CD59 pro- tect cells against lysis by the complement complex [24], and it has been demonstrated that when expressed on exosomes, these molecules protect against complement-mediated vesi- cle lysis [25]. The expression of CD47, which is extensively expressed on red blood cells, is considered to prevent recog- nition by macrophages, and it is referred to as a “don’t-eat- me” signal or “marker of self” [26]. The presence of these three proteins on the plasma-derived EVs suggests that they are at least partly protected from rapid consumption in the circulation thus allowing for prolonged circulation time. The concentration of albumin in plasma ranges between 30 and 50 mg/mL [27], and albumin is, therefore, the main contaminant when plasma EVs are isolated by ultracentrifu- gation, irrespectively of washings [28]. Therefore, several previous studies determining the proteome of plasma EVs by mass spectrometry have used other isolation methods such as SEC, commercial columns, density gradients, and immune- affinity capture [29, 30]. Looze et al. were the first to use mass spectrometry on EVs isolated from human plasma [31]. Although an ambitious approach including both gel exclusion chromatography and rate zonal centrifugation was applied to isolate vesicles, only 66 proteins could be identi- fied [31]. Kalra et al. identified 213 proteins using three dif- ferent isolation methods for plasma EVs [29]. In addition, de Menezes-Neto et al. used different isolation methods and identified 330 proteins [30]. Protein analysis of plasma‑derived extracellular vesicles Together, these data show that the majority of the plasma EVs originated from erythrocytes and plate- lets; however, even though we were unable to detect other neurons, glia cells, and skeletal muscle cells can also express CD56, which is why it cannot be concluded that theses EVs really originate from NK cells. The expression of the follow- ing specific cell type proteins was not detectable in plasma EVs: CD326 (EpCAM) for epithelial cells, CD146 for endothelial cells, CD45 for leukocytes, CD66b for granulo- cytes, CD14 and CD33 for macrophages/monocytes, CD34 neurons, glia cells, and skeletal muscle cells can also express CD56, which is why it cannot be concluded that theses EVs really originate from NK cells. The expression of the follow- ing specific cell type proteins was not detectable in plasma EVs: CD326 (EpCAM) for epithelial cells, CD146 for endothelial cells, CD45 for leukocytes, CD66b for granulo- cytes, CD14 and CD33 for macrophages/monocytes, CD34 1 3 2883 Detailed analysis of the plasma extracellular vesicle proteome after separation from… Table 1 Identification of common EV proteins Proteins highlighted in bold are the proteins that have not been previously identified in plasma EVs with mass spectrometry or SOMAscan® [29–31, 33] ND not detected Protein group Proteins Rabs Rab-1A, -1B, -2A, -2B, -4A, -4B, -5A, -5B, -5C, -6A, -6B, -7a, -8A, -8B Rab-10, -11B, -14, -18 Rab-21, -27B Rab-30, -32, -33A, -35, -37, -38 Annexins Annexin A2, Annexin A4, Annexin A7, Annexin A11 Tetraspanins CD9, CD63, CD81, CD82, CD151, TSPAN2, TSPAN14, TSPAN32 Common EV markers MHC class I, Ezrin, Flotillin-1, Flotillin-2, Cofilin-1, Profilin-1, CD59, 14-3-3 protein (beta/alpha, epsilon, eta, gamma, sigma, theta, zeta/delta) Heat shock proteins Heat shock 70 kDa protein 1A/1B Heat shock cognate 71 kDa Heat shock protein 75 kDa, mitochondrial Heat shock protein beta-1 Heat shock protein HSP 90-alpha and beta ESCRT ESCRT-0—ND ESCRT-I—VPS-28, VPS-37B ESCRT-II—ND ESCRT-III—CHMP4B, CHMP6 ESCRT accessory—Clathrin, Alix 2883 Detailed analysis of the plasma extracellular vesicle proteome after separation from… ND not detected cell-specific markers in our present analysis, these other cells might still contribute to the mixture of circulating EVs. human plasma [29–32]. However, only half of the identified proteins were observed in more than one study, demonstrat- ing a large heterogeneity between studies [30]. Furthermore, relatively few specific vesicle markers were identified, dem- onstrating how difficult it is to isolate and separate EVs from abundant plasma proteins [30]. In addition, Welton et al. Protein analysis of plasma‑derived extracellular vesicles In total, these studies together identified approximately 400 proteins in EVs isolated from When we compare our data set to these previously pub- lished data sets, it is clear that the overlap is very small and that we have identified almost 800 novel proteins previ- ously not identified in plasma EVs (Fig. 5e). Several of these proteins have been identified in previous proteomic stud- ies on EVs from other sources such as cell lines (Table 1, highlighted in bold) and some of these EV proteins, such as CD63, CD9, and CD81, have previously been identified in plasma EVs with Western blot, electron microscopy, or flow cytometry [15, 29, 30, 33, 34], but so far never with mass spectrometry. Thus, our current study has identified many 1 3 2884 N. Karimi et al. [38, 39]), the particles/vesicles floating at a density as low as < 1.025 g/cm3 should be composed of > 90% lipids and have very low concentrations of both RNA and protein. We, therefore, analysed the RNA content in the high-density and low-density bands to determine whether the lipoprotein particles with lower density also contain RNA. Bioanalyzer analysis showed that the EV-enriched band (high-density band) contained significantly more RNA than the lipoprotein-enriched band (low-density band) (Fig. 6a). The high-density band was then separated further by SEC, and the collected fractions were pooled (fractions 1–6, 7–12, 13–18, 19–24, and 25–30) prior to ultracentrifugation and RNA analysis. The RNA in the high-density band was mainly present in SEC fractions 7–12 (Fig. 6b), which again were the fractions containing most of the EVs and no lipoprotein particles (Figs. 4d, e, 5a–d). These data strongly suggest that most of the circu- lating RNA is, indeed, packed in the EVs and that lipopro- tein particles, except HDL, and a potential subpopulation of low-density EVs contain little or no RNA. novel plasma EV proteins, far beyond what has previously been reported. This highlights two important conclusions from our study. First, a combination of several purification steps is required to generate a sample that is pure enough for mass spectrometry. Second, several particles observed in plasma are, indeed, not vesicles, and it is, therefore, most likely not yet possible to isolate sufficient EVs from a small volume of plasma (< 10 mL) for subsequent analyses using the current state-of-the-art techniques. RNA analysis of plasma‑derived extracellular vesicles Both EVs [7, 35] and HDL [36] have been shown to con- tain RNA. Interestingly, it has been shown that EVs of different densities differ in RNA content [35, 37]. Thus, the amount of and characteristics of the RNA cargo could be partly responsible for the different densities, with EVs of higher density containing more RNA [35]. Based on the densities for different biomolecules (protein (1.35 g/ cm3), RNA and DNA (1.7 g/cm3), and (lipids ~ 1 g/cm3) 1 25 200 500 1000 2000 4000 [nt] Low-density band High-density band SEC Fraction 1-6 SEC Fraction 7-12 SEC Fraction 13-18 SEC Fraction 19-24 SEC Fraction 25-30 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 25 0 5 10 15 20 [FU] 25 0 5 10 15 20 [FU] 25 0 5 10 15 20 [FU] 25 30 A B g. 6 RNA isolation from EV-enriched and lipoprotein- nriched fractions. a Six millilitres of plasma were loaded onto a 0%/30%/10% iodixanol density cushion, and the low-density and igh-density bands were isolated. RNA was isolated with a miR- URY RNA Isolation Kit—Cell and Plant (Exiqon) directly from 00 µL of the high-density and low-density bands and analysed with Bioanalyzer® (Agilent). N = 4–6, and the results are presented as he average ± SEM. p value < 0.01. LD low-density, HD high- density. b Fifty-eight millilitres of plasma were ultracentrifuged, and the pellets from a 16,500×g and 118,000×g spin were re-suspended in PBS, mixed, loaded on top of a density cushion (50%/30%/10% iodixanol) and centrifuged. The band between 30 and 10% was subse- quently loaded onto a 10 mL Sepharose CL-2B column and 30 frac- tions of 500 µL each were collected in pools of 6 fractions (3 mL in total/pool). RNA analysis of plasma‑derived extracellular vesicles The sample pools were ultracentrifuged, and RNA was isolated and analysed as in a 25 200 500 1000 2000 4000 [nt] Low-density band High-density band 25 200 500 1000 2000 4000 [nt] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] A A A A 25 200 500 1000 2000 4000 [nt] Low-density band High-density band 25 200 500 1000 2000 4000 [nt] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 25 200 500 1000 2000 4000 [nt] Low-density band High-density band SEC Fraction 1-6 SEC Fraction 7-12 SEC Fraction 13-18 SEC Fraction 19-24 SEC Fraction 25-30 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 25 0 5 10 15 20 [FU] 25 0 5 10 15 20 [FU] 25 0 5 10 15 20 [FU] 25 30 A B Fig. 6 RNA isolation from EV-enriched and lipoprotein- enriched fractions. a Six millilitres of plasma were loaded onto a 50%/30%/10% iodixanol density cushion, and the low-density and high-density bands were isolated. RNA was isolated with a miR- CURY RNA Isolation Kit—Cell and Plant (Exiqon) directly from 300 µL of the high-density and low-density bands and analysed with density. b Fifty-eight millilitres of plasma were ultracentrifuged, a the pellets from a 16,500×g and 118,000×g spin were re-suspend in PBS, mixed, loaded on top of a density cushion (50%/30%/1 iodixanol) and centrifuged. References 1. Anderson NL, Anderson NG (2002) The human plasma proteome: history, character, and diagnostic prospects. Mol Cell Proteom 1(11):845–867 2. Perakis S, Speicher MR (2017) Emerging concepts in liquid biop- sies. BMC Med 15(1):75 3. Lässer C (2015) Exosomes in diagnostic and therapeutic applica- tions: biomarker, vaccine and RNA interference delivery vehicle. Expert Opin Biol Ther 15(1):103–117 4. Yanez-Mo M, Siljander PR, Andreu Z et al (2015) Biological properties of extracellular vesicles and their physiological func- tions. J Extracell Vesicles 4:27066 5. Shelke GV, Jang SC, Yin Y et al (2016) Human mast cells release extracellular vesicle-associated DNA. Matters. https:// doi.org/10.19185/matters.201602000034f 6. Lazaro-Ibanez E, Sanz-Garcia A, Visakorpi T et al (2014) Differ- ent gDNA content in the subpopulations of prostate cancer extra- cellular vesicles: apoptotic bodies, microvesicles, and exosomes. Prostate 74(14):1379–1390 In conclusion, by combining a density cushion and SEC, we could for the first time isolate EVs from plasma with high purity. Furthermore, we show that the majority of particles detected in plasma are not EVs, which empha- sises the relevance of stringent isolation methods prior to the analysis of EV cargo and function to avoid studying non-EV-associated features. 7. Valadi H, Ekström K, Bossios A et al (2007) Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 9(6):654–659 8. Skog J, Wurdinger T, van Rijn S et al (2008) Glioblastoma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers. Nat Cell Biol 10(12):1470–1476 Acknowledgements The authors acknowledge the assistance of the Krefting Research Centre clinical research nurses for sample collection and the participants for donating their blood and their time. We thank the Proteomics Core Facility at Sahlgrenska Academy, Gothenburg University, for performing the proteomic analysis. We acknowledge the Centre for Cellular Imaging at the University of Gothenburg and the National Microscopy Infrastructure, NMI (VR-RFI 2016-00968) for providing assistance in microscopy. The authors also thank the cost action BM1202 MEHAD for the Short Term Scientific Missions (STSMs) Grant [40] to AC to visit RN’s laboratory. 9. Logozzi M, De Milito A, Lugini L et al (2009) High levels of exosomes expressing CD63 and caveolin-1 in plasma of mela- noma patients. PLoS One 4(4):e5219 10. Eldh M, Olofsson Bagge R, Lasser C et al (2014) MicroRNA in exosomes isolated directly from the liver circulation in patients with metastatic uveal melanoma. BMC Cancer 14:962 11. Compliance with ethical standards Conflict of interest JL and SCJ have written several patents in the field of extracellular vesicles as therapeutics and are currently or have previously been employees of Codiak BioSciences Inc. CL and AC are co-inventors on a patent using extracellular vesicles as diagnostic tools in diseases. The other authors declare that there are no financial, personal, or professional interests that could be construed to have in- fluenced the paper. Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creativeco mmons.org/licenses/by/4.0/), which permits unrestricted use, distribu- tion, 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. RNA analysis of plasma‑derived extracellular vesicles The band between 30 and 10% was sub quently loaded onto a 10 mL Sepharose CL-2B column and 30 fr tions of 500 µL each were collected in pools of 6 fractions (3 mL SEC Fraction 1-6 SEC Fraction 7-12 S 25 2 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 25 0 5 10 15 20 [FU] 25 30 B B SEC Fraction 7-12 SEC Fraction 13-18 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] [nt] 0 5 10 15 20 [FU] 0 5 10 15 20 [FU] 25 SEC Fraction 19-24 25 200 500 1000 2000 4000 [nt] 0 5 10 15 20 [FU] 25 SEC Fraction 19-24 SEC Fraction 25-30 25 200 500 1000 2000 4000 [nt] 25 200 500 1000 2000 4000 [nt] 0 5 10 15 20 [FU] 25 0 5 10 15 20 [FU] 25 SEC Fraction 25-30 25 200 500 1000 2000 4000 [nt] 0 5 10 15 20 [FU] 25 Fig. 6 RNA isolation from EV-enriched and lipoprotein- enriched fractions. a Six millilitres of plasma were loaded onto a 50%/30%/10% iodixanol density cushion, and the low-density and high-density bands were isolated. RNA was isolated with a miR- CURY RNA Isolation Kit—Cell and Plant (Exiqon) directly from 300 µL of the high-density and low-density bands and analysed with a Bioanalyzer® (Agilent). N = 4–6, and the results are presented as the average ± SEM. p value < 0.01. LD low-density, HD high- density. b Fifty-eight millilitres of plasma were ultracentrifuged, and the pellets from a 16,500×g and 118,000×g spin were re-suspended in PBS, mixed, loaded on top of a density cushion (50%/30%/10% iodixanol) and centrifuged. The band between 30 and 10% was subse- quently loaded onto a 10 mL Sepharose CL-2B column and 30 frac- tions of 500 µL each were collected in pools of 6 fractions (3 mL in total/pool). The sample pools were ultracentrifuged, and RNA was isolated and analysed as in a 3 Detailed analysis of the plasma extracellular vesicle proteome after separation from… 2885 Conclusion Funding This work was funded by Grants from the Swedish Research Council (Grant no. K2014-85X-22504-01-3), the Swedish Cancer Foundation, VBG Group Herman Krefting Foundation for Asthma and Allergy Research, and the Swedish Heart and Lung Foundation (Grant nos. CAN2014/844, 20150588). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Isolation of EVs from bio-fluids for downstream analysis is still problematic, because many body fluids have a complex biochemical and physical composition. This is particularly true for plasma and serum, making it difficult to obtain pure EV isolates from such fluids. This study demonstrates that a two-step isolation procedure, combining density cush- ion separation followed by SEC, isolates EVs from human plasma, and efficiently separates EVs from the main con- taminants lipoproteins and plasma proteins. By floating the vesicles on a density cushion, the EVs could be separated from the chylomicrons and other lipoproteins with lower densities than EVs, and by loading the floated EVs on an SEC column, the EVs could be further separated from solu- ble proteins and lipoproteins with a smaller size than EVs. With this isolation approach, plasma EVs of previously unat- tained purity could be identified by electron microscopy and Western blot, and in total, 1187 proteins could be identified in the EV isolates with mass spectrometry, without excessive contamination of plasma proteins and lipoprotein particles. Several of these proteins have been identified previously in EVs isolated from different sources, but they have not previ- ously been detected with mass spectrometry using plasma EVs isolated by other methods. Again, the results presented here support the feasibility of the combined use of flotation and SEC to isolate highly pure EVs from blood. However, the need for these highly purified blood-derived EVs might depend on the downstream analyses. As the lipoprotein par- ticles with low density only contained low or no amounts of RNA, the requirement of purification might be less if it is the RNA that will be analysed. Furthermore, if a focused analysis such as ELISA or the SOMAscan® assay is used, the problem of contaminating protein is probably smaller; however, for mass spectrometry analysis, for example, this approach is crucial as well as for functional studies. References Feingold KR, Grunfeld C (2015) Introduction to lipids and lipo- proteins. In: De Groot LJ, Chrousos G, Dungan K, Feingold KR, Grossman A, Hershman JM, Koch C, Korbonits M, McLachlan R, 1 3 2886 N. Karimi et al. New M et al (eds) Endotext [Internet]. MDText.com, Inc, South Dartmouth (MA) 27. Executive Committee of the German Medical Association on the Recommendation of the Scientific Advisory Board (2016) Cross-sectional guidelines for therapy with blood components and plasma derivatives, Chap 5 Human Albumin—Revised. Transfus Med Hemother 43(3):223–232 12. Yuana Y, Levels J, Grootemaat A et al. (2014) Co-isolation of extracellular vesicles and high-density lipoproteins using den- sity gradient ultracentrifugation. J Extracell Vesicles. https://doi. org/10.3402/jev.v3.23262 28. Baranyai T, Herczeg K, Onodi Z et al (2015) Isolation of exosomes from blood plasma: qualitative and quantitative com- parison of ultracentrifugation and size exclusion chromatography methods. PLoS One 10(12):e0145686 g j 13. Sodar BW, Kittel A, Paloczi K et al (2016) Low-density lipopro- tein mimics blood plasma-derived exosomes and microvesicles during isolation and detection. Sci Rep 6:24316 g p 14. Welton JL, Webber JP, Botos LA et al (2015) Ready-made chro- matography columns for extracellular vesicle isolation from plasma. J Extracell Vesicles 4:27269 29. Kalra H, Adda CG, Liem M et al (2013) Comparative proteomics evaluation of plasma exosome isolation techniques and assess- ment of the stability of exosomes in normal human blood plasma. Proteomics 13(22):3354–3364 p 15. Boing AN, van der Pol E, Grootemaat AE et al. (2014) Single-step isolation of extracellular vesicles by size-exclusion chromatogra- phy. J Extracell Vesicles. https://doi.org/10.3402/jev.v3.23430 30. de Menezes-Neto A, Saez MJ, Lozano-Ramos I et al (2015) Size- exclusion chromatography as a stand-alone methodology identifies novel markers in mass spectrometry analyses of plasma-derived vesicles from healthy individuals. J Extracell Vesicles 4:27378i 16. Gross JC, Chaudhary V, Bartscherer K et al (2012) Active Wnt proteins are secreted on exosomes. Nat Cell Biol 14(10):1036–1045 31. Looze C, Yui D, Leung L et al (2009) Proteomic profiling of human plasma exosomes identifies PPARgamma as an exo- some-associated protein. Biochem Biophys Res Commun 378(3):433–438 ( ) 17. Kim DK, Lee J, Kim SR et al (2015) EVpedia: a community web portal for extracellular vesicles research. Bioinformatics 31(6):933–939 32. Bastos-Amador P, Royo F, Gonzalez E et al (2012) Proteomic analysis of microvesicles from plasma of healthy donors reveals high individual variability. J Proteom 75(12):3574–3584 18. References van Niel G, Bergam P, Di Cicco A et al (2015) Apolipoprotein E regulates amyloid formation within endosomes of pigment cells. Cell Rep 13(1):43–51 p 19. Thery C, Ostrowski M, Segura E (2009) Membrane vesicles as conveyors of immune responses. Nat Rev Immunol 9(8):581–593 33. Welton JL, Brennan P, Gurney M et al (2016) Proteomics analy- sis of vesicles isolated from plasma and urine of prostate cancer patients using a multiplex, aptamer-based protein array. J Extracell Vesicles 5:31209 20. Marzesco AM, Janich P, Wilsch-Brauninger M et al (2005) Release of extracellular membrane particles carrying the stem cell marker prominin-1 (CD133) from neural progenitors and other epithelial cells. J Cell Sci 118(Pt 13):2849–2858 34. Lässer C, Alikhani VS, Ekström K et al (2011) Human saliva, plasma and breast milk exosomes contain RNA: uptake by mac- rophages. J Transl Med 9:9 p 21. Didiot MC, Hall LM, Coles AH et al (2016) Exosome-mediated delivery of hydrophobically modified siRNA for huntingtin mRNA silencing. Mol Ther J Am Soc Gene Ther 24(10):1836–1847 Ö 35. Lässer C, Shelke GV, Yeri A et al (2017) Two distinct extracellular RNA signatures released by a single cell type identified by micro- array and next-generation sequencing. RNA Biol 14(1):58–72 22. Hagel L, Östberg M, Andersson T (1996) Apparent pore size distributions of chromatography media. J Chromatogr A 743(1):33–42 36. Vickers KC, Palmisano BT, Shoucri BM et al (2011) MicroRNAs are transported in plasma and delivered to recipient cells by high- density lipoproteins. Nat Cell Biol 13(4):423–433 23. van der Pol E, Coumans FA, Grootemaat AE et al (2014) Particle size distribution of exosomes and microvesicles determined by transmission electron microscopy, flow cytometry, nanoparticle tracking analysis, and resistive pulse sensing. J Thromb Haemost 12(7):1182–1192 37. Willms E, Johansson HJ, Mager I et al (2016) Cells release sub- populations of exosomes with distinct molecular and biological properties. Sci Rep 6:22519 38. Riva S, Barrai I, Cavalli-Sforza L et al (1969) Dependence of the buoyant density of single-stranded DNA on base composition. J Mol Biol 45(2):367–374 24. Raposo G, Stoorvogel W (2013) Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol 200(4):373–383 25. Colombo M, Raposo G, Thery C (2014) Biogenesis, secretion, and intercellular interactions of exosomes and other extracellular vesicles. Annu Rev Cell Dev Biol 30:255–289 39. Fischer H, Polikarpov I, Craievich AF (2004) Average protein density is a molecular-weight-dependent function. Protein Sci Publ Protein Soc 13(10):2825–2828 26. References Kalra H, Drummen GP, Mathivanan S (2016) Focus on extracel- lular vesicles: introducing the next small big thing. Int J Mol Sci 17(2):170 40. O’Driscoll L, Stoorvogel W, Thery C et al (2017) European net- work on microvesicles and exosomes in health and disease (ME- HaD). Eur J Pharm Sci Off J Eur Fed Pharm Sci 98:1–3 1 3
https://openalex.org/W2467222603	https://ueaeprints.uea.ac.uk/id/eprint/59833/1/Published_manuscript.pdf	English	null	Connective Tissue Degeneration: Mechanisms of Palmar Fascia Degeneration (Dupuytren’s Disease)	Current molecular biology reports	2,016	cc-by	6,609	Curr Mol Bio Rep DOI 10.1007/s40610-016-0045-3 Curr Mol Bio Rep DOI 10.1007/s40610-016-0045-3 ROGENITOR CELLS BIOLOGY AND REGENERATION (I KALAJZIC, SECTION EDITOR) Connective Tissue Degeneration: Mechanisms of Palmar Fascia Degeneration (Dupuytren’s Disease) S. Karkampouna1,9 & M. Kreulen2 & M. C. Obdeijn3 & P. Kloen4 & A. L. Dorjée5 & F. Rivellese5,6 & A. Chojnowski7 & I. Clark8 & Marianna Kruithof-de Julio1,9 # The Author(s) 2016. This article is published with open access at Springerlink.com Abstract Dupuytren’s disease is a connective tissue disorder of the hand causing excessive palmar fascial fibrosis with associated finger contracture and disability. The aetiology of the disease is heterogeneous, with both genetic and environ- mental components. The connective tissue is abnormally infiltrated by myofibroblasts that deposit collagen and other extracellular matrix proteins. We describe the clinical profile of Dupuytren’s disease along with current therapeutic schemes. Recent findings on molecular and cellular parameters that are dysregulated in Dupuytren’s disease, which may contribute to the onset of the disease, and the role of resident inflammation promoting fibrosis, are highlighted. We review recent literature focusing on non-myofibroblast cell types (stem cell-like cells), their pro-inflammatory and pro-fibrotic role that may account for abnormal wound healing response. Keywords Inflammation . Dupuytren’s . Fibrosis . Regeneration . Contracture Keywords Inflammation . Dupuytren’s . Fibrosis . Regeneration . Contracture This article is part of the Topical Collection on Stem/Progenitor Cells Biology and Regeneration * Marianna Kruithof-de Julio marianna.kruithofdejulio@dkf.unibe.ch S. Karkampouna sofia.karkampouna@dkf.unibe.ch M. Kreulen kreulen@mac.com M. C. Obdeijn m.c.obdeijn@amc.uva.nl P. Kloen p.kloen@amc.uva.nl A. L. Dorjée A.L.Dorjee@lumc.nl F. Rivellese rivelles@gmail.com A. Chojnowski adrian.chojnowski@nnuh.nhs.uk I. Connective Tissue Degeneration: Mechanisms of Palmar Fascia Degeneration (Dupuytren’s Disease) Clark I.Clark@uea.ac.uk This article is part of the Topical Collection on Stem/Progenitor Cells Biology and Regeneration 1 Department of Urology, Leiden University Medical Center, Albinusdreef 2, Leiden, ZA 2333, The Netherlands 2 Department of Plastic Surgery, Rode Kruis Ziekenhuis, Vondellaan 13, Beverwijk 1942, LE, The Netherlands 3 Department of Plastic Reconstructive and Hand Surgery, Academic Medical Center, Meibergdreef 9, Amsterdam 1100, DD, The Netherlands 4 Department of Orthopedic Surgery, Academic Medical Center, Meibergdreef 9, Amsterdam 1100, DD, The Netherlands 5 Department of Rheumatology, Leiden University Medical Center, Albinusdreef 2, Leiden 2333, ZA, The Netherlands 6 Centre for Experimental Medicine and Rheumatology, William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK 7 Institute of Orthopaedics, Norfolk and Norwich University Hospital, Norwich, UK 8 Biomedical Research Centre, School of Biological Sciences, University of East Anglia, Norwich, UK 9 Urology Research Laboratory, Department of Urology and Department of Clinical Research, University of Bern, Murtenstrasse 35, Bern 3008, Switzerland 1 Department of Urology, Leiden University Medical Center, Albinusdreef 2, Leiden, ZA 2333, The Netherlands 1 Department of Urology, Leiden University Medical Center, Albinusdreef 2, Leiden, ZA 2333, The Netherlands * Marianna Kruithof-de Julio marianna.kruithofdejulio@dkf.unibe.ch 2 Department of Plastic Surgery, Rode Kruis Ziekenhuis, Vondellaan 13, Beverwijk 1942, LE, The Netherlands S. Karkampouna sofia.karkampouna@dkf.unibe.ch 3 Department of Plastic Reconstructive and Hand Surgery, Academic Medical Center, Meibergdreef 9, Amsterdam 1100, DD, The Netherlands 3 Department of Plastic Reconstructive and Hand Surgery, Academic Medical Center, Meibergdreef 9, Amsterdam 1100, DD, The Netherlands M. Kreulen kreulen@mac.com M. C. Obdeijn m.c.obdeijn@amc.uva.nl P. Kloen p.kloen@amc.uva.nl A. L. Dorjée A.L.Dorjee@lumc.nl F. Rivellese rivelles@gmail.com A. Chojnowski adrian.chojnowski@nnuh.nhs.uk I. Clark I.Clark@uea.ac.uk 4 Department of Orthopedic Surgery, Academic Medical Center, Meibergdreef 9, Amsterdam 1100, DD, The Netherlands 4 Department of Orthopedic Surgery, Academic Medical Center, Meibergdreef 9, Amsterdam 1100, DD, The Netherlands 5 Department of Rheumatology, Leiden University Medical Center, Albinusdreef 2, Leiden 2333, ZA, The Netherlands 5 Department of Rheumatology, Leiden University Medical Center, Albinusdreef 2, Leiden 2333, ZA, The Netherlands 6 Centre for Experimental Medicine and Rheumatology, William Harvey Research Institute, Barts and The London School of Medicine and Dentistry, Queen Mary University of London, London, UK Minimally Invasive (Percutaneous) Treatment Minimally invasive (or percutaneous) techniques have be- come more popular as can be performed in the outpatient setting. Principally, they involve either collagenase injections or needle fasciotomy. The goal is to rupture palpable cords causing digital flexion contracture. Local triamcinolone acetonide injections do not release DD contractures. In the case of Dupuytren’s disease (DD), although it is not clear whether its pathogenesis is of mechanical or biochemical nature, the net result is the same: excess production of matrix proteins and excessive accumulation of extracellular matrix (scarring) which changes tissue architecture and causes digital contraction. Perhaps we should re-evaluate DD not only as excess scarring but also as a condition of abnormal tissue regeneration. Collagenase is an enzyme solution (derived from Clostridium histolyticum), which is injected directly into the DD cord. The cord will weaken due to enzymatic digestion and rupture when manipulated over the next few days. This technique (FDA-ap- proved [8]) tends to be effective for MCP contractures, though with certain complications, and is considered a promising alter- native for less severe contractures. It is shown to be of particular value in MCP contractures [9]. However, recurrences are report- ed and complications of oedema, tendon rupture, pain and lymphadenopathy are described [5]. In this review, we will discuss the recent research findings on DD focusing on the role of non-fibroblastic cell populations (immune cells, vascular cells or potential stem/progenitor cells). 7 Institute of Orthopaedics, Norfolk and Norwich University Hospital, Norwich, UK 7 Institute of Orthopaedics, Norfolk and Norwich University Hospital, Norwich, UK 8 Biomedical Research Centre, School of Biological Sciences, University of East Anglia, Norwich, UK 9 Urology Research Laboratory, Department of Urology and Department of Clinical Research, University of Bern, Murtenstrasse 35, Bern 3008, Switzerland Curr Mol Bio Rep Introduction joints might lead to a positive Btabletop test^ (the hand cannot lie flat on a tabletop), but few patients experience functional limitations. The indications for treatment are governed by func- tional loss and progression and may be subject to local healthcare system guidelines. An extension deficit greater than 30° may lead to contracture of the accessory collateral ligaments and the palmar plate of the PIP joint. PIP joint contractures are generally regarded as more difficult to treat than MCP contrac- tures because of the secondary joint contracture and weakening of the extensor mechanism caused by palmar fibrosis. Wound repair and tissue regeneration after injury is widely accepted to occur during the adult life of large mammals. In humans, liver regeneration after partial hepatectomy [1] or gum regeneration [2] is a complete and restorative process. However, skin wounds, incisions or excisions lead to forma- tion of scar tissue, which does not resolve for a long term, thus, the cell-mediated tissue regeneration is incomplete while scar tissue persists. Embryonic skin wounds lead to scar-free, completely regenerated tissues, while the majority of mechan- ical injuries during adult life lead to scar tissue formation [3]. Thus, there might exist a Bco-dependent^ link between tissue regeneration, cell replenishment and scarring. The early phases of the wound healing response are dependent on in- flammation and fibrogenesis, recruitment of platelets, immune cell and fibroblast invasion, pro-inflammatory cytokine secre- tion, differentiation of fibroblasts to myofibroblasts and fibrin clot formation. If the damaging stimuli are repetitive, this will lead to persistent inflammation; higher levels of interleukins, tumor necrosis factor alpha (TNFα) and pro-fibrogenic transforming growth factor beta (TGFβ) and therefore scar- ring. It has been proposed that the same signals that regulate scar-free embryonic regeneration also regulate the adult wound healing response. These cellular processes might be controlled by the levels and/or localization of those same sig- nals as well as of the (extra)cellular context, developmental stage, tissue specificity and repetitive versus acute injury. Awide range of treatment options are available [6] generally involving mechanical release or excision of excessive fibrotic tissue. However, recurrence rates are high, ranging from 8 to 66 % (average 33 %) [7]. The surgeon and patient should be aware that there is no curative treatment for the disease. Non-operative Treatment Non-operative treatments such as physiotherapy, splinting and local radiotherapy may affect disease progression, but long- term efficacy is unclear. Clinical Problems of DD and Current Therapeutic Possibilities Dupuytren’s disease typically presents in the fourth and fifth decade of life with thickening and nodule formation of the affected fascial structures in the hand (Fig. 1a) [4]. Percutaneous needle fasciotomy has been popularized in recent decades. Cutting the cord with a needle is effective for solitary central cords in the palm of the hand to release MCP contractures. More distal, the effectiveness of the release de- creases and the risk of iatrogenic neurovascular damage in- creases. Also, high recurrence rates up to 60 % within 3 years are reported [10]. These less invasive techniques may be a useful tool in the sicker patients with co-morbidities who cannot undergo surgery or in those who want immediate improvement. Disease progression leads to longitudinally oriented cord- like structures that limit extension of the involved fingers and ultimately to metacarpophalangeal (MCP) and interphalangeal joint (PIP or DIP) contractures [5]. Younger patients often have a more aggressive disease progression. Most patients with a first presentation of DD do not have pain or functional disability and require no treatment. A small extension deficit of the MCP Curr Mol Bio Rep Fig. 1 a Clinical presentation of Dupuytren’s disease; preoperative rigid contracture, surgical incision during palmar fasciectomy with prevalent collagen cord, resected nodule and cord specimen. b Immune cell types (leukocytes, monocytes, B and T cells) residing in nodules from DD patient material (FACS analysis, N = 3). c Immunofluorescence of CD3, alpha smooth muscle actin (αSMA), tryptase and CD68 expression in Dupuytren’s nodules. DAPI (nuclei). d Ex vivo culture of Dupuytren’s nodules and treatments with mast cell stabilizer chromolyn. Immunofluorescence for αSMA (myofibroblasts) and tryptase expression (mast cells). e Ex vivo culture of Dupuytren’s nodules and treatment with anti-TNFa antibody (golimumab) and control IgG. Immunofluorescence for αSMA (myofibroblasts) and CD68 expression (macrophages) Inflammation and Fibrosis A recent genome-wide analysis study using exon arrays indicated a variety of genes that are differentially expressed in DD patient fibroblasts compared with control (thigh skin punch form unaffected individuals) fibroblasts [28]. Among the list of genes are ECM and tissue re-modelling genes sug- gesting aberrant matrix synthesis and turnover. Top hits are the matrix metalloproteinase-1 (MMP-1), MMP-3 and MMP-16, which have decreased expression in DD fibroblasts, though other studies have implicated MMP-2 and MMP-14 [29]. MMP-1, MMP-14 and to some extend MMP-2 are collagen- degrading enzymes, whilst MMP-3 and MMP-16 activate such enzymes. ADAM15, ADAMTS10, ADAMTS2 and ADAMTS3 showed increased expression in DD patient sam- ples. ADAMTS2 and ADAMTS3 are procollagen propeptidases, involved in collagen biosynthesis. More than 20 collagen genes showed upregulated expression in DD es- pecially those of the COL1, COL3, COL4 and COL5 cluster. Regulation of cell surface proteins involved in interaction with ECM, such as the integrin family, is disrupted, showing either significantly higher (ITGA11) or lower expression (ITGA2, ITGA6 and ITGA4). Members of the TGFβ and WNT pathway are also dysregulated e.g. follistatin, BMP4, inhibin subunit INHBA, WNT2, frizzled 4, and RSPO3. A previous genome-wide association study has shown several genes of the WNT signalling pathway to be dysregulated (WNT2, WNT4, WNT7B, RSPO2, SFRP4, SULF1) [27]. Further characterization has indicated decreased WNT2 and increased b-catenin and WNT7B in the DD nodules along with increased ACTA2 (α-smooth muscle actin, a myofibroblast marker) and COL1A1 and COL3A1 expression [30••]. A recent genome-wide analysis study using exon arrays indicated a variety of genes that are differentially expressed in DD patient fibroblasts compared with control (thigh skin punch form unaffected individuals) fibroblasts [28]. Among the list of genes are ECM and tissue re-modelling genes sug- gesting aberrant matrix synthesis and turnover. Top hits are the matrix metalloproteinase-1 (MMP-1), MMP-3 and MMP-16, which have decreased expression in DD fibroblasts, though other studies have implicated MMP-2 and MMP-14 [29]. MMP-1, MMP-14 and to some extend MMP-2 are collagen- degrading enzymes, whilst MMP-3 and MMP-16 activate such enzymes. ADAM15, ADAMTS10, ADAMTS2 and ADAMTS3 showed increased expression in DD patient sam- ples. ADAMTS2 and ADAMTS3 are procollagen propeptidases, involved in collagen biosynthesis. More than 20 collagen genes showed upregulated expression in DD es- pecially those of the COL1, COL3, COL4 and COL5 cluster. The majority of knowledge on aspects of the wound healing response has derived by studies of acute skin injuries. Inflammation and Fibrosis Resident and inflammatory cells (e.g. mast cells, leukocytes) release growth factors, proteases and prostaglandins that are essential for removal of damaged epithelial cells, protection from infec- tious factors and activation of fibroblasts into myofibroblasts that form the fibrous scar tissue and have mechanical properties to mediate wound closure. Although acute tissue injury is re- solved completely, repetitive chronic injury, the addition of other factors such as age or diabetes or chronic inflammation have the potential to interfere with the correct remodelling of tissue and are contributing factors to persistent scarring (e.g. hypertrophic scars) [36, 37]. For instance, macrophage and leukocyte-depleted transgenic mice (PU.1 null) have rapid skin wound repair with reduced fibrosis [38]. The mechanism by which inflammation influences fibrosis remains elusive. Similarly to skin wound repair, connective tissue diseases such as DD or Peyronie’s disease are likely to be affected by inflam- mation. Despite the first study that reported macrophages and leukocytes around the DD nodules [39], the role of immune cells in DD fibrosis had not been characterized until recently. A limitation that accounts for this was the overlapping recognition of fibroblasts by anti-macrophage antibodies (Mac-3, CD68, MHC class II, CD45), which may have led to misidentification of pure fibroblast and macrophage cell populations [40]. p y Regulation of cell surface proteins involved in interaction with ECM, such as the integrin family, is disrupted, showing either significantly higher (ITGA11) or lower expression (ITGA2, ITGA6 and ITGA4). Members of the TGFβ and WNT pathway are also dysregulated e.g. follistatin, BMP4, inhibin subunit INHBA, WNT2, frizzled 4, and RSPO3. A previous genome-wide association study has shown several genes of the WNT signalling pathway to be dysregulated (WNT2, WNT4, WNT7B, RSPO2, SFRP4, SULF1) [27]. Further characterization has indicated decreased WNT2 and increased b-catenin and WNT7B in the DD nodules along with increased ACTA2 (α-smooth muscle actin, a myofibroblast marker) and COL1A1 and COL3A1 expression [30••]. High levels of inflammatory cytokines were detected in tis- sue from DD patients along with CD68+ monocytes and clas- sically activated M1 (pro-inflammatory) and alternatively acti- vated M2 (regenerative) macrophages [41]. The same study showed that pro-inflammatory factor TNFα promotes DD fi- brosis, although IL-6 and IL-1b did not have the same effect, via activation of WNT signalling [41]. Bianchi et al. have also reported increased expression of IL-6 and IL-1b cytokines in DD as well as presence of CD68-positive cells [42]. Surgical Techniques is considered the gold standard of surgical treatment for functionally disabling DD [11]. Surgical excision of diseased tissue is called fasciectomy and can vary from limited to radical excision. For skin incisions, there are multiple options ranging from midline Y-V advancement flaps, Z-plasties and zigzag (Bruner- type) incisions [11]. Radical fasciectomy has high complica- tion rate without a significantly decreased recurrence rate. Dermofasciectomy involves removal of the diseased fascia including the overlying skin. This technique, if used radically, may reduce long-term recurrence rates. Smaller or Bfirebreak^ skin grafts probably do not improve recurrence rates over fasciectomy alone. Partial fasciectomy After removal of the diseased tissue (fasciectomy), addi- tional procedures might be necessary to correct capsular joint contractures (PIP joint capsulolysis) or to reconstruct a skin defect (local skinflap or graft). Postoperative splinting of the hand may not improve medium-term outcome [5]. Complications of open surgery include delayed wound healing, skin flap necrosis, digital nerve and vessel injury, joint stiffness, hematoma and pain issues [12]. Complex regional pain syndrome (CRPS) can be a devastating compli- cation of surgery prolonging recovery and requiring long-term hand therapy support and chronic pain treatments [13]. Curr Mol Bio Rep Molecular and Cellular Alterations in DD Degeneration microRNAs (miRNAs) are implicated in many biological processes and have been associated with several diseases; however, few studies have investigated the potential role of miRNAs in DD fibrosis. The first microarray studies identi- fied unique profiles of miRNAs in control versus DD fascia palmaris [33, 34]. Mosakhani et al. reported that some miRNAs (e.g. miR29-C, miR29-130b, miR29-101) are pre- dicted to regulate both the WNT and TGFβ pathways [34]. RNA sequencing analysis of DD and the control fascia mate- rial shows a unique enrichment of over 70 miRNAs in the DD and a distinct, smaller subset enriched in control fascia [35]. Target prediction analysis indicated that anti-fibrotic miRNAs targeting collagen mRNAs, which are present in the normal fascia, are in fact depleted in DD patient material [35]. Thus, these studies add to our understanding of DD pathogenesis. The main characteristic of DD is accumulation of extracellular matrix proteins, which form an abnormal connective tissue of the palmar fascia mainly containing collagen (mostly collagen type I and type III) and myofibroblasts [14]. Several studies have implicated the TGFβ and WNT pathways as drivers of fibrosis in DD (reviewed in [15–17]). Expression analyses in DD (by e.g. microarray or q-RT-PCR) [18–27] have shown that these key pathways are indeed modulated, but do not always yield the same results: either different set of hits or different mode of expression (downregulation versus upregu- lation). Considering the heterogeneity among individuals and additional factors including biopsy material or comparison with carpal tunnel-derived fascia palmaris or Bnormal^ adja- cent tissue from DD patients, as well as the derivation of tissue fibroblasts, there is high variability in the outcomes. Contribution of Stem/Progenitor Cells in DD Mast cells are granulated tissue-resident cells of the innate immunity, best known for their involvement in al- lergic disorders, but also playing a role in several autoimmune conditions [45], where they can have both pro- and anti-in- flammatory/immunomodulatory effects [46, 47]. In the con- text of Dupuytren’s inflammation, mast cell activation could contribute to the modulation of the inflammatory response leading to fibrosis. For example, tryptase-initiated signalling influences neutrophil and monocyte recruitment, muscle tis- sue regeneration [48], fibroblast proliferation [49, 50] and activation of latent TGFβ [51]. Interestingly, the number of mast cells has been shown to be increased in Dupuytren’s contracture in comparison with normal fascia tissue [52]. Inhibition of mediator release by the mast cell stabilizer com- pound chromolyn (chromoglicic acid) is used as anti-allergy treatment for asthma, conjunctivitis and food allergies [53]. As a proof of principle, we tested the effect of inhibition of mast cells ex vivo and their potential influence upon fibro- blasts on our ex vivo human tissue culture method [54••, 55]. We have exposed DD nodule-derived slices to chromolyn for 72 h; our preliminary data suggest a decrease in tryptase and alpha smooth muscle actin (αSMA) expression (Fig. 1d). Co- labelling of CD68 and αSMA in DD specimens indicates the presence of macrophages around the microvessel clusters (Fig. 1e, left panel), while ex vivo treatment of tissue with neutralizing TNFα antibody (golimumab) has little effect at the lowest concentration (1 μg/ml) but leads to reduced CD68 expression when used at a higher concentration (10 μg/ml). p g Cell replacement treatments using MSCs for organ fibrosis, such as liver [62], lung [63] or heart [64], have been attempted in many studies in order to improve cell replenishment and tissue regeneration. MSCs are multipotent stromal cells which differentiate into distinct cell lineages: osteoblasts, chondrocytes, adipose cells, muscle cells as well as tenocytes, skin cells and differentiated stromal cells of connective tissue (fibroblast phenotype) [65]. However, the ability of MSCs to give rise to fibroblasts is often overlooked, along with the subsequent effects on exacerbating instead of ameliorating fibrosis. MSCs may differentiate into highly specified fibro- blastic populations such as inflammatory fibroblasts or mye- loid fibroblasts. In DD, palmar fascia tissue recent studies have reported and characterized the presence of resident MSCs or adipose stem cells [66••, 67••], or a stem cell-like subpopulation of Thy1 (CD90)-positive cells has been identi- fied [68••]. Contribution of Stem/Progenitor Cells in DD fibrogenic factor TGFβ, released by platelets, fibroblasts and macrophages, also mediates inflammation-related signalling (such as p38 activation) also in DD fibroblasts [45, 44]. The field of DD research is mainly focused on the mecha- nisms of deregulated proliferation of myofibroblasts and their matrix-producing properties. However, this may be at the end stage of the disease and not necessarily during its onset. The switch to excessive fibrosis may be indeed controlled by other cell types such as the infiltrating immune cells, vascular smooth cells, endothelial cells, pericytes [57], fibrocytes orig- inating from the bone marrow or possibly multilineage pro- genitors that give rise to myofibroblasts [58] (mesenchymal stem/stromal cells). Pericytes, the endothelium-covering cells, have been attributed with stem cell properties in several or- gans [59], which is likely to be the case also in DD fibrosis. Vessel structures that contain high levels of laminins, a key basal membrane component of the connective tissue, facilitate proliferation of myofibroblasts evident by proliferating cen- tres in the vicinity of these vessels [60••]. Endothelial cell and mesenchymal stem cell (MSC)-enriched protein CD105 (type III receptor, endoglin) has been found to be expressed near these proliferation centres. Vessel structures in DD specimens appear abnormally large or with fused vessels and are located distinctly from the myofibroblast-enriched area. It is highly likely that these structures are sweat glands from subcutane- ous dermis that seem to be encapsulated within the fibrotic nodules as also suggested by Viil et al. The presence of stem cells has been reported in the vicinity of cutaneous sweat glands [61]; thus, we hypothesize that similar stem/progenitor cell populations may exist in DD. Given the high proliferative properties of these myofibroblasts, such an assumption seems probable; however, detailed investigations are needed. Our unpublished observations demonstrate the presence of a small number of immune cell types (leukocytes, monocytes, B and T cells) residing in nodules from DD patient material (Fig. 1b). Immunofluorescence of fixed material immediately after surgical removal of nodules confirmed the presence of CD3-positive T cells along with CD68-positive macrophages (Fig. 1c) similar to the study of Verjee et al. In addition, we detected tryptase, a well-known specific alpha marker of mast cells (Fig. 1c). Inflammation and Fibrosis The pro- Novel players have been implicated in the pathogenesis of DD such as Wilm’s tumour protein-1 (WT1) [31•] and YAP [32•]. YAP1, a member of the Hippo pathway, which has re- cently been shown to promote differentiation of fibroblasts into myofibroblasts, potentially acts downstream of TGFβ [32•]. Curr Mol Bio Rep Conclusions 2. Park YJ, Cha S, Park YS. Regenerative applications using tooth derived stem cells in other than tooth regeneration: a literature re- view. Stem Cells Int. 2016;2016:9305986. In this review, we have discussed the clinical problems associ- ated with DD and the current therapeutic tools currently avail- able. In addition, we have reviewed recent publications on stud- ies regarding novel regulator genes or cellular processes in- volved in DD pathogenesis. We have discussed how fibrosis and tissue regeneration are inter-dependent processes, for in- stance deregulated tissue regeneration or Bre-cellularization^ leads to overstimulation of fibrosis. One of the underlying fac- tors regulating fibrosis and regeneration is inflammation, while the role of inflammation in DD is only beginning to be unravelled. The latest research studies introduce the possibility that DD may be an autoimmune disease, which however re- quires more investigation. Given the high demand for safe and effective anti-fibrotic drugs for DD, we propose that novel com- pounds that inhibit both myofibroblasts and immune cells (mast, B and T cells) or a combination of anti-fibrotic and anti-inflammatory compounds are promising candidates. From a different point of view, we discuss the possibility of resident stem/progenitor cells to be a pool for myofibroblasts or inhibit their fibrogenic activity based on recent publications. The potential of MSC differentiation to the fibroblast lineage or stromal precursor cells has not yet been characterized in DD. 3. Ferguson MW, O’Kane S. Scar-free healing: from embryonic mechanisms to adult therapeutic intervention. Philos Trans R Soc Lond B Biol Sci. 2004;359(1445):839–50. 4. Rayan GM. Dupuytren disease: anatomy, pathology, presentation, and treatment. J Bone Joint Surg Am. 2007;89(1):189–98. 5. Eaton C. Evidence-based medicine: Dupuytren contracture. Plast Reconstr Surg. 2014;133(5):1241–51. 5. Eaton C. Evidence-based medicine: Dupuytren contracture. Plast Reconstr Surg. 2014;133(5):1241–51. 6. Desai SS, Hentz VR. The treatment of Dupuytren disease. J Hand Surg Am. 2011;36(5):936–42. 6. Desai SS, Hentz VR. The treatment of Dupuytren disease. J Hand Surg Am. 2011;36(5):936–42. 7. Van Giffen N, Degreef I, De Smet L. Dupuytren’s disease: outcome of the proximal interphalangeal joint in isolated fifth ray involve- ment. Acta Orthop Belg. 2006;72(6):671–7. 8. Gilpin D, Coleman S, Hall S, Houston A, Karrasch J, Jones N. Injectable collagenase Clostridium histolyticum: a new nonsurgical treatment for Dupuytren’s disease. J Hand Surg Am. 2010;35(12): 2027–38.e1. 9. Hurst LC, Badalamente MA, Hentz VR, Hotchkiss RN, Kaplan FTD, Meals RA, et al. Injectable collagenase Clostridium histolyticum for Dupuytren’s contracture. References cells are of the mononuclear cell lineages sharing properties with fibroblasts and mesenschymal stromal cells (CD45RO, 25F9 and MRP8/14). Treament of fibrocytes with serum am- yloid P and FDA-approved Xiapex collagenase inhibited the expansion of fibrocytes in vitro, a promising finding regarding the extended use of Xiapex [70••]. cells are of the mononuclear cell lineages sharing properties with fibroblasts and mesenschymal stromal cells (CD45RO, 25F9 and MRP8/14). Treament of fibrocytes with serum am- yloid P and FDA-approved Xiapex collagenase inhibited the expansion of fibrocytes in vitro, a promising finding regarding the extended use of Xiapex [70••]. Papers of particular interest, published recently, have been highlighted as: • Of importance • Of importance •• Of major importance •• Of major importance 1. Michalopoulos GK. Liver regeneration. J Cell Physiol. 2007;213(2):286–300. Conclusions N Engl J Med. 2009;361(10):968–79. 10. Watt AJ, Curtin CM, Hentz VR. Collagenase injection as nonsur- gical treatment of Dupuytren’s disease: 8-year follow-up. J Hand Surg Am. 2010;35(4):534–9, 9 e1. 11. Rodrigues JN, Becker GW, Ball C, Zhang W, Giele H, Hobby J, et al. Surgery for Dupuytren’s contracture of the fingers. Cochrane Database Syst Rev. 2015;12:CD010145. 12. Mavrogenis AF, Spyridonos SG, Ignatiadis IA, Antonopoulos D, Papagelopoulos PJ. Partial fasciectomy for Dupuytren’s contrac- tures. J Surg Orthop Adv. 2009;18(2):106–10. Contribution of Stem/Progenitor Cells in DD Cell-cell contact of DD myofibroblasts with adi- pose stem cells resulted in inhibition of contractility and smooth muscle actin expression of myofibroblasts [69••]. A recent study has identified an increased resident and circulat- ing fibrocyte population in DD tissue compared to control tissue [70••]. Fibrocyte characterization showed that these A recent study has characterized the immune response in a large subset of DD specimens and, similar to our unpublished observations, has detected the infiltration of DD tissue by sev- eral immune cells. After extensive characterization of Tcells and cytokine profiling, the authors suggest that T cells may contrib- ute to the development of DD possibly via (auto)antigen-driven processes possibly due to microvascular damage [56••]. Overall, the fibrotic response in DD is being recognized as an immune-mediated response, with an important involve- ment of different immune cells. Our preliminary observations suggest that mast cells could also be involved in the inflam- matory response leading to the development of fibrosis, an intriguing observation warranting additional investigations on the role of mast cells and other immune cells in DD. Curr Mol Bio Rep Compliance with Ethical Standards 13. Bulstrode NW, Jemec B, Smith PJ. The complications of Dupuytren’s contracture surgery. J Hand Surg Am. 2005;30(5):1021–5. Conflict of Interest Sofia Karkampouna, Michiel Kreulen, Miryam Obdeijn, Peter Kloen, Annemarie Dorjée, Felice Rivellese, Adrian Chojnowski, Ian Clark and Marianna Krutihof-de Julio declare that they have no conflict of interest. 14. Bazin S, Le Lous M, Duance VC, Sims TJ, Bailey AJ, Gabbiani G, et al. Biochemistry and histology of the connective tissue of Dupuytren’s disease lesions. Eur J Clin Invest. 1980;10(1):9–16. 15. Tomasek JJ, Gabbiani G, Hinz B, Chaponnier C, Brown RA. Myofibroblasts and mechano-regulation of connective tissue re- modelling. Nat Rev Mol Cell Biol. 2002;3(5):349–63. Human and Animal Rights and Informed Consent This article con- tains studies on human subjects; Dupuytren’s connective tissue nodules were collected during partial fasciectomy procedure. Confirmation that the Medical Research Involving Human Subjects Act (WMO) does not apply to the present study was obtained by the local ethics committee (reference number W12_245 # 12.17.0279) since the research was per- formed on Bwaste^ material. 16. Bowley E, O’Gorman DB, Gan BS. β-catenin signaling in fibroproliferative disease. J Surg Res. 2007;138(1):141–50. 17. Nunn AC, Schreuder FB. Dupuytren’s contracture: emerging in- sight into a Viking disease. Hand Surg. 2014;19(3):481–90. 18. Shih B, Watson S, Bayat A. Whole genome and global expression profiling of Dupuytren’s disease: systematic review of current find- ings and future perspectives. Ann Rheum Dis. 2012;71(9):1440–7. 19. Qian A, Meals RA, Rajfer J, Gonzalez-Cadavid NF. Comparison of gene expression profiles between Peyronie’s disease and Dupuytren’s contracture. Urology. 2004;64(2):399–404. 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 appro- priate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. 20. Rehman S, Salway F, Stanley JK, Ollier WER, Day P, Bayat A. Molecular phenotypic descriptors of Dupuytren’s disease defined using Informatics analysis of the transcriptome. J Hand Surg Am. 2008;33(3):359–72. Curr Mol Bio Rep 38. Martin P, D’Souza D, Martin J, Grose R, Cooper L, Maki R, et al. Wound healing in the PU.1 null mouse—tissue repair is not depen- dent on inflammatory cells. Curr Biol. 2003;13(13):1122–8. 21. Zhang AY, Fong KD, Pham H, Nacamuli RP, Longaker MT, Chang J. Gene expression analysis of Dupuytren’s disease: the role of TGF- β2. J Hand Surg Eur Vol. 2008;33(6):783–90. 22. Johnston P, Chojnowski AJ, Davidson RK, Riley GP, Donell ST, Clark IM. A complete expression profile of matrix-degrading me- talloproteinases in Dupuytren’s disease. J Hand Surg Am. 2007;32(3):345–51. 39. Andrew JG, Andrew SM, Ash A, Turner B. An investigation into the role of inflammatory cells in Dupuytren’s disease. J Hand Surg Br. 1991;16(3):267–71. 40. Inoue T, Plieth D, Venkov CD, Xu C, Neilson EG. Antibodies against macrophages that overlap in specificity with fibroblasts. Kidney Int. 2005;67(6):2488–93. 23. Lee LC, Zhang AY, Chong AK, Pham H, Longaker MT, Chang J. Expression of a novel gene, MafB, in Dupuytren’s disease. J Hand Surg Am. 2006;31(2):211–8. 41. Verjee LS, Verhoekx JSN, Chan JKK, Krausgruber T, Nicolaidou V, Izadi D, et al. Unraveling the signaling pathways promoting fibrosis in Dupuytren’s disease reveals TNF as a therapeutic target. Proc Natl Acad Sci U S A. 2013;110(10):E928–37. 24. Shih B, Wijeratne D, Armstrong DJ, Lindau T, Day P, Bayat A. Identification of biomarkers in Dupuytren’s disease by comparative analysis of fibroblasts versus tissue biopsies in disease-specific phe- notypes. J Hand Surg Am. 2009;34(1):124–36. 42. Bianchi E, Taurone S, Bardella L, Signore A, Pompili E, Sessa V, et al. Involvement of pro-inflammatory cytokines and growth factors in the pathogenesis of Dupuytren’s contracture: a novel target for a possible future therapeutic strategy? Clin Sci. 2015;129(8):711–20. 25. Pan D, Watson HK, Swigart C, Thomson JG, Honig SC, Narayan D. Microarray gene analysis and expression profiles of Dupuytren’s contracture. Ann Plast Surg. 2003;50(6):618–22. 26. Furniss D, Dolmans GH, Hennies HC. Genome-wide association scan of Dupuytren’s disease. J Hand Surg Am. 2011;36(4):755–6. author reply 6. 43. Bujak M, Ratkaj I, Markova-Car E, Jurišić D, Horvatić A, Vučinić S, et al. Inflammatory gene expression upon TGF-β1-induced p38 acti- vation in primary Dupuytren’s disease fibroblasts. Front Mol Biosci. 2015;2:68. 27. Dolmans GH, Werker PM, Hennies HC, Furniss D, Festen EA, Franke L, et al. Wnt signaling and Dupuytren’s disease. N Engl J Med. 2011;365(4):307–17. 44. Krause C, Kloen P, ten Dijke P. Elevated transforming growth factor β and mitogen-activated protein kinase pathways mediate fibrotic traits of Dupuytren’s disease fibroblasts. Fibrogenesis Tissue Repair. 2011;4(1):1–15. 28. Forrester HB, Temple-Smith P, Ham S, de Kretser D, Southwick G, Sprung CN. Genome-wide analysis using exon arrays demonstrates an important role for expression of extracellular matrix, fibrotic control and tissue remodelling genes in Dupuytren’s disease. PLoS One. 2013;8(3):e59056. 45. Suurmond J, van der Velden D, Kuiper J, Bot I, Toes RE. Mast cells in rheumatic disease. Eur J Pharmacol. 2015;778:116–24. 46. Suurmond J, Rivellese F, Dorjee AL, Bakker AM, Rombouts YJ, Rispens T, et al. Toll-like receptor triggering augments activation of human mast cells by anti-citrullinated protein antibodies. Ann Rheum Dis. 2015;74(10):1915–23. 29. Wilkinson JM, Davidson RK, Swingler TE, Jones ER, Corps AN, Johnston P, et al. MMP-14 and MMP-2 are key metalloproteases in Dupuytren’s disease fibroblast-mediated contraction. Biochim Biophys Acta. 2012;1822(6):897–905. 47. Rivellese F, Suurmond J, Habets K, Dorjee AL, Ramamoorthi N, Townsend MJ, et al. Ability of Interleukin-33- and immune complex- triggered activation of human mast cells to down-regulate monocyte- mediated immune responses. Arthritis Rheum. 2015;67(9):2345–53. 30.•• van Beuge MM, ten Dam E-JPM, Werker PMN, Bank RA. Wnt pathway in Dupuytren disease: connecting profibrotic signals. Transl Res. 2015;166(6):762–71.e3. First study demonstrating a role for noncanonical Wnt signaling pathway in DD. 48. Duchesne E, Bouchard P, Roussel MP, Cote CH. Mast cells can regulate skeletal muscle cell proliferation by multiple mechanisms. Muscle Nerve. 2013;48(3):403–14. 31.• Crawford J, Raykha C, Charles D, Gan BS, O’Gorman DB. WT1 expression is increased in primary fibroblasts derived from Dupuytren’s disease tissues. J Cell Commun Signal. 2015;9(4): 347–52. Identification of WT1 as important factor in DD as shown by its increased expression in DD fibroblasts. 49. Frungieri MB, Albrecht M, Raemsch R, Mayerhofer A. The action of the mast cell product tryptase on cyclooxygenase-2 (COX2) and subsequent fibroblast proliferation involves activation of the extra- cellular signal-regulated kinase isoforms 1 and 2 (erk1/2). Cell Signal. 2005;17(4):525–33. 32.• Piersma B, de Rond S, Werker PMN, Boo S, Hinz B, van Beuge MM, et al. YAP1 Is a driver of myofibroblast differentiation in normal and diseased fibroblasts. Am J Pathol. 2015;185(12): 3326–37. First study demonstrating an involvement of YAP sig- nalling pathway in DD. 50. Abe M, Kurosawa M, Ishikawa O, Miyachi Y, Kido H. Mast cell tryptase stimulates both human dermal fibroblast proliferation and type I collagen production. Clin Exp Allergy. 1998;28(12):1509–17. 33. Satish L, LaFramboise WA, Johnson S, Vi L, Njarlangattil A, Raykha C, et al. Fibroblasts from phenotypically normal palmar fascia exhibit molecular profiles highly similar to fibroblasts from active disease in Dupuytren’s Contracture. BMC Med Genomics. 2012;5(1):1–14. 51. Tatler AL, Porte J, Knox A, Jenkins G, Pang L. Tryptase activates TGFβ in human airway smooth muscle cells via direct proteolysis. Biochem Biophys Res Commun. 2008;370(2):239–42. 52. Schubert TE, Weidler C, Borisch N, Schubert C, Hofstadter F, Straub RH. Dupuytren’s contracture is associated with sprouting of substance P positive nerve fibres and infiltration by mast cells. Ann Rheum Dis. 2006;65(3):414–5. 34. Mosakhani N, Guled M, Lahti L, Borze I, Forsman M, Paakkonen V, et al. Unique microRNA profile in Dupuytren’s contracture supports deregulation of β-catenin pathway. Mod Pathol. 2010;23(11):1544–52. 35. Riester SM, Arsoy D, Camilleri ET, Dudakovic A, Paradise CR, Evans JM, et al. RNA sequencing reveals a depletion of collagen targeting microRNAs in Dupuytren’s disease. BMC Med Genomics. 2015;8(1):1–14. Extensive analysis of microRNA se- quencing in DD and implications of differentially regulated collagen-targeting microRNAs in the pathogenesis of DD. 53. Borriello F, Granata F, Varricchi G, Genovese A, Triggiani M, Marone G. Immunopharmacological modulation of mast cells. Curr Opin Pharmacol. 2014;17:45–57. 54.•• Karkampouna S, Kruithof BP, Kloen P, Obdeijn MC, van der Laan AM, Tanke HJ, et al. Novel ex vivo culture method for the study of Dupuytren’s disease: effects of TGFβ type 1 receptor modulation by antisense oligonucleotides. Mol Ther Nucleic Acids. 2014;3: e142. Ex vivo maintenance of DD tissue and pharmacological treatment with small molecule inhibitor and antisense oligonu- cleotides targeting TGFβ signalling. 36. Wong VW, Paterno J, Sorkin M, Glotzbach JP, Levi K, Januszyk M, et al. Mechanical force prolongs acute inflammation via T-cell- dependent pathways during scar formation. FASEB J. 2011;25(12): 4498–510. 37. Eming SA, Martin P, Tomic-Canic M. Wound repair and regenera- tion: mechanisms, signaling, and translation. Sci Transl Med. 2014;6(265):265sr6–sr6. 55. Karkampouna S, Kloen P, Obdeijn MC, Riester SM, van Wijnen AJ, Kruithof-de JM. Human Dupuytren’s ex vivo culture for the 55. Karkampouna S, Kloen P, Obdeijn MC, Riester SM, van Wijnen AJ, Kruithof-de JM. Human Dupuytren’s ex vivo culture for the Curr Mol Bio Rep study of myofibroblasts and extracellular matrix interactions. J Vis Exp. 2015;98:e52534. 65. Wang Y, Chen X, Cao W, Shi Y. Plasticity of mesenchymal stem cells in immunomodulation: pathological and therapeutic implica- tions. Nat Immunol. 2014;15(11):1009–16. 56.•• Mayerl C, Del Frari B, Parson W, Boeck G, Piza-Katzer H, Wick G, et al. Characterisation of the inflammatory response in Dupuytren’s disease. J Plast Surg Hand Surg. 2016;50(3):171–9. This study thoroughly identified the presence of different immune cell types in blood and tissue of DD patients. 66.•• Hindocha S, Iqbal SA, Farhatullah S, Paus R, Bayat A. Characterization of stem cells in Dupuytren’s disease. Br J Surg. 2011;98(2):308–15. First study showing expression of adipose and skin stem cell markers in DD tissue. 67.•• Iqbal SA, Manning C, Syed F, Kolluru V, Hayton M, Watson S, et al. Identification of mesenchymal stem cells in perinodular fat and skin in Dupuytren’s disease: a potential source of myofibroblasts with impli- cations for pathogenesis and therapy. Stem Cells Dev. 2012;21(4):609– 22. Identification of mesenchymal stem cells in primary DD cul- tures that retain their trilineage differentiation capacity in vitro. 57. Crisan M, Yap S, Casteilla L, Chen C-W, Corselli M, Park TS, et al. A perivascular origin for mesenchymal stem cells in multiple hu- man organs. Cell Stem Cell. 2008;3(3):301–13. 58. Prockop DJ. Inflammation, fibrosis, and modulation of the process by mesenchymal stem/stromal cells. Matrix Biol. 2016;51:7–13. 59. Birbrair A, Zhang T, Wang Z-M, Messi Maria L, Mintz A, Delbono O. Pericytes at the intersection between tissue regeneration and pathology. Clin Sci. 2015;128(2):81–93. 68.•• Ratkaj I, Bujak M, Jurišić D, Baus Lončar M, Bendelja K, Pavelić K, et al. Microarray analysis of Dupuytren’s disease cells: the profibrogenic role of the TGF-β inducible p38 MAPK pathway. Cell Physiol Biochem. 2012;30(4):927–42. Microarray analysis indicated p38 MAPK pathway and CD90-stem cell markers as pathogenic factors in DD. 60.•• Viil J, Maasalu K, Mäemets-Allas K, Tamming L, Lõhmussaar K, Tooming M, et al. Laminin-rich blood vessels display activated growth factor signaling and act as the proliferation centers in Dupuytren’s contracture. Arthritis Res Ther. 2015;17(1):1–10. Highlights the importance of vasculature as a niche of prolifer- ating myofibroblasts. 69.•• Verhoekx JS, Mudera V, Walbeehm ET, Hovius SE. Adipose- derived stem cells inhibit the contractile myofibroblast in Dupuytren’s disease. Plast Reconstr Surg. 2013;132(5):1139–48. This study suggested that adipose stem cells may deliver antifibrotic signals upon myofibroblasts instead of providing a source of (myo)fibroblasts. 61. Gao Y, Li M, Zhang X, Bai T, Chi G, Liu JY, et al. Isolation, culture and phenotypic characterization of human sweat gland epithelial cells. Int J Mol Med. 2014;34(4):997–1003. 62. Fiore EJ, Mazzolini G, Aquino JB. Mesenchymal stem/ stromal cells in liver fibrosis: recent findings, old/new caveats and future perspectives. Stem Cell Rev. 2015;11(4):586–97. Á 70.•• Iqbal SA, Hayton MJ, Watson JS, Szczypa P, Bayat A. First iden- tification of resident and circulating fibrocytes in Dupuytren’s dis- ease shown to be inhibited by serum amyloid P and Xiapex. PLoS ONE. 2014;9(6):e99967. Bone marrow derived monocytes dif- ferentiate into fibrocytes which are found for the first time in DD tissues; the differentiation of fibrocytes or their profibrotic properties are prone to inhibition when exposed to high levels of SAP and Xiapex collagenase treatment. 63. Álvarez D, Levine M, Rojas M. Regenerative medicine in the treat- ment of idiopathic pulmonary fibrosis: current position. Stem Cells Cloning. 2015;8:61–5. 64. Trial J, Entman ML, Cieslik KA. Mesenchymal stem cell-derived inflammatory fibroblasts mediate interstitial fibrosis in the aging heart. J Mol Cell Cardiol. 2016;91:28–34.
https://openalex.org/W2807308836	https://escholarship.org/content/qt1t3546rm/qt1t3546rm.pdf?t=qaf4pg	English	null	Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10	Virus evolution	2,018	cc-by	4,220	UCLA UCLA Previously Published Works Title Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10 Permalink https://escholarship.org/uc/item/1t3546rm Journal Virus Evolution, 4(1) ISSN 2057-1577 Authors Suchard, Marc A Lemey, Philippe Baele, Guy et al. Publication Date 2018 DOI 10.1093/ve/vey016 Peer reviewed UCLA UCLA Previously Published Works Title Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10 Permalink https://escholarship.org/uc/item/1t3546rm Journal Virus Evolution, 4(1) ISSN 2057-1577 Authors Suchard, Marc A Lemey, Philippe Baele, Guy et al. Publication Date 2018 DOI 10.1093/ve/vey016 Peer reviewed UCLA UCLA Previously Published Works Title Bayesian phylogenetic and phylodynamic data integ Permalink https://escholarship.org/uc/item/1t3546rm Journal Virus Evolution, 4(1) ISSN 2057-1577 Authors Suchard, Marc A Lemey, Philippe Baele, Guy et al. Publication Date 2018 DOI 10.1093/ve/vey016 Peer reviewed UCLA UCLA Previously Published Works Title Bayesian phylogenetic and phylodynamic data integration using Permalink https://escholarship.org/uc/item/1t3546rm Journal Virus Evolution, 4(1) ISSN 2057-1577 Authors Suchard, Marc A Lemey, Philippe Baele, Guy et al. Publication Date 2018 DOI 10.1093/ve/vey016 Peer reviewed UCLA UCLA Previously Published W Title Bayesian phylogenetic and phylodynamic Permalink https://escholarship.org/uc/item/1t3546rm Journal Virus Evolution, 4(1) ISSN 2057-1577 Authors Suchard, Marc A Lemey, Philippe Baele, Guy et al. Publication Date 2018 DOI 10.1093/ve/vey016 Peer reviewed Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10 Marc A. Suchard,1,2,3,,† Philippe Lemey,4,‡ Guy Baele,4,§ Daniel L Alexei J. Drummond,6,7, and Andrew Rambaut8,,* Marc A. Suchard,1,2,3,,† Philippe Lemey,4,‡ Guy Baele,4,§ Daniel L. Ayres,5 Alexei J. Drummond,6,7, and Andrew Rambaut8,,* 1Department of Biomathematics, David Geffen School of Medicine, University of California, Los Angeles, 621 Charles E. Young Dr., South, Los Angeles, CA, 90095 USA, 2Department of Biostatistics, Fielding School of Public Health, University of California, Los Angeles, 650 Charles E, Young Dr., South, Los Angeles, CA, 90095 USA, 3Department of Human Genetics, David Geffen School of Medicine, University of California, Los Angeles 695 Charles E. Young Dr., South, Los Angeles, CA, 90095 USA, 4Department of Microbiology and Immunology, Rega Institute, KU Leuven, Herestraat 49, 3000 Leuven, Belgium, 5Center for Bioinformatics and Computational Biology, University of Maryland, College Park, 125 Biomolecular Science Bldg #296, College Park, MD 20742 USA, , 6Department of Computer Science, University of Auckland, 303/38 Princes St., Auckland, 1010 NZ, 7Centre for Computational Evolution, University of Auckland, 303/38 Princes St., Auckland, 1010 NZ and 8Institute of Evolutionary Biology, University of Edinburgh, Ashworth Laboratories, Edinburgh, EH9 3FL UK -mail: msuchard@ucla.edu (M.A.S.); alexei@cs.auckland.ac.nz (A.J.D.); a.rambaut@ed.ac.uk (A.R.) Corresponding author: E-mail: msuchard@ucla.edu (M.A.S.); alexei@cs.auckland.ac.nz (A.J.D.); a.rambaut@ed.ac.uk (A.R.) †http://orcid.org/0000-0001-9818-479X ‡http://orcid.org/0000-0003-2826-5353 §http://orcid.org/0000-0002-1915-7732 http://orcid.org/0000-0003-4337-3707 ‡http://orcid.org/0000-0003-2826-5353 §http://orcid.org/0000-0002-1915-7732 http://orcid.org/0000-0003-4337-3707 *http://orcid.org/0000-0003-4337-3707 Abstract The Bayesian Evolutionary Analysis by Sampling Trees (BEAST) software package has become a primary tool for Bayesian phylogenetic and phylodynamic inference from genetic sequence data. BEAST uniﬁes molecular phylogenetic reconstruc- tion with complex discrete and continuous trait evolution, divergence-time dating, and coalescent demographic models in an efﬁcient statistical inference engine using Markov chain Monte Carlo integration. A convenient, cross-platform, graphical user interface allows the ﬂexible construction of complex evolutionary analyses. Key words: phylogenetics; phylodynamics; Bayesian inference; Markov chain Monte Carlo. ancient DNA, and the phylodynamics and molecular epidemiol- ogy of infectious disease (Drummond et al. 2012). BEAST’s spe- cific focus on time-scaled trees, and the evolutionary analyses dependent on them, has given it a unique place in the toolbox of molecular evolution and phylogenetic researchers. Since in- ception, a strong motivation for BEAST development has been Powered by the California Digital Library University of California eScholarship.org Virus Evolution, 2018, 4(1): vey016 doi: 10.1093/ve/vey016 Resources doi: 10.1093/ve/vey016 Resources V C The Author(s) 2018. Published by Oxford University Press. 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 reuse, distribution, and reproduction in any medium, provided the original work is properly cited. V C The Author(s) 2018. Published by Oxford University Press. 3. Flexible model design BEAST’s companion graphical user interface program, BEAUti, allows the user to import data, select models, choose prior dis- tributions, and specify the settings for both Bayesian inference and marginal likelihood estimation. Our efforts on BEAUti 1.10 have focused on allowing the user to easily link or unlink substi- tution, clock and tree models across multiple partitions as well as linking individual parameters to provide considerable adapt- ability in model design. Additionally, BEAUti can also group var- ious parameters in a hierarchical phylogenetic model prior (Suchard et al. 2003), which allows parameters to take different values but be linked by a common distribution, the parameters of which can then be inferred. For example, flexible codon model parameterizations, using hierarchical phylogenetic mod- els (Baele et al. 2016b) and incorporating a range of potential predictive variables for substitution behaviour (Bielejec et al. 2016a), provide insight into the tempo and mode of pathogen evolution. Multivariate continuous traits are incorporated using phylo- genetic Brownian diffusion processes, modelling the shared an- cestral dependence across taxa and the correlations between these variables. Such continuous models have most frequently been applied to diffusion on a geographical landscape with the traits representing coordinates and the phylogeny reconstruct- ing the epidemiological process within the host population (Lemey et al. 2010). The landscapes can also represent other spaces, and integration of antibody binding assay data have ex- tended ‘antigenic cartography’ (Smith et al. 2004) approaches to model simultaneous antigenic and genetic evolution and infer the viral trajectories in the immunological space generated by the host population (Bedford et al. 2014). Marginal likelihood estimation to compare models using Bayes factors has become common practice in Bayesian phylo- genetic inference. BEAST 1.10 now features marginal likelihood estimation (Baele et al. 2012), using path sampling (Gelman and Meng 1998; Lartillot and Philippe 2006) and stepping-stone sam- pling (Xie et al. 2011), as well as the recently developed general- ized stepping-stone sampling (Fan et al. 2011; Baele et al. 2016a) that offers increased accuracy and improved numerical stability by employing the concept of ‘working distributions’, i.e. distri- butions with known normalizing constants and parameterized using samples from the posterior distribution. Standard Brownian diffusion processes that assume a zero- mean displacement along each branch may however be unreal- istic for many evolutionary problems (including geographical reconstruction). 2 \| Virus Evolution, 2018, Vol. 4, No. 1 2 \| Virus Evolution, 2018, Vol. 4, No. 1 of) continuous, binary and discrete traits (Cybis et al. 2015), as demonstrated by applications to flower morphology, antibiotic resistance, and viral epitope evolution. To infer correlations be- tween high-dimensional traits computationally efficiently, a novel phylogenetic factor analysis approach assumes that a small unknown number of independent evolutionary factors evolve along the phylogeny and generate clusters of dependent traits at the tips (Tolkoff et al. 2018). the rapid growth of pathogen genome sequencing as part of public health responses to infectious diseases (Grenfell et al. 2004). In particular, fast evolving viruses can now be tracked in near real-time (see, e.g. Quick et al. 2016) to understand their ep- idemiology and evolutionary dynamics. In BEAST version 1.10, we have introduced a series of advan- ces with a particular focus on delivering accurate and informa- tive insights for infectious disease research through the integration of diverse data sources, including phenotypic and epidemiological information, with molecular evolutionary models. These advances fall into three broad themes—the inte- gration of diverse sources of extrinsic information as covariates of evolutionary processes, the increased flexibility and modula- rization of the model design process with robust and accurate model testing methods, and substantial improvements on the speed and efficiency of the statistical inference. Further extending the data integration approach, BEAST 1.10 includes a flexible framework for incorporating time-varying covariates of the effective population size over time. This uses Gaussian Markov random fields to reconstruct smoothed effec- tive population size trajectories while simultaneously estimat- ing to what extent predictor variables (e.g. fluctuations in climatic factors, host mobility, or vector density) may have driven the dynamics (Gill et al. 2016). Using a similar general- ized linear modeling (GLM) approach, classical epidemiological time-series data such as case counts (Gill et al. 2016) can be inte- grated with pathogen genome sequence data to provide joint in- ference of important epidemiological parameters. 2. Data integration Many traits in phylogenetics are represented as or partitioned into a finite number of discrete values, with geographical loca- tion standing out as a popular example. Because BEAST is dedi- cated to sampling time-scaled phylogenies, new developments of discrete character mapping enable the reconstruction of timed viral dispersal patterns while accommodating phyloge- netic uncertainty. By extending the discrete diffusion models to incorporate empirical data as covariates or predictors of transi- tion rates, BEAST can simultaneously test and quantify a range of potential predictive variables of the diffusion process (Lemey et al. 2014). Further, realizations of the trait transition process can also be efficiently produced, to pinpoint the nature and tim- ing of changes in evolutionary history beyond ancestral node state reconstruction (termed Markov jumps), or to infer the time spent in a particular state (Markov rewards) (Minin and Suchard 2008). For molecular data, fast stochastic mapping approaches are also employed to obtain site-specific dN=dS estimates, inte- grating over the posterior distribution of phylogenies and an- cestral reconstructions to quantify uncertainty on these measures of the selective forces on individual codons (Lemey et al. 2012). Finally, recent host-transmission models allow the integra- tion of complete or partial knowledge of a pathogen’s transmis- sion history, enabling the simultaneous inference of within- host population dynamics, viral evolutionary processes, and transmission times and bottlenecks (Vrancken et al. 2014). Likewise, other priors enable the reconstruction of transmission trees of infectious disease epidemics and outbreaks, while ac- commodating phylogenetic uncertainty and employ a newly designed set of phylogenetic tree proposals that respect node partitions (Hall et al. 2015). 1. Introduction First released over 14 years ago, the Bayesian Evolutionary Analysis by Sampling Trees (BEAST) software package has be- come firmly established in a broad diversity of biological fields from phylogenetics and paleontology, population dynamics, ( ) y y 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 reuse, distribution, and reproduction in any medium, provided the original work is properly cited. 1 1 3. Flexible model design A recently developed relaxed directional ran- dom walk allows the diffusion processes to take on different di- rectional trends in different parts of the phylogeny while preserving model identifiability (Gill et al. 2017) and opens up these processes for a wide range of applications. BEAST 1.10 also extends multivariate phylogenetic diffusion to latent liabil- ity model formulations in order to assess correlations between traits of different data types, including (various combinations M. A. Suchard et al. \| 3 Figure 1. Phylodynamic analysis of the 2013–2016 West African Ebola virus epidemic, encompassing simultaneous estimation of sequence and discrete (geographic) trait data with a GLM ﬁtted to the discrete trait model in order to establish potential predictors of viral transition between locations. Plotted are a snapshot of geo- graphic spread using SpreaD3 (Bielejec et al. 2016b), the maximum clade credibility tree, the posterior estimates of the GLM coefﬁcients for seven possible predictors for Ebola virus spread (Bayes Factor support values of 3, 20, and 150 are indicated by vertical lines) and the effective population size through time, estimated by incorpo- rating case counts. Figure 1. Phylodynamic analysis of the 2013–2016 West African Ebola virus epidemic, encompassing simultaneous estimation of sequence and discrete (geographic) trait data with a GLM ﬁtted to the discrete trait model in order to establish potential predictors of viral transition between locations. Plotted are a snapshot of geo- graphic spread using SpreaD3 (Bielejec et al. 2016b), the maximum clade credibility tree, the posterior estimates of the GLM coefﬁcients for seven possible predictors for Ebola virus spread (Bayes Factor support values of 3, 20, and 150 are indicated by vertical lines) and the effective population size through time, estimated by incorpo- rating case counts. 4. Performance and efficiency advances, BEAST 1.10 can yield a sizeable increase in effectively independent posterior samples per unit-time over previous software versions. For the example data described below, we see a 5- to 25-fold improvement depending on the model pa- rameter, using an NVIDIA Titan V. Increasing model complexity and sequence availability in modern-day analyses have stretched the computational demands of Bayesian phylogenetic inference. To improve effi- ciency for large-scale sequence data, BEAST 1.10 uses the BEAGLE library (Ayres et al. 2012) that provides access to mas- sive parallelization on a range of computing architectures. In particular, the combination of BEAST 1.10 with BEAGLE 3.0 (Ayres et al., under review) allows multiple data partitions to be parallelized across a single high-performance device (i.e. a GPGPU graphics board) allowing for the utilization of the full ca- pacity of these devices, reducing the computational overheads. As the complexity of phylogenetic model designs increase, con- comitant with the surge in scale of genomic data, updating only a parameter associated with a single data partition limits the occupation of the massively multicore devices. To address this we have developed an adaptive multivariate transition kernel that simultaneously updates parameters across all the parti- tioned data, making more efficient use of available hardware (Baele et al. 2017). Through a combination of these two 5.1 Availability BEAST 1.10 is open source under the GNU lesser general public license and available at https://beast-dev.github.io/beast-mcmc for cross-platform compiled programs and https://github.com/ beast-dev/beast-mcmc for software development and source code. It requires Java version 1.6 or greater. Documentation, tutorials, and help are available at http://beast.community and many users actively discuss BEAST usage and development in the ‘beast-users’ GoogleGroup discussion group (http://groups. google.com/group/beast-users). We also host an expanding suite of R tools—designed for posterior analyses using BEAST (https://github.com/beast-dev/RBeast). & , Rambaut, A., and Suchard, M. A. (2017) ‘Adaptive MCMC in Bayesian Phylogenetics: An Application to Analyzing Partitioned Data in BEAST’, Bioinformatics, 33: 1798–805. & , and Suchard, M. A. (2016a) ‘Genealogical Working Distributions for Bayesian Model Testing with Phylogenetic Uncertainty’, Systematic Biology, 65: 250–64. , Suchard, M. A., Bielejec, F., and Lemey, P. (2016b) ‘Bayesian Codon Substitution Modeling to Identify Sources of Pathogen Evolutionary Rate Variation’, Microbial Genomics, 2: e00005. Bedford, T., Suchard, M. A., Lemey, P., Dudas, G., Gregory, V., Hay, A. J., McCauley, J. W., Russell, C. A., Smith, D. J., and Rambaut, A. (2014) ‘Integrating Inﬂuenza Antigenic Dynamics with Molecular Evolution’, eLife, 3: e01914. 4.1 Example Figure 1 presents a spatiotemporal reconstruction of Ebola virus evolution and spread during the 2013–2016 West African epi- demic, highlighting several aspects of phylodynamic data inte- gration. The estimates are based on a large data set of 1,610 genomes that represent over 5 per cent of the known cases (Dudas et al. 2017). Administrative regions (n ¼ 56) are included as discrete sampling locations to estimate viral dispersal through time while testing the contribution of a set of potential covariates to the pattern of spread using a GLM parameteriza- tion of phylogeographic diffusion (Lemey et al. 2014). This indi- cates, for example, the importance of population sizes and geographic distance to explain viral dispersal intensities. 4 \| Virus Evolution, 2018, Vol. 4, No. 1 5. Relationship to BEAST2 and other software References A range of other software focusing on phylodynamic analy- ses of fast-evolving pathogens has been described since the last version of BEAST was published. Of particular note are LSD (To et al. 2016), TreeDater (Volz and Frost 2017), and TreeTime (Sagulenko et al. 2018). These programs use least-squares algo- rithms (LSD) or maximum likelihood inference (TreeDater, TreeTime) and provide rapid analysis on large data sets for a subset of the models that BEAST provides. However, the former program implements very limited phylodynamic models and the latter two programs require a phylogenetic tree, inferred us- ing other software, as input data, conditioning parameter esti- mates on this single tree. Ayres, D. L., Cummings M. P., et al. ‘Under review. BEAGLE 3.0: Improved Usability for a High-Performance Computing Library for Statistical Phylogenetics’, Systematic Biology [WorldCat] , Darling, A., Zwickl, D. J., Beerli, P., Holder, M. T., Lewis, P. O., Huelsenbeck, J. P., Ronquist, F., Swofford, D. L., Cummings, , Darling, A., Zwickl, D. J., Beerli, P., Holder, M. T., Lewis, P. O., Huelsenbeck, J. P., Ronquist, F., Swofford, D. L., Cummings, M. P., Rambaut, A., and Suchard, M. A. (2012) ‘BEAGLE: An Application Programming Interface and High-Performance Computing Library for Statistical Phylogenetics’, Systematic Biology, 61: 170–3. Baele, G., Lemey, P., Bedford, T., Rambaut, A., Suchard, M. A., and Alekseyenko, A. V. (2012) ‘Improving the Accuracy of Demographic and Molecular Clock Model Comparison While Accommodating Phylogenetic Uncertainty’, Molecular Biology and Evolution, 29: 2157–67. 5. Relationship to BEAST2 and other software KU Leuven, OT/14/115), and the Research Foundation— Flanders (‘Fonds voor Wetenschappelijk Onderzoek— Vlaanderen’, G066215N, G0D5117N and G0B9317N). GB acknowledges support from the Interne Fondsen KU Leuven/Internal Funds KU Leuven. DLA is supported by NSF grant DBI 1661443. We gratefully acknowledge support from NVIDIA Corporation with the donation of parallel comput- ing resources used for this research. Distinct from BEAST 1.10 described here, BEAST2 is an indepen- dent project (Bouckaert et al. 2014) intended as a platform that more readily facilitates the development of packages of models and analyses by other researchers. Although both projects share many of the same models and the underlying inference frame- work, BEAST has increasingly focused on the analysis of rapidly evolving pathogens and their evolution and epidemiology. We affirm that BEAST will continue to be developed in parallel to the BEAST2. While these projects share a recent common origin, each now aims to foster complementary research domains. Acknowledgements (2016) ‘Real-Time, Portable Genome Sequencing for Ebola Surveillance’, Nature, 530: 228–32. G., Sabeti, P. C., Sall, A., Stro¨her, U., Wurie, I., Suchard, M. A., Lemey, P., and Rambaut, A. (2017) ‘Virus Genomes Reveal Factors That Spread and Sustained the Ebola Epidemic’, Nature, 544: 309–15. Sagulenko, P., Puller, V., and Neher, R. A. (2018) ‘Treetime: Maximum-Likelihood Phylodynamic Analysis’, Virus Evolution, 4: vex042. Fan, Y., Wu, R., Chen, M. H., Kuo, L., and Lewis, P. O. (2011) ‘Choosing among Partition Models in Bayesian Phylogenetics’, Molecular Biology and Evolution, 28: 523–32. Smith, D. J., Lapedes, A. S., de Jong, J. C., Bestebroer, T. M., Rimmelzwaan, G. F., Osterhaus, A. D. M. E., and Fouchier, R. A. Smith, D. J., Lapedes, A. S., de Jong, J. C., Bestebroer, T. M., Rimmelzwaan, G. F., Osterhaus, A. D. M. E., and Fouchier, R. A. M. (2004) ‘Mapping the Antigenic and Genetic Evolution of Inﬂuenza Virus’, Science, 305: 371–6. Gelman, A., and Meng, X.-L. (1998) ‘Simulating Normalizing Constants: From Importance Sampling to Bridge Sampling to Path Sampling’, Statistical Science, 13: 163–85. M. (2004) ‘Mapping the Antigenic and Genetic Evolution of Inﬂuenza Virus’, Science, 305: 371–6. Suchard, M. A., Kitchen, C. M. R., Sinsheimer, J. S., and Weiss, R. E. (2003) ‘Hierarchical Phylogenetic Models for Analyzing Multipartite Sequence Data’, Systematic Biology, 52: 649–64. Gill, M. S., Ho, T., Si, L., Baele, G., Lemey, P., and Suchard, M. A. (2017) ‘A Relaxed Directional Random Walk Model for Phylogenetic Trait Evolution’, Systematic Biology, 66: 299–319. , Lemey, P., Bennett, S. N., Biek, R., and Suchard, M. A. (2016) ‘Understanding past Population Dynamics: Bayesian Coalescent-Based Modeling with Covariates’, Systematic Biology, 65: 1041–56. To, T.-H., Jung, M., Lycett, S., and Gascuel, O. (2016) ‘Fast Dating Using Least-Squares Criteria and Algorithms’, Systematic Biology, 65: 82–97. Tolkoff, M. R., Alfaro, M. E., Baele, G., Lemey, P., and Suchard, M. A., 2018. ‘Phylogenetic Factor Analysis’, Systematic Biology, 67: 384–99. Grenfell, B. T., Pybus, O. G., Gog, J. R., Wood, J. L. N., Daly, J. M., Mumford, J. A., and Holmes, E. C. (2004) ‘Unifying the Epidemiological and Evolutionary Dynamics of Pathogens’, Science, 303: 327–32. Volz, E., and Frost, S. (2017) ‘Scalable Relaxed Clock Phylogenetic Dating’, Virus Evolution, 3: vex025. Volz, E., and Frost, S. (2017) Scalable Relaxed Clock Phylogenetic Dating’, Virus Evolution, 3: vex025. Vrancken, B., Rambaut, A., Suchard, M. A., Drummond, A., Baele, Hall, M., Woolhouse, M., and Rambaut, A. Acknowledgements M., Gire, S., Gladden-Young, A., Gnirke, A., Goba, A., Grant, D. S., Haagmans, B. L., Hiscox, J. A., Jah, U., Kargbo, B., Kugelman, J. R., Liu, D., Lu, J., Malboeuf, C. M., Mate, S., Matthews, D. A., Matranga, C. B., Meredith, L. W., Qu, J., Quick, J., Pas, S. D., Phan, M. V. T., Pollakis, G., Reusken, C. B., Sanchez-Lockhart, M., Schaffner, S. F., Schieffelin, J. S., Sealfon, R. S., Simon- Loriere, E., Smits, S. L., Stoecker, K., Thorne, L., Tobin, E. A., Vandi, M. A., Watson, S. J., West, K., Whitmer, S., Wiley, M. R., Winnicki, S. M., Wohl, S., Wo¨lfel, R., Yozwiak, N. L., Andersen, K. G., Blyden, S. O., Bolay, F., Carroll, M. W., Dahn, B., Diallo, B., Formenty, P., Fraser, C., Gao, G. F., Garry, R. F., Goodfellow, I., Gu¨ nther, S., Happi, C. T., Holmes, E. C., Keı¨ta, S., Kellam, P., Koopmans, M. P. G., Kuhn, J. H., Loman, N. J., Magassouba, N., Naidoo, D., Nichol, S. T., Nyenswah, T., Palacios, G., Pybus, O. G., Sabeti, P. C., Sall, A., Stro¨her, U., Wurie, I., Suchard, M. A., Lemey, P., and Rambaut, A. (2017) ‘Virus Genomes Reveal Factors That Spread and Sustained the Ebola Epidemic’, Nature, 544: 309–15. Amino Acid Sites under Positive Selection’, Bioinformatics, 28: 3248–56. Amino Acid Sites under Positive Selection’, Bioinformatics, 28: 3248–56. , Rambaut, A., Bedford, T., Faria, N., Bielejec, F., Baele, G., Russell, C. A., Smith, D. J., Pybus, O. G., Brockmann, D. et al. (2014) ‘Unifying Viral Genetics and Human Transportation Data to Predict the Global Transmission Dynamics of Human Inﬂuenza H3N2’, PLoS Pathogens, 10: e1003932. , Rambaut, A., Bedford, T., Faria, N., Bielejec, F., Baele, G., Russell, C. A., Smith, D. J., Pybus, O. G., Brockmann, D. et al. Inﬂuenza H3N2’, PLoS Pathogens, 10: e1003932. & , Welch, J., and Suchard, M. (2010) ‘Phylogeography Takes a Relaxed Random Walk in Continuous Space and Time’, Molecular Biology and Evolution, 27: 1877–85. Minin, V. N., and Suchard, M. A. (2008) ‘Fast, Accurate and Simulation-Free Stochastic Mapping’, Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 363: 3985–95. Gu¨ nther, S., Happi, C. T., Holmes, E. C., Keı¨ta, S., Kellam, P., Koopmans, M. P. G., Kuhn, J. H., Loman, N. J., Magassouba, N., Naidoo, D., Nichol, S. T., Nyenswah, T., Palacios, G., Pybus, O. Quick, J., Loman, N., Duraffour, S., Simpson, J. et al. Acknowledgements We would like to thank the many developers and contribu- tors to BEAST 1.10, including: Alex Alekseyenko, Trevor Bedford, Filip Bielejec, Erik Bloomquist, Luiz Carvalho, Gabriela Cybis, Gytis Dudas, Roald Forsberg, Mandev Gill, Matthew Hall, Joseph Heled, Sebastian Hoehna, Denise Kuehnert, Wai Lok Sibon Li, Gerton Lunter, Sidney Markowitz, Vladimir Minin, Julia Palacios, Michael Defoin Platel, Oliver Pybus, Beth Shapiro, Korbinian Strimmer, Max Tolkoff, Chieh-Hsi Wu, and Walter Xie. This work was sup- ported in part by the European Union Seventh Framework Programme for research, technological development and demonstration under Grant Agreement no. 278433- PREDEMICS and no. 725422-ReservoirDOCS. The VIROGENESIS project receives funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 634650. The Artic Network receives funding from the Wellcome Trust through project 206298/Z/17/Z. MAS is partly supported by NSF grant DMS 1264153 and NIH grants R01 HG006139, R01 AI107034 and U19 AI135995. PL acknowledges support by the Special Research Fund, KU Leuven (‘Bijzonder Onderzoeksfonds’, Bielejec, F., Baele, G., Rodrigo, A. G., Suchard, M. A., and Lemey, P. (2016a) ‘Identifying Predictors of Time-Inhomogeneous Viral Evolutionary Processes’, Virus Evolution, 2: vew023. & , Vrancken, B., Suchard, M. A., Rambaut, A., and Lemey, P. (2016b) ‘SpreaD3: Interactive Visualization of Spatiotemporal History and Trait Evolutionary Processes’, Molecular Biology and Evolution, 33: 2167–9. Bouckaert, R., Heled, J., Ku¨ hnert, D., Vaughan, T., Wu, C.-H., Xie, , , , J , , , g , , , , , D., Suchard, M. A., Rambaut, A., and Drummond, A. J. (2014) ‘BEAST 2: A Software Platform for Bayesian Evolutionary Analysis’, PLoS Computational Biology, 10: e1003537. Cybis, G. B., Sinsheimer, J. S., Bedford, T., Mather, A. E., Lemey, P., and Suchard, M. A. (2015) ‘Assessing Phenotypic Correlation through the Multivariate Phylogenetic Latent Liability Model’, The Annals of Applied Statistics, 9: 969. Drummond, A. J., Suchard, M. A., Xie, D., and Rambaut, A. (2012) ‘Bayesian Phylogenetics with BEAUti and the BEAST 1.7’, Molecular Biology and Evolution, 29: 1969–73. Dudas, G., Carvalho, L. M., Bedford, T., Tatem, A. J., Baele, G., Faria, N. R., Park, D. J., Ladner, J. T., Arias, A., Asogun, D., Bielejec, F., Caddy, S. L., Cotten, M., D’Ambrozio, J., Dellicour, S., Caro, A. D., Diclaro, J. D., II, Durrafour, S., Elmore, M. J., M. A. Suchard et al. \| 5 Fakoli, L. S., III, Faye, O., Gilbert, M. L., Gevao, S. Acknowledgements (2015) ‘Epidemic Reconstruction in a Phylogenetics Framework: Transmission Trees as Partitions of the Node Set’, PLoS Computational Biology, 11: e1004613. Vrancken, B., Rambaut, A., Suchard, M. A., Drummond, A., Baele, G., Derdelinckx, I., Van Wijngaerden, E., Vandamme, A.-M., Van Laethem, K., and Lemey, P. (2014) ‘The Genealogical Population Dynamics of HIV-1 in a Large Transmission Chain: Bridging within and among Host Evolutionary Rates’, PLoS Computational Biology, 10: e1003505. Lartillot, N., and Philippe, H. (2006) ‘Computing Bayes Factors Using Thermodynamic Integration’, Systematic Biology, 55: 195–207. p gy Xie, W., Lewis, P. O., Fan, Y., Kuo, L., and Chen, M. H. (2011) ‘Improving Marginal Likelihood Estimation for Bayesian Phylogenetic Model Selection’, Systematic Biology, 60: 150–60. Lemey, P., Minin, V. N., Bielejec, F., Pond, S. L. K., and Suchard, M. A. (2012) ‘A Counting Renaissance: Combining Stochastic Mapping and Empirical Bayes to Quickly Detect
https://openalex.org/W2901458052	https://europepmc.org/articles/pmc6267161?pdf=render	English	null	Advancing sustainability science for the SDGs	Sustainability science	2,018	cc-by	4,223	SPECIAL FEATURE: EDITORIAL Integrated Knowledge Generation for Transformations towards Sustainability from Local to Global Scales Mark Stafford Smith1 · Christina Cook2 · Youba Sokona3 · Thomas Elmqvist4 · Kensuke Fukushi5 · Wendy Broadgate6 · Marcin Pawel Jarzebski5 Mark Stafford Smith1 · Christina Cook2 · Youba Sokona3 · Thomas Elmqvist4 · Kensuke Fukushi5 · Wendy Broadgate6 · Marcin Pawel Jarzebski5 Published online: 14 November 2018 © The Author(s) 2018 From the perspective of being “indivisible”, there is a growing appreciation of the interactions and dependencies among the goals, both in terms of substance and for coherent policy alignment. A key area that exemplifies the need to manage interactions relates to progress on climate change, the subject of SDG13 as well as the UN Framework Con- vention on Climate Change. The latest scientific findings on climate change show that the hoped-for plateauing of growth in greenhouse gas emissions has not yet occurred (Jackson et al. 2017), and global warming in 2017 reached 1 °C above pre-industrial conditions (WMO 2018). Research reaffirms that further warming will occur, which will cause coupled changes in all components of the climate system and amplify existing risks faced by natural and human systems (IPCC Working Group I 2013). Reversing this trend will require a major departure from business-as-usual and a change from incremental to exponential reductions in greenhouse gas emissions (e.g. Rockström et al. 2016). However, success on climate change will depend on aligning with other SDGs (e.g. SDG12 on responsible consumption and production); and success in meeting other SDGs will in turn depend on controlling climate change (e.g. hunger, SDG2). A key role of the research community is to provide knowledge to under- stand synergies and trade-offs so that challenging decisions can be made to maximise the synergies and minimise trade- offs (Griggs et al. 2014). Sustainability Science (2018) 13:1483–1487 https://doi.org/10.1007/s11625-018-0645-3 Sustainability Science (2018) 13:1483–1487 https://doi.org/10.1007/s11625-018-0645-3 Introduction In September 2015, the world’s leaders agreed to the United Nations’ Agenda 2030 for sustainable development (UN 2015), including 17 sustainable development goals (SDGs). The SDGs provide a remarkable common global vision towards a safe, just and sustainable world for all human beings to thrive on the planet. The goals are seen as ambi- tions, and challenges, for all countries of all income lev- els, without exception. Between them, they provide a more resolved view of sustainable development than simply talk- ing about economic, social and environmental dimensions; indeed, most goals are at least partially integrated across these dimensions. As such, the SDGs provide a normative framing that is “indivisible and universal” for the emerging discipline of sustainability science. Mark Stafford Smith: former Chair of Future Earth Science Committee; Christina Cook: Independent Consultant, Canada, former Future Earth Secretariat Global Hub; Youba Sokona: former Future Earth Science Committee Member. * Mark Stafford Smith Mark.Staffordsmith@csiro.au Marcin Pawel Jarzebski marcin.p.jarzebski@gmail.com 1 CSIRO Land and Water, Canberra, Australia 2 Montreal, Canada 3 South Centre, Geneva, Switzerland 4 Stockholm Resilience Centre, Stockholm University, Stockholm, Sweden 5 Integrated Research System for Sustainability Science (IR3S), The University of Tokyo Institute for Advanced Study (UTIAS), The University of Tokyo, Tokyo, Japan 6 Future Earth Secretariat Global Hub, ℅ Royal Swedish Academy of Sciences, Stockholm, Sweden Mark Stafford Smith: former Chair of Future Earth Science Committee; Christina Cook: Independent Consultant, Canada, former Future Earth Secretariat Global Hub; Youba Sokona: former Future Earth Science Committee Member. Mark Stafford Smith: former Chair of Future Earth Science Committee; Christina Cook: Independent Consultant, Canada, former Future Earth Secretariat Global Hub; Youba Sokona: former Future Earth Science Committee Member. * Mark Stafford Smith Mark.Staffordsmith@csiro.au Marcin Pawel Jarzebski marcin.p.jarzebski@gmail.com f Although the SDGs are conceived as “universal”, that is, applying to all nations, it is also challenging that nations (including various levels of government, businesses and organisations) will need to take deeply differentiated and context-specific actions to achieve the objectives of Agenda 2030. Despite the need for global outcomes, most imple- mentation will be local. For low-income countries, the main concerns are to bring national development objectives focused on all aspects of poverty alleviation together with issues of economic growth and the enhancement of human well-being. Introduction Meanwhile, higher income countries must 1 CSIRO Land and Water, Canberra, Australia 2 Montreal, Canada 5 Integrated Research System for Sustainability Science (IR3S), The University of Tokyo Institute for Advanced Study (UTIAS), The University of Tokyo, Tokyo, Japan 6 Future Earth Secretariat Global Hub, ℅ Royal Swedish Academy of Sciences, Stockholm, Sweden (0121 3456789) 3 Sustainability Science (2018) 13:1483–1487 1484 recognise that they are often ‘under-developed’ in terms of SDG targets associated with issues such as reducing waste, obesity, greenhouse gas emissions and overall resource use. In these countries, the onus will be on delivering signifi- cant reductions in the impacts of resource use, nowhere bet- ter illustrated than with regard to reducing greenhouse gas emissions. Yet in both lower and higher income countries, the temptation will be to cherry-pick goals that are easily achieved. Earth 2013), and addressing this challenge is fundamental to delivering the ‘indivisibility’ of the SDGs. Future Earth was created to address this structural challenge, with a specific mandate to integrate across disciplines and to incorporate knowledge from beyond the bounds of academia to address the pressing problems created by global environmental change. Since its inception in 2015, Future Earth has created Knowledge-Action Networks of people and organisations collaborating to build the knowledge and tools needed to tackle the greatest sustainability challenges of our time. Future Earth works to strengthen and expand these networks by (i) building communities and mobilising capacity to col- laborate on research and innovation in each network’s scope (e.g. by hosting conferences and workshops, supporting fel- lowships, and facilitating strategic collaborations); and (ii) facilitating, co-designing and synthesising research to scale solutions across sectors and geographies (e.g. by seeding projects to catalyse transformations, highlighting priorities to funders, and co-disseminating new knowledge to drive action). The SDGs thus offer an opportunity to align the impera- tive of climate change action and sustainable development goals at local, national and global scales, at the same time defining key challenges for sustainability science in its support of policy. These challenges also resonate with the agenda of Future Earth, a major initiative on research and innovation for global sustainability, which aims to develop scientific knowledge and systemic solutions towards achiev- ing both the SDGs and a healthy planet. This Special Feature thus explores some aspects of how sustainability science can support the transformations needed to achieve sustainable development. Introduction The papers are based on presentations delivered at the 7th International Conference on Sustainability Science (‘ICSS 2017’) “Global Goals—New Approaches to Knowledge Generation—Chal- lenges and Solutions from Local to Global Scales”, hosted by Future Earth, the University of Tokyo and the Stockholm Resilience Centre in August 2017. The conference brought together members of Future Earth’s emerging new Knowl- edge-Action Networks and Innovation Labs, both briefly described below. Accordingly, this editorial provides the background to these activities: the second section describes Future Earth’s Knowledge-Action Networks, the next section outlines the SDG labs, and then the following section intro- duces the remaining papers in the Special Feature and draws conclusions for both the progress of sustainability science as a discipline, and also the policy process around the SDGs. Future Earth recognises that networks alone will not over- come the more intransigent obstacles of knowledge frag- mentation that partially led to its creation. Still, the develop- ment of this particular form of network aims to mobilise an emergent organisational innovation to address global issues. Future Earth also hosts a series of global research projects, networks of more disciplinary communities which have been driving global environmental change research for three dec- ades. The newer Knowledge-Action Networks aim to lever- age the fundamental science and knowledge produced by the Global Research Projects, and work toward Future Earth’s goal to “encourage co-design and co-production (…) by researchers in collaboration with various stakeholders in governments, industry and business, international organi- sations, and civil society” (Future Earth 2013). Future Earth’s Knowledge-Action Networks (KANs) focus on themes such as health, cities and oceans. Integra- tion across the KANs and the understanding of trade-offs and synergies between SDGs are tackled in an integrated way (cf. Nilssen et al. this issue) in the Future Earth work on science for Earth targets (for a full list of the networks, see http://www.futureearth.org). Several of the papers in this issue draw on the initial experiences of or contribute to these knowledge-action networks (e.g. Harms et al. this issue). Future Earth and its Knowledge‑Action Networks The process of integrating knowledge—and approaches to creating knowledge—across different academic disciplines, epistemologies, and societal sectors generally happens in an ad hoc and incomplete manner. Indeed, in academia, efforts to integrate social sciences with natural sciences and humanities, and link the result with non-scientific knowl- edge are less rewarded or supported than detailed investi- gations within narrow disciplinary confines (e.g. Bromham et al. 2016; Haider et al. 2017; Rivera-Ferre et al. 2013). The challenge of diffuse and fragmented knowledge is perva- sive in global environmental change research (e.g. see Future SDG labs The outcomes of this initial set of 16 SDG labs were pre- sented at ICSS 2017, and covered ideas such as integrated solutions for water in Indonesia (‘Water Warriors’), bringing indigenous knowledge into decision-making on health in the Pacific Islands, flooding and water issues as well as engaging youth in Africa, and to design how to map knowledge about the SDGs in Australia. The results of the SDG Labs are pre- sented in a companion e-book (Springer, under development), where more details about the process may be found; one exam- ple is described in this Special Feature (Maher et al. 2018). Lindgren et al. (2018) discuss sustainable food systems from the health perspective. The authors discuss “opportuni- ties for and challenges to sustainable food systems from a human health perspective by making the case for avoiding the transition to unhealthy less sustainable diets (using India as an exemplar), reducing food waste by changing consumer behaviour (with examples from Japan), and using innova- tions and new technologies to reduce the environmental impact of healthy food production”. Their paper touches on “two of the challenges to achieving healthy sustainable diets for a global population, i.e. reduction on the yield and nutri- tional quality of crops (in particular vegetables and fruits) due to climate change; and trade-offs between food produc- tion and industrial crops”. The SDG Lab process was unashamedly experimen- tal within the framing of sustainability science, exploring the challenges of creating local solutions for sustainable development whilst considering how to scale towards more universal outcomes. The labs raised issues such as barri- ers to systemic change, power (im)balances, the selection of participants and partners, opportunism and timing, the importance of agents of change, and the need for ‘heartware’ to engage people; these issues are not new, but contribute to a growing set of context-sensitive case studies aimed at the SDGs. Many activities around the world are contribut- ing to such experimentation and it is important that sustain- ability science continues to synthesise and learn from these efforts. In the Future Earth community, the Lab concept is being picked up across the networks as one process to stim- ulate collaboration between researchers and innovators on context-specific but scalable solutions. For example, SDG Labs were also developed in association with the Seedbeds of Transformation Conference held in South Africa in 2018 (http://www.seedbeds.futureearth.org). SDG labs The Social Innovation Lab concept, developed by the Uni- versity of Waterloo for the Rockefeller Foundation (West- ley et al. 2015), has been used around the world to cata- lyse change. Ahead of ICSS 2017, Future Earth used this 1 3 3 Sustainability Science (2018) 13:1483–1487 1485 approach to initiate a number of innovation labs focused on SDG implementation, under the title ‘SDG Labs’. The design was deliberately experimental but empowering, and partici- pants were provided with a set of guidelines that Future Earth and the Stockholm Resilience Centre had developed from the Social Innovation Lab approach. The approach focused on stimulating local solutions across the globe towards meeting the SDGs. The labs brought together a range of research- ers from different disciplines with other sectors of society to develop solutions to complex local problems. By focussing on prototype or scalable solutions that could lead to sustain- ability transformations, the Labs offered inspiration to the communities involved and others learning from them. maps the sustainability science of the Future Earth commu- nity—to support politicians charged with implementing the SDGs by giving them a navigation tool for using the relevant sustainability science. To accomplish this objective of the conference, eight sessions were organised. In this Special Feature, seven articles are included to reflect each session and discussion conducted on its theme. Nilsson et al. (2018) in their paper explore the mapping of interactions between the sustainable development goals, drawing on “a major international research study applied to the SDGs on health, energy and the ocean, it analyses how interactions depend on key factors such as geographical con- text, resource endowments, time horizon and governance”. Nilsson et al. (2018) synthesise “experiences and insights from the application of a new conceptual framework for mapping and assessing SDG interactions using a defined typology and characterization approach”. Nilsson et al. (2018) examine “the future potential, barriers and oppor- tunities for applying the approach in scientific research, in policy-making and in bridging the two through a global SDG Interactions Knowledge Platform as a key mechanism for assembling, systematising and aggregating knowledge on interactions”. SDG labs As such, the aim is to evolve a positive tool to help catalyse change and bring about momentum in transformations supported by research). Martinez-Harms et al. (2018) discuss natural assets, and the concept and activities of the Natural Assets KAN. Their paper frames “Future Earth around natural assets emphasising the recognition of pluralism and identifying the challenges of translating different visions about the role of natural assets, including via policy formulation, for local to global sustainability challenges”. The discussion by Martinez-Harms et al. (2018) will be useful in developing inter- and transdisciplinary solutions for human–environ- mental problems. Bengtsson et al. (2018) examine the transformation of systems of consumption and production for the purpose of achieving the SDGs. They show that “while the efficiency approach contains essential elements of a transition to sus- tainability, it is by itself highly unlikely to bring about sus- tainable outcomes. Concomitantly, research also finds that Building on these issues (2018) discuss design principles and oppor- tunities for integrating design thinking with sustainability science towards achieving the SDGs. Maher et al. (2018) “examine the process of designing MetaMAP: an interactive graphic tool for collaborating to understand social–ecologi- cal systems and design well-integrated solutions. MetaMAP was created using Research through Design methods which integrate creative and scientific thinking. By applying design thinking, researchers and practitioners from different back- grounds undertook multiple cycles of problem framing, solution development, testing and reflection”. Maher et al. (2018) discuss design principles and oppor- tunities for integrating design thinking with sustainability science towards achieving the SDGs. Maher et al. (2018) “examine the process of designing MetaMAP: an interactive graphic tool for collaborating to understand social–ecologi- cal systems and design well-integrated solutions. MetaMAP was created using Research through Design methods which integrate creative and scientific thinking. By applying design thinking, researchers and practitioners from different back- grounds undertook multiple cycles of problem framing, solution development, testing and reflection”. • Emphasise the understanding of how to ‘tinker’ locally with an eye on how context-specific solutions may be used in other parts of the world • Support solutions and scaling by encouraging and docu- menting an explosion of experiments involving research- ers and stakeholders such as the SDG Labs within a con- text of broader networksf • Consider and where appropriate integrate different types of knowledge—indigenous, practitioner, policy, aca- demic—and create tools that help practitioners to benefit equitably from that knowledgel l Other key threads that emerged during ICSS 2017 included: meeting the SDGs in an integrated way (‘indivisi- ble’), how approaches to this could be scaled up to transform society (‘universal’), and how all of this could be facilitated by the evolving, even maturing, discipline of sustainability science with the newly expanding suite of tools it has availa- ble, as exemplified in the Special Feature papers that follow. l Other key threads that emerged during ICSS 2017 included: meeting the SDGs in an integrated way (‘indivisi- ble’), how approaches to this could be scaled up to transform society (‘universal’), and how all of this could be facilitated by the evolving, even maturing, discipline of sustainability science with the newly expanding suite of tools it has availa- ble, as exemplified in the Special Feature papers that follow. • Recognise reflexively that “sustainability is itself an open-ended social learning process”, in the words of Francesca Farioli at the conference. Building on these issues The principle objective of the ICSS 2017 is to feed into the High-Level Political Forum of the UN SDG process. The conference aims to create an output that showcases and 1 1 3 3 Sustainability Science (2018) 13:1483–1487 1486 volumes of consumption and production are closely associ- ated with environmental impacts, indicating a need to curtail these volumes in ways that safeguard social sustainability, which is unlikely to be possible without a restructuring of existing socioeconomic arrangements”. “Based on this determination, this paper provides some suggestions on how governments and other actors involved in SDGs operation- alisation could more effectively pursue SCP from a systemic standpoint and use the transformation of systems of con- sumption and production as a lever for achieving multiple sustainability objectives”. agenda, Bengtsson et al. show that the SDG targets are not generally expressed in the system-wide way that is needed to ensure global outcomes. Then, Elmqvist et al. explore tinkering as a tool that is sensitive to conditions of uncertainty and complexity in urban sustainability. Maher et al. show that bringing tools such as design thinking into sustainability research may assist the transdisciplinarity which is needed.fl Van der Leeuw (2018) rounds off the collection by reflect- ing on these indicators of the revitalisation and maturing of sustainability science, whilst highlighting some chal- lenges—our need to focus on a society subject to perma- nent change, understanding and stimulating innovation and change beyond a stability-focused paradigm, emphasising relationships as much as entities, rethinking the “idea of progress” in western cultures, and achieving modesty in the (still important but changing) role of research in society. Elmqvist et al. (2018) discuss ‘urban tinkering’. They define tinkering as “a mode of operation, encompassing policy, planning and management processes, that seeks to transform the use of existing and design of new urban sys- tems in ways that diversify their functions, anticipate new uses and enhance adaptability, to better meet the social, eco- nomic and ecological needs of cities under conditions of deep uncertainty about the future”. This approach has the potential to substantially complement and augment conven- tional urban development. Overall, ICSS 2017 and this Special Feature show how sustainability researchers now need to: • Enrich their theory and practice of deep interdisciplinar- ity and transdisciplinarity, to support, the indivisibility and universality of sustainable development as framed by the SDGs Maher et al. Building on these issues At the same time, key messages emerged for policy-makers: i Nilsson et al. open by documenting a formal approach to assessing interactions among SDGs that argues for a systematic way of collating and quantifying benefits in this regard, following the reality that no individual SDG can be met without meeting all the others (noting that the balance and priorities may be very specific to individual countries). There follow a series of papers exploring issues of integration, transformation and scaling in more-or-less specific areas of the SDGs. Lindgren et al. emphasise links among targets for health and food. Harms et al. discuss the need to manage natural assets within a pluralist vision that recognises the importance of collaboration, equity and power. While the Sustainable Consumption and Production aspects of the SDGs identify a potentially transformative • The indivisible intent of the SDGs requires a continued emphasis on maintaining coherence among policy areas at all levels of governance, in the face of the reality that many institutional incentives promote fragmentation: a powerful area for the practical expression of this princi- ple is in ensuring the close alignment between work on climate change and the rest of the SDGs • Ensuring the universality of Agenda 2030 requires a bal- anced focus on locally and nationally appropriate action, whilst understanding how this contributes the global out- comes that affect us all; allowing either side of the bal- 1 3 3 Sustainability Science (2018) 13:1483–1487 1487 integrated framework for sustainable development goals. Ecol Soc 19(4):Art. 49. https://doi.org/10.5751/es-07082-190449 ance to dominate will undermine global sustainability and human well-being. Haider LJ, Hentati-Sundberg J, Giusti M, Goodness J, Hamann M, Masterson VA, Meacham M, Merrie A, Ospina D, Schill C, Sinare H (2017) The undisciplinary journey: early-career perspectives in sustainability science. Sustain Sci 13:191. https://doi.org/10.1007/ s11625-017-0445-1 • These priorities can be brought together in practice by endorsing and encouraging many local action-learning experiments that link benefits across many SDGs, and which are actively networked and tracked to enable global learning IPCC Working Group I (2013) Working Group I Contribution to the IPCC Fifth Assessment Report. Climate change 2013: the physical science basis—summary for policymakers Jackson RB, Le Quere C, Andrew RM, Canadell JG, Peters GP, Roy J, Wu L (2017) Warning signs for stabilizing global CO2 emissions. Environ Res Lett 12(11):110202. Building on these issues https://doi.org/10.1088/1748- 9326/aa9662 Acknowledgements This Special Feature contains contributions from the 7th International Conference on Sustainability Science (ICSS 2017) held on 24–26 August 2017 in Stockholm, Sweden. This event was organised jointly by the University of Tokyo (IR3S) and Future Earth Secretariat (Global Hubs in Sweden and Japan), Japan Science and Technology Agency, and the University of Stockholm (Stockholm Resilience Centre). The University Tokyo supported expenses for open access of this Special Feature. OECD has co-funded this conference through “Co-Operative Research Programme Managing Natural Capi- tal for the Future”. We would like to express our gratitude to Dr. Pri- mal Silva, OECD Representative, for supervising the conference. Mark Stafford Smith and Youba Sokona co-authored this Editorial but were not involved in the editing process of the Special Feature. Christina Cook co-authored this Editorial and was involved in the early editing process stage of the Special Feature. Lindgren E, Harris F, Dangour AD, Gasparatos A, Hiramatsu M, Javadi F, Loken B, Murakami T, Scheelbeek P, Haines A (2018) Sustain- able food systems—a health perspective. Sustain Sci. https://doi. org/10.1007/s11625-018-0586-x g Maher R, Maher M, Mann S, McAlpine CA (2018) Integrating design thinking with sustainability science: a research through design approach. Sustain Sci. https://doi.org/10.1007/s11625-018-0618-6 Martinez-Harms MJ, Gelcich S, Krug RM, Maseyk JFF, Moers- berger H, Rastogi A, Wambugu G, Krug CB, Spehn EM, Pas- cual U (2018) Framing natural assets for advancing sustainability research: translating different perspectives into actions. Sustain Sci. https://doi.org/10.1007/s11625-018-0599-5 Nilsson M, Chisholm E, Griggs D, Howden-Chapman P, McCollum D, Messerli P, Neumann B, Stevance A, Visbeck M, Stafford-Smith M (2018) Mapping interactions between the sustainable develop- ment goals: lessons learned and ways forward. Sustain Sci. https ://doi.org/10.1007/s11625-018-0604-z Open Access This article is distributed under the terms of the Crea- tive Commons Attribution 4.0 International License (http://creativeco mmons.org/licenses/by/4.0/), which permits unrestricted use, distribu- tion, 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. Rivera-Ferre MG, Pereira L, Karpouzoglou T, Nicholas KA, Onzere S, Waterlander W, Mahomoodally F, Vrieling A, Babalola FD, Ummenhofer CC, Dogra A, de Conti A, Baldermann S, Evoh C, Bollmohr S (2013) A vision for transdisciplinarity in Future Earth: perspectives from young researchers. J Agric Food Syst Community Dev 3(4):249–260. https://doi.org/10.5304/jafsc d.2013.034.031f References Rockström J, Gaffney O, Rogelj J, Meinshausen M, Nakicenovic N, Schellnhuber HJ (2016) A roadmap for rapid decarbonisation. Science 355(6331):1269–1271. https://doi.org/10.1126/scien ce.aah3443 Bengtsson M, Alfredsson E, Cohen M, Lorek S, Schroeder P (2018) Transforming systems of consumption and production for achiev- ing the sustainable development goals: moving beyond efficiency. Sustain Sci. https://doi.org/10.1007/s11625-018-0582-1 UN (2015) Transforming our world: the 2030 agenda for sustainable development. United Nations, New York Bromham L, Dinnage R, Hua X (2016) Interdisciplinary research has consistently lower funding success. Nature 534:684–687. https:// doi.org/10.1038/nature18315 van der Leeuw S (2018) Closing remarks: novel approaches to com- plex societal change and sustainability. Sustain Sci. https://doi. org/10.1007/s11625-018-0581-2 Elmqvist T, Siri J, Andersson E, Anderson P, Bai X, Das PK, Gatere T, Gonzalez A, Goodness J, Handel SN, Hermansson Török E, Kavonic J, Kronenberg J, Lindgren E, Maddox D, Maher R, Mbow C, McPhearson T, Mulligan J, Nordenson G, Spires M, Stenkula U, Takeuchi K, Vogel C (2018) Urban tinkering. Sustain Sci. 51:2. https://doi.org/10.1007/s11625-018-0611-0 Westley F, Laban S, Rose C, McGowan K, Robinson K, Tjornbo O, Tovey M (2015) Social innovation lab guide. Waterloo Institute for Social Innovation and Resilience, Waterloo WMO (2018) WMO statement on the state of the global climate in 2017. World Meteorological Organisation, Geneva Future Earth (2013) Future earth initial design: report of the transition team. International Council for Science (ICSU), Parisf Öf Griggs D, Stafford Smith M, Rockström J, Öhman MC, Gaffney O, Glaser G, Kanie N, Noble I, Steffen W, Shyamsundar P (2014) An 1 1 3 3
https://openalex.org/W3195339082	https://link.springer.com/content/pdf/10.1007/s00158-021-03027-6.pdf	English	null	Surface smoothing for topological optimized 3D models	Structural and multidisciplinary optimization	2,021	cc-by	12,742	Abstract The topology optimization methodology is widely applied in industrial engineering to design lightweight and efficient components. Despite that, many techniques based on structural optimization return a digital model that is far from being directly manufactured, mainly because of surface noise given by spikes and peaks on the component. For this reason, mesh post-processing is needed. Surface smoothing is one of the numerical procedures that can be applied to a triangulated mesh file to return a more appealing geometry. In literature, there are many smoothing algorithms available, but especially those based on the modification of vertex position suffer from high mesh shrinkage and loss of important geometry features like holes and surface planarity. For these reasons, an improved vertex-based algorithm based on Vollmer’s surface smoothing has been developed and introduced in this work along with two case studies included to evaluate its performances compared with existent algorithms. The innovative approach herein developed contains some sub-routines to mitigate the issues of common algorithms, and confirms to be efficient and useful in a real-life industrial context. Thanks to the developed functions able to recognize the geometry feature to be frozen during the smoothing process, the user’s intervention is not required to guide the procedure to get proper results. Keywords Additive manufacturing · Topology optimization · Surface smoothing · Mesh processing · Structural manufacturing Antonio Bacciaglia1 · Alessandro Ceruti1 · Alfredo Liverani1 Received: 4 January 2021 / Revised: 16 July 2021 / Accepted: 24 July 2021 / Published online: 17 August 2021 © The Author(s) 2021 1 Department of Industrial Engineering (DIN), University of Bologna, Bologna, Italy Structural and Multidisciplinary Optimization (2021) 64:3453–3472 https://doi.org/10.1007/s00158-021-03027-6 Structural and Multidisciplinary Optimization (2021) 64:3453–3472 https://doi.org/10.1007/s00158-021-03027-6 Structural and Multidisciplinary Optimization (2021) 64:3453–3472 https://doi.org/10.1007/s00158-021-03027-6 RESEARCH PAPER 1 Introduction 2015). AM is known for the high freedom of shaping, time reduction in the design-to-manufacturing cycle, the capa- bility of building complex biomimetic shapes with a high strength to weight ratio, called lattice structures (Savio et al. 2019), and reduction of parts thus avoiding bolted connec- tions or welding. The design workflow belonging to AM is well summarized by the expression: "What You See Is What You Build" (Gibson etal. 2015).i Topology optimization (TO) is a numerical design technique that allows designing efficient and lightweight components (Bendsøe and Sigmund 2011). This design methodology is used especially in the automotive and aerospace industries. In these applications, the light-weighting research is taken to extremes and reflects on better performances, reduction of manufacturing and maintenance costs and emissions. On the one hand, due to the resulting complex and intricated solution coming from TO analyses, the 3D models can’t be directly manufactured by traditional processes based on chip removal. On the other hand, the complexity of the structure well matches with additive manufacturing (AM) techniques which are based on adding material layer by layer (Gao et al. Due to the aforementioned benefits, the AM and TO cou- pling is becoming a recurring theme in the research com- munity, especially in the last few years when there has been a positive trend to bring AM to the small industry and even to the single consumer (Bacciaglia et al. 2020). Several con- tributions combining TO and AM are already available in the literature (Gaynor et al. 2014; Rezaie et al. 2013), but the overall design workflow is still not user-friendly and far from being direct. (Zegard and Paulino 2016) try to fill the gap by proposing a simple methodology to streamline the last step of making manufacturable the 3D models coming from structural optimization, especially for voxel-based models. They describe the importance of intermediate steps before the manufacturing process, even if AM shows outstanding 1 Department of Industrial Engineering (DIN), University of Bologna, Bologna, Italy (0123 1 3456789) 3 3454 A. Bacciaglia et al. detailed way the consistency of the 3D model volume and does not offer a strong edge and corner preserving capabil- ity. To fill this technological gap, the work herein described aims at developing a smoothing methodology for triangu- lated mesh based on vertex position for 3D models coming from TO analysis whose external surface is discretized with tetrahedral elements. 1 Introduction The developed methodology is easy to use, maintains important features of the no-smoothing space and addresses all the known problems which affect the vertex-based smoothing methods as volume shrinkage and vertex drifting. The new algorithm has been implemented in Matlab, and applied to different 3D models, characterized by complex shapes and containing no-smoothing spaces. These models come from a topology optimization code embed- ded in FreeCAD, an open-source Computer-Aided Design (CAD) software that allows the development of new work- benches. The innovative algorithm has been compared to well established vertex-based methodologies available in literature to evaluate its performances during the smooth- ing process and to demonstrate its advantages.l properties for the production of TO models. For this reason, they developed a tool called TOPslicer useful to analyze and improve optimized models based on voxels. Voxelization is a representation method that uses hexahedral elements to discretize a control volume in which the 3D model is con- tained (Jense 1989) and has many applications in the AM field (Bacciaglia et al., 2019). i However, there is still a gap in the design-to-manufac- turing workflow for topologically optimized structures dis- cretized by tetrahedral mesh elements. As stated before, the Standard Triangulation Language (STL) file of the 3D mod- els coming from TO needs post-processing routines before the production phase to analyze and repair non-manifold edges, cracks and peaks which may originate from the opti- mization. The external surface smoothing is inspired by the image denoising techniques aiming at removing the noise that affects the pixels in an image and uniformly deters their information (Buades et al. 2005), (Ming Zhang and Gun- turk 2008). The same basic concept is applied to surface smoothing which is a numerical method useful to detect and remove noise and spikes from the surface model, returning a more appealing geometry by evolving the surface iteratively (Desbrun et al. 1999). The manuscript is organized as follows. Section 2 briefly describes the structural optimization methodologies avail- able in literature and describes an analysis where our meth- odology can be applied along with a list of available smooth- ing algorithms. Section 3 embeds a detailed description of the innovative post-processing methodology where several sub-routines are used to get the optimal result. Examples and applications for the developed methodology are shown in Sect. 4 along with a discussion of the results. Finally, the work is summarized in Sect. 1 Introduction 5 and provides conclusions and possible future developments of the established research topic. Literature has several approaches for surface fairing based on the modification of mesh using the position of vertices (Sorkine 2005), using local curvature of neighbour faces (Belyaev and Ohtake 2003), filters based on patch normal (Wei et al. 2019) or by filtering the surface with a frequency- based approach (Taubin 1995). Each approach shows advan- tages and disadvantages, depending on the specific appli- cation, but the methodologies based on vertex position are known to be easy to implement, fast and with reasonable performances. However, some issues as volume shrinkage need to be considered and solved. Despite all these efforts, there is a lack of user-friendly methodology to post-process tetrahedral 3D models using smoothing algorithms based on mesh modification of the position of vertices. In particular, a good smoothing framework fitting the AM and TO require- ments should carry out efficiently the following tasks: 1 3 2.1 Topology optimization TO is a numerical design methodology that can guaran- tee the best material distribution, by assigning material or void to all the elements of the discretized volume without forcing the algorithm to pre-designed shapes. This design freedom gives the possibility to obtain innovative and high- performance shapes reducing the material and the structure weight but maintaining the same degree of functionality. In literature, there are different TO numerical techniques, a non-inclusive list includes: Whatever it is the optimization approach chosen by the designer, the common optimization methods are related to the quality of the component mesh of the finite element model. In this case, the optimal solution can be far from the manufacturable status, because of the presence of peaks, cracks or non-manifold edges in the mesh that discretizes the 3D shape. At this point there are two possibilities of the design process: (a) use post-processing approaches applied directly on the optimal solution or (b) re-design the compo- nent sketching it from scratch taking inspiration from the optimal result of the previous step. The research community is pushing towards the first solution to accelerate the design- to-manufacturing cycle, decrease the cost and increase the design workflow efficiency, especially in industrial contests. For example, there are some recent contributions that cou- ple the TO method with the Non-Uniform Rational Basis Spline (NURBS) hyper-surfaces framework (Costa et al. 2018). This combination provides CAD-compatible descrip- tors of the topology of the structure, that are not related to the mesh quality of the finite element model; moreover, the boundaries’ reconstruction becomes a straightforward task thanks to the NURBS implementation (Costa et al. 2021). However, this approach highly depends on the designer’s experience when NURBS discrete parameters should be set: after the optimization process, the overall structure external smoothing highly depends on the NURBS weights. Lastly, a higher amount of NURBS control point reflects not only on improved performances but also on long computational time. • SIMP (Solid Isotropic Material with Penalization) this method is mainly used for the minimum compliance problem. It’s a gradient-based approach that updates the 3D model at each iteration after the structural analy- sis using a continuous distribution of material density (Bendsøe 1989). • ESO (Evolutionary Structural Optimization method) this approach uses a fully dense control volume and at each iteration subtracts unneeded material until reaching an optimal structure (Xie and Steven 1996). 2.1 Topology optimization • BESO (Bi-directional ESO) this numerical method is based on the ESO approach but has also the capability of adding material if it is necessary to reach the optimum (Li et al. 2001). In the following, we will refer only to the SIMP approach in which the design variable ( 휌e ) is the density of the mate- rial of a discrete element e. Its name comes from the depend- ency of the single e-th element stiffness tensor ( Ee ) from the material density by the power-law (Eq. 1): (1) E퐞= E(휌e ) = 휌p eE퐨, with, 휌e ∈[휌min, 1 ] (1) To overcome the aforementioned issues, this work tries to contribute in the same direction to propose a general- purpose post-processing approach for surface smoothing. Indeed, the developed approach aims at covering a wider number of circumstances compared to the TO-NURBS approach, since it can smooth the external surface of meshes coming from different sources such as topology optimiza- tion or reverse-engineering from points clouds obtained through 3D scanners and photogrammetry. In this specific research, the developed methodology is applied to finite ele- ment based optimized structures without the need of rede- signing them within a CAD system. Aim of this work, only 3D models coming from TO analyses will be considered in the following, and the same process can be applied for any kind of optimization methodology or engineering design approach as long as the 3D model can be exported as an STL surface mesh. where E퐨 is the assigned isotropic material stiffness tensor and p is the penalization factor (usually higher than 3). 휌min the minimum allowable relative density value for empty ele- ments that are greater than zero. This density value ensures the numerical stability of the FEM (Finite Element Method) analyses. The TO problem is known to be not well-posed because the solution is mesh-dependent. To reduce this dependency, the TO problem needs to be restricted using some methods such as the density filter (Bourdin 2001) or the sensitivity one (Sigmund 1997). For a more detailed description of the TO methodology, please refer to (Bendsøe and Sigmund 2011). The case studies that will be shown in the following to test the smoothing algorithm are obtained by an own TO framework, called ToOp, embedded in FreeCAD and based on Python macros. 2 Design and surface smoothing for unconventional structures Topology optimization, ground structure method and gener- ative design are the main approaches for structure optimiza- tion used in industrial applications as aerospace (Wong et al. 2018), automotive (Mantovani et al. 2020) and biomedicine (Machado and Trabucho 2004).i • improve the external shape of the model both in the case of voxel and surface mesh. TO minimizes a fitness function that in most cases is rep- resented by the overall structural compliance, maximizing the global stiffness. To solve the problem it is mandatory to have information about the boundary conditions, the load case applied to a predefined working volume, the presence of passive elements (e.g. holes) and the maximum material volume fraction needed to avoid a fully dense solution (Sig- mund 1997). The ground structure method approximates a truss-like structure with a finite number of beam elements removing unnecessary elements from a connected truss structure while freezing the nodal positions (Ohsaki 2011). • limit the volume shrinkage during the smoothing itera- tions. • reduce or completely avoid the loss of features of TO models during the numerical process (e.g. holes or flat surfaces), so that they do not need to be post-processed (these regions will be referred to as "no-smoothing- space"). Some researches aim at preserving mesh features in denoising processes as done in (Lee and Wang 2005). How- ever, the approach therein described does not mention in a 3 3455 Surface smoothing for topological optimized 3D models Generative design is an iterative process that, knowing the boundary conditions, will generate a certain number of pos- sible solutions that meet the initial constraints; thanks to the designer intervention, the best solution will be chosen (Krish 2011). 2.2 Surface smoothing Nowadays, lots of methods useful to optimize the external surface of mesh files based on the manipulation of different data of the STL files are available. STL file contains the coordinates of the vertices composing the mesh, the IDs of the vertices that compose a triangular facet and the compo- nents of the normal vector for each facet. These are the infor- mation that a smoothing approach can manipulate taking as input the STL file format, whatever is the AM process and the material selected to manufacture the object. An improvement of the previously cited approach is the Scale-Dependent Laplacian smoothing in which the Lapla- cian operator uses weights, called Scale-Dependent Umbrella (SDU) or Fujiwara weights, which are propor- tional to the relative distance between the vertices wij = 1 \|eij\| (Desbrun et al. 1999). This feature preserves the size of the triangles and decreases the vertex drifting. However, mesh shrinkage is still a critical issue. Besides, SDU weights introduce the Laplacian operator’s dependency on the solu- tion of the previous iteration, because eij must be updated at each mesh change. However, (Desbrun et al. 1999) shows that keeping constant the Fujiwara weights produces a neg- ligible error. In the following, the attention will be directed to vertex- based approaches available in literature: they are the simplest and the easiest to be implemented even if they suffer from critical problems as high volume reduction during the itera- tions. As the name says, these approaches use neighbour- hood information in terms of spatial position to update the mesh. The largest part of the vertex-based approaches takes inspiration from the Laplacian smoothing (Sorkine 2005) whose operation can be modelled as a diffusion problem. This mathematical problem shows two desirable properties: the mesh connectivity is maintained and only the position of the vertices changes; each vertex is moved using only the information about its neighbours. The above-mentioned dif- fusion equation can be expressed as (Sorkine 2005): Another vertex-based approach taken under consideration is the Improved Laplacian smoothing, also called HC-algo- rithm (HC stands for Humphrey’s Classes) (Vollmer et al. 1999). This methodology aims at improving the Laplacian approach and decreases volume shrinkage. 2.1 Topology optimization N1(i) represents the 1-ring-neighbourhood vertex set which consists of all vertices that are connected to the i-th vertex by one edge, while xi is the vector of coordinates of the spatial position of the i-th vertex. Often the TO solutions show an external surface that is far from being smooth and ready to be manufactured. This comes directly from the TO process that assigns material or void to the tetrahedral elements which compose the discre- tized control volume according to the sensitivity analysis to make the resulting structure more efficient. For this reason, the solution is characterized by an external surface made of spikes and peaks which are undesired in the final 3D model. Noisy 3D optimized models come from the ToOp own-built framework as will be seen in the case studies included in this research. This is the reason why a post-processing algorithm for external surface smoothing is essential to make the TO framework useful in a design context where the optimized solution should be directly ready to be manufactured through AM processes due to the high complexity of the resulting shapes. To comply with this request, a smoothing process based on vertices position modification is developed in Mat- lab, but before describing in detail the new algorithm, it is worth understanding which are the available smoothing approaches in literature. (3) L ( xi ) = ∑ j휖N1(i) wij(xj −xi) (3) The standard Laplacian smoothing replaces a mesh vertex with the average position of its neighbours wij = 1 n where n is the number of the one-ring neighbours. This smooth- ing method shows the advantage of being very simple and computationally fast. However, it is affected by a strong ver- tex drifting (vertex movement not along the surface normal direction) and mesh shrinkage (mesh volume reduction) as the number of iterations grows. 2.1 Topology optimization It is based on a SIMP approach using a sensitivity filter to make the problem well-posed. Differently from other TO open-source codes available in literature, our 1 3 3456 A. Bacciaglia et al. framework is capable of returning an optimized structure after a TO analysis using a user-friendly GUI (Graphic User Interface) and an easy workflow from the design of the con- trol volume, the simulation settings through the meshing and FEM analysis directly to the post-processing of the 3D model using the same software. 휕X 휕t = 휆L(X), (2) 휕X 휕t = 휆L(X), (2) 휕t where X is a tensor that embodies the vertices of the mesh that will change during the iterative process, L is the Lapla- cian function, λ is a weight factor between 0 and 1 repre- senting the diffusion speed and 휕t embodies the variation of the surface mesh during the iterative process. Assuming a Laplacian’s operator linearization, an explicit or implicit solution scheme can be used to find the evolving surface mesh during the iterations. To reduce complexity, in the fol- lowing, only an explicit solving scheme will be used. Given Eq. 2, the available algorithms mainly vary among them by a different expression of the Laplacian operator that in the linearized form is represented by Eq. 3. N1(i) represents the 1-ring-neighbourhood vertex set which consists of all vertices that are connected to the i-th vertex by one edge, while xi is the vector of coordinates of the spatial position of the i-th vertex. 휕t where X is a tensor that embodies the vertices of the mesh that will change during the iterative process, L is the Lapla- cian function, λ is a weight factor between 0 and 1 repre- senting the diffusion speed and 휕t embodies the variation of the surface mesh during the iterative process. Assuming a Laplacian’s operator linearization, an explicit or implicit solution scheme can be used to find the evolving surface mesh during the iterations. To reduce complexity, in the fol- lowing, only an explicit solving scheme will be used. Given Eq. 2, the available algorithms mainly vary among them by a different expression of the Laplacian operator that in the linearized form is represented by Eq. 3. 2.2 Surface smoothing These two values should be chosen to satisfy the following mathematical expression: 0.01 < 1 𝜆+ 1 𝜇< 0.1. The following steps involve the introduction of two own programmed sub-routines which are used to recognize fea- tures of the 3D model, the so-called no-smoothing-space. Following the flowchart shown in Fig. 1, a Matlab func- tion called detect_flat_surface is used to find the vertices belonging to a flat surface by studying the components of the normal vector of the selected facet and the neighbour’s ones with a more detailed description in Sect. 3.1. The sec- ond subroutine is used to find holes that may be present in the digital model, independently from the shape of the cavity and will be described in detail in Sect. 3.2. To fulfil this capability, a function called detect_holes_edges searches for closed-loop sharp edges which belong to the summit of holes, taking inspiration from the methodology presented in (Qu and Stucker 2005) for different purposes. Both the cited sub-routines return the IDs of the vertices that belong to a flat surface or an edge of a hole. By the union of these two ID lists, the algorithm obtains an array containing the vertices belonging to the no-smoothing space that is passed to the core function of the algorithm. Though, the available algorithms are still non-optimized for complex shapes coming from TO analyses where the designer wants to freeze important features such as holes or surfaces that should be kept planar in the ready-to-be- manufactured digital model. To fill this technological gap, the authors developed the Optimized Humphrey’s Classes— Scale-Dependent Umbrella algorithm (in the following Optimized HC-SDU algorithm) that combines the SDU and a modified version of the HC-algorithms to exploit their advantages, also including several sub-routines to solve the problems addressed before. 2.2 Surface smoothing This is obtained by adding to a first step (called push-forward and consists of the application of the classic Laplacian operator), a second step (push-back) to partially push towards the old position the vertices by a value that is the average of its own and its 1 3 3 Surface smoothing for topological optimized 3D models 3457 neighbours’ difference position vectors weighted by a factor β. Moreover, combined with the above-mentioned strategy to decrease the volume shrinkage, the new position of a vertex is evaluated considering not only the neighbour vertices but also the central vertex position. Indeed, the original vertex position is weighted by a factor α and included to help the algorithm to converge easily. The variables 훼 and 훽 should be set by the user to obtain 𝛼> 0 and 𝛽> 0.5 . Thanks to all these improvements, the HC algorithm preserves better the mesh features and size during the iterations, even if a bit of shrinkage is still present. To examine in depth the mathematical background of this approach, the readers are referred to (Vollmer et al. 1999). models coming from TO analysis with the scope of main- taining important features without suffering from volume shrinkage. The flowchart containing the overall methodol- ogy is shown in Fig. 1. The innovative approach takes as an input an STL mesh file usually coming from a TO analysis. The file is analyzed and the topology information pieces (vertices V, facets F and normal components N) are saved in matrices. In the following, the algorithm asks the user to insert a numerical value for the four parameters needed for the smoothing, which are: 훼 and 훽 coming from the HC-algo- rithm, 휆 that controls the diffusion speed of the process and itermax which controls the maximum number of iterations the algorithm can do before stopping if the convergence is not reached (the difference between two consecutive solutions should be < 0.01). Pre-set values are suggested: 훼=0.27, β = 0.51, λ = 0.6307 and itermax=150. Taubin’s algorithm is one of the best smoothing algo- rithms present in literature. This approach is similar to the HC one because of the implementation of a two-step smoothing (forward and backwards) to correct the shrinkage. However, Taubin allows fine-tuning of both the steps by set- ting two scalar values (λ and μ) so that they balance each other (Taubin 1995). 3 Optimized Humphrey’s Classes— Scale‑Dependent Umbrella algorithm The core of the process is made by the smoothing algo- rithm itself based on the HC methodology with a small change in the forward step where an SDU weight scheme is adopted instead of the classic Laplacian one to reduce the vertex drifting. Another change occurs in the inclusion of This section contains a description of the innovative vertex- based smoothing algorithm developed. The scope of this methodology is to satisfy the necessity to post-process 3D Fig. 1 Optimized HC-SDU flowchart explaining the methodology to obtain a smooth STL file mesh Fig. 1 Optimized HC-SDU flowchart explaining the methodology to obtain a smooth STL file mesh 1 3 3 3458 A. Bacciaglia et al. the original mesh: the original version of the HC algorithm uses a weighted original vertex position to evaluate the rela- tive position vector. However, in the Optimized HC-SDU algorithm, the mean position between the original and the current mesh will be used to compute the difference vector. This is done to delete the background noise of smoothed models by the original HC algorithm, which is a behaviour that affects the latter smoothing method. Moreover, at the end of each iteration, the volume of the smoothed mesh is compared to the initial volume and rescaled according to a modified version of (Desbrun et al. 1999) formulation. Vol- ume rescaling is necessary to avoid high mesh shrinkage due to the diffusion process that models the surface smoothing for vertex-based approaches. More details will be given in Sects. 3.3 and 3.4. the matrices containing the facet topologies (F), the normal vector components for each triangular face (N) and a scalar value (L) that will be explained later on. The function scrolls each facet of matrix F and compares its components of the normal vector with the neighbour triangles (IDs saved in vector q). The sub-routine is capable of counting the number of neighbour facets with the same normal (saved in the sca- lar variable z). Assuming that the i-th facet has x neighbours, if the number of the facets with the same normal to the i-th facet is higher than x−L, then the i-th facet belongs to a pla- nar surface and the three vertices are saved in an array ( a1 ). 3 Optimized Humphrey’s Classes— Scale‑Dependent Umbrella algorithm A threshold value (L) is imposed on the function to capture facets that belong to a planar surface that are near a sharp edge of the component. Indeed, assuming L = 0, many facets belonging to planar surfaces are lost during the process in the transition regions where the surface curvature suddenly changes (sharp edges). After several trials, a threshold value of L = 2 is chosen for the case studies that will be presented in Sect. 4. In fact, by choosing L > 2, the function starts to select triangles that do not belong to the planar surface any- more. Figure 2 shows the capability of recognizing planar surfaces (in yellow colour) from an STL model. 3.1 Flat surface detection Following the flowchart shown in Fig. 1, the first sub-routine implements a function that recognizes flat surfaces of the 3D model (pseudo-code is available in the following, where the symbol stands for’is a function of’). The function inputs are Fig. 2 Detect_flat_surface function applied to a component with L = 2. Front and rear views show the good capability of recognizing flat sur- faces highlighted in yellow colour. (Color figure online) Fig. 2 Detect_flat_surface function applied to a component with L = 2. Front and rear views show the good capability of recognizing flat sur- faces highlighted in yellow colour. (Color figure online) 1 3 3459 Surface smoothing for topological optimized 3D models developed in (Qu and Stucker 2005) where circular holes are detected for path-planning in CNC (Computerized Numeri- cal Control) machine context. A pseudo-code explaining the methodology behind this function is available in the follow- ing (where the ← symbol stands for ‘is a function of’) 3.2 Holes detection The second add-on included in the smoothing process is a function, called detect_holes_edges, whose aim is to recog- nize the presence of holes and cavities in the digital model. The procedure follows a technique similar to the algorithm 1 3 A. Bacciaglia et al. 3460 Fig. 3 A flowchart that describes the steps the function detect_holes_edges follows to return the vertices belonging to holes Fig. 3 A flowchart that describes the steps the function detect_holes_edges follows to return the vertices belonging to holes Fig. 4 Detect_holes_edges function applied to a component with many through-holes of different shapes. On the left, all the sharp edges are recognized by the function in red, while on the right, only Fig. 4 Detect_holes_edges function applied to a component with many through-holes of different shapes. On the left, all the sharp edges are recognized by the function in red, while on the right, only those belonging to simple closed loops show the good capability to recognize holes in the 3D model those belonging to simple closed loops show the good capability to recognize holes in the 3D model Fig. 4 Detect_holes_edges function applied to a component with many through-holes of different shapes. On the left, all the sharp edges are recognized by the function in red, while on the right, only those belonging to simple closed loops show the good capability to recognize holes in the 3D model Firstly, all the sharp edges (S_E) are selected by the func- tion being available the matrices of mesh faces F, mesh nor- mal N, mesh edges E and face adjacencies saved in matrix ADJ. An edge is defined as sharp when it is in common between two adjacent facets (common_edge), which have an angle θ between the two normal vectors, that is between 85 and 95°. Up to now, only holes perpendicular to an external surface can be captured by the developed methodology, but in further studies, this limitation will be mitigated. Then the add-on investigates each edge of the mesh: if the i-th edge is sharp (check_edge = 0) then it’s saved as the first element of a loop, otherwise, the following edges are investigated. If the edge is sharp, the endpoint is saved as the initial starting point in the variable SPinitial. Next, that endpoint is used as a new starting point (SP) and all the edges linked to SP are investigated (link_edge). 3.3 Volume rescaling As mentioned before, volume shrinkage is the main issue in all the smoothing processes based on the modification of the vertex position of the surface mesh. This is a critical issue that drove the research community to find alternative approaches to smooth the external surface. However, (Des- brun et al. 1999) shows that it is possible to rescale at each smoothing iteration the matrix describing the position of the vertices by a scalar value 훾 which is given by the comparison of the STL original mesh volume Vol0 with the volume of the i-th iteration Voli by the Eq. 4: (4) 훾= 3 √ Vol0 Voli 훾= 3 √ Vol0 Voli (4) Each vertex position is then multiplied by the 훾 factor at the end of each iteration if the two volumes are differ- ent. In this way, the initial volume value is constant and the shrinkage is avoided. However, it is important to note that this scheme where V is just multiplied by a scalar value will not preserve the location of geometric constraints such as the size of the bounding box or the prescribed location of supports. To overcome this issue, the volume rescaling is applied by multiplying the matrix of the vertices by B defined as an identity matrix multiplied by the factor 훾 of Eq. 4. However, in the main diagonal, there are some iden- tity elements, i.e. 퐁(i, i) = 1 , if the index i is a member of vector a (array that contains the IDs of all the nodes belong- ing to holes or flat surfaces that do not need a smoothing process) (Eq. 5). In this way, the i-th node will not undergo the volume rescaling process, guarantying the preservation of constraining positions. In the following, this nomenclature will be respected: the positioning vector of the i-th vertex in the original noisy mesh will be denoted as oi, the positioning vector of the vertex that is still not modified by the current iteration of the smooth- ing algorithm is called ci. Lastly, the vertex belonging to the smoothed mesh will be defined with si. As prescribed in the HC algorithm, two steps are used to smooth the mesh. Though, a major difference involves the push-forward step which is characterized by the implementation of the SDU weights to decrease the vertex drifting instead of the classic Laplacian. 3.4 Core of the algorithm After the description of all the sub-routines implemented to fulfil the research goal, in this section, the core of the devel- oped approach is described. As the name of the algorithm introduces, the developed smoothing methodology is based on the coupling of the SDU weight functions with the HC algorithm. For the sake of clarity, the former is a Laplacian smoothing approach where the relative weight functions of mesh vertices are defined to be proportional to the relative distance between the vertices wij = 1 \|eij\| . This feature main- tains the size of the triangles and decreases the vertex drifting. The latter is a smoothing approach with a first classic Lapla- cian step and a second step to moderately push towards the old position of the node by a value that depends on the relative difference position vectors of all the neighbours. In addition, in the original HC approach, the new position of a vertex depends also on the original central vertex position. This dependency is modified in the developed algorithm compared to the original HC. In the latter, the relative position vector (diffi) is a function of the original mesh by a scalar weight. In the former, the relative position vector depends on the mean position between the original and the current mesh, weighted by the same scalar value α. This is done to alleviate the main disadvantage of the HC algorithm, which is the mitigation of the biggest mesh peaks and surface noise, while light back- ground noise is still present on the smoothed model. Figure 4 shows the performance of holes detection of the described functionality in a model with several through- holes of different shapes. 3.2 Holes detection The function counts the number of sharp edges connected to SP. Usually, holes are determined by a simple closed loop of sharp edges meaning that no intersections are present. For this reason, on the one hand, if more than one sharp edge is found to be linked with SP, 1 3 3 Surface smoothing for topological optimized 3D models 3461 (5) 퐁= ⎡ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢ ⎢⎣ 훾0 0 0 훾0 0 0 1 ⋯ 0 ⋮ ⋱ ⋮ 0 ⋯ 훾0 0 0 1 0 0 0 훾 ⎤ ⎥ ⎥ ⎥ ⎥ ⎥ ⎥ ⎥⎦ the function clears the k-th loop, clears the variable SPinitial and investigates the following edge of the mesh. On the other hand, if only a new sharp edge is connected with the previ- ous one, the new edge is saved in the k-th closed-loop, SP is updated with the new endpoint and compared SPinitial. If SP and SPinitial. are equivalent, then the loop is closed, and the next edge is investigated to look for other loops. If SP is not equal to SPinitial, then the chain continues including more sharp edges in the k-th loop until the first edge is found to close the actual loop. In the end, the function returns the number of closed loops made of sharp edges found in the model and an array containing the IDs of the vertices touched by the selected sharp edges (a). Fig. 3 shows the flowchart of the methodology to understand if a sharp edge belongs to the summit of a hole. (5) 4.1 Cantilevered beam with passive elements A 100 × 30 × 10 mm cantilever beam with a shear load of 100 N applied on the free-end is optimized by the ToOp environment and an STL model, made of 11,369 vertices and 22,798 facets, is used as an input file to test the devel- oped smoothing algorithm. For the sake of clarity, in the TO analysis, a volume fraction of 0.4, a penalization factor of 3, a volume mesh size of 1 mm and the AlMg3 material are set. The peculiarity of this geometry is the presence of two passive element regions (two 10 mm diameter circular through-holes) that were not optimized by the TO algorithm and that should be maintained even after the surface post- processing, as it may happen in a real-life context in indus- trial engineering where cables and other structural elements may cross a structure. In the following case studies, 훼 and 훽 (scalar variables affecting the HC and Optimized HC-SDU algorithms) have been set after a sensitivity analysis to find the optimal val- ues. The best condition is defined as the one for which the algorithm converges, meaning that the distance between two consecutive meshes is lower than 0.01 without reaching the maximum number of iterations itermax , and the total change is maximized. 훼 and 훽 are ranged following literature sugges- tions: Vollmer states that the HC-algorithm converges if 0 < α < 1 and 0.5 < β < 1 (Vollmer et al. 1999). Several simu- lations are performed to study how the developed algorithm performs: α is changed using a 0.09 step from 0.18 to 0.72 because extreme values badly perform, while β is changed using a 0.07 step from 0.51 to 1. The diffusion speed value 휆 is chosen from literature ( 휆= 0.6307) because it represents a good trade-off value useful to better preserve the mesh volume using a limited number of iterations (Desbrun et al. 1999). Finally, itermax is chosen to be a compromise between a good-quality solution and low computational effort for the smoothing process and is arbitrary fixed to be 150 as the first attempt. All the simulations are performed by running The chosen input values for the variables involved in the process are 훼=0.8, 훽 . (=0.51, 휆=0.6307 and itermax=150. 3.3 Volume rescaling From a mathematical perspective, the forward step is characterized by the evaluation of the temporary smoothed position of the i-th vertex; along with si , in this step, the vec- tor containing the relative distance positioning vector to the original position by the 훼 weight is estimated: 1 3 3462 A. Bacciaglia et al. (6) si = ci + 2휆 edgeij ∑ j∈neighbors(i) cj−ci edgeij 퐝퐢퐟퐟i = si −훼 2(oi + ci) + (1 −훼)ci The overall algorithm is repeated until a satisfactory result is obtained (‘smooth enough’) with a while cycle. This verbatim collects the mathematical condition for which the difference in terms of distance between two consecutive solutions should be < 0.01. (6) Then a push-back step is characterized by the same approach developed for the HC algorithm to calculate the final smoothed position of the i-th vertex: By introducing the sub-routines described before, the overall smoothing algorithm can be described thanks to the following pseudo-code (where the ← symbol stands for ‘is a function of’): (7) si = si −훽퐝퐢퐟퐟i + 1 −훽 neighbor.size(i) ∑ j∈neighbors(i) 퐝퐢퐟퐟j (7) si = si −훽퐝퐢퐟퐟i + 1 −훽 neighbor.size(i) ∑ j∈neighbors(i) 퐝퐢퐟퐟j (7) Table 1 3 × 3 sensitivity matrix for α and β parameters of the HC- SDU algorithm for the cantilevered beam case study (λ = 0.6307, itermax = 150). On the top, the 3 × 3 matrix of the mesh distance between the last two smoothing iterations; on the bottom, the 3 × 3 matrix with the total change of the overall model. The optimum input configuration is represented by the combination of input values giv- ing the bold measure g ; , Mesh distance β 0.5 0.51 0.56 α 0.75 0.3229 0.1229 0.0442 0.8 0.0218 0.0096 0.0100 0.85 0.0095 0.0092 0.0088 Total change β 0.5 0.51 0.56 α 0.75 255541 218848 187604 0.8 213786 182158 140849 0.85 155704 125084 100261 1 Surface smoothing for topological optimized 3D models 3463 Fig. 5 Automatic detection of the no-smoothing space in the optimized cantilever beam: a detection of flat surfaces in yellow, b detection of 4 closed-loops where the circular holes appear in red. (Color figure online) Fig. 5 Automatic detection of the no-smoothing space in the optimized cantilever beam: a detection of flat surfaces in yellow, b detection of 4 closed-loops where the circular holes appear in red. (Color figure online) 4 Case studies the Matlab programmed codes on a workstation with 32 GB RAM and an Intel Zeon CPU @ 3.50 GHz. This section describes two case studies that have been used to evaluate the performances of the developed algorithm and compare it with algorithms available in literature. The performances are compared by the evaluation of the mesh volume variation during the iterations and the Total Change of the STL model. The change of the model is defined as the Euclidian distance between the original vertex position and the smoothed one. The Total Change is defined as the sum of all the changes for the overall surface (Gostler 2015). For benchmarking proposes, the Optimized HC-SDU algo- rithm is compared with the classic Laplacian smoothing, the Laplacian smoothing using SDU weights, the HC-algorithm and Taubin approach (μ will be set to − 0.53 for the follow- ing simulations, as literature suggests). 4.1 Cantilevered beam with passive elements A 3 × 3 sensitivity matrix is obtained by varying α and β around the optimum value by a step forward and backwards (Table 1) procedure. The optimum value is defined as the input configuration where the total change is maximized while the distance between two consecutive meshes is lower than 0.01 (convergence criteria). The other parameters such as λ and itermax are chosen conveniently: the diffusion speed matches literature benchmarks, while the maximum number of iteration is set to limit the computational time and cost. The best input configuration is highlighted in bold in the sensitivity matrices. Following the methodology shown in Fig. 1, the two functions detect_flat_surfaces and detect_holes_edges are run to detect automatically the no-smoothing space. The results are shown in Fig. 5. 1 3 A. Bacciaglia et al. 3464 Fig. 6 Smoothed models using from the top on the left: Laplacian smoothing with SDU weights, Laplacian smoothing, Taubin, HC- algorithm and Optimized HC-SDU algorithm. On the right, two detailed views of HC and HC-SDU results for a better visual evalua- tion with main differences highlighted with arrows Fig. 6 Smoothed models using from the top on the left: Laplacian smoothing with SDU weights, Laplacian smoothing, Taubin, HC- algorithm and Optimized HC-SDU algorithm. On the right, two detailed views of HC and HC-SDU results for a better visual evalua- tion with main differences highlighted with arrows the absolute nodal displacement from the original to the smoothed model is plotted. After the surface fairing, the 3D models are also visu- ally compared (Fig. 6). Moreover, a quantitative perfor- mance evaluation is done studying the behaviour of the mesh Total Volume and Total Change during the iterative process (Fig. 7). Table 2 collects the computational time needed to run the smoothing process, the required iterations to reach the convergence and quantitative comparison of the dimensions of the features of the model. Indeed, thanks to the modification of the volume rescaling proposed by Des- brun by a matrix multiplication, the position of geometric constraints is guaranteed, as it can be seen in Fig. 8 where 3 4.2 General Electric bracket The same approach has been applied to the model of a jet engine bracket (‘General Electric Jet Engine Bracket Chal- lenge’) used in a real-life industrial application. The model is firstly optimized within the ToOp environment: the four holes in the base of the component have been constrained and a shear load of 4525 N has been applied on the two 3 Surface smoothing for topological optimized 3D models 3465 Fig. 7 a behaviour of total change vs iterations for the cantilever beam case study, b behaviour of total volume vs iterations for the cantilever beam case study Fig. 7 a behaviour of total change vs iterations for the cantilever beam case study, b behaviour of total volume vs iterations for the cantilever beam case study Table 2 Dimensional and computational comparison of different smoothing algorithms applied on the cantilever beam example (percentage error in round brackets) Table 2 Dimensional and computational comparison of different smoothing algorithms applied on the cantilever beam example (percentage error in round brackets) Original Laplacian SDU HC Taubin Optimized HC-SDU Volume [mm3] 10330 3096 (− 70%) 2848 (− 72%) 10230 (− 1%) 8607 (− 17%) 10330 (0%) L [mm] 100 92.58 (− 7.4%) 92.57 (− 7.4%) 99.74 (− 0.3%) 98.93 (− 1.1%) 100 (0%) H [mm] 30 26.58 (− 11%) 26.36 (− 12%) 29.73 (− 1%) 27.86 (− 7%) 30 (0%) Holes diameter [mm] 10 6.19 (− 38%) 5.82 (− 42%) 9.96 (− 0.4%) 9.60 (− 4%) 10 (0%) Smoothing time [s] – 252 273 38 43 39 Iterations – 150 150 12 68 96 iterations to reach the convergence, and a comparison with the model’s dimensions. upper wings at 45° to the basement; a volume fraction of 50%, an initial volume mesh size of 2 mm and the Ti6Al4V material are chosen, while the penalization factor is set to 3. The resulting geometry is shown in Fig. 9. After the opti- mization, an STL model made of 7358 vertices and 14,738 facets is used as an input file to the smoothing algorithms. For this case study, the chosen values for the variables of the smoothing process are 훼=0.27, 훽=0.51, 휆=0.6307 and itermax =150. Even in this case, a 3 × 3 sensitivity matrix is built to demonstrate that this setup is the best one, as done for the previous case study where the best input configuration is highlighted in bold (Table 3). 4.3 Discussion of the results From the results shown in the previous section, it can be said that the Laplacian smoothing with classic weights and with SDU weights reaches poor results in quality for both case studies. As it can be seen in Figs. 6 and 11, they both suffer from high volume shrinkage (average reduction of 76% and 49%, respectively) even if high values of total changes are reached for both geometries. Moreover, they both do not reach convergence after 150 iterations, meaning that the dis- tance between two consecutive solutions is higher than 0.01; this reflects on the highest values of computational time. Figure 9 shows also the no-smoothing spaces which are automatically detected by the developed functions. As done for the previous component, the quantitative (Fig. 10) and qualitative (Fig. 11) results of the smoothing process are shown. Table 4 collects the computational time needed to run all the smoothing algorithms, the required l During the smoothing process, the no-smoothing spaces of the digital models are highly modified, making them 1 3 3466 A. Bacciaglia et al. Fig. 8 Absolute nodal dis- placement plot defined as the distance from the initial to the smoothed model: the values go from 0 (in blue) meaning no displacement to 1 (in yellow) of the absolute maximum displace- ment. (Color figure online) Fig. 9 Automatic detection of the no-smoothing space of the optimized engine bracket: a detection of flat surfaces in yellow, b detection of 12 closed-loops where the circular holes appear in red. (Color figure online) completely unrecognizable. Just to provide some numeri- cal values the classic Laplacian reduces the hole by 38% worse behaviour in the bracket case study. In general, this smoothing approach highly appreciated in literature con- A. Bacciaglia et al. 3466 Fig. 8 Absolute nodal dis- placement plot defined as the distance from the initial to the smoothed model: the values go from 0 (in blue) meaning no displacement to 1 (in yellow) of the absolute maximum displace- ment. (Color figure online) Fig. 8 Absolute nodal dis- placement plot defined as the distance from the initial to the smoothed model: the values go from 0 (in blue) meaning no displacement to 1 (in yellow) of the absolute maximum displace- ment. (Color figure online) Fig. 1 3 4.3 Discussion of the results 10 a the behaviour of total change vs iterations for the selected algorithms for the bracket case study, b the behaviour of total volume vs iterations for the selected algorithms for the bracket case study Lastly, Optimized HC-SDU algorithms decisively per- form better. Looking at the choice of the input parameters Surface smoothing for topological optimized 3D models 3467 Table 3 3 × 3 sensitivity matrix for α and β parameters of the HC- SDU algorithm for the GE bracket case study (λ = 0.6307. iter- max = 150). On the top, the 3 × 3 matrix of the mesh distance between the last two smoothing iterations; on the bottom, the 3 × 3 matrix with the total change of the overall model. The optimum input configura- tion is represented by the combination of input values giving the bold measure Total change β 0.5 0.51 0.56 α 0.18 0.1640 0.1595 0.1350 0.27 0.1230 0.0099 0.0090 0.36 0.0090 0.0087 0.0086 Total change β 0.5 0.51 0.56 α 0.18 878550 866824 805846 0.27 624070 607834 562232 0.36 383020 377251 342130 Fig. 10 a the behaviour of total change vs iterations for the selected algorithms for the bracket case study, b the behaviour of total volume vs iterations for the selected algorithms for the bracket case study Lastly, Optimized HC-SDU algorithms decisively per- form better. Looking at the choice of the input parameters addressed with the sensitivity matrices (Tables 1 and 3), it has been found that fixing β = 0.51 (near the lower boundary of β suggested by Vollmer) reflects satisfactory results, with maximization of total change and matching of convergence criteria. However, the choice of α is not straightforward as the previous parameter, because two different values have been set for the case studies presented in this work. It has been found that α depends on the number of vertices belong- ing to the no-smoothing-space: on the one hand, the beam example has almost 50% of vertices of the overall STL file belonging to flat surfaces or holes edges and a higher α value is needed to give more importance on the original mesh topology. On the other hand, the bracket mesh model has only 30% of nodes that belongs to the no-smoothing space, reflecting on a lower α input value to satisfy convergence criteria but reaching at the same time high levels of total change. 4.3 Discussion of the results 9 Automatic detection of the no-smoothing space of the optimized engine bracket: a detection of flat surfaces in yellow, b detection of 12 closed-loops where the circular holes appear in red. (Color figure online) Fig. 9 Automatic detection of the no-smoothing space of the optimized engine bracket: a detection of flat surfaces in yellow, b detection of 12 closed-loops where the circular holes appear in red. (Color figure online) completely unrecognizable. Just to provide some numeri- cal values, the classic Laplacian reduces the hole by 38% on average, while the SDU of more than 50%. For all the mentioned reasons, these 2 smoothing approaches are con- sidered not applicable in a real-life context to post-process complex geometries which come from TO analyses before manufacturing them. worse behaviour in the bracket case study. In general, this smoothing approach, highly appreciated in literature con- tributions, is fast, reaches convergence criteria but suffers from a sensitive amount of volume shrinkage (25% mean reduction between the two case studies) and the features are not preserved.f A different discussion involves the HC algorithm which always converges very rapidly but with small changes and no significant improvements on the final mesh geometry. For this reason, the volume shrinkage level and the feature degradations can be considered negligible. Classic Laplacian and SDU approaches poorly post-pro- cess the 3D models, while Taubin’s algorithm has a mid- field behaviour. On the one hand, it performs well in the cantilever beam example, but on the other hand, it shows 1 3 3467 Surface smoothing for topological optimized 3D models only 30% of nodes that belongs to the no-smoothing space, reflecting on a lower α input value to satisfy convergence Table 3 3 × 3 sensitivity matrix for α and β parameters of the HC- SDU algorithm for the GE bracket case study (λ = 0.6307. iter- max = 150). On the top, the 3 × 3 matrix of the mesh distance between the last two smoothing iterations; on the bottom, the 3 × 3 matrix with the total change of the overall model. The optimum input configura- tion is represented by the combination of input values giving the bold measure Total change β 0.5 0.51 0.56 α 0.18 0.1640 0.1595 0.1350 0.27 0.1230 0.0099 0.0090 0.36 0.0090 0.0087 0.0086 Total change β 0.5 0.51 0.56 α 0.18 878550 866824 805846 0.27 624070 607834 562232 0.36 383020 377251 342130 Fig. 4.3 Discussion of the results Focusing the attention on the smoothed results of the developed methodology, it can be said that it reaches con- vergence, reflecting on a faster smoothing process on the same order of magnitude of Taubin’s approach. For this reason, it is arguable that the developed functions of fea- ture detection and volume rescaling that force the algorithm during the iterations do not affect the computational cost. Looking at the values of the mesh volumes (Tables 2 and 4), the Optimized HC-SDU algorithm, thanks to the volume rescaling sub-routine, perfectly maintains the initial volume value. The developed algorithm behaves better than Taubin 1 A. Bacciaglia et al. 3468 Fig. 11 Smoothed models using a Laplacian smoothing, b Laplacian smoothing with SDU weights, c Taubin, d HC-algorithm and e Optimized HC-SDU Fig. 11 Smoothed models using a Laplacian smoothing, b Laplacian smoothing with SDU weights, c Taubin, d HC-algorithm and e Optimized HC-SDU Table 4 Dimensional and computational comparison of different smoothing algorithms applied on the bracket example (percentage error in round brackets) Original Laplacian SDU HC Taubin Optimized HC-SDU Volume [mm3] 117700 20940 (− 82%) 87420 (− 26%) 112600 (− 4%) 79402 (− 33%) 117700 (0%) 2 loaded Holes diameter [mm] 18 11.43 (− 37%) 19.73 (+ 10%) 17.44 (− 3%) 21.51 (+ 19%) 18 (0%) 4 Holes in the base diameter [mm] 6 Not evaluable 1.62 (− 73%) 5.88 (− 2%) 4.56 (− 24%) 6 (0%) Smoothing time [s] – 145 139 74 22 19 Table 4 Dimensional and computational comparison of different smoothing algorithms applied on the bracke round brackets) Table 4 Dimensional and computational comparison of different smoothing algorithms applied on the bracket example (percentage error in round brackets) Original Laplacian SDU HC Taubin Optimized HC-SDU Volume [mm3] 117700 20940 (− 82%) 87420 (− 26%) 112600 (− 4%) 79402 (− 33%) 117700 (0%) 2 loaded Holes diameter [mm] 18 11.43 (− 37%) 19.73 (+ 10%) 17.44 (− 3%) 21.51 (+ 19%) 18 (0%) 4 Holes in the base diameter [mm] 6 Not evaluable 1.62 (− 73%) 5.88 (− 2%) 4.56 (− 24%) 6 (0%) Smoothing time [s] – 145 139 74 22 19 Iterations – 150 150 40 91 94 Fig. 12 Comparison of a detailed view of the GE bracket mesh: a original model, b Taubin’s algorithm, c Taubin’s algorithm coupled with the no-smoothing-space detection, and d Optimized HC-SDU Fig. 4.3 Discussion of the results 12 Comparison of a detailed view of the GE bracket mesh: a original model, b Taubin’s algorithm, c Taubin’s algorithm coupled with the no-smoothing-space detection, and d Optimized HC-SDU 1 3 Surface smoothing for topological optimized 3D models 3469 when the task to maintain model features is addressed: the dimensions of the hole match perfectly between the original and the optimized model while Taubin’s algorithm shows a reduction in holes size (4% diameter reduction on the beam model, 24% on the bracket) which is an undesired effect in a real-life application. The connections between many components in a complex real-life assembly could be a straightforward example: if some holes used to connect components are modified in shape or reduced in diameter, the assembly can’t be completed and the object has to be discarded and re-designed. In other applications, holes may require further machining to respect GD&T (Geometrical Dimensioning and Tolerances) prescription where errors in diameter and more dramatically in the position may lead to discard the component. Optimized HC-SDU approach is the best one compared to the methods taken under consideration in this work where both high frequency and background noises are smoothed. A parameter that helps to summarize the results is the total change behaviour represented in Figs. 7 and 10. Its amount needs to be a trade-off between two diverging demands: a good-looking model (highly smoothed), and a model which does not collapse on itself (high shrinkage). Therefore, the ranking of the suitability of the algorithms for the smoothing of TO analysed components have been set by looking at the same time to the total change and the total volume plots: the best performances are characterized by the coupling of high total change (good smoothing) and low or null total volume decrease. Using this evaluation scale, the Optimized HC-SDU algorithm shows better performances compared to the other ones with Taubin’s algorithm that is not far away, even if small model degradation occurs. The developed algorithm maintains the initial mesh volume and preserves the features the designer would like to maintain still showing a high total change value. The Scale-Dependent Laplacian and the classic Laplacian obtain high values of total change and a smoothed external surface, but as evident from Figs. 6 and 11, the models collapsed on themselves and the shrinkage effect is dramatic. 4.3 Discussion of the results The automatic detection procedure for holes proved to be reliable and useful to detect features to keep unsmoothed without human intervention. diameter and more dramatically in the position may lead discard the component. From the previous results and comparison, it can be sa that Taubin and Optimized HC-SDU algorithms can be con sidered the two best approaches considered in this researc Looking closely at a particular region of the GE brack smoothed model (Fig. 12), it can be said that both algorithm perform roughly similarly, with the biggest mesh peaks an surface noise that is mitigated compared to the original mes (Fig. 12a). However, as previously said, Taubin’s approac does not preserve the no-smoothing space as the flat surface and the dimensions of the holes (Fig. 12b). Referring to th developed methodology, the reader’s concern may be th it is difficult to assess if the main improvement is due t the application of the detect_flat_surface and detect_holes edges functions before the surface smoothing or the actu Optimized HC-SDU algorithm itself in its completenes However, this doubt can be easily solved because, as can b seen in Fig. 12c, Taubin’s algorithm coupled with the tw developed functions to isolate the no-smoothing space, doe not produce satisfactory results. Moreover, thanks to add tional simulations that are not included here for the sake o brevity, it was noticed that the modified version of Taubin slower and does not reach the levels of Total Change reache by the Optimized HC-SDU approach. Due to brevity, on Taubin’s approach coupled with the developed functions detect the no-smoothing-space is shown, but similar discu sions could be applied to the other methodologies compare in this research. The final geometry is distorted and muc worse than the overall developed methodology (Fig. 12d Thanks to these sets of results it can be noticed that th Table 5 Structure compliance comparison and % error estimatio between noisy and smooth mesh for the GE bracket case study Noisy mesh Smooth me Structure compliance [N∙mm] 1.9239e + 09 1.9593e + 0 Error [%] – + 1.8410 From the previous results and comparison, it can be said that Taubin and Optimized HC-SDU algorithms can be con- sidered the two best approaches considered in this research. Looking closely at a particular region of the GE bracket smoothed model (Fig. 4.3 Discussion of the results 12), it can be said that both algorithms perform roughly similarly, with the biggest mesh peaks and surface noise that is mitigated compared to the original mesh (Fig. 12a). However, as previously said, Taubin’s approach does not preserve the no-smoothing space as the flat surfaces and the dimensions of the holes (Fig. 12b). Referring to the developed methodology, the reader’s concern may be that it is difficult to assess if the main improvement is due to the application of the detect_flat_surface and detect_holes_ edges functions before the surface smoothing or the actual Optimized HC-SDU algorithm itself in its completeness. However, this doubt can be easily solved because, as can be seen in Fig. 12c, Taubin’s algorithm coupled with the two developed functions to isolate the no-smoothing space, does not produce satisfactory results. Moreover, thanks to addi- tional simulations that are not included here for the sake of brevity, it was noticed that the modified version of Taubin is slower and does not reach the levels of Total Change reached by the Optimized HC-SDU approach. Due to brevity, only Taubin’s approach coupled with the developed functions to detect the no-smoothing-space is shown, but similar discus- sions could be applied to the other methodologies compared in this research. The final geometry is distorted and much worse than the overall developed methodology (Fig. 12d). Thanks to these sets of results it can be noticed that the As mentioned in this section, the performances of the innovative methodology seem satisfactory from the denois- ing point of view. However, a detailed discussion should focus on how the smoothing process impacts the structure performances in loading conditions, such as the final struc- ture compliance and how the smoothing post-processing could be optimized to limit the impact on structure perfor- mance changes. This point will be analysed in detail in fur- ther researches. To provide the reader with some numerical data, the compliance of the smoothed structure of the GE bracket has been computed and compared to the compliance value of the corresponding noisy model. Let c be the com- pliance of the structure, U the nodal generalized displace- ment, K the global stiffness matrix and F is the matrix of the nodal generalized external forces, 휌e the element density and 퐊퐞 the element stiffness matrix, following the approach described in (Costa et al. 4.3 Discussion of the results 2018), c can be computed with: (8) ⎧ ⎪ ⎨ ⎪⎩ c = 퐔T퐊퐔, with 퐊= Ne∑ i=1 휌e퐊퐞 퐊퐔= 퐅 (8) Table 5 Structure compliance comparison and % error estimation between noisy and smooth mesh for the GE bracket case study Noisy mesh Smooth mesh Structure compliance [N∙mm] 1.9239e + 09 1.9593e + 09 Error [%] – + 1.8410 Table 5 Structure compliance comparison and % error estimation between noisy and smooth mesh for the GE bracket case study Noisy mesh Smooth mesh Structure compliance [N∙mm] 1.9239e + 09 1.9593e + 09 Error [%] – + 1.8410 Table 5 Structure compliance comparison and % error estimation between noisy and smooth mesh for the GE bracket case study Combining both Eq. 8, it can be seen that the compli- ance depends on the displacement field and the nodal exter- nal forces. Assuming a mesh with n nodes, both U and F have dimensions [ nx3 ]; this implies that c has dimensions 1 3470 A. Bacciaglia et al. drifting. Moreover, a volume rescaling formula is used to resize the mesh at each iteration to maintain the initial vol- ume of the digital volume; this is done to avoid the shrinkage effect which affects the majority of available algorithms. Besides, two functions are developed to automatically rec- ognize features like holes and planar surfaces which should be maintained during the smoothing process to be present either in the final digital model. Two case studies included in the work show the efficiency of the developed method- ology compared to existent ones. The results suggest that the main goals that pushed the algorithm design have been addressed properly. Specifically, volume shrinkage is com- pletely deleted during the process with a satisfactory total change value and computational time compared to the other approaches. Moreover, features that should be frozen during the process are perfectly recognized and maintained in the final 3D models. [3 × 3]. However, just to have a scalar and comparable value, the same approach used in the Top3D software, developed by (Liu and Tovar 2014) is used, where the 9 elements are summed up. Therefore, it is possible to estimate the overall structure compliance values which are reported in Table 5: the dis- placement values available from the topology optimization analysis have been used, together with the load conditions already described for the GE bracket. 4.3 Discussion of the results In the following, the structure compliance of the noisy mesh is assumed as the benchmarking value, since it comes directly from the topol- ogy optimization analysis. Indeed, structure compliance is the fitness function which the SIMP TO process minimizes during the optimization runs. After these computations, it is possible to compare the structural performance of the smoothed structure to the noisy 3D model: from Table 5, it can be seen that the overall compliance increases, going in the opposite direction compared to the aim of the topol- ogy optimization. However, the increment is limited to 2%, which is very close to the value of the unsmoothed part. On the one hand, the structure compliance slightly increases, but on the other hand, the smoothing approach returns a 3D model based on the results of the TO which is ready to be manufactured without the need to model from scratch in CAD software the optimized component. This accelerates the design-to-manufacturing cycle and reduces the design- er’s workload, even if a small approximation in terms of compliance should be accepted (Table 5). i In the future, the methodology will be improved to detect every type of orientation of the holes: up to now only holes perpendicular to the body surface can be recognized. For the preliminary study, the input parameters are set after a trial and error approach and sensitivity matrices are built to show the optimum input setup in terms of scalar coefficients for the HC algorithm. The diffusion speed and the maximum number of iterations are chosen following literature sug- gestions and a good trade-off between computational cost and accuracy of the smoothed model. In further studies, the convergence parameters as well as the mesh distance and the maximum number of iterations will be changed to study how these parameters may affect the final result. Moreo- ver, performance studies will be carried out to compare the structure compliance between the optimized noisy model and the smoothed version. Additionally, it’s straightforward to note that the proposed rescaling scheme may not preserve orientation near blocked nodes even if this has not been an issue in the case studies shown in this research; further stud- ies with different geometries must better address this issue. 4.3 Discussion of the results Supplementary studies should be carried out involving both the developed algorithm, the Laplace–Beltrami operator, the methodologies which consider local curvatures and the iso- surface approaches to investigate the results and compare the efficiency of the algorithms. To sum up, the methodology does not require complex mathematical operations and could be easily integrated into commercial Topology Optimization software to improve the final results. The Optimized HC-SDU algorithm requires the setting of four parameters, but average values could be used as a default: in this way, the algorithm shows encouraging results and the end-user is not required to do a lot of tests to set the values. References Krish S (2011) A practical generative design method. Comput Aided Des 43(1):88–100. https://doi.org/10.1016/j.cad.2010.09.009 Li Q, Steven GP, Xie YM (2001) A simple checkerboard suppression algorithm for evolutionary structural optimization. Struct Multi- discip Optim 22(3):230–239. https://doi.org/10.1007/s001580100 140 Bacciaglia A, Ceruti A, Liverani A (2019) A systematic review of vox- elization method in additive manufacturing. Mech Ind 20(6):630. https://doi.org/10.1051/meca/2019058 Bacciaglia A, Ceruti A, Liveran A (2020) Additive Manufacturing Challenges and Future Developments in the Next Ten Years. In: Rizzi C, Oreste Andrisano A, Leali F, Gherardini F, Pini F, Verg- nano A (eds) Design tools and methods in industrial engineer- ing, In: Caterina Rizzi, Angelo Oreste Andrisano. Lecture Notes in Mechanical Engineering. . Springer International Publishing, Cham, pp 891–902 Liu K, Tovar A (2014) An efficient 3D topology optimization code written in matlab. Struct Multidiscip Optim 50(6):1175–1196. https://doi.org/10.1007/s00158-014-1107-x Machado G, Trabucho L (2004) Some results in topology optimization applied to biomechanics. Comput Struct 82(17–19):1389–1397. https://doi.org/10.1016/j.compstruc.2004.03.034 Mantovani S, Barbieri S, Giacopini M, Croce A, Sola A, Bassoli E (2020) Synergy between Topology optimization and additive manufacturing in the automotive field. Proc Inst Mech Eng Part B. https://doi.org/10.1177/0954405420949209i pp Belyaev A, Ohtake Y (2003) A comparison of mesh smoothing meth- ods. In:Israel-Korea Bi-national conference on geometric mod- eling and computer graphics, pp 83–87. Tel Aviv University, Tel Aviv Zhang M, Gunturk BK (2008) Multiresolution bilateral filtering for image denoising. IEEE Trans Image Process 17(12):2324–2333. https://doi.org/10.1109/TIP.2008.2006658i Bendsøe MP (1989) Optimal shape design as a material distribution problem. Struct Optim 1(4):193–202. https://doi.org/10.1007/ BF01650949 Ohsaki M (2011) Optimization of finite dimensional structures. CRC Press/Taylor & Francis, Boca Raton Bendsøe MP, Sigmund O (2011) Topology optimization: theory, meth- ods, and applications. 2nd edn, Corrected printing. Engineering Online Library. Springer, Berlin Qu X, Stucker B (2005) Circular hole recognition for STL-based tool- path generation. Rapid Prototyp J 11(3):132–139. https://doi.org/ 10.1108/13552540510601255 Bourdin B (2001) Filters in topology optimization. Int J Numer Meth Eng 50(9):2143–2158. https://doi.org/10.1002/nme.116 Rezaie R, Badrossamay M, Ghaie A, Moosavi H (2013) Topology opti- mization for fused deposition modeling process. Procedia CIRP 6:521–526. https://doi.org/10.1016/j.procir.2013.03.098 Buades A, Coll B, Morel JM (2005) A review of image denoising algorithms, with a new one. Multisc Model Simul 4(2):490–530. https://doi.org/10.1137/040616024 Savio G, Meneghello R, Concheri G (2019) Design of variable thickness triply periodic surfaces for additive manufacturing. Progress Addit Manuf 4(3):281–290. https://doi.org/10.1007/ s40964-019-00073-x Costa G, Montemurro M, Pailhès J (2018) A 2D topology optimisa- tion algorithm in NURBS framework with geometric constraints. Conflict of interest The authors declare no conflict of interest. Conflict of interest The authors declare no conflict of interest. Gao W, Zhang Y, Ramanujan D, Ramani K, Chen Y, Williams CB, Wang CCL, Shin YC, Zhang S, Zavattieri PD (2015) The status, challenges, and future of additive manufacturing in engineering. Comput Aided Des 69(12):65–89. https://doi.org/10.1016/j.cad. 2015.04.001 Replication of results All the information required to replicate the results found in Sects. 3–4 is fully disclosed in this paper. The same results will be obtained by implementing the equations and process presented here. The algorithms discussed in this paper are implemented in Matlab and Python by the authors, which cannot be shared due to further researches. Gaynor AT, Meisel NA, Williams CB, Guest JK (2014) Multiple- material topology optimization of compliant mechanisms cre- ated via polyjet three-dimensional printing. J Manuf Sci Eng 136(6):061015. https://doi.org/10.1115/1.4028439 ‘GE Jet Engine Bracket Challenge’. n.d. Accessed 23 April 2020. https://grabcad.com/challenges/ge-jet-engine-bracket-challenge. 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/. Gibson I, Rosen D, Stucker B (2015) Additive manufacturing technolo- gies. Springer, New York Gostler A (2015) Denoising medical surface meshes for 3D-printing. Bachelor Thesis, TU Wien-Faculty of Informatics, Institute of Visual Computing & Human-Centered Technology. https://www. cg.tuwien.ac.at/research/publications/2015/gostler-2015-3dp/gostl er-2015-3dp-Thesis.pdf Jense GJ (1989) Voxel-based methods for CAD. Comput Aided Des 21(8):528–533. https://doi.org/10.1016/0010-4485(89)90061-4 Lee KW, Wang WP (2005) Feature-preserving mesh denoising via bilateral normal filtering. In: Ninth international conference on computer aided design and computer graphics (CAD-CG’05). pp 275–80 IEEE, Hong Kong 5 Conclusions Mesh post-processing techniques are crucial in the design- to-manufacturing cycle especially in those contexts where TO design methodologies are employed. Indeed, the result- ing geometry is far from being manufacturable as it is and for this reason, smoothing algorithms need to be applied to return a more appealing external surface. Several procedures are already available in literature, however, a large part of them show important weaknesses as volume shrinkage and loss of model features when applied, especially for vertex- based approaches. To bridge the gap, the authors developed a smoothing methodology based on the HC-Laplacian one. At first, the Scale-Dependent-Umbrella weights are used instead of the classic Laplacian ones to decrease the vertex fi To sum up, thanks to the proposed results, it is possible to say that the Optimized HC-SDU method is suitable for implementation in commercial software and could help in obtaining structures ready for Additive Manufacturing just as the output of the Topology Optimization process. In this way, the painstakingly CAD sketching from scratch of the optimized component where the output of TO is imitated is avoided, with a strong gain in time to market, reduction of operator efforts, and increase in precision. 3 3 3471 Surface smoothing for topological optimized 3D models Funding This research did not receive any specific grant from fund- ing agencies in the public, commercial, or not-for-profit sectors. Open access funding provided by Alma Mater Studiorum - Università di Bologna within the CRUI-CARE Agreement. Costa G, Montemurro M, Pailhès J (2021) NURBS hyper-surfaces for 3D topology optimization problems. Mech Adv Mater Struct 28(7):665–684. https://doi.org/10.1080/15376494.2019.1582826 Desbrun, Mathieu, Mark Meyer, Peter Schröder, and Alan H. Barr. 1999. Implicit fairing of irregular meshes using diffusion and curvature flow. In: Proceedings of the 26th Annual Conference on Computer Graphics and Interactive Techniques - SIGGRAPH ’99, pp 317–24. ACM Press, Not Known Declarations Conflict of interest The authors declare no conflict of interest. Wong J, Ryan L, Kim IY (2018) Design optimization of aircraft landing gear assembly under dynamic loading. Struct Multidiscip Optim 57(3):1357–1375. https://doi.org/10.1007/s00158-017-1817-y Xie YM, Steven GP (1996) Evolutionary structural optimization for dynamic problems. Comput Struct 58(6):1067–1073. https://doi. org/10.1016/0045-7949(95)00235-9 Zegard T, Paulino GH (2016) Bridging topology optimization and additive manufacturing. Struct Multidiscip Optim 53(1):175–192. https://doi.org/10.1007/s00158-015-1274-4 References Int J Mech Mater Des 14(4):669–696. https://doi.org/10.1007/ s10999-017-9396-z 1 3 3472 A. Bacciaglia et al. Wong J, Ryan L, Kim IY (2018) Design optimization of aircraft landing gear assembly under dynamic loading. Struct Multidiscip Optim 57(3):1357–1375. https://doi.org/10.1007/s00158-017-1817-y Sigmund O (1997) On the design of compliant mechanisms using topology optimization*. Mech Struct Mach 25(4):493–524. https://doi.org/10.1080/08905459708945415 ( ) p g y Xie YM, Steven GP (1996) Evolutionary structural optimization for dynamic problems. Comput Struct 58(6):1067–1073. https://doi. org/10.1016/0045-7949(95)00235-9 p g Sorkine O (2005) Laplacian mesh processing. Eurographics 2005-state of the art reports, 18 p. https://doi.org/10.2312/EGST.20051044. p p p g Taubin G (1995) a signal processing approach to fair surface design. In: Proceedings of the 22nd annual conference on computer graphics and interactive techniques - SIGGRAPH ’95. pp 351–58. ACM Press g Zegard T, Paulino GH (2016) Bridging topology optimization and additive manufacturing. Struct Multidiscip Optim 53(1):175–192. https://doi.org/10.1007/s00158-015-1274-4 q pp Vollmer J, Mencl R, Muller H (1999) Improved laplacian smoothing of noisy surface meshes. Comput Graph Forum 18(3):131–138. https://doi.org/10.1111/1467-8659.00334s Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Wei M, Huang J, Xie X, Liu L, Wang J, Qin J (2019) Mesh denoising guided by patch normal co-filtering via kernel low-rank recovery. IEEE Trans Visual Comput Graph 25(10):2910–2926. https://doi. org/10.1109/TVCG.2018.2865363 1 3 3
https://openalex.org/W4395002516	https://bmcmededuc.biomedcentral.com/counter/pdf/10.1186/s12909-024-05365-7	English	null	Artificial intelligence and medical education: application in classroom instruction and student assessment using a pharmacology & therapeutics case study	BMC medical education	2,024	cc-by	11,097	RESEARCH Open Access (2024) 24:431 (2024) 24:431 BMC Medical Education Sridharan and Sequeira BMC Medical Education (2024) 24:431 https://doi.org/10.1186/s12909-024-05365-7 Sridharan and Sequeira BMC Medical Education https://doi.org/10.1186/s12909-024-05365-7 © The Author(s) 2024. 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. Abstract Background Artificial intelligence (AI) tools are designed to create or generate content from their trained parameters using an online conversational interface. AI has opened new avenues in redefining the role boundaries of teachers and learners and has the potential to impact the teaching-learning process. Methods In this descriptive proof-of- concept cross-sectional study we have explored the application of three generative AI tools on drug treatment of hypertension theme to generate: (1) specific learning outcomes (SLOs); (2) test items (MCQs- A type and case cluster; SAQs; OSPE); (3) test standard-setting parameters for medical students. Results Analysis of AI-generated output showed profound homology but divergence in quality and responsiveness to refining search queries. The SLOs identified key domains of antihypertensive pharmacology and therapeutics relevant to stages of the medical program, stated with appropriate action verbs as per Bloom’s taxonomy. Test items often had clinical vignettes aligned with the key domain stated in search queries. Some test items related to A-type MCQs had construction defects, multiple correct answers, and dubious appropriateness to the learner’s stage. ChatGPT generated explanations for test items, this enhancing usefulness to support self-study by learners. Integrated case-cluster items had focused clinical case description vignettes, integration across disciplines, and targeted higher levels of competencies. The response of AI tools on standard-setting varied. Individual questions for each SAQ clinical scenario were mostly open-ended. The AI-generated OSPE test items were appropriate for the learner’s stage and identified relevant pharmacotherapeutic issues. The model answers supplied for both SAQs and OSPEs can aid course instructors in planning classroom lessons, identifying suitable instructional methods, establishing rubrics for grading, and for learners as a study guide. Key lessons learnt for improving AI-generated test item quality are outlined. Conclusions AI tools are useful adjuncts to plan instructional methods, identify themes for test blueprinting, generate test items, and guide test standard-setting appropriate to learners’ stage in the medical program. However, experts need to review the content validity of AI-generated output. We expect AIs to influence the medical education Full list of author information is available at the end of the article Full list of author information is available at the end of the article Kannan Sridharan1* and Reginald P. Sequeira1 Kannan Sridharan1* and Reginald P. Sequeira1 Study design and ethicsh The present study is observational, cross-sectional in design, conducted in the Department of Pharmacology & Therapeutics, College of Medicine and Medical Sciences, Arabian Gulf University, Kingdom of Bahrain, between April and August 2023. Ethical Committee approval was not sought given the nature of this study that neither had any interaction with humans, nor collection of any per sonal data was involved. Pharmacology and Therapeutics (P & T) is a core dis cipline embedded in the undergraduate medical cur riculum, mostly in the pre-clerkship phase. However, the application of therapeutic principles forms one of the key learning objectives during the clerkship phase of the undergraduate medical career. Student assess ment in pharmacology & therapeutics (P&T) is with test items such as multiple-choice questions (MCQs), inte grated case cluster questions, short answer questions (SAQs), and objective structured practical examination (OSPE) in the undergraduate medical curriculum. It has been argued that AIs possess the ability to communicate an idea more creatively than humans [7]. It is impera tive that with access to billions of trillions of datasets the AI platforms hold promise in playing a crucial role in the conception of various test items related to any of the disciplines in the undergraduate medical curricu lum. Additionally, AIs provide an optimized curriculum for a program/course/topic addressing multidimensional problems [8], although robust evidence for this claim is lacking.h Backgroundi analytics, lack of high-quality firm evidence favoring the utility of AIs in medical education, and lack of funding [9]. Open-access AI platforms are available free to users without any restrictions. Hence, as a proof-of-concept, we chose to explore the utility of three AI platforms to identify specific learning objectives (SLOs) related to pharmacology discipline in the management of hyper tension for medical students at different stages of their medical training. Artificial intelligence (AI) has great potential to revo lutionize the field of medical education from curricular conception to assessment [1]. AIs used in medical edu cation are mostly generative AI large language mod els that were developed and validated based on billions to trillions of parameters [2]. AIs hold promise in the incorporation of history-taking, assessment, diagnosis, and management of various disorders [3]. While applica tions of AIs in undergraduate medical training are being explored, huge ethical challenges remain in terms of data collection, maintaining anonymity, consent, and owner ship of the provided data [4]. AIs hold a promising role amongst learners because they can deliver a personal ized learning experience by tracking their progress and providing real-time feedback, thereby enhancing their understanding in the areas they are finding difficult [5]. Consequently, a recent survey has shown that medical students have expressed their interest in acquiring com petencies related to the use of AIs in healthcare during their undergraduate medical training [6].h © The Author(s) 2024. 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. Sridharan and Sequeira BMC Medical Education (2024) 24:431 (2024) 24:431 Page 2 of 13 Sridharan and Sequeira BMC Medical Education landscape to empower learners, and to align competencies with curriculum implementation. AI literacy is an essential competency for health professionals. Keywords Medical education, Pharmacology, Therapeutics, Assessment, SLOs, OSPE, MCQs Study procedure We conducted the present study in May-June 2023 with the Poe© chatbot interface created by Quora© that pro vides access to the following three AI platforms: • Sage Poe [10]: A generative AI search engine developed by Anthropic© that conceives a response based on the written input provided. Quora has renamed Sage Poe as Assistant© from July 2023 onwards. • Claude-Instant [11]: A retrieval-based AI search engine developed by Anthropic© that collates a response based on pre-written responses amongst the existing databases. g • ChatGPT version 3.5 [12]: A generative architecture- based AI search engine developed by OpenAI© trained on large and diverse datasets. We queried the chatbots to generate SLOs, A-type MCQs, integrated case cluster MCQs, integrated SAQs, and OSPE test items in the domain of systemic hyper tension related to the P&T discipline. Separate prompts were used to generate outputs for pre-clerkship (preclini cal) phase students, and at the time of graduation (before starting residency programs). Additionally, we have also evaluated the ability of these AI platforms to estimate the proportion of students correctly answering these test The existing literature has evaluated the knowledge, attitude, and perceptions of adopting AI in medical edu cation. Integration of AIs in medical education is the need of the hour in all health professional education. However, the academic medical fraternity facing chal lenges in the incorporation of AIs in the medical curricu lum due to factors such as inadequate grounding in data Page 3 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 (2024) 24:431 Sridharan and Sequeira BMC Medical Education items. We used the following queries for each of these objectives: Integrated case cluster MCQs I. Write 20 integrated case cluster MCQs with 2 questions in each cluster with 5 choices for undergraduate medical students during the pre- clerkship phase integrating pharmacology and physiology related to systemic hypertension with a case vignette. Specific learning objectives I. Can you generate specific learning objectives in the pharmacology discipline relevant to undergraduate medical students during their pre-clerkship phase related to anti-hypertensive drugs? II. Write 20 integrated case cluster MCQs with 2 questions in each cluster with 5 choices for undergraduate medical students during the pre- clerkship phase integrating pharmacology and physiology related to systemic hypertension with a case vignette. Please do not include ‘none of the above’ as the choice. (This modified search query was used because test items with ‘None of the above’ option were generated with the previous search query). II. Can you generate specific learning objectives in the pharmacology discipline relevant to undergraduate medical students at the time of graduation related to anti-hypertensive drugs? A-type MCQs In the initial query used for A-type of item, we specified the domains (such as the mechanism of action, pharma cokinetics, adverse reactions, and indications) so that a sample of test items generated without any theme-related clutter, shown below: III. Write 20 integrated case cluster MCQs with 2 questions in each cluster with 5 choices for undergraduate medical students at the time of graduation integrating pharmacology and physiology related to systemic hypertension with a case vignette. I. Write 20 single best answer MCQs with 5 choices related to anti-hypertensive drugs for undergraduate medical students during the pre-clerkship phase of which 5 MCQs should be related to mechanism of action, 5 MCQs related to pharmacokinetics, 5 MCQs related to adverse reactions, and 5 MCQs should be related to indications. Specific learning objectivesh The pre-clerkship phase SLOs identified by Sage Poe, Claude-Instant, and ChatGPT are listed in the electronic supplementary materials 1–3, respectively. In general, a broad homology in SLOs generated by the three AI plat forms was observed. All AI platforms identified appro priate action verbs as per Bloom’s taxonomy to state the SLO; action verbs such as describe, explain, recognize, discuss, identify, recommend, and interpret are used to state the learning outcome. The specific, measurable, achievable, relevant, time-bound (SMART) SLOs gener ated by each AI platform slightly varied. All key domains of antihypertensive pharmacology to be achieved during the pre-clerkship (pre-clinical) years were relevant for graduating doctors. The SLOs addressed current JNC Treatment Guidelines recommended classes of antihy pertensive drugs, the mechanism of action, pharmaco kinetics, adverse effects, indications/contraindications, dosage adjustments, monitoring therapy, and principles of monotherapy and combination therapy.h II. Can you generate 5 OSPE pharmacology and therapeutics prescription writing exercises containing appropriate instructions for the patients for the assessment of undergraduate medical students during their pre-clerkship phase related to anti-hypertensive drugs? III. Can you generate 5 OSPE pharmacology and therapeutics prescription writing exercises containing appropriate instructions for the patients for the assessment of undergraduate medical students at the time of graduation related to anti- hypertensive drugs? Both authors independently evaluated the AI-generated outputs, and a consensus was reached. We cross-checked the veracity of answers suggested by AIs as per the Joint National Commission Guidelines (JNC-8) and Goodman and Gilman’s The Pharmacological Basis of Therapeutics (2023), a reference textbook [13, 14]. Errors in the A-type Both authors independently evaluated the AI-generated outputs, and a consensus was reached. We cross-checked the veracity of answers suggested by AIs as per the Joint National Commission Guidelines (JNC-8) and Goodman and Gilman’s The Pharmacological Basis of Therapeutics (2023), a reference textbook [13, 14]. Errors in the A-type The SLOs to be achieved by undergraduate medi cal students at the time of graduation identified by Sage Poe, Claude-Instant, and ChatGPT listed in electronic supplementary materials 4–6, respectively. The identi fied SLOs emphasize the application of pharmacology knowledge within a clinical context, focusing on compe tencies needed to function independently in early resi dency stages. These SLOs go beyond knowledge recall and mechanisms of action to encompass competencies related to clinical problem-solving, rational prescrib ing, and holistic patient management. Integrated short answer questions I. Write a short answer question scenario with difficult questions based on the theme of a newly diagnosed hypertensive patient for undergraduate medical students with the main objectives related to the physiology of blood pressure regulation, risk factors for systemic hypertension, pathophysiology of systemic hypertension, pathological changes in the systemic blood vessels in hypertension, pharmacological management, and non- pharmacological treatment of systemic hypertension. The MCQs generated with the above search query were not based on clinical vignettes. We queried again to gen erate MCQs using clinical vignettes specifically because most medical schools have adopted problem-based learning (PBL) in their medical curriculum. II. Write 20 single best answer MCQs with 5 choices related to anti-hypertensive drugs for undergraduate medical students during the pre-clerkship phase using a clinical vignette for each MCQ of which 5 MCQs should be related to the mechanism of action, 5 MCQs related to pharmacokinetics, 5 MCQs related to adverse reactions, and 5 MCQs should be related to indications. II. II. Write a short answer question scenario with moderately difficult questions based on the theme of a newly diagnosed hypertensive patient for undergraduate medical students with the main objectives related to the physiology of blood pressure regulation, risk factors for systemic hypertension, pathophysiology of systemic hypertension, pathological changes in the systemic blood vessels in hypertension, pharmacological management, and non-pharmacological treatment of systemic hypertension. We attempted to explore whether AI platforms can pro vide useful guidance on standard-setting. Hence, we used the following search query. III. Write a short answer question scenario with questions based on the theme of a newly diagnosed hypertensive patient for undergraduate medical students at the time of graduation with the main objectives related to the physiology of blood pressure regulation, risk factors for systemic hypertension, III. Can you do a simulation with 100 undergraduate medical students to take the above questions and let me know what percentage of students got each MCQ correct? Sridharan and Sequeira BMC Medical Education (2024) 24:431 Page 4 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 pathophysiology of systemic hypertension, pathological changes in the systemic blood vessels in hypertension, pharmacological management, and non-pharmacological treatment of systemic hypertension. pathophysiology of systemic hypertension, pathological changes in the systemic blood vessels in hypertension, pharmacological management, and non-pharmacological treatment of systemic hypertension. MCQs were categorized as item construction defects, multiple correct answers, and uncertain appropriateness to the learner’s level. Specific learning objectivesh The SLOs gener ated require higher cognitive ability of the learner: action verbs such as demonstrate, apply, evaluate, analyze, develop, justify, recommend, interpret, manage, adjust, educate, refer, design, initiate & titrate were frequently used. Table 1 Preliminary conceptual framework for establishing content validity of AI-generated test items Table 1 Preliminary conceptual framework for establishing content validity of AI-generated test items Table 1 Preliminary conceptual framework for establishing content validity of AI-generated test items Domains Items assessed Technical • Are the test items/explanation technically accurate and free from empirical or clinical mistakes? Comprehensiveness • Do the test items /explanations suffi ciently address relevant topics/subtopics? • Is the within-topic variation (range of examples, patient characteristics, scenario descriptions) at the desired level? Education level • Are the test items /model answers appro priate to the education level of the learner? • Is the structure of the questions/explana tions aligned to learning outcomes (as per Bloom’s taxonomy)? Free of construction defects • Are the test items /answers framed in a way to present a clear “best response” with appropriate and unambiguous distractors? • Does the test item/explanation avoid therapeutic controversies. • Are the test items integrated with the case vignette (without being standalone)? Table 1 Preliminary conceptual framework for establishing content validity of AI-generated test items Domains Items assessed Technical • Are the test items/explanation technically accurate and free from empirical or clinical mistakes? Comprehensiveness • Do the test items /explanations suffi ciently address relevant topics/subtopics? • Is the within-topic variation (range of examples, patient characteristics, scenario descriptions) at the desired level? Education level • Are the test items /model answers appro priate to the education level of the learner? • Is the structure of the questions/explana tions aligned to learning outcomes (as per Bloom’s taxonomy)? Free of construction defects • Are the test items /answers framed in a way to present a clear “best response” with appropriate and unambiguous distractors? • Does the test item/explanation avoid therapeutic controversies. • Are the test items integrated with the case vignette (without being standalone)? OSPEs I. Can you generate 5 OSPE pharmacology and therapeutics prescription writing exercises for the assessment of undergraduate medical students at the time of graduation related to anti-hypertensive drugs? Integrated short answer questions Test items in the integrated case cluster MCQs, SAQs and OSPEs were evaluated with the Preliminary Conceptual Framework for Establishing Content Validity of AI-Generated Test Items based on the following domains: technical accuracy, comprehen siveness, education level, and lack of construction defects (Table 1). The responses were categorized as complete and deficient for each domain. and non-pharmacological treatm hypertension. OSPEs I. Can you generate 5 OSPE pharm therapeutics prescription writing assessment of undergraduate me the time of graduation related to drugs? II. Can you generate 5 OSPE pharm therapeutics prescription writing containing appropriate instructio for the assessment of undergradu students during their pre-clerksh anti-hypertensive drugs? III. Can you generate 5 OSPE pharm therapeutics prescription writing containing appropriate instructio for the assessment of undergradu students at the time of graduatio hypertensive drugs? Both authors independently evaluate outputs, and a consensus was reached the veracity of answers suggested by National Commission Guidelines (JN and Gilman’s The Pharmacological B (2023), a reference textbook [13, 14]. Table 1 Preliminary conceptual framewor content validity of AI-generated test items Domains Items assessed Technical • Are the test items accurate and free f mistakes? Comprehensiveness • Do the test items ciently address rele • Is the within-topic examples, patient c descriptions) at the Education level • Are the test items priate to the educa • Is the structure of tions aligned to lea Bloom’s taxonomy) Free of construction defects • Are the test items way to present a cl appropriate and un • Does the test item therapeutic contro • Are the test items case vignette (with A-type MCQsh It is important to align student assessment to the cur riculum; in the PBL curriculum, MCQs with a clinical vignette are preferred. The modification of the query specifying the search to generate MCQs with a clinical vignette on domains specified previously gave appropri ate output by all three AI platforms evaluated (Sage Poe; Claude- Instant; Chat GPT). The scenarios generated had a good clinical fidelity and educational fit for the pre- clerkship student perspective.h The errors observed with AI outputs on the A-type MCQs are summarized in Table 2. No significant pattern was observed except that Claude-Instant© generated test items in a stereotyped format such as the same choices for all test items related to pharmacokinetics and indica tions, and all the test items in the ADR domain are linked to the mechanisms of action of drugs. This illustrates the importance of reviewing AI-generated test items by content experts for content validity to ensure alignment with evidence-based medicine and up-to-date treatment guidelines.h Almost similar themes under each domain were identi fied by the Claude-Instant AI platform with few notable exceptions: hydrochlorothiazide (instead of clonidine) in MOA and pharmacokinetics domains, respectively; under the ADR domain ankle edema/ amlodipine, sexual dysfunction and fatigue in male due to alpha-1 receptor blocker; under clinical indications the best initial mono therapy for clinical scenarios such as a 55-year old male with Stage-2 hypertension; a 75-year-old man Stage 1 hypertension; a 35-year-old man with Stage I hyperten sion working on night shifts; and a 40-year-old man with stage 1 hypertension and hyperlipidemia. g The test items generated by ChatGPT had the advan tage of explanations supplied rendering these more useful for learners to support self-study. The following examples illustrate this assertion: “A patient with hyper tension is started on a medication that works by blocking beta-1 receptors in the heart (metoprolol)”. Metoprolol is a beta blocker that works by blocking beta-1 receptors in the heart, which reduces heart rate and cardiac output, resulting in a decrease in blood pressure. However, this explanation is incomplete because there is no mention of other less important mechanisms, of beta receptor block ers on renin release. Also, these MCQs were mostly recall type: Which of the following medications is known to have a significant first-pass effect? A-type MCQsh The MCQs for the pre-clerkship phase identified by Sage Poe, Claude-Instant, and ChatGPT listed in the electronic supplementary materials 7–9, respectively, and those identified with the search query based on the • Are the test items integrated with the case vignette (without being standalone)? Sridharan and Sequeira BMC Medical Education (2024) 24:431 Page 5 of 13 (2024) 24:431 Sridharan and Sequeira BMC Medical Education clinical vignette in electronic supplementary materials (10–12). is also associated with constipation, albeit to a lesser extent, compared to verapamil. In the clinical indication domain, the case description asking “most commonly used in the treatment of hypertension and heart failure” is controversial because the options listed included losar tan, ramipril, and hydrochlorothiazide but the suggested correct answer was ramipril. This is a good example to stress the importance of vetting the AI-generated MCQ by experts for content validity and to assure robust psy chometrics. The MCQ on the most used drug in the treatment of “hypertension and diabetic nephropathy” is more explicit as opposed to “hypertension and diabetes” by Claude-Instant because the therapeutic concept of reducing or delaying nephropathy must be distinguished from prevention of nephropathy, although either an ACEI or ARB is the drug of choice for both indications. ( ) All MCQs generated by the AIs in each of the four domains specified [mechanism of action (MOA); phar macokinetics; adverse drug reactions (ADRs), and indi cations for antihypertensive drugs] are quality test items with potential content validity. The test items on MOA generated by Sage Poe included themes such as renin- angiotensin-aldosterone (RAAS) system, beta-adren ergic blockers (BB), calcium channel blockers (CCB), potassium channel openers, and centrally acting anti hypertensives; on pharmacokinetics included high oral bioavailability/metabolism in liver [angiotensin recep tor blocker (ARB)-losartan], long half-life and renal elimination [angiotensin converting enzyme inhibitors (ACEI)-lisinopril], metabolism by both liver and kid ney (beta-blocker (BB)-metoprolol], rapid onset- short duration of action (direct vasodilator-hydralazine), and long-acting transdermal drug delivery (centrally act ing-clonidine). Regarding the ADR theme, dry cough, angioedema, and hyperkalemia by ACEIs in susceptible patients, reflex tachycardia by CCB/amlodipine, and orthostatic hypotension by CCB/verapamil addressed. Clinical indications included the drug of choice for hypertensive patients with concomitant comorbidity such as diabetics (ACEI-lisinopril), heart failure and low ejection fraction (BB-carvedilol), hypertensive urgency/ emergency (alpha cum beta receptor blocker-labetalol), stroke in patients with history recurrent stroke or tran sient ischemic attack (ARB-losartan), and preeclampsia (methyldopa). A-type MCQsh The explanation reads: pro pranolol is known to have a significant first pass-effect, meaning that a large portion of the drug is metabolized by the liver before it reaches systemic circulation. Losar tan, amlodipine, ramipril, and hydrochlorothiazide do As with Claude-Instant AI, ChatGPT-generated test items on MOA were mostly similar. However, under the pharmacokinetic domain, immediate- and extended- release metoprolol, the effect of food to enhance the oral bioavailability of ramipril, and the highest oral bioavail ability of amlodipine compared to other commonly used antihypertensives were the themes identified. Whereas the other ADR themes remained similar, constipation due to verapamil was a new theme addressed. A-type MCQsh Notably, in this test item, amlodipine was an option that increased the difficulty of this test item because amlodipine therapy Sridharan and Sequeira BMC Medical Education (2024) 24:431 Page 6 of 13 Sridharan and Sequeira BMC Medical Education Table 2 Comparison of types of errors in the A-type MCQs between the AI platforms in pre-clerkship phase and at graduation Types of errors Sage Poe© Chart GPT© Claude-Instant© Pre-clerkship (n = 5) At gradua tion (n = 5) Pre-clerkship (n = 5) At gradua tion (n = 5) Pre-clerkship (n = 5) At grad uation (n = 5) Pharmacokinetics Item construction defect 5b, c, d, e, f 4 ac, ad, ae, ag None None 5 au None More than one correct option None 1s 1s None None None Appropriateness to learners’ level controversial 3 b, c, e 1ac 2 r, t 1an None None Mechanisms of action Item construction defect None None 2 q, u 1aL None None More than one correct option 1a 1ab None 3am, ao, ap None None Appropriateness to learners’ level controversial None None 1v None None None Adverse drug reactions Item construction defect None None None None None None More than one correct option 1g 1ah 1z 2aq, ar None None Appropriateness to learners’ level controversial 1h None None None None None Indications Item construction defect None None 1z None 1 p None More than one correct option 3 i, k, l 2 a.i., ak 2 y, aa 1as 4 m, n, o, p None Appropriateness to learners’ level controversial 2 j, k 1aj None None 1 n None a-lisinopril (test item #1); b-bioavailability and metabolized by liver (item #6); c-renal elimination and long half-life (item# 7); d-metabolized by both liver and kidney (item #8); e-rapid onset of action and short duration (item #9); f- transdermal patch and duration of action (item #10); g-lisinopril and losartan (item #13); h-verapamil and orthostatic hypotension (item #15); i-lisinopril and losartan (item # 16); j-carvedilol for heart failure with reduced ejection fraction (item #17); k-lisinopril and losartan for stroke prevention (item #18); L-hydralazine and methyldopa (item #20); m-lisinopril and losartan (item #17); n-HT Rx in night shift worker (item #18); o- treatment of hypertension and chronic kidney disease (item #19); p- treatment of hypertension and hyperlipidemia (item #20); q-metoprolol and its effects on blood vessels (item #4); R-longest half-life among antihypertensives (item # 6); s- four possible correct answers (item #7); t-immediate and extended- release preparations (item # 8); u- antihypertensive drug to be taken with food (item #9); v-highest bioavailable antihypertensive drug (item #10); x-losartan and ramipril cause hyperkalemia (item#11); y-losartan and ramipril can be used for treating hypertension and heart failure (item #16); z- hypertension with angina and amlodipine (item #17); aa-losartan and ramipril in hypertension and diabetic Nephropathy (item #18); ab-lisinopril and losartan (item #4); ac-high bioavailability and metabolized by liver (item #6); ad-eliminated by kidneys and long half-life (item #7); ae-metoprolol metabolized by both liver and kidneys (item #8); af-rapid onset and short duration (item #9); ag-transdermal patch and longer duration (item #10); ah-lisinopril and losartan (item #13); ai-hydrochlorothiazide, lisinopril and losartan are indicated for hypertension and diabetes mellitus (item #16); aj-losartan as antihypertensive drug for preventing recurrent stroke (item #19); ak- methyldopa and hydralazine can be used in preeclampsia (item #20); al-ramipril inhibits RAAS (item #2); am-losartan and candesartan blocks type II receptors (item #3); an-doxazosin with longest duration of action (item #6); ao-ramipril and losartan contra-indicated in renal impairment (item #8); ap-losartan, ramipril and spironolactone cause hyperkalemia (item # 13); ar- hydrochlorothiazide and hydrochlorothiazide/triamterene can result in photosensitivity (item # 15); as-losartan and ramipril reduce morbidity and mortality in heart failure (item #16); and au-The question should be what type of pharmacokinetic characteristics best describes the respective drug (items #6 to #10) a-lisinopril (test item #1); b-bioavailability and metabolized by liver (item #6); c-renal elimination and long half-life (item# 7); d-metabolized by both liver and kidney (item #8); e-rapid onset of action and short duration (item #9); f- transdermal patch and duration of action (item #10); g-lisinopril and losartan (item #13); h-verapamil and orthostatic hypotension (item #15); i-lisinopril and losartan (item # 16); j-carvedilol for heart failure with reduced ejection fraction (item #17); k-lisinopril and losartan for stroke prevention (item #18); L-hydralazine and methyldopa (item #20); m-lisinopril and losartan (item #17); n-HT Rx in night shift worker (item #18); o- treatment of hypertension and chronic kidney disease (item #19); p- treatment of hypertension and hyperlipidemia (item #20); q-metoprolol and its effects on blood vessels (item #4); R-longest half-life among antihypertensives (item # 6); s- four possible correct answers (item #7); t-immediate and extended- release preparations (item # 8); u- antihypertensive drug to be taken with food (item #9); v-highest bioavailable antihypertensive drug (item #10); x-losartan and ramipril cause hyperkalemia (item#11); y-losartan and ramipril can be used for treating hypertension and heart failure (item #16); z- hypertension with angina and amlodipine (item #17); aa-losartan and ramipril in hypertension and diabetic Nephropathy (item #18); ab-lisinopril and losartan (item #4); ac-high bioavailability and metabolized by liver (item #6); ad-eliminated by kidneys and long half-life (item #7); ae-metoprolol metabolized by both liver and kidneys (item #8); af-rapid onset and short duration (item #9); ag-transdermal patch and longer duration (item #10); ah-lisinopril and losartan (item #13); ai-hydrochlorothiazide, lisinopril and losartan are indicated for hypertension and diabetes mellitus (item #16); aj-losartan as antihypertensive drug for preventing recurrent stroke (item #19); ak- methyldopa and hydralazine can be used in preeclampsia (item #20); al-ramipril inhibits RAAS (item #2); am-losartan and candesartan blocks type II receptors (item #3); an-doxazosin with longest duration of action (item #6); ao-ramipril and losartan contra-indicated in renal impairment (item #8); ap-losartan, ramipril and spironolactone cause hyperkalemia (item # 13); ar- hydrochlorothiazide and hydrochlorothiazide/triamterene can result in photosensitivity (item # 15); as-losartan and ramipril reduce morbidity and mortality in heart failure (item #16); and au-The question should be what type of pharmacokinetic characteristics best describes the respective drug (items #6 to #10) not have significant first-pass effect. A-type MCQsh While I cannot perform a full simulation, here are some consider ations for administering MCQ questions to medical students: provide details on the purpose and instruc tions for the test to ensure students understand what is expected out of them; consider including some easier questions in addition to more challenging ones to boost students’ confidence and motivation; avoid ambiguous wording and ensure each ques tion has only one unambiguously correct answer option; when analyzing results, look for questions factors that may have impacted performance, like unclear wording or answer options that are too simi lar. Focusing on weak questions can help improve future tests; compare performance across different student groups to identify areas where more teach ing may be needed; provide detailed answer expla nations and review sessions after the test to help students learn from any errors and reinforce correct concepts”. However, ChatGPT offered the simulation result estimates for the 20 MCQ on antihypertensive drugs: mechanism of action ranged from 67 to 92%, pharmacokinetics 63–86%, adverse effects 65–82%, and clinical indications 64–89%. Furthermore, it also stated that “Overall, the performance of the stu dents was quite good, with most questions having a response rate of over 70%. However, there were some questions where the projected correct response rate was rather low, such as Question #5 (Mechanism of Action of Hydrochlorothiazide; see Electronic Sup plementary Material 12) and Question 18 (Indica tions for Verapamil; see Electronic Supplementary Material 10). This may suggest areas where students need more focused education or review.” especially in an integrated PBL curriculum. Both the AI assistants generated excellent clinical vignettes and themes; however, most of the cluster MCQs by Chat GPT had “None of the above” as an option, which is often considered a test item construction flaw. Notwithstand ing these limitations, case cluster integrated test items are valuable for learners to integrate their knowledge of different basic medical sciences and their application to clinical sciences. This integrated approach can be used for both instructional and student assessment purposes to make the course more meaningful. Indeed, one of the basic tenets of PBL is curriculum integration. In the next query, we asked AI assistants to write integrated case cluster MCQs with 2 test items in each cluster with 5 options for undergraduate medical stu dents at the time of graduation integrating pharmacol ogy and physiology of hypertension with a case vignette. A-type MCQsh However, it is also important to extend the explanation further by stating that the first-pass effect of propranolol does not lead to total loss of pharmacological activity because the metab olite hydroxy propranolol also has potent beta-blocking activity. Another MCQ test item had a construction defect: “A patient with hypertension is started on a medi cation that can cause photosensitivity. Which of the fol lowing medications is most likely responsible?” Options included: losartan, amlodipine, ramipril, hydrochloro thiazide, hydrochlorothiazide/triamterene. The explana tion states that hydrochlorothiazide (HCTZ) can cause photosensitivity, which is increased sensitivity to sunlight that can cause skin rashes and sunburns. Losartan, amlo dipine, ramipril, and HCTZ/triamterene are less likely to cause photosensitivity. However, it is irrational to claim that HCTZ/triamterene, one of the wrong options, is less likely to cause photosensitivity. The difficulty index of such test items is likely unacceptable in high-stake tests to make equitable psychometric decisions. The responses from Sage Poe, Claude-Instant, and ChatGPT on guidance for standard-setting methods are listed in the electronic supplementary materials 13–15. Sage Poe offered this information with the caveat that “the results of this simulation were purely hypotheti cal and do not reflect the actual performance of any real students”. The simulation estimates of examinees cor rectly answering test items ranged from 60 to 90% for the mechanism of action, 65–85% for pharmacokinetics, 60–80% for adverse reactions, and 75–90% for clinical indications of antihypertensives. In contrast, Claude-Instant admitted that “Perform ing a full simulation with 100 medical students is beyond the scope of what I can do as an AI assistant. I can generate sample MCQ questions and answers, but I do not have access to a population of medical students to administer the questions. A true simula tion that administers questions to real students and analyzes the results would require access to a data Page 7 of 13 Page 7 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 (2024) 24:431 Sridharan and Sequeira BMC Medical Education base of medical students, a platform to deliver the questions, and code to analyze the responses. As an AI assistant, I lack these capabilities. A-type MCQsh The responses are listed in the electronic supplementary materials 19–21. Claude-Instant has provided only par tial responses to both the queries. Overall, the informa tion provided in case vignettes requires the learner to interpret the clinical information and make therapeutic decisions. Such test items are appropriate for assessing the minimum competency of medical school graduates to start internship or residency programs; these MCQs are generally comparable to test items used in USMLE Part 2. The evaluation of the integrated case cluster test items provided by all the AI platforms is summarized in Table 3. In general, the AI outputs had at least one con struction defect, and did not include all the core sub-top ics to a sufficient degree. fi When asked to simulate the above-generated test items to 100 undergraduate medical students at the time of graduation and provide a percentage of students who are likely to correctly answer each item, the AI assistant responded: “As an AI language model, I cannot simulate the responses of 100 undergraduate medical students likely to obtain the correct answer for each question. However, I can provide some general information about the diffi culty level of each question based on the complexity of the concepts being tested and the level of integration between pharmacology and physiology”. Integrated case cluster MCQs We asked AI assistants to generate 20 integrated case cluster MCQs with 2 test items in each cluster with five options for undergraduate medical students in the pre- clerkship phase integrating pharmacology and physiology related to systemic hypertension with a case vignette and the responses by Sage Poe, Claude-Instant, and Chat GPT are listed in the electronic supplementary mate rials (16–18). In all instances, the test items generated had focused case descriptions in the form of a clinical vignette, and horizontal integration across the patho physiology of hypertension and pharmacology of anti hypertensive drugs. These test items mostly targeted the ‘knows (knowledge)’ or ‘knows how (competence)’ level on Miller’s pyramid and are suitable for assessing the clinical competence of pre-clerkship medical students, AI assistant went on to clarify that questions that require knowledge of basic pharmacology principles, such as the mechanism of action of specific drugs, are likely to be easier for students to answer correctly. Test items that require an understanding of the physiologi cal mechanisms underlying hypertension and correlat ing with symptoms are likely to be more challenging for students. The AI assistant sorted these test items into two categories accordingly. Overall, the difficulty level of the test item is based on the level of integration between pharmacology and pathophysiology. Short answer questionsh The responses to a search query on generating SAQs appropriate to the pre-clerkship phase Sage Poe, Claude- Instant, and ChatGPT generated items are listed in the electronic supplementary materials 22–24 for difficult questions and 25–27 for moderately difficult questions. fi It is apparent from these case vignette descriptions that the short answer question format varied. Accordingly, the scope for asking individual questions for each scenario is open-ended. In all instances, model answers are supplied which are helpful for the course instructor to plan class room lessons, identify appropriate instructional meth ods, and establish rubrics for grading the answer scripts, and as a study guide for students. The responses of Sage Poe, Claude-Instant, and Chat GPT for the search query to generate SAQs at the time of graduation are listed in the electronic supplementary materials 28–30. It is interesting to note how AI assis tants considered the stage of the learner while generat ing the SAQ. The response by Sage Poe is illustrative for comparison. “You are a newly graduated medical student who is working in a hospital” versus “You are a medical student in your pre-clerkship.” We then wanted to see to what extent AI can differenti ate the difficulty of the SAQ by replacing the search term “difficult” with “moderately difficult” in the above search prompt: the changes in the revised case scenarios are substantial. Perhaps the context of learning and practice (and the level of the student in the MD/medical program) may determine the difficulty level of SAQ generated. It is worth noting that on changing the search from cardi ology to internal medicine rotation in Sage Poe the case description also changed. Thus, it is essential to select an appropriate AI assistant, perhaps by trial and error, to generate quality SAQs. Most of the individual questions tested stand-alone knowledge and did not require stu dents to demonstrate integration. Some questions were retained, deleted, or modified to align with competency appropriate to the context (Elec tronic Supplementary Materials 28–30). Overall, the test items at both levels from all AI platforms were techni cally accurate and thorough addressing the topics related to different disciplines (Table 3). The differences in learn ing objective transition are summarized in Table 4. A comparison of learning objectives revealed that almost all objectives remained the same except for a few (Table 5). Integrated case cluster MCQs Test items that Page 8 of 13 Page 8 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 Sridharan and Sequeira BMC Medical Education Table 3 Assessment of test items using the preliminary conceptual framework for establishing content validity of AI-generated test items (integrated case clusters, SAQs and OSPEs) Types of errors Sage Poe© Chart GPT© Claude-Instant© Pre-clerkship At graduation Pre-clerkship At graduation Pre-clerkship At graduation Integrated case cluster Technical accuracy Complete Complete Complete Complete Deficient Complete a Comprehensiveness Deficient Deficient Deficient Deficient Deficient Deficient a Education level Complete Deficient Deficient Deficient Complete Complete a Free of construction defects Deficient Deficient Deficient Deficient Deficient Complete a Short answer questions Technical accuracy Complete Complete Complete Complete Complete Complete Comprehensiveness Complete Complete Complete Complete Complete Complete Education level Complete Deficient Complete Deficient Complete Complete Free of construction defects Deficient Deficient Deficient Deficient Deficient Complete OSPEs Technical accuracy Not Available Complete Complete Complete Complete Comprehensiveness Complete Complete Complete Complete Education level Complete Complete Complete Complete Free of construction defects Complete Complete Complete Complete a- Only a small portion of the requested test items were provided by the concerned AI tool require an understanding of both pharmacological and physiological mechanisms are likely to be more challeng ing for students requiring a strong foundation in both pharmacology and physiology concepts to be able to cor rectly answer integrated case-cluster MCQs. Table 4 Comparison of the SAQ test items generated by Sage Poe for pre-clerkship phase and graduating students Table 4 Comparison of the SAQ test items generated by Sage Poe for pre-clerkship phase and graduating students Pre-clerkship phase At graduation Physiological mechanisms that regulate blood pressure in the body. Pathophysiology of sys temic hypertension. Risk factors for systemic hypertension. Potential complications of untreated hypertension. Diuretics to lower blood pressure. ACEI to lower blood pressure. Frequency of checking blood pressure in patients with hypertension. Recommended blood pressure target for pa tients with hypertension. ACEI– Angiotensin-converting enzyme inhibitors At graduation How does sympathetic nervous system activation affect blood pressure, and how can medications targeting this sys tem be used to manage hypertension? What are pathological changes that occur in the systemic blood vessels in hypertension, and how do these changes contribute to cardiovascular complications? How does sympathetic ner vous system activation affect blood pressure? How does sympathetic ner vous system activation affect blood pressure? How does sympathetic nervous system activation affect blood pressure, and how can medications targeting this sys tem be used to manage hypertension? What are pathological changes that occur in the systemic blood vessels in hypertension? What are pathological changes that occur in the systemic blood vessels in hypertension? What are pathological changes that occur in the systemic blood vessels in hypertension, and how do these changes contribute to cardiovascular complications? What are non-pharmaco logical treatment options in hypertension? What are non-pharmacological treat ment options in hypertension, and how effective are they in managing this condition? What are the pharmacological treat ment options for hypertension, and how do they work? vertical integration of basic sciences and clinical sciences (Table 6). Taken together, these in-depth qualitative comparisons suggest that AI assistants such as Sage Poe and ChatGPT consider the learner’s stage of training in designing test items, learning outcomes, and answers expected from the examinee. It is critical to state the search query explicitly to generate quality output by AI assistants. The responses from Claude-Instant and ChatGPT for the search query related to generating OSPE test items at the time of graduation are listed in electronic supplemen tary materials 35 and 36. In contrast to the pre-clerkship phase, OSPEs generated for graduating doctors’ compe tence assessed more advanced drug therapy comprehen sion. For example, writing a prescription for: Short answer questionsh A similar trend was apparent with test items gener ated by other AI assistants, such as ChatGPT. The con trasting differences in questions are illustrated by the Page 9 of 13 Page 9 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 Sridharan and Sequeira BMC Medical Education Table 5 Comparison of learning objectives in SAQ generated for pre-clerkship phase and graduating students Pre-clerkship phase At graduation Mechanism of action of diuretics in hyperten sion treatment. Mechanism of action of ACEI in hyperten sion treatment. Frequency of blood pressure checks for patients with hypertension. Blood pressure targets for patients with hypertension. ACEI– Angiotensin-converting enzyme inhibitors a 45-year-old man, and gestational hypertension at 32 weeks in a 35-year-old (Claude-Instant AI). Incorporat ing appropriate instructions facilitates the learner’s abil ity to educate patients and maximize safe and effective therapy. The OSPE item required students to write a pre scription with guidance to start conservatively, choose an appropriate antihypertensive drug class (drug) based on the patients’ profile, specifying drug name, dose, dos ing frequency, drug quantity to be dispensed, patient name, date, refill, and caution as appropriate, in addi tion to prescribers’ name, signature, and license num ber. In contrast, ChatGPT identified clinical scenarios to include patients with hypertension and CKD, hyperten sion and bronchial asthma, gestational diabetes, hyper tension and heart failure, and hypertension and gout (ChatGPT). Guidance for dosage titration, warnings to be aware, safety monitoring, and frequency of follow-up and dose adjustment. These test items are designed to assess learners’ knowledge of P & T of antihypertensives, as well as their ability to provide appropriate instructions to patients. These clinical scenarios for writing prescrip tions assess students’ ability to choose an appropriate drug class, write prescriptions with proper labeling and dosing, reflect drug safety profiles, and risk factors, and make modifications to meet the requirements of special populations. The prescription is required to state the drug name, dose, dosing frequency, patient name, date, refills, and cautions or instructions as needed. A conser vative starting dose, once or twice daily dosing frequency based on the drug, and instructions to titrate the dose slowly if required. Table 6 Comparison of the SAQ test items generated by ChatGPT for pre-clerkship phase and graduating students Discussion In the present study, AI tools have generated SLOs that comply with the current principles of medical education [15]. AI tools are valuable in constructing SLOs and so are especially useful for medical fraternities where train ing in medical education is perceived as inadequate, more so in the early stages of their academic career. Data suggests that only a third of academics in medical schools have formal training in medical education [16] which is a limitation. Thus, the credibility of alternatives, such as the AIs, is evaluated to generate appropriate course learning outcomes. (5) A 35-year-old pregnant woman with preeclampsia at 29 weeks require doubling methyldopa dose and con sider adding labetalol or nifedipine based on severity and educate on signs of worsening and to follow-up immedi ately for any concerning symptoms. These case scenarios are designed to assess the abil ity of the learner to comprehend the complexity of anti hypertensive regimens, make evidence-based regimen adjustments, prescribe multidrug combinations based on therapeutic response and tolerability, monitor complex patients for complications, and educate patients about warning signs and follow-up. We observed that the AI platforms in the present study generated quality test items suitable for different types of assessment purposes. The AI-generated outputs were similar with minor variation. We have used generative AIs in the present study that could generate new con tent from their training dataset [17]. Problem-based and interactive learning approaches are referred to as “bot tom-up” where learners obtain first-hand experience in solving the cases first and then indulge in discussion with the educators to refine their understanding and critical thinking skills [18]. We suggest that AI tools can be use ful for this approach for imparting the core knowledge and skills related to Pharmacology and Therapeutics to undergraduate medical students. A recent scoping review evaluating the barriers to writing quality test items based on 13 studies has concluded that motivation, time con straints, and scheduling were the most common [19]. AI tools can be valuable considering the quick generation of quality test items and time management. However, as observed in the present study, the AI-generated test items nevertheless require scrutiny by faculty members for content validity. Moreover, it is important to train faculty in AI technology-assisted teaching and learning. The General Medical Council recommends taking every opportunity to raise the profile of teaching in medical schools [20]. OSPEsh (1) A 65-year- old male with resistant hypertension and CKD stage 3 to optimize antihypertensive regi men required the answer to include starting ACEI and diuretic, titrating the dosage over two weeks, considering adding spironolactone or substituting ACEI with an ARB, and need to closely monitor serum electrolytes and kid ney function closely. The OSPE test items generated by Claude-Instant and ChatGPT appropriate to the pre-clerkship phase (without mentioning “appropriate instructions for the patients”) are listed in the electronic supplementary materials 31 and 32 and with patient instructions on the electronic supplementary materials 33 and 34. For rea sons unknown, Sage Poe did not provide any response to this search query.hi (2) A 55-year-old woman with hypertension and parox ysmal arrhythmia required the answer to include switch ing ACEI to ARB due to cough, adding a CCB or beta blocker for rate control needs, and adjusting the dosage slowly and monitoring for side effects. The five OSPE items generated were suitable to assess the prescription writing competency of pre-clerkship medical students. The clinical scenarios identified by the three AI platforms were comparable; these scenarios include patients with hypertension and impaired glu cose tolerance in a 65-year-old male, hypertension with chronic kidney disease (CKD) in a 55-year-old woman, resistant hypertension with obstructive sleep apnea in f (3) A 45-year-old man with masked hypertension and obstructive sleep apnea require adding a centrally act ing antihypertensive at bedtime and increasing dosage as f (3) A 45-year-old man with masked hypertension and obstructive sleep apnea require adding a centrally act ing antihypertensive at bedtime and increasing dosage as Sridharan and Sequeira BMC Medical Education (2024) 24:431 Sridharan and Sequeira BMC Medical Education (2024) 24:431 Page 10 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 needed based on home blood pressure monitoring and refer to CPAP if not already using one. refined further by providing a focused case history along with relevant clinical and laboratory data to enhance clinical fidelity and bring a closer fit to the competency framework. (4) A 75-year-old woman with isolated systolic hyper tension and autonomic dysfunction to require stopping diuretic and switching to an alpha blocker, upward dos age adjustment and combining with other antihyperten sives as needed based on postural blood pressure changes and symptoms. Strengths and limitations To the best of our knowledge, this is the first study evalu ating the utility of AI platforms in generating test items related to a discipline in the undergraduate medical cur riculum. We have evaluated the AI’s ability to generate outputs related to most types of assessment in the under graduate medical curriculum. The key lessons learnt for improving the AI-generated test item quality from the present study are outlined in Table 7. We used a struc tured framework for assessing the content validity of the test items. However, we have demonstrated using a single case study (hypertension) as a pilot experiment. We chose to evaluate anti-hypertensive drugs as it is a core learning objective and one of the most common disorders relevant to undergraduate medical curricula worldwide. It would be interesting to explore the output from AI platforms for other common (and uncommon/region-specific) dis orders, non-/semi-core objectives, and disciplines other than Pharmacology and Therapeutics. An area of inter est would be to look at the content validity of the test items generated for different curricula (such as problem- based, integrated, case-based, and competency-based) during different stages of the learning process. Also, we did not attempt to evaluate the generation of flowcharts, algorithms, or figures for generating test items. Another potential area for exploring the utility of AIs in medical education would be repeated procedural practices such as the administration of drugs through different routes by trainee residents [27]. Several AI tools have been iden tified for potential application in enhancing classroom instructions and assessment purposes pending validation in prospective studies [28]. Lastly, we did not administer the AI-generated test items to students and assessed their performance and so could not comment on the validity of test item discrimination and difficulty indices. Addition ally, there is a need to confirm the generalizability of the findings to other complex areas in the same discipline as well as in other disciplines that pave way for future stud ies. The conceptual framework used in the present study for evaluating the AI-generated test items needs to be validated in a larger population. Future studies may also AI platforms may not always have access to all stan dard treatment guidelines. Discussion Hence, both the academic faculty and the institution must consider investing resources in AI train ing to ensure appropriate use of the technology [21]. g g p A similar output was provided by ChatGPT, with clini cal scenarios such as prescribing for patients with hyper tension and myocardial infarction; hypertension and chronic obstructive pulmonary airway disease (COPD); hypertension and a history of angina; hypertension and a history of stroke, and hypertension and advanced renal failure. In these cases, wherever appropriate, pharma cotherapeutic issues like taking ramipril after food to reduce side effects such as giddiness; selection of the most appropriate beta-blocker such as nebivolol in patients with COPD comorbidity; the importance of tak ing amlodipine at the same time every day with or with out food; preference for telmisartan among other ARBs in stroke; choosing furosemide in patients with hyper tension and edema and taking the medication with food to reduce the risk of gastrointestinal adverse effect are stressed.h The AI outputs on OSPE test times were observed to be technically accurate, thorough in addressing core sub- topics suitable for the learner’s level and did not have any construction defects (Table 3). Both AIs provided the model answers with explanatory notes. This facilitates the use of such OSPEs for self-assessment by learners for formative assessment purposes. The detailed instructions are helpful in creating optimized therapy regimens, and designing evidence-based regimens, to provide appro priate instructions to patients with complex medical histories. One can rely on multiple AI sources to iden tify, shortlist required case scenarios, and OSPE items, and seek guidance on expected model answers with explanations. The model answer guidance for antihy pertensive drug classes is more appropriate (rather than a specific drug of a given class) from a teaching/learn ing perspective. We believe that these scenarios can be The AI outputs assessed in the present study had errors, particularly with A-type MCQs. One notable observation was that often the AI tools were unable to differentiate the differences between ACEIs and ARBs. AI platforms access several structured and unstructured data, in addition to images, audio, and videos. Discussion Hence, the AI platforms can commit errors due to extracting details from unauthenticated sources [22] created a framework identifying 28 factors for reconstructing the path of AI Sridharan and Sequeira BMC Medical Education (2024) 24:431 Page 11 of 13 (2024) 24:431 Sridharan and Sequeira BMC Medical Education failures and for determining corrective actions. This is an area of interest for AI technical experts to explore. Also, this further iterates the need for human examination of test items before using them for assessment purposes.h include AI- directed (learner as recipient), AI-supported (learner as collaborator), and AI-empowered (learner as leader) that are based on Behaviorism, Cognitive-Social constructivism, and Connectivism-Complex adaptive systems, respectively [25]. AI techniques have potential to stimulate and advance instructional and learning sci ences. More recently a three- level model that synthesizes and unifies existing learning theories to model the roles of AIs in promoting learning process has been proposed [26]. The different components of our study rely upon these paradigms and learning theories as the theoretical underpinning. There are concerns that AIs can memorize and pro vide answers from their training dataset, which they are not supposed to do [23]. Hence, the use of AIs-generated test items for summative examinations is debatable. It is essential to ensure and enhance the security features of AI tools to reduce or eliminate cross-contamination of test items. Researchers have emphasized that AI tools will only reach their potential if developers and users can access full-text non-PDF formats that help machines comprehend research papers and generate the output [24]. Strengths and limitations However, in the present study, it was observed that all three AI platforms gen erally provided appropriate test items regarding the choice of medications, aligning with recommendations from contemporary guidelines and standard textbooks in pharmacology and therapeutics. The prompts used in the study were specifically focused on the pre-clerkship phase of the undergraduate medical curriculum (and at the time of their graduation) and assessed fundamental core concepts, which were also reflected in the AI out puts. Additionally, the recommended first-line antihy pertensive drug classes have been established for several decades, and information regarding their pharmacoki netics, ADRs, and indications is well-documented in the literature. Different paradigms and learning theories have been proposed to support AI in education. These paradigms Table 7 Key take home messages for improving AI-generated test item quality Key take home messages • Compare multiple AI platforms to evaluate the output fidelity. • Link course syllabus, SLOs, expected competency, and learner’s stage in the program. • Use unambiguous and specific search prompts to refine the search iteration strategy. • Decide whether test items sought are for formative or summative purpose. • Clarify the expected test items match on Bloom’s taxonomy. • Seek high fidelity clinical vignette to promote context-based learning. • Define the level of integration appropriate to learner’s stage in the program. • Integrate the complexity of OSPE clinical scenarios to patient-instructions. • Recognize the limitations of AIs such as a limited access to all treat ment guidelines. • Ensure the validity of AI generated test items by content experts. • Evaluate simulation- based standard setting guidance offered by AIs to real world situation. Table 7 Key take home messages for improving AI-generated test item quality k h Key take home messages • Evaluate simulation- based standard setting guidance offered by AIs to real world situation. Sridharan and Sequeira BMC Medical Education (2024) 24:431 Page 12 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 4. Masters K. Ethical use of Artificial Intelligence in Health Professions Education: AMEE Guide 158. Med Teach. 2023;45(6):574–84. 4. Masters K. Ethical use of Artificial Intelligence in Health Professions Education: AMEE Guide 158. Med Teach. 2023;45(6):574–84. try to evaluate the variations in the AI outputs with rep etition of the same queries. try to evaluate the variations in the AI outputs with rep etition of the same queries. 5. Nagi F, Salih R, Alzubaidi M, Shah H, Alam T, Shah Z, Househ M. Applications of Artificial Intelligence (AI) in Medical Education: a scoping review. Stud Health Technol Inf. 2023;305:648–51. 5. Nagi F, Salih R, Alzubaidi M, Shah H, Alam T, Shah Z, Househ M. Applications of Artificial Intelligence (AI) in Medical Education: a scoping review. Stud Health Technol Inf. 2023;305:648–51. Author contributions 14. Eschenhagen T. Treatment of hypertension. In: Brunton LL, Knollmann BC, editors. Goodman & Gilman’s the pharmacological basis of therapeutics. 14th ed. New York: McGraw Hill; 2023.f RPS– Conceived the idea; KS– Data collection and curation; RPS and KS– Data analysis; RPS and KS– wrote the first draft and were involved in all the revisions. 15. Shabatura J. September. Using Bloom’s taxonomy to write effective learning outcomes. https://tips.uark.edu/using-blooms-taxonomy/ (Accessed on 19th 2023). Received: 26 September 2023 / Accepted: 28 March 2024 24. Brainard J. November. As scientists face a flood of papers, AI developers aim to help. Science, 21 2023. doi.10.1126/science.adn0669.i to help. Science, 21 2023. doi.10.1126/science.adn0669. Data availability All the data included in this study are provided as Electronic Supplementary Materials. 17. Boscardin C, Gin B, Golde PB, Hauer KE. ChatGPT and generative artificial intelligence for medical education: potential and opportunity. Acad Med. 2023. https://doi.org/10.1097/ACM.0000000000005439. (Published ahead of print). Funding None Trainor A, Richards JB. Training medical educators to teach: bridging the 16. Trainor A, Richards JB. Training medical educators to teach: bridging the gap between perception and reality. Isr J Health Policy Res. 2021;10(1):75.i Conclusion 6. Mehta N, Harish V, Bilimoria K, et al. Knowledge and attitudes on artificial intelligence in healthcare: a provincial survey study of medical students. MedEdPublish. 2021;10(1):75. 6. Mehta N, Harish V, Bilimoria K, et al. Knowledge and attitudes on artificial intelligence in healthcare: a provincial survey study of medical students. MedEdPublish. 2021;10(1):75. Notwithstanding ongoing discussions and controver sies, AI tools are potentially useful adjuncts to optimize instructional methods, test blueprinting, test item gen eration, and guidance for test standard-setting appropri ate to learners’ stage in the medical program. However, experts need to critically review the content validity of AI-generated output. These challenges and caveats are to be addressed before the use of widespread use of AIs in medical education can be advocated. 7. Mir MM, Mir GM, Raina NT, Mir SM, Mir SM, Miskeen E, Alharthi MH, Alamri MMS. Application of Artificial Intelligence in Medical Education: current scenario and future perspectives. J Adv Med Educ Prof. 2023;11(3):133–40.i 7. Mir MM, Mir GM, Raina NT, Mir SM, Mir SM, Miskeen E, Alharthi MH, Alamri MMS. Application of Artificial Intelligence in Medical Education: current scenario and future perspectives. J Adv Med Educ Prof. 2023;11(3):133–40. fi 8. Garg T. Artificial Intelligence in Medical Education. Am J Med. 2020;133(2):e68.i 9. Matheny ME, Whicher D, Thadaney IS. Artificial intelligence in health care: a report from the National Academy of Medicine. JAMA. 2020;323(6):509–10. 9. Matheny ME, Whicher D, Thadaney IS. Artificial intelligence in health care: a report from the National Academy of Medicine. JAMA. 2020;323(6):509–10. 10. Sage Poe. Available at: https://poe.com/Assistant (Accessed on. 3rd June 2023). 10. Sage Poe. Available at: https://poe.com/Assistant (Accessed on. 3rd June 2023). 11. Claude-Instant: Available at: https://poe.com/Claude-instant (Accessed on 3rd. June 2023). 11. Claude-Instant: Available at: https://poe.com/Claude-instant (Accessed on 3rd. June 2023). Competing interests The authors declare no competing interests. 21. Cooper A, Rodman A. AI and Medical Education - A 21st-Century Pandora’s Box. N Engl J Med. 2023;389(5):385–7. Author details 1 1Department of Pharmacology & Therapeutics, College of Medicine & Medical Sciences, Arabian Gulf University, Manama, Kingdom of Bahrain 22. Chanda SS, Banerjee DN. Omission and commission errors underlying AI failures. AI Soc. 2022;17:1–24. 23. Narayanan A, Kapoor S. ‘GPT-4 and Professional Benchmarks: The Wrong Answer to the Wrong Question’. Substack newsletter. AI Snake Oil (blog). https://aisnakeoil.substack.com/p/gpt-4-and-professional-benchmarks (Accessed on 19th September 2023). Received: 26 September 2023 / Accepted: 28 March 2024 Consent for publication Not applicable. 20. Developing teachers and trainers in undergraduate medical education. Advice supplementary to Tomorrow’s Doctors. (2009). https://www.gmc-uk. org/-/media/documents/Developing_teachers_and_trainers_in_undergrad uate_medical_education___guidance_0815.pdf_56440721.pdf (Accessed on 19th September 2023). Declarations 18. Duong MT, Rauschecker AM, Rudie JD, Chen PH, Cook TS, Bryan RN, Mohan S. Artificial intelligence for precision education in radiology. Br J Radiol. 2019;92(1103):20190389. Ethics approval and consent to participate Ethics approval and consent to participate Not applicable as neither there was any interaction with humans, nor any personal data was collected in this research study. 19. Karthikeyan S, O’Connor E, Hu W. Barriers and facilitators to writing quality items for medical school assessments - a scoping review. BMC Med Educ. 2019;19(1):123. Supplementary Information 12. ChatGPT: Available at: https://poe.com/ChatGPT (Accessed on 3rd. June 2023). 12. ChatGPT: Available at: https://poe.com/ChatGPT (Accessed on 3rd. June 2023). The online version contains supplementary material available at https://doi. org/10.1186/s12909-024-05365-7. The online version contains supplementary material available at https://doi. org/10.1186/s12909-024-05365-7. 13. James PA, Oparil S, Carter BL, Cushman WC, Dennison-Himmelfarb C, Handler J, Lackland DT, LeFevre ML, MacKenzie TD, Ogedegbe O, Smith SC Jr, Svetkey LP, Taler SJ, Townsend RR, Wright JT Jr, Narva AS, Ortiz E. 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA. 2014;311(5):507–20. Supplementary Material 1 Supplementary Material 1 References 25. Ouyang F, Jiao P. Artificial intelligence in education: the three paradigms. Computers Education: Artif Intell. 2021;2:100020. 1. Tolsgaard MG, Pusic MV, Sebok-Syer SS, Gin B, Svendsen MB, Syer MD, Brydges R, Cuddy MM, Boscardin CK. The fundamentals of Artificial Intel ligence in medical education research: AMEE Guide 156. Med Teach. 2023;45(6):565–73. 2. Sriwastwa A, Ravi P, Emmert A, Chokshi S, Kondor S, Dhal K, Patel P, Chepelev LL, Rybicki FJ, Gupta R. Generative AI for medical 3D printing: a compari son of ChatGPT outputs to reference standard education. 3D Print Med. 2023;9(1):21. 3. Azer SA, Guerrero APS. The challenges imposed by artificial intelligence: are we ready in medical education? BMC Med Educ. 2023;23(1):680. 1. Tolsgaard MG, Pusic MV, Sebok-Syer SS, Gin B, Svendsen MB, Syer MD, Brydges R, Cuddy MM, Boscardin CK. The fundamentals of Artificial Intel ligence in medical education research: AMEE Guide 156. Med Teach. 2023;45(6):565–73. 1. Tolsgaard MG, Pusic MV, Sebok-Syer SS, Gin B, Svendsen MB, Syer MD, Brydges R, Cuddy MM, Boscardin CK. The fundamentals of Artificial Intel ligence in medical education research: AMEE Guide 156. Med Teach. 2023;45(6):565–73. 26. Gibson D, Kovanovic V, Ifenthaler D, Dexter S, Feng S. Learning theories for artificial intelligence promoting learning processes. Br J Edu Technol. 2023;54(5):1125–46. 2. Sriwastwa A, Ravi P, Emmert A, Chokshi S, Kondor S, Dhal K, Patel P, Chepelev LL, Rybicki FJ, Gupta R. Generative AI for medical 3D printing: a compari son of ChatGPT outputs to reference standard education. 3D Print Med. 2023;9(1):21. 2. Sriwastwa A, Ravi P, Emmert A, Chokshi S, Kondor S, Dhal K, Patel P, Chepelev LL, Rybicki FJ, Gupta R. Generative AI for medical 3D printing: a compari son of ChatGPT outputs to reference standard education. 3D Print Med. 2023;9(1):21. 27. Guerrero DT, Asaad M, Rajesh A, Hassan A, Butler CE. Advancing Surgical Education: the Use of Artificial Intelligence in Surgical Training. Am Surg. 2023;89(1):49–54. 3. Azer SA, Guerrero APS. The challenges imposed by artificial intelligence: are we ready in medical education? BMC Med Educ. 2023;23(1):680. 28. Lee S. AI tools for educators. EIT InnoEnergy Master School Teachers Confer ence. 2023. https://www.slideshare.net/ignatia/ai-toolkit-for-educators?from_ action=save (Accessed on 24th September 2023). Page 13 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 28. Lee S. AI tools for educators. EIT InnoEnergy Master School Teachers Confer ence. 2023. https://www.slideshare.net/ignatia/ai-toolkit-for-educators?from_ action=save (Accessed on 24th September 2023). Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Page 13 of 13 Sridharan and Sequeira BMC Medical Education (2024) 24:431 Page 13 of 13 Page 13 of 13
https://openalex.org/W2078501371	https://bmcmededuc.biomedcentral.com/counter/pdf/10.1186/1472-6920-13-158	English	null	Comparison of tutored group with tutorless group in problem-based mixed learning sessions: a randomized cross-matched study	BMC medical education	2,013	cc-by	5,093	© 2013 Hayashi 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. * Correspondence: sho5-884@umin.ac.jp 1Medical Education Center, Aichi Medical University School of Medicine, Nagakute, Aichi 480-1195, Japan Full list of author information is available at the end of the article Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 Open Access Comparison of tutored group with tutorless group in problem-based mixed learning sessions: a randomized cross-matched study Shogo Hayashi1, Koji Tsunekawa2, Chikako Inoue3 and Yoshitaka Fukuzawa1 Abstract Background: Problem-based learning (PBL) involves discussions among students who resolve loosely-structured problems to facilitate learning. In the PBL curriculum, faculty tutors are employed as facilitators for small groups of students. Because of lack of time and staff shortage, the effectiveness of tutorless PBL has been discussed as an alternate option. Methods: Sessions in which tutored and tutorless PBL groups are mixed were presented by 1st-year medical students, who experienced both tutored and tutorless groups alternately in the two sessions of a year. To examine the effectiveness of tutored and tutorless PBL, written examination scores (WES) and self-contentment scores (SCS) were statistically analysed. Results: WES averages did not significantly differ between the tutored and tutorless groups; however, a significantly greater variation was observed in WES in the tutorless group. SCS averages tended to be higher in the tutored PBL than in tutorless PBL groups. Conclusions: Students in these tutorless PBL groups performed well in their written examinations, whereas those in the tutored PBL groups, achieved this and reported better self-contentment with their learning experience. Tutorless PBL sessions were considered to be comparable to tutored PBL sessions at least in the early stages. Keywords: Problem-based learning, Tutorless group, Curriculum, Large class, Learning strategy expertise, characteristics and behaviour are believed to influence both the process perspective and learning outcomes [4,5]. Methods At Aichi Medical University, the PBL course comprises the following two units: PBL1 for first year medical stu- dents and PBL2 for third and-forth year medical students. The goal of PBL1 is introduction and early exposure of clinical medicine to students, and the role of PBL2 is to train students on the art of clinical problem solving. Be- tween 2007 and 2008, as part of the curriculum develop- ment of the existing PBL1, we conducted a randomised cross-matched study on the learning outcomes involved in PBL1 (2007, N = 102; 2008, N = 100). The research design is shown in Figure 1. The study was performed in accord- ance with the provisions of the Declaration of Helsinki. The executive council of the Medical Education Center Aichi Medical University first reviewed the detail protocol, including the future plan of submission. After the review by this committee, the academic affairs department and The aim of this study was to examine the effectiveness of tutorless PBL by comparing learning outcomes be- tween tutorless and tutored groups. Roberts et al. [11] and Kaliyadan et al. [14] reported their experiences with tutorless PBL and concluded that there were no signifi- cant differences in learning outcomes between tutored and tutorless PBL. However, in the abovementioned re- ports, several important differences such as group mem- ber characteristics, scenarios or learning materials that were present between the tutored and tutorless PBL conditions were observed. Nicholl and Lou [15] recently reported on tutorless PBL using a model for a large class facilitated by one instructor; they argued that students could achieve the required learning outcomes with tutor- less PBL. Moreover, these reports do not only compare Guidance & informed consent (N=100-102) Group A (N=50-51) Group B (N=50-51) Randomized grouping Tutored PBL (7 small groups) Randomized subgrouping Session 1 (4 SGD in a week) Tutorless PBL (7 small groups) Tutored PBL (7 small groups) Randomized re-subgrouping Written examination & Questionnaire (last day in a week) Session 2 (4 SGD in a week) Written examination & Questionnaire (last day in a week) Tutorless PBL (7 small groups) Figure 1 Chronological list of problem-based learning (PBL) undertaken by students in this study. Group A in the left row and Group B in the right row. In the middle row, sessions experienced by all students conducted in a large classroom are chronologically displayed. Background g In the mid-1960s, problem-based learning (PBL) was adopted as a new approach for medical education at McMaster University, Ontario, Canada. PBL has been de- scribed as ‘a learning that results from the process of work- ing towards understanding or resolution of a problem’ [1]. PBL not only facilitates the acquisition of knowledge but also that of other generic desirable attributes such as effect- ive communication skills, ability to work in a team (team work), problem-solving skills, self-directed learning ability, ability to share information, appreciate other points of view and identification of personal strengths and weak- nesses [2]. Because many of these skills are related to the tutorial process and group dynamics [3], the tutors’ In the approach adopted by the McMaster University medical school, a faculty tutor was present during all group activities to monitor, assess and provide immediate input i. e. each group was tutored [6]. Providing a facilitator for PBL can be a problem during times of faculty/staff short- age [6-16]. Therefore, many schools have tried other PBL formats in an attempt to reduce the demands on faculty/ staff time and resources; examples of these formats in- clude student-tutored PBL [17] and tutorless PBL [6]. In student-tutored PBL, one student studies the prob- lem in advance and then takes on the role as the tutor of the group instead of the faculty tutor. In tutorless PBL, neither the student tutor nor the faculty tutor is present. There have been many reports suggesting that student- tutored PBL can be just as effective as faculty-tutored Correspondence: sho5-884@umin.ac.jp 1Medical Education Center, Aichi Medical University School of Medicine, Nagakute, Aichi 480-1195, Japan Full list of author information is available at the end of the article Page 2 of 7 Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 learning outcomes between tutored and tutorless PBL be- cause there are many other factors influencing the study and results. To the best of our knowledge, this is the first randomised cross-matched study comparing tutored and tutorless PBL. PBL with regard to learning outcomes of student-tutored PBL. These sessions can be conducted by senior stu- dents [17-19] or also by peer-level students from the same class [3,9,10]. Similarly, a meta-analysis by Leary et al. [20] indicated that student tutors were equally ef- fective when compared with faculty tutors. However, there is limited information on learning outcomes of tutorless PBL [11,14,15]. Table 1 Student results on the self-evaluation in each session Table 1 Student results on the self-evaluation in each session 2007 Session 1 Session 2 Group A: Group B: p value Group B: Group A: p value Tutored Tutorless Tutored Tutorless (n = 51) (n = 50) (n = 51) (n = 51) average SD average SD average SD average SD Self-directed learning 4.19 0.72 3.94 0.71 0.08 4.25 0.74 4.06 0.86 0.22 Activeness 4.13 0.77 4.06 0.71 0.60 4.35 0.72 4.20 0.83 0.31 Scientific basis 3.96 0.77 3.74 0.72 0.14 4.00 0.80 3.89 0.69 0.36 Group dynamics 4.31 0.79 3.88 0.96 0.02 4.57 0.61 4.27 0.72 0.03 Attentiveness 4.20 0.69 4.28 0.67 0.54 4.39 0.72 4.27 0.70 0.40 2008 Session 1 Session 2 Group A: Group B: p value Group B: Group A: p value Tutored Tutorless Tutored Tutorless (n = 46) (n = 45) (n = 45) (n = 46) average SD average SD average SD average SD Self-directed Learning 4.41 0.75 4.42 0.70 0.96 4.51 0.59 4.24 0.66 0.037 Activeness 4.35 0.82 4.50 0.71 0.34 4.67 0.52 4.24 0.72 0.001 Scientific basis 4.20 0.78 4.34 0.66 0.33 4.44 0.62 4.14 0.70 0.027 Group dynamics 4.60 0.65 4.56 0.73 0.73 4.56 0.62 4.40 0.78 0.28 Attentiveness 4.50 0.72 4.58 0.57 0.55 4.60 0.50 4.40 0.67 0.19 2007 or standard deviation between both groups. Therefore, it was concluded that Groups A and B matched in each grade. combination of the percentage of attendance and writ- ten examination scores (WES). Daily reports, tutor evalu- ation, self-evaluation and answers to the questionnaire were not included in the summative evaluation. While comparing sessions, the average ± standard de- viation of WES in 2007 was 64.98 ± 18.74 and 68.68 ± 14.91 for sessions 1 and 2, respectively. In 2008, WES was 76.80 ± 19.76 and 75.25 ± 19.26 for sessions 1 and 2, respectively. The two-way ANOVA recognised a signifi- cant effect associated with the year, but no significant ef- fects according to sessions were observed. Furthermore, no significant difference was recognised in the Tukey– Kramer HSD test between sessions 1 and 2 in any year. The average score significantly differed between 2007 and 2008; however, because no difference was observed between the sessions in each year, results of sessions 1 and 2 for each year were considered to be reproducible. Table 1 Student results on the self-evaluation in each session Therefore, it was decided that results for each year will be analysed by combining the results of sessions 1 and 2 in groups A and B, respectively, to form the tutored group and those of sessions 1 and 2 in groups B and A, respectively, to form the tutorless group. The average value of WES and self-contentment scores (SCS) for each session was examined with the two-way analysis of variance (ANOVA) and uniformity of exami- nations was tested with the Tukey–Kramer’s honesty sig- nificant difference (HSD) test. The validity of grouping during each year was examined using the unpaired t-test and Bartlett’s test by comparing the total scores of Groups A and B. With regard to WES and self-evaluation results, the average values for the t-test and dispersion of the F-test were compared between the tutored and tutor- less PBL groups. The value of p<0.05 (two-tailed) was considered statistically significant. Furthermore, all statis- tical analyses were conducted using JMP 8.0.1 (SAS insti- tute Inc., Cary, North Carolina, USA) and Prism 6.0b (Graphpad software, Inc., San Diego, CA, USA). Results Validity of matching and reproducibility of the result In 2007, the average ± standard deviation of WES in Groups A and B was 131.45 ± 22.32 and 135.86 ± 17.91, respectively. In 2008, WES in Groups A and B was 154.69 ± 29.85 and 135.86 ± 17.91, respectively. No significant difference was observed in the average value Methods Guidance & informed consent (N=100-102) Randomized grouping Tutored PBL (7 small groups) Tutorless PBL (7 small groups) Tutored PBL (7 small groups) Tutorless PBL (7 small groups) Figure 1 Chronological list of problem-based learning (PBL) undertaken by students in this study. Group A in the left row and Group B in the right row. In the middle row, sessions experienced by all students conducted in a large classroom are chronologically displayed. Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 Page 3 of 7 scenario of chief complaints such as abdominal pain and cough, i.e. four SGDs per session. In group A, SGDs were tutored in the first session and tutorless in the sec- ond session. In group B, SGDs were tutorless in the first session and tutored in the second session. Each session was designed such that the schedule for lectures or la- boratory practices did not coincide, except for a daily short lecture related to the scenario of each day. Both groups completed a daily report on every SGDs and noted details of any self-learning. On the last day, a writ- ten examination (full marks, 100 points) on the contents of each session was conducted. A questionnaire includ- ing a 5-tiered self-evaluation on self-contentment and other items shown in Table 1 was simultaneously dis- tributed. An overall evaluation was conducted using a the faculty council of our school approved this curriculum. The Institutional Review Board exempted the study from review. These details were explained to all students prior to the start of PBL1, and they were advised that self- evaluation and a questionnaire were voluntary. Students were given the option of not participating in this re- search, but they all decided to participate. Students were randomly divided into two groups of equal numbers, Group A and Group B. Each group was randomly reor- ganised into seven small groups comprising seven or eight students for each session. All students attended two sessions in a year; every session covered a four day period (Figure 2). Small group discussions (SGDs) were conducted every day in each session about recurring Figure 2 Timetable of each problem-based learning (PBL) session. Figure 2 Timetable of each problem-based learning (PBL) session. Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 Page 4 of 7 Written examination scores and SCS in both years (2007, r = 0.023, p = 0.74; 2008, r = −0.021 p = 0.76). 74.33 ± 21.28 in the tutored and tutorless groups, respect- ively (Figure 3). In the tutorless groups, a tendency towards higher average scores during both years was observed, but this dif- ference was not significant (2007, p = 0.25; 2008, p = 0.20). However, in 2008, variances tended to be significantly larger than the average score in the tutorless group (2007, p = 0.058; 2008, p = 0.039). Self-evaluation Self-evaluation results, excluding SCS, for each session are shown in Table 1. In the tutored group of 2007, there was a tendency for high self-evaluation. In particular, self-evaluation of group dynamics were significantly dif- ferent. However, inconsistent trends were recognised in 2008. In each student during both years, there was a ten- dency for high self-evaluation in the tutored PBL session but no significant differences were found. When the re- sults of sessions 1 and 2 for both years were combined, no significant differences were found between the tutored and tutorless groups. Self-contentment scores h In 2007, the average ± standard deviation for SCS was 4.08 ± 0.78 and 3.88 ± 0.88 in the tutored and tutorless groups, respectively. In 2008, SCS was 4.37 ± 0.76 and 4.29 ± 0.74 in the tutored and tutorless groups, respect- ively (Figure 3). In both years, the average score tended to be lower in the tutorless group but this was not sig- nificant (2007, p = 0.092; 2008, p = 0.42). A tendency in the variations from the average scores was inconsistent, and no significant differences were found (2007, p = 0.23; 2008, p = 0.75). No correlation was observed between WES Written examination scores The average ± standard deviation of WES in 2007 was 67.83 ± 11.11 and 65.82 ± 13.42 in the tutored and tutorless groups, respectively. In 2008, WES was 77.85 ± 17.30 and 74.33 ± 21.28 in the tutored and tu ively (Figure 3). In the tutorless groups, a ten average scores during both years wa ference was not significant (2007, p However, in 2008, variances ten larger than the average score in th p = 0.058; 2008, p = 0.039). Self-contentment scores In 2007, the average ± standard 4.08 ± 0.78 and 3.88 ± 0.88 in the groups, respectively. In 2008, SC 4.29 ± 0.74 in the tutored and tut ively (Figure 3). In both years, th to be lower in the tutorless group nificant (2007, p = 0.092; 2008, p the variations from the average s and no significant differences were 2008, p = 0.75). No correlation was Hayashi et al. BMC Medical Education 2013 http://www.biomedcentral.com/1472-692 74.33 ± 21.28 in the tutored and tutorless groups, res ively (Figure 3). In the tutorless groups, a tendency towards h average scores during both years was observed, but th ference was not significant (2007, p = 0.25; 2008, p = However, in 2008, variances tended to be signific larger than the average score in the tutorless group ( p = 0.058; 2008, p = 0.039). Self-contentment scores In 2007, the average ± standard deviation for SCS 4.08 ± 0.78 and 3.88 ± 0.88 in the tutored and tuto groups, respectively. In 2008, SCS was 4.37 ± 0.76 4.29 ± 0.74 in the tutored and tutorless groups, res ively (Figure 3). In both years, the average score te to be lower in the tutorless group but this was no nificant (2007, p = 0.092; 2008, p = 0.42). A tenden the variations from the average scores was inconsi and no significant differences were found (2007, p = 2008, p = 0.75). No correlation was observed between Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 Page 5 of 7 and SCS in both years (2007, r = 0.023, p = 0.74; 2008, r = −0.021 p = 0.76). and SCS in both years (2007, r = 0.023, p = 0.74; 2008, r = −0.021 p = 0.76). Discussion In the present study, a mixed course comprising tutored and tutorless PBL was undertaken by first-year medical students, and the learning outcomes were analysed. The results indicate that students undertaking tutorless Figure 3 Students’ performance on the written examination score and self-contentment score during each year. The error bars represent the standard deviations. Figure 3 Students’ performance on the written examination score and self-contentment score during each year. The error bars represent Figure 3 Students’ performance on the written examination score and self-contentment score during each year. The error bars represent the standard deviations. Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 Page 6 of 7 A possible reason why WES and SCS did not signifi- cantly differ between the tutorless and tutored groups may be because students in the tutorless group commu- nicated with those in the tutored group after every SGD, thereby allowing students in the tutorless group to get some information from those in the tutored group. This helped in decreasing the gap in learning outcomes be- tween both groups. This may also prompt peer-assisted learning for students in both groups. Ross and Cameron [23] argued that peer-assisted learning is an efficient and effective way of preparing medical students for their fu- ture role as educators. Although little attention has been paid to the effects of peer-assisted learning in medical schools [24], the positive effects of peer-assisted learning in medical education is gaining notoriety [24,25]. Al- though there have been no studies comparing tutorless PBL to student-tutored PBL, it appears that the effect of peer-assisted learning in student-tutored PBL is higher than that of tutored and tutorless PBL; however, a mixed course, including tutorless and tutored PBL, has unique characteristics as well as the potential to provide good learning outcomes. PBL were adequately prepared for written examinations, whereas tutored PBL prepares students for written exam- inations and also increases self-contentment with their learning experience. Tutorless PBL is an efficient way to reduce demands on faculty time and resources [6,8,11,12,14,15]. Discussion However, tutorless PBL may give rise to several problems that may impede learning [21], and according to Duncan-Hewitt [8] these are as follows: (1) students’ emotions can inter- fere with their willingness to participate and decrease their quality of learning, (2) misapprehensions and weak thinking, group and problem-solving skills can cause students to become engrossed in the problem-solving process, and the problem-solving skills can cause students to concentrate more on the problem-solving process and (3) there is no opportunity to directly examine students’ abilities and skills, which can be systematically improved. Despite these problems, we did not provide any special intervention for students in the tutorless group; however, sufficient guidance was provided to all students in both the tutorless and tutored groups prior to PBL. In addition, following reports regarding SGD and self-learning, we provided formative evaluation and feedback every day. With regard to the fact that WES and SCS did not significantly differ between the tutorless and tutored groups, it appears that these traditional complementary methods may have affected the learning outcomes of the tutorless group. There are two primary limitations to the current study. First, we were unable to develop other methods of meas- uring the effects of group learning and could measure only self-evaluation of group dynamics and attentiveness. The other types of evaluations such as peer evaluation may be more important in the tutorless group. Second limitation relates to the setting (four SGDs in a week session) and period (pre-clinical students), in which this study was conducted. Usually PBL tutorials are con- ducted in the first and second sessions with two to three days time for self study. A curriculum in which the SGD frequency is reduced during each session would be re- quired. Moust et al. [26] reported that students in faculty-led PBL performed better than those in peer- facilitated groups during essay examinations designed to assess higher-order cognitive skills. Thus, we completed PBL2 for the third- and forth-year students as a tutored PBL course. More research on the cognitive effects of tutorless PBL for medical students during their clinical years is required. In contrast, variances in WES scores were considerably greater in the tutorless than tutored groups. Moreover, the average SCS in the tutored group tended to be higher than that in the tutorless group. Discussion The daily report also appeared to show that the quality and quantity of SGD and self-learning varied widely in the tutorless group. These results imply that it is unavoidable that tutorless PBL may give rise to some students who do not learn well without a structured process, thereby re- ceiving lower scores. Although the daily short lectures directed the students to study, Lee et al. [22] reported that these were not correlated with perceived self di- rected learning ability in their case-oriented problem- stimulated mixed PBL curriculum for the 1st and 2nd- year medical students. Nicholl and Lou [15] argued that in their tutorless PBL model, it was important to provide as many opportunities as possible for formative assess- ment in order to monitor and adjust the development of the tutorless groups. We agree with their argument that ongoing and immediate formative assessment is valuable in tutorless PBL. Recent case reports have suggested that the use of pre-set cues, particularly pictures and videos [14] or e-learning resources [11] can help conduct effect- ive tutorless PBL. To improve tutorless PBL, these new learning resources can positively contribute to students’ self-contentment with their learning experience. Competing interests g 20. Leary HM, Walker AE, Fitt MH, Shelton BE: Expert Versus Novice Tutors: Impacts on Student Outcomes in Problem-Based Learning. Paper presented at the Annual Meeting of the American Educational Research Association, San Diego, CA; 2009. http://digitalcommons.usu.edu/itls_facpub/6/. The authors declare that they have no competing interests. Authors’ contributions SH d d ll d SH conducted all studies, performed statistical analyses and drafted the manuscript. KT and CI helped draft and critically appraise the manuscript. YF was involved in the conceptualisation of the study and participated in its design and coordination. All authors read and approved the final manuscript. 21. Duek JE, Wilkerson L, Adinolfi T: Learning issues identified by students in tutorless problem-based tutorials. Adv Health Sci Educ 1996, 1:29–40. 22. Lee Y-M, Mann KV, Frank BW: What drives students’ self-directed learning in a hybrid PBL curriculum. Adv Health Sci Educ 2010, 15:425–437. 23. Ross MT, Cameron HS: Peer assisted learning: a planning and implementation framework: AMEE Guide no. 30. Med Teach 2007, 29:527–545. Acknowledgements The authors wish to thank Dr. Munekazu Naito, for his comments on the preliminary draft and Ms. Akemi Mitsuya, for the excellent secretarial assistance. The authors wish to thank Dr. Munekazu Naito, for his comments on the preliminary draft and Ms. Akemi Mitsuya, for the excellent secretarial assistance. 24. Peets AD, Coderre S, Wright B, Jenkins D, Burak K, Leskosky S, McLaughlin K: Involvement in teaching improves learning in medical students: a randomized cross-over study. BMC Med Educ 2009, 9:55. Author details 1 25. Ten Cate O, Durning S: Peer teaching in medical education: twelve reasons to move from theory to practice. Med Teach 2007, 29:591–599. Shuman S: structure, mechanism, and evolution of the mRNA capping apparatus. Prog Nucleic Acid Res Mol Biol 2000, 66:1–40. 1Medical Education Center, Aichi Medical University School of Medicine, Nagakute, Aichi 480-1195, Japan. 2Department of Pharmacology, Aichi Medical University School of Medicine, Nagakute, Aichi, Japan. 3Aichi Medical University Graduate School of Medicine, Nagakute, Aichi, Japan. 1Medical Education Center, Aichi Medical University School of Medicine, Nagakute, Aichi 480-1195, Japan. 2Department of Pharmacology, Aichi 26. Moust JHC, DeVolder ML, Nuy HJP: Peer teaching and higher level cognitive learning outcomes in problem-based learning. High Educ 1989, 18:737–742. Received: 16 January 2013 Accepted: 15 November 2013 Received: 16 January 2013 Accepted: 15 November 2013 Published: 1 December 2013 doi:10.1186/1472-6920-13-158 Cite this article as: Hayashi et al.: Comparison of tutored group with tutorless group in problem-based mixed learning sessions: a randomized cross-matched study. BMC Medical Education 2013 13:158. Conclusions Tutorless PBL can potentially produce learning outcomes that are comparable to tutored PBL; however, tutorless PBL is different from faculty/staff-tutored PBL and student-tutored PBL. Tutorless PBL has been used when PBL conducted in large classrooms [6,8,11,12,15]. How- ever, tutorless PBL should not be easily used in the same way as student-tutored PBL because of the difficulty in maintaining faculty tutors or learning rooms. An appropri- ate and effective curriculum can be administered in every school by combining tutored PBL and student-tutored PBL Hayashi et al. BMC Medical Education 2013, 13:158 http://www.biomedcentral.com/1472-6920/13/158 Page 7 of 7 or tutorless PBL. We encourage the implementation of PBL in schools because this will potentially lead to further developments in the area of PBL. An Evaluation of its Present Status. Edited by Nooman Z, Schmidt H, Ezzat E. New York, USA: Springer; 1990:123–134. An Evaluation of its Present Status. Edited by Nooman Z, Schmidt H, Ezzat E. New York, USA: Springer; 1990:123–134. An Evaluation of its Present Status. Edited by Nooman Z, Schmidt H, Ezzat E. New York, USA: Springer; 1990:123–134. 18. Johansen ML, Martenson DF, Bircher J: Students as tutors in problem-based learning: does it work? Med Educ 1992, 26:163–165. learning: does it work? Med Educ 1992, 26:163–165 19. Sorbal D: Peer tutoring and student outcomes in a problem-based learning course. Med Educ 1994, 28:284–289. References 1. Barrows HS, Tamblyn RW: Problem-based learning. New York: Springer; 1980. 1. Barrows HS, Tamblyn RW: Problem-based learning. New York: Springer; 1980. 2. Holen A: The PBL group: self-reflections and feedback for improved learning and growth. Med Teach 2000, 22:485–488. 2. Holen A: The PBL group: self-reflections and feedback for improved learning and growth. Med Teach 2000, 22:485–488. learning and growth. Med Teach 2000, 22:485–488. 3. Kassab S, Abu-Hijleh MF, Al-Shboul Q, Hamdy H: Student-led tutorials in problem-based learning: educational outcomes and students’ perceptions. Med Teach 2005, 27:521–526. 4. Dolmans DH, Gijselaers WH, Moust JH, de Grave WS, Wolfhagen IH, van der Vleuten CP: Trends in research on the tutor in problem-based learning: conclusions and implications for educational practice and research. Med Teach 2002, 24:173–180. 5. Chng E, Yew EH, Schmidt HG: Effects of tutor-related behaviours on the process of problem-based learning. Adv Health Sci Educ Theory Pract 2011, 16:491–503. 6. Woods DR, Hall FL, Eyels CH, Hrymak AN, Duncan-Hewitt WC: Tutored versus tutorless groups in problem-based learning. Am J Pharm Educ 1996, 60:231–238. 7. Moust JC, Schmidt HG: Effects of staff and student tutors on student achievement. High Educ 1994, 28:471–482. 8. Duncan-Hewitt WC: A Focus on process improves problem-based learning outcomes in large classes. Am J Pharm Educ 1996, 60:408–416. 8. Duncan-Hewitt WC: A Focus on process improves problem-bas 9. Steele DJ, Medder JD, Turner P: A comparison of learning outcomes and attitudes in student- versus faculty-led problem-based learning: an experimental study. Med Educ 2000, 34:23–29. 9. Steele DJ, Medder JD, Turner P: A comparison of learning outcomes and attitudes in student- versus faculty-led problem-based learning: an experimental study. Med Educ 2000, 34:23–29. 10. Solomon P, Crowe J: Perceptions of student peer tutors in a problem-based learning programme. Med Teach 2001, 23:181–186. 10. Solomon P, Crowe J: Perceptions of student peer tutors in a problem-based learning programme. Med Teach 2001, 23:181–186. 11. Roberts C, Lawson M, Newble D, Self A, Chan P: The introduction of large class problem-based learning into an undergraduate medical curriculum: an evaluation. Med Teach 2005, 27:527–533. 11. Roberts C, Lawson M, Newble D, Self A, Chan P: The introduction of large class problem-based learning into an undergraduate medical curriculum: an evaluation. Med Teach 2005, 27:527–533. References 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: 12. Pastirik PJ: Using problem-based learning in a large classroom. Nurse Educ Pract 2006, 6:261–267. 12. Pastirik PJ: Using problem-based learning in a large classroom. Nurse Educ Pract 2006, 6:261–267. 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 13. Kingsbury MP, Lymn JS: Problem-based learning and larger student groups: mutually exclusive or compatible concepts – a pilot study. BMC Med Educ 2008, 8:35. 13. Kingsbury MP, Lymn JS: Problem-based learning and larger student groups: mutually exclusive or compatible concepts – a pilot study. BMC Med Educ 2008, 8:35. • Convenient online submission • Thorough peer review 14. Kaliyadan F, Amri M, Dhufiri M, Amin TT, Khan MA: Effectiveness of a modified tutorless problem-based learning method in dermatology – a pilot study. J Eur Acad Dermatol Venereol 2012, 26:111–113. y 15. Nicholl TA, Lou K: A model for small-group problem-based learning in a large class facilitated by one instructor. Am J Pharm Educ 2012, 76:117. 16. Young L, Papinczak T: Strategies for sustaining quality in PBL facilitation for large student cohorts. Adv Health Sci Educ Theory Pract 2012, 21:1–9. 17. De Grave W, De Volder M, Gijselaers W, Damoieaux V: Peer teaching and problem-based learning: tutor characteristics, tutor functioning, group functioning, and student achievement. In Innovation in Medical Education: 17. De Grave W, De Volder M, Gijselaers W, Damoieaux V: Peer teaching and problem-based learning: tutor characteristics, tutor functioning, group functioning, and student achievement. In Innovation in Medical Education:
https://openalex.org/W4393988831	https://phsreda.com/e-articles/10586/Action10586-110638.pdf	Russian	null	Subject Olympiads as a tool for career guidance work of universities	null	2,024	cc-by	1,296	DOI 10.31483/r-110638 DOI 10.31483/r-110638 Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) Publishing house "Sreda" ПРЕДМЕТНЫЕ ОЛИМПИАДЫ КАК ИНСТРУМЕНТ ПРОФОРИЕНТАЦИОННОЙ РАБОТЫ ВУЗОВ Аннотация: широкомасштабная профориентационная работа должна определяться состоянием рынка труда с одной стороны, и с осознанным вы- бором абитуриента – с другой. При подготовке к поступлению в вуз особое преимущество имеет участие в профильных олимпиадах. Олимпиада дает учащемуся дополнительные баллы при поступлении в вуз и в свою очередь мо- жет рассматриваться как форма привлечения талантливой молодежи в выс- шую школу. Ключевые слова: предметные олимпиады, профориентация, довузовская подготовка, вуз. Современное общество требует от университетов соответствия социаль- ным запросам и вызовам XXI века, так как заинтересовано в высоком профес- сионализме деятельно-ориентированных молодых людей. Работать над привле- чением абитуриентов необходимо с учетом базисных принципов системности образовательного цикла, который способствовал бы обеспечению тесного вза- имодействия между кластерами основного (общего) и высшего образования. Широкомасштабная профориентационная работа должна определяться состоя- нием рынка труда с одной стороны, и с осознанным выбором абитуриента – с другой. Возникающая необходимость тесного сотрудничества разноуровневых ступеней образования, начиная с организаций общего, среднего профессио- Издательский дом «Среда» нального и дополнительного образования до высшей школы, требует подклю- чения представителей органов власти, руководителей системы образования му- ниципального и областного значений [1]. нального и дополнительного образования до высшей школы, требует подклю- чения представителей органов власти, руководителей системы образования му- ниципального и областного значений [1]. Интеллектуальным центром в этом сотрудничестве становится универси- тет, где возможно пересечение, и в конечном итоге – достижение общих целей, положительно отражающихся в развитии муниципальных образований [2]. Развитие и реализация профориентационной деятельности через довузов- ское образование поддерживает направление модернизации отечественной высшей школы. Предметные олимпиады, в данном случае, выступают элемен- том системы инициирующей среди учащихся компетентностную ориентацию с формированием активного мотивационного потенциала к будущей профес- сии [4]. Олимпиады, сегодня, более надежны, чем высокие баллы ЕГЭ, и являются гарантией поступления в сильнейшие вузы страны на самые востребованные специальности. Кроме дополнительных баллов, олимпиады дают и другие пре- имущества, в том числе, мотивацию расширять предметный кругозор. Подготовиться к олимпиаде невозможно в рамках отведенных школьной программой часов. Объем работы, по извлечению информации из периодиче- ских и непериодических литературных источников – колоссален и может быть осуществлен только в том случае, когда это становится образом жизни. Необ- ходимо одним словом «гореть». Участвовать в конкурсно-олимпиадном движении под силу только насто- ящему эрудиту. Подготовка к олимпиаде по конкретной дисциплине, требует от школьника высокого уровня знания смежных дисциплин, например, участие в олимпиаде по биологии подразумевает подготовку по химии, физике и даже математике. Участие в олимпиадах способствует развитию креативного мышления. Оригинальность и сложность олимпиадных заданий в разы превышает таковую типичных экзаменов и контрольных. ttps://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) 2 https://phsreda.com ПРЕДМЕТНЫЕ ОЛИМПИАДЫ КАК ИНСТРУМЕНТ ПРОФОРИЕНТАЦИОННОЙ РАБОТЫ ВУЗОВ Возникает потребность нестандартно мыслить и уметь взглянуть на вопрос под неочевидным углом. Регулярная тре- 2 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) Publishing house "Sreda" нировка в данной области неизменно приводит к формированию творческого подхода к решению проблем, что находит отражение в профессиональной кон- курентоспособности. Отдельного внимания заслуживает способность к организации времени. Постоянная работа и затрачиваемые усилия по совмещению процесса подго- товки к олимпиадам с учебой и другими отраслями жизни, требует от школьни- ка соблюдения жесткого тайминга. Регулярные занятия по заранее составлен- ному графику, позволяют олимпиаднику освоить навыки управления своим временем, стать более эффективным не только в организации учебного процес- са, но и в самой продуктивности. Способность к постановке целей и задач, со- действует равномерному распределению нагрузок и движению к поставленно- му результату последовательно. Грамотное планирование деятельности находит отражение в успешном решении актуальных жизненных ситуаций во время учебы в вузе, на работе, и, в целом, в любой сфере [5]. Процесс подготовки будущего участника олимпиады намного конструк- тивнее механической отработки, так называемого «натаскивания», на решение типовых задач единого государственного экзамена. Старшеклассник маши- нально, и в тоже время, направленно повышает уровень подготовки, отражени- ем которого является высокая позиция в рейтинге учащихся. Еще одно несо- мненное преимущество олимпиады проявляется в психологической устойчиво- сти к условиям стресса государственного экзамена, базирующейся на полной уверенности в своих силах. В России ежегодно организуется проведение большого количества олим- пиад. Среди предлагаемого многообразия необходимо определиться с выбором олимпиады, отвечающей требованиям выпускника [3]. Среди критериев выбора можно выделить следующие: − заинтересованность предметом; − заинтересованность предметом; − цели и задачи, устанавливаемые абитуриентом; − задания олимпиад прошлых лет; − задания олимпиад прошлых лет; − место проведения. − место проведения. 3 Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) Издательский дом «Среда» Издательский дом «Среда» Направленность олимпиад обычно совпадают со школьными предметами. Среди них: обществознание, русский язык, математика, физика, химия, биоло- гия, информатика, иностранные языки, безопасность жизнедеятельности. В ка- честве примера приведены олимпиады, где возможно участие по профилям: − Всероссийская олимпиада школьников по профилям: экология, астроно- мия, искусство (мировая художественная культура), технология, основы без- опасности жизнедеятельности; − Всероссийская олимпиада школьников по профилям: экология, астроно- мия, искусство (мировая художественная культура), технология, основы без- опасности жизнедеятельности; − олимпиада школьников «Ломоносов», организатор МГУ им. М.В. Ломоносова: астрономия, робототехника, право, экология; − НИУВШЭ «Высшая проба» – китайский и японский языки, финансовая грамотность и экономика, психология, право, искусство, технология. ttps://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) 4 https://phsreda.com ПРЕДМЕТНЫЕ ОЛИМПИАДЫ КАК ИНСТРУМЕНТ ПРОФОРИЕНТАЦИОННОЙ РАБОТЫ ВУЗОВ − НИУВШЭ «Высшая проба» – китайский и японский языки, финансовая грамотность и экономика, психология, право, искусство, технология. Планировать участие в олимпиадах необходимо не только со среднего школьного звена, но даже с младшего. Это позволит углубить знания не по од- ному предмету, а по нескольким одновременно. Такой принцип поможет буду- щему олимпиаднику применять междисциплинарный подход в решении зада- ний, а также не ограничивать свое участие одним профилем. Желание поступить на бюджетные места престижных вузов России, спо- собствует выбору высокоуровневых олимпиад, таких Всероссийская олимпиада школьников, олимпиады, ведущих вузов Москвы, Санкт-Петербурга, Казани, Новосибирска и других крупных регионов страны [4]. Тщательное изучение профилей и привилегий вузовских олимпиад должно стать инструментом успешного поступления в выбранный вуз. Одновременное участие в олимпиаде по нескольким профилям – расширяют возможности аби- туриента. Особое место занимают внутривузовские олимпиады, организаторами ко- торых выступают региональные высшие учебные заведения. Прямым назначе- нием таких мероприятий является работа по выявлению одаренных учащихся. С целью приобщения школьников к олимпиадному движению необходимо ак- тивно привлекать их к научно-исследовательской деятельности, в частности, на уровне отдельного учебного заведения. Повысить привлекательность вуза, в 4 https://phsreda.com Содержимое доступно по лицензии Creative Commons Attribution 4.0 license (CC-BY 4.0) Publishing house "Sreda" условиях высокой конкуренции возможно посредством доступности информа- ции об олимпиадных мероприятиях. Региональным университетам необходимо создавать на базе школ и среднеспециальных учебных заведений филиалы ка- федр университета; включать специализированные школьные научно- исследовательские кружки в секции конференций, проводимых на базе универ- ситета. Эффективная довузовская работа, в том числе, посредством проведения внутренних олимпиад, позволит увеличить процент одиннадцатиклассников, которые становятся студентами этого же вуза. Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) Список литературы 1. Гапонюк П.Н. Довузовская подготовка как компонент непрерывного об- разования / П.Н. Гапонюк // Педагогика. – 2011. – №9. – С. 122–124. 1. Гапонюк П.Н. Довузовская подготовка как компонент непрерывного об- разования / П.Н. Гапонюк // Педагогика. – 2011. – №9. – С. 122–124. 2. Казакова О.Н. Система выявления и поддержки талантливых детей в университетском образовательном кластере / О.Н. Казакова, М.С. Пашкевич, Е.Н. Диденко // Университетский комплекс как региональный центр образова- ния, науки и культуры: материалы Всероссийской научно-методической конфе- ренции (с международным участием); Оренб. гос. ун-т. – Оренбург: Универси- тет, 2013. – С. 2966–2975. EDN REGKXT 3. Перечневые олимпиады [Электронный ресурс]. – Режим доступа: https://propostuplenie.ru/article/perechnevye-olimpiady-shkolnikov-i-vseros-2020/ (дата обращения:15.09.2020). 4. О плюсах и минусах Всероссийских олимпиад школьников [Электрон- ный ресурс]. – Режим доступа: https://www.gazeta.ru/science/ 2016/05/16_a_8242919.shtml (дата обращения: 15.09.2020). 5. Казакова О.Н. Довузовское образование в системе современной профес- сиональной подготовки / О.Н. Казакова [Электронный ресурс]. – Режим досту- па: https://cyberleninka.ru/article/n/dovuzovskoe-obrazovanie-v-sisteme- sovremennoy-professionalnoy-podgotovki (дата обращения: 16.09.2020). 5 5 Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0) Content is licensed under the Creative Commons Attribution 4.0 license (CC-BY 4.0)
https://openalex.org/W2295894593	https://periodicos.uff.br/contracampo/article/download/17353/10990	Portuguese	null	Nas brechas do real esfacelado	Contracampo/Revista Contracampo	2,003	cc-by	2,959	RESENHA RESENHA * Mano Bruno é filósofo, doutor em Teoria Psicanalítica pela UFRJ e Professor do Instituto de Letras da UERJ Nas brechas do real esfacelado Mano Bru no(UERJ» Uma escrita impecável, com palavras estudadas, belas imagens quase apolíneas - não obstante, o que mais nos toca no livro Tela atravessada, do jovem André Queiroz, é esta espécie de "dandysnio" discreto, um inegável retraimento da presença que contrasta com suas formulações provocantes. ágeis e coreográficas. Trata-se de um ensaio sobre cinema e filosofia, onde a convicção transmitida ao corpo das palavras explode, através de diversos filmes, em fotografias do mundo contemporâneo que perdeu suas referências (éticas, políticas e etc.) Esse parece o traço invisível que costura a seleção dos filmes: resistir às armadilhas do "vazio do sentido" promover na esfera desse vazio a moldura para uni outro pensar, um pensar pesado, indo dos simulacros atuais à loucura da corporeidade esgarçada do sem dentro algum da obra de Artaud. 153 A descorporificação do presente André Queiroz não abandona a estratégia dupla comum a muitos livros: ter dois títulos. Assim escreve Tela atravessada (é um livro sobre cinema), mas o subtítulo nos diz que isso ocorre a propósito da relação entre a filosofia e cinema. E tendo em conta a diversidade dos filmes, quais questões filosóficas sustentam a vocação deambulatória dos diferentes ensaios presentes nessa obra? A primeira é com certeza o esfacelamento do real a partir de uma fractalização de valores no mundo contemporâneo. "Pós-orgia", este termo, retomado por André, é uma expressão criada por Jean Baudrillard ao traçar uni quadro mais ou menos desolador dos tempos em que vivemos. André caminha junto com Baudrillard para pensar, no limite de todos os desmantelamentos presentes nos dias atuais, um deserto que parece crescer pelas bordas. E com certeza o cinema tem conseguido expressá-lo. A filmografia escolhida permite ao livro Tela atravessada que não se detenha apenas numa constatação perturbada perante os sinais aterradores de niilismo e barbárie que nos são contemporâneos. Procura um fio, ainda tênue, para nos conceber uma arte alegria trágica, mesmo que CONTRACAMPO suspeitando do vazio de idéias deste nosso virar de século. Talvez o termo "pós-orgia" não tenha a força de um conceito filosófico, mas serve para nomear uma descorporificação do presente, uma desmaterialização do real. É em torno dessa descorporificação que podemos ieunir o cinema de diretores tão diferentes, entre eles Costa-Gravas, Peter Weir, Danny Boyle, Walter Salies Jr. (e Danielia Thomas)? , y y , ( ) Poder-se-á suspeitar, quando se fala em "vazio", o que se lamenta é isto: não existem hoje idéias que salvem nem idéias que fündamentem. Com certeza, Baudrillard tem levado esse problema contemporâneo às ultimas conseqüências. Desde seus trabalhos iniciais, ele tem associado o fetichismo a um desejo perverso de código que expulsa as contradições pertencentes aos processos reais. Segundo Baudrillard, a mercadoria invade campos novos, distantes do valor específico da troca econômica. A lógica da mercadoria é uma sistematização progressiva atuando num sistema virtualmente total capaz de substituir todos os valores por valores signos. A fetichização da mercadoria é um trabalho de significação e codificação. Deste modo, esvazia-se a corporeidade do real, tornando-o investido de desejos errantes. A descorporificação do presente Já em textos relativamente recentes, é caso de A transparência do mal (1990), Baudrillard aposta numa visão hiperbólica do sistema de equivalências, no qual o "sistema funciona não tanto pela mais- valia da mercadoria mas pela mais-valia estética do signo" (BAUDRILLARD, J. (1990) p.23.), é a era da "pós-orgia", ou do "estádio fractal" de todos os valores: 1. 54 A lei que nos é importante é a da confusão dos gêneros. Tudo é sexual. Tudo é político. Tudo é estético. [ ... I - o grau Xérox da cultura. Cada categoria é levada a seu mais alto grau de generalização e, por isso, perde toda a especificidade e se desfaz em todas as outras. Quando tudo é político, nada mais é político, e a palavra já não tem sentido, Quando tudo é sexual, nada mais é sexual, e o sexo perde toda a determinação. Quando tudo é estético, nada mais é belo nem feio, e a própria arte desaparece (Ibidem,p. 15e16.). Era do "pós-orgia" significa a perda de qualquer referência, num grau máximo de generalidade os signos todos se equivalem e circulam soltos, nada os fundamenta. É nesse mundo sem referência que André Queiroz situa a temática de O quarto poder, filme de Costa-Gravas. Num presente descorporificado refém e seqüestrador se equivalem. A função! acontecimento (seqüestro) pode ir passando de mão em mão por sua precariedade. E mais: o repórter torna-se o grande seqüestrador porque conhece o caráter auto-referente da imagem. g O Show de Truman, de Peter Weir, torna essa desmaterializáção do real mais evidente. Trata-se dos limites de uma cidade-cenário que padece de tempo real. CONrRAcAMP0 Completando esse quadro temos. em Tela atravessada, a interpretação do filme Trainspottine: sem limites de Danny Boyle. O filme apresenta vidas num estilo bordeline (indo da euforia à overdose). É o mundo do pós-orgia onde o tempo, abolido de todo acontecimento, anuncia o fim do político, do sexual, do estético. Diz-nos André: "o vazio e a desesperança típicos do pós-orgia, seriam suficientes para condenar de antemão quaisquer agenciamentos coletivos de desejo" (QUEIRÓZ, A. (2001) p.78). Resta o mais importante: o que fazer no após a orgia, quando todos os signos se equivalem? Na visão de André, essa é a questão que aparece em Terra estrangeira, de Walter Salies Jr e Daniella Thomas: a fuga de uni país de terceiro mundo, que massacra todas as possibilidades de singularização. A descorporificação do presente força o ingresso numa europa xenófoba onde as ofertas subjetivas não chegam a permitir linhas de fuga que ativem no corpo um contorno alegre. Imagens não-subjetivas, acontecimentos pré-individuais Poderá o leitor ser sensível a uma segunda vertente filosófica em atravessada. Não é mais o nome de Baudrillard que salta aos olhos e sim o de GuIes Deleuze. O grande mérito consiste em não procurar uma "mensagem" deleuziana, mas um "movimento" que cria a partir das imag seu próprio campo operatório, resgatando nos filmes a mobilidade vertiginosa de cada um (a fidelidade aí é fundamental, são filmes vertiginosos). 1 55 g ) Diretores tão diferentes, Kiarostami, Bergman, Kieslowski, Resnais. o que os aproxima? André parece responder essa pergunta a partir de Gosto de cereia, de Kiarostami, há algo de trágico nesse filme: o apagamento da individuação para a determinação do singular. E a questão se desdobra: assim podemos dizer que Never e Hiroshima, em Hiroshima mon amour, de Resnais, são nomes de singularidades, de estados intensos que ultrapassam os limites individualizados dos corpos das personagens ou das cidades. E o que são esses estados afetivos intensos senão a experimentação do "outro" que quebra a "ideologia da segurança material". E isso nos remete ao problema central do filme de Bergman, Cenas de um casamento: a singularização "deve ser catalizadora de práticas de 'outramenlos' que nos desestabilize de nossas certezas, que nos diferencie do que nós somos, e que nos desautorize de nós mesmos" (ibidem. p87). q q p E claro que Tela jatravessada não se deixa inserir numa quadrícula apertada de explícitação uma teoria da diferença no mundo contemporâneo através do cinema. Há neste livro uma tentativa de manter a distância justa eníre r. teoria a as imagens. g Seguindo unia fundamentação deleuziana, o autor de Tela atravessada percorre. através de uma filmografia bem selecionada, o universo dos confrontos entre a asfixia sem linhas de fuga do mundo atual e o campo CONTRACAMP() processual dos afetos com seus riscos desterritorializantes. Com isso vemos a "estrangeiridade" da trilogia de Kieslowski (Trois Couleurs) que chama a todo instante por planos de consistência, mas também vemos Na companhia dos homens, de Labute, com seus vetores de microfacismos achatando singularizações préindividuais, num universo under2round feito de trapaças. Uma face bem próxima disso André encontra, também, nofilme Ódio, de Kassovitz, através do problema dos skinheads e do medo crescente de terceiro mundialização dos países desenvolvidos. O acaso singular sobresi mesmo A terceira influência em Tela atravessada é de Carlos Henrique Escobar. Na verdade trata-se de um "anexo". Neste ponto assumo uma primeira pessoa do singular para uma confissão: leio Escobar com certo temor. Medo de entrar numa matéria espessa, desgrenhada, grumosa, em que me perco e me escapa. Mas o artigo "Nietzsche/Artaud e lugar do pensamento- outro"nos vem em auxílio, ajuda a tornar as coisas um pouco mais claras, dando uma gravidade extrema às questões. A referência é a Cena - Nietzsche atravessando a loucura sem interioridade de Van Gogh e Artaud. O que se convencionou chamar de Razão entra em confronto com a radical abismal idade de um "pensar pesado do pensamento" que pensa a partir do "acaso singular sobre si mesmo"; Este artigo evoca a porção de loucura na fragilidade dos olhares e a exterioridade mesma do "além homem". O que está em questão é o trágico da dor, da loucura e da morte (nessa errança da cena/corpo de Nietzsche a sussurrar o pensamento do eterno retorno). Salta-nos aos olhos a solidão Nietzsche/Zaràtustra experimentada onde ela se mostra mais espessa. E nesse "pensar pesado" os conceitos se avolumam, seja na cena de uma crueldade, seja na "corporeidade suja" do "pensamento-corpo". Embora não concordemos com a idéia de "destragicização" que Escobar vê em Deleuze, destacamos:a beleza da plasticidade do texto de André que enquanto pensador da diferença nos apresenta esse Outro: espécie de pensamento corporalizado em que "o Eu se decompõe numa lucidez maior e mais breve" (KLOSSOWSKI, P. (2000) p.5 1). 156 CONTRACAMPO CONTRACAMPO Submissão de artigos Contracampo acolhe colaborações de autores do Brasil edo exterior, pertencentes ou não a programas de Comunicação. Os trabalhos submetidos são avaliados por dois pareceristas do Conselho Editorial e publicados após recomendação de ambos, obedecendo às datas de fechamento dos números semestrais. SOMENTE ARTIGOS INÉDITOS PODERÃO SER SUBMETIDOS A CONTRACAMPO. Formato e preparação de artigos para submissão Os trabalhos submetidos devem possuir entre 5000 a 7000 palavras, ou de 13 a 15 laudas, digitadas em Word (ou similar), fonte Times New Roman, corpo 12. O formato da página adotado para os trabalhos é o A4 (21 x 29.7cm) com as seguintes medidas: 2cm de margem superior: 3cm de margem inferior, 5cm de margem esquerda, esquerda, 3cm de margem direita e medianiz 0. Cabeçalho de 1.2 cm e rodapé de 1,6 cm. Alinhamento pela esquerda somente. O espaço entre linhas deve ser simples para o corpo do texto, e o espacejamento do parágrafo deve ser de 12 antes e 6 pontos depois de cada parágrafo, com recuo direito e esquerdo de valor 0. 157 Todos os parágrafos devem ser indentados em 0.5cm. NORMAS PARA PUBLICAÇÃO Submissão de artigos Título e autoria Fonte Anal, negrito, corpo 12, centrado e espacejamento entre linhas de 1,5 linhas. Usar caixa-alta (letra maiúscula) apenas para a primeira letra do título do artigo, exceto para nomes próprios ou palavras que exijam o uso de caixa-alta por razões gramaticais. p g O nome do autor deve ser colocado abaixo da última linha do título, centrado, em Anal, itálico, corpo 10. Se o artigo possuir mais de um autor, separar os nomes destes com um vírgula e espaço. Se o número de autores exceder uma linha, não dividir o nome de um autor: coloque-o na linha seguinte. O nome de cada autor deve estar na seguinte ordem: primeiro nome, segundo nome ou inicial (se for o caso) e o sobrenome. g ( ) Incluir no final da primeira página do artigo nota de rodapé (com entrada após o sobrenome de cada autor, numeração personalizada por asterisco) com dados biográficos sobre o(s) autor(es), constante de filiação institucional, titulação, etc. com cerca de 50 palavras, em fonte Anal, regular, corpo 9. CONTRACAMPO Palavràschave e resumos As palavras-chaves e resumo em português devem vir após o título e nome dos autores. Usar 3 a 5 palavras-chaves, fonte Anal, corpo 10. Estas devem ser colocadas antes do resumo e serem antecedidas da palavra: "Palavras- chaves", alinhada à esquerda, em Anal, negrito, corpo 10. O resumo deve ter aproximadamente 150 palavras. Usar fonte Anal, corpo 10 para o resumo. As especificações para parágrafo e coluna são as mesmas do corpo do texto (siga as instruções dadas para margens, espacejamento, indentação, coluna e alinhamento). O resumo deve ter como título a palavra "Resumo", alinhada à esquerda, em Anal negrito, corpo 10. Na linha seguinte coloque o texto do resumo. As palavras-chaves e resumo em inglês devem observar o mesmo formato para o português, sendo, porém, em itálico. Devem vir após o resumo em português. Substituir os títulos: "Palavras-chaves" por "Keywords" e "Resumo" por "Absfract" , para as palavras-chaves e resumo em inglês. O texto do artigo deve ser iniciado na página seguinte às palavras- chaves (keywords) e ao resumo (abstract) em inglês. Para tanto, inserir quebra de página entre a página com título, autoria, resumo e abstract e corpo do artigo. 158 Não inserir numeração nas páginas. Referências As referências bibliográficas devem vir logo após a seção de notas e seguir a N3R 6023 DA ABNT, observado o formato de margens e espaços definidos acima. Todas as referências devem vir listadas alfabeticamente e cronologicamente na seção de referências. Sol) o título "Referências" (ver especificações para títulos de primeiro nível). Para formatar as referências use fonte Times New Roman, regular, corpo 12, alinhamento à esquerda. espacejamento simples entre linhas, recuo especial de 0,5cm da segunda tinha em diante e espaço de parágrafo de 6 pontos após cada referência. 159 Subtítulos Sugerimos o uso de não mais de três níveis de subtítulos para seu artigo, os quais devem seguir estas especificações: primeiro nível de subtítulo: Anal, negrito, corpo 11, alinhado à esquerda, espaço de parágrafo de 24 pontos antes do subtítulo e de 6 pontos após o subtítulo; segundo nível de subtítulo: Anal, negrito, corpo 10, indentado em 0.5cm, espaço de parágrafo de 12 pontos antes do subtítulo e de 6 pontos após o subtítulo; terceiro nível de subtítulo: Anal, negrito, itálico, corpo 10, indentado em 1.Ccm, espaço de parágrafo de 12 pontos antes do subtítulo e de 6 pontos após o subtítulo. Usar caixa-alta (letra maiúscula) apenas para a primeira letra do subtítulo do artigo, exceto para nomes próprios ou palavras que exijam ouso de caixa-alta por razões gramaticais. Inserir um parágrafo (Enter) entre o texto que antecede o subtítulo e este. CONTRÂCAMPO CONTRÂCAMPO Imagens Para o envio de material fotográfico. os editores devem ser previamente contatados para maiores especificações. Notas Usar notas de final de documento apenas, com exceção da nota biográfica do(s) autor, que aparece na primeira página do artigo. Usar fonte Anal, regular, corpo 9 e alinhar o texto pela esquerda. Usar notas de rodapé quando absolutamente necessário e evitar notas longas. As notas devem ser numeradas consecutivamente ao longo do texto. Citações Para citar um autor no corpo do texto use 'aspas simples'. Citações que excedam três linhas devem ser separadas do corpo do texto (aperte a tecla 'enter' unia vez). Para citações, usar fonte Anal, regular, corpo 9, alinhamento à esquerda. Espacejaniento entre linhas deve ser simples, recuo de parágrafo de 0.5cm, e espaço de 6 pontos após cada parágrafo de sua citação (se for o caso). Considerações sobre direitos autorais Os artigos submetidos devem vir acompanhados de autorização de publicação pelo autor. ç p Contracampo detém o copyrlg/it sobre o conteúdo da publicação aceita. p ç p Contracampo detém o copyrlg/it sobre o conteúdo da publicação aceita. Fotocópias de artigos são autorizadas mediante solicitação aos editores e crédito rias fontes. Para evitar violação das leis de direitos autorais, favor não utilizar longas e muitas citaçó..s de unia mesma fonte, ou figuras publicadas previamente sem um documento de autorização de uso dos direitos autorais. isto também se refere a imagens produzidas pelo autor, publicadas CONTRACAMPO em outro veículo, cujo direito autoral tenha sido transferido à editora referente ao veículo anterior. Autores que não fornecerem documentos de autorização de uso de direitos autorais terão seus artigos devoIvdos. Envio de artigos: Os originais deverão ser enviados em três cópias impressas, acompanhadas de disquete, para o seguinte endereço: Rua Tiradentes, 148, Ingá! Niterói - CEP: 24270-240 Rio de Janeiro- RJ Os originais deverão ser enviados em três cópias impressas, acompanhadas de disquete, para o seguinte endereço: Rua Tiradentes, 148, Ingá! Niterói - CEP: 24270-240 Rio de Janeiro- RJ Os originais deverão ser enviados em três cópias imp acompanhadas de disquete, para o seguinte endereço: Os originais deverão ser enviados em três cópias impressas, acompanhadas de disquete, para o seguinte endereço: Rua Tiradentes, 148, Ingá! Niterói - CEP: 24270-240 Rio de Janeiro- RJ Rio de Janeiro- RJ Contracampo Uma revista financiada pelo programa PROAP/CAPES. e-mail para contato: sibonei@gbl.com.br 160 CONTRACAMPO 161 (()N 1 RA(AMI() 162 De como Antonio Banderas perdeu o sotaque - e outros fenômenos transnacionais c Ar D+ rkr r!r' Ppov!' L4iic P p!rFc Irnrv -- r --------- Política da Representaç: Programas de desemprego prograrnac A construção de uma história do cinema brasileiro: política estatal e cinema alternativo nos anos Embrafilme r,-rr A proteção à infância na televisão européia A TV. os Literatos e as Massas no Brasil 4 T i i-1 Ismail Xavier: o cinema e os filmes ou doze temas em torno da imagem RESENHA 3 :dJc- CON SE De como Antonio Banderas perdeu o sotaque - e outros fenômenos transnacionais c Ar D+ rkr r!r' Ppov!' L4iic P p!rFc Irnrv -- r --------- Política da Representaç: Programas de desemprego prograrnac A construção de uma história do cinema brasileiro: política estatal e cinema alternativo nos anos Embrafilme r,-rr A proteção à infância na televisão européia A TV. os Literatos e as Massas no Brasil 4 T i i-1 Ismail Xavier: o cinema e os filmes ou doze temas em torno da imagem RESENHA 3 :dJc- C r,-rr 3 :dJc-
W2600511152.txt	https://revista.univap.br/index.php/revistaunivap/article/download/815/480	pt		O USO DA TECNOLOGIA EM SALA DE AULA POR PROFESSORES DE UMA ESCOLA PÚBLICA DO MUNICÍPIO DE ALEGRE-ES	Revista UniVap	2,016	cc-by	303	127 O USO DA TECNOLOGIA EM SALA DE AULA POR PROFESSORES DE UMA ESCOLA PÚBLICA DO MUNICÍPIO DE ALEGRE-ES Vanessa Fagundes Campos1 Caroline Palacio de Araujo2 Aline Tintori Mantovani3 Kalia Dável Grecco4 Janete Valani5 Kênia Rezende6 Elias Terra Werner7 Resumo: As tecnologias se difundiram em todas as esferas da sociedade, possibilitando um vasto acesso à informação para grande parte da população. Na educação, a inserção da tecnologia como ferramenta de ensino enfrenta várias dificuldades, na qual é usada de forma limitada na sala de aula. Esta pesquisa teve como objetivo identificar os principais obstáculos enfrentados pelos professores na utilização de recursos tecnológicos como ferramentas de ensino, permitindo a compreensão das mudanças necessárias para que ocorra a implementação efetiva desses recursos durante a ministração em suas aulas. O estudo foi realizado com professores do ensino fundamental e médio de uma escola pública do município de AlegreES, onde foi aplicado um questionário como instrumento para coleta de dados. Como resultado, nota-se que os professores entrevistados não têm dificuldade na utilização de tais recursos e que a escola oferece todo suporte a eles. Entretanto, estes enfrentam diariamente muitos desafios, tais como, infraestrutura e excesso de alunos, que dificultam o processo de ensino, o que leva a pouca utilização desses recursos pelos professores questionados. Palavras-chave: Ensino-aprendizagem; Educação; Recursos tecnológicos. 1 2 3 4 5 6 7 Ciências Biológicas/Universidade Federal do Espírito Santo, Brasil. E-mail: vanessafcampos@outlook.com. Ciências Biológicas/Universidade Federal do Espírito Santo, Brasil. E-mail: carolinepalacio@yahoo.com.br. Ciências Biológicas/Universidade Federal do Espírito Santo, Brasil. E-mail: alinetintori@hotmail.com. Ciências Biológicas/Universidade Federal do Espírito Santo, Brasil. E-mail: kaliagrecco@hotmail.com. Ciências Biológicas/Universidade Federal do Espírito Santo, Brasil. E-mail: janetevalani@gmail.com. Ciências Biológicas/Universidade Federal do Espírito Santo, Brasil. E-mail: kenia.rezende@yahoo.com.br. Departamento de Biologia/Universidade Federal do Espírito Santo. Brasil. E-mail: eliaswerner12@gmail.com. Revista Univap – revista.univap.br São José dos Campos-SP-Brasil, v. 22, n. 40, Edição Especial 2016. ISSN 2237-1753
https://openalex.org/W2995258867	https://europepmc.org/articles/pmc7023398?pdf=render	English	null	Utilizing Liposomal Quercetin and Gallic Acid in Localized Treatment of Vaginal Candida Infections	Pharmaceutics	2,019	cc-by	13,054	Utilizing Liposomal Quercetin and Gallic Acid in Localized Treatment of Vaginal Candida Infections Keywords: vaginal infection; liposomes; Candida; polyphenols; quercetin; gallic acid pharmaceutics pharmaceutics pharmaceutics pharmaceutics Utilizing Liposomal Quercetin and Gallic Acid in Localized Treatment of Vaginal Candida Infections Barbara Giordani 1,2 , Purusotam Basnet 3,4 , Ekaterina Mishchenko 5, Barbara Luppi 1 and Nataša Škalko-Basnet 2,* Barbara Giordani 1,2 , Purusotam Basnet 3,4 , Ekaterina Mishchenko 5, Barbara Luppi 1 and Nataša Škalko-Basnet 2,* 1 Department of Pharmacy and Biotechnology, University of Bologna, Via San Donato 19/2, 40127 Bologna Italy; barbara.giordani4@unibo.it (B.G.); barbara.luppi@unibo.it (B.L.) y; g ( ); pp ( ) 2 Drug Transport and Delivery Research Group, Department of Pharmacy, Faculty of Health Sciences, University of Tromsø The Arctic University of Norway, Universitetsveien 57, 9037 Tromsø, Norway 3 IVF Clinic, Department of Obstetrics and Gynecology, University Hospital of North Norway, Sykehusvegen 38, 9019 Tromsø, Norway; purusotam.basnet@uit.no 4 Women’s Health and Perinatology Research Group, Department of Clinical Medicine, University of Tromsø The Arctic University of Norway, Universitetsveien 57, 9037 Tromsø, Norway 5 Department of Medical Biology, Faculty of Health Sciences, University of Tromsø The Arctic University of Norway, Sykehusveien 44, 9037 Tromsø, Norway; ekaterina.mishchenko@uit.no * Correspondence: natasa.skalko-basnet@uit.no; Tel.: +47-7764-6640 Received: 19 November 2019; Accepted: 17 December 2019; Published: 20 December 2019 Received: 19 November 2019; Accepted: 17 December 2019; Published: 20 December 2019 Abstract: Vulvovaginal candidiasis (VVC) is a widely spread fungal infection that causes itching, pain and inﬂammation at the vaginal site. Although common, currently available treatment suﬀers from limited eﬃcacy and high recurrence. In addition, the growing problem of resistance to azole drugs used in current treatments emphasizes the need for superior treatment options. Antimicrobial polyphenols are an attractive approach oﬀering multitargeting therapy. We aimed to develop novel liposomes for simultaneous delivery of two polyphenols (quercetin, Q, and gallic acid, GA) that, when released within the vaginal cavity, act in synergy to eradicate infection while alleviating the symptoms of VVC. Q was selected for its anti-itching and anti-inﬂammatory properties, while GA for its reported activity against Candida. Novel liposomes containing only Q (LP-Q), only GA (LP-GA) or both polyphenols (LP-Q+GA) were in the size range around 200 nm. Q was eﬃciently entrapped in both LP-Q and in LP-Q+GA (85%) while the entrapment of GA was higher in LP-Q+GA (30%) than in LP-GA (25%). Liposomes, especially LP-Q+GA, promoted sustained release of both polyphenols. Q and GA acted in synergy, increasing the antioxidant activities of a single polyphenol. Polyphenol-liposomes were not cytotoxic and displayed stronger anti-inﬂammatory eﬀects than free polyphenols. Finally, LP-GA and LP-Q+GA considerably reduced C. albicans growth. 1. Introduction Vulvovaginal candidiasis (VVC) is an ever-living problem aﬀecting 70–75% of women of reproductive age at least once during their life. Around 40–50% of them will experience a recurrence, and especially serious is VVC among pregnant women. Candida albicans, and other non-albicans related species, are the major causative agents of VVC [1,2]. Although C. albicans is a commensal microorganism, the perturbation of vaginal homeostasis could facilitate the overgrowth of this opportunistic fungus and the onset of symptomatic candidiasis. Even if VVC is not a life-threatening problem, it can impair the quality of life of patients leading to several physical and sexual impediments, with economic costs estimated at one billion dollars per year [1–3]. Candidiasis clinically manifests mostly through Pharmaceutics 2020, 12, 9; doi:10.3390/pharmaceutics12010009 www.mdpi.com/journal/pharmaceutics 2 of 21 Pharmaceutics 2020, 12, 9 itching, pruritus, irritation, burning, pain, white cheesy discharges, vulvar and vaginal erythema, and edema [4]. Fidel et al. [5] proved that the pathogenesis of VVC has a prominent immunological component, involving the recruitment of neutrophils to the vaginal mucosa and activation of related proinﬂammatory cytokine and chemokines. This strong inﬂammatory response limits possibilities to control the Candida burden, exacerbates the tissue damage and discomfort, and can trigger chronic infections [4]. Therapeutic approaches for VVC comprise both local and oral administration of diﬀerent azole drugs, such as ﬂuconazole, ketoconazole and clotrimazole [6], that are able to reduce symptoms of an initial infection in a large number of patients [3]. Unfortunately, since all azoles have a similar fungistatic eﬀect on Candida spp., the cells exposed repetitively to these antifungals may adapt to drug pressure and became resistant [7]. Taking into consideration the emerging problem of drug resistance, as well as the high incidence of VVC, it is clear that new therapeutic strategies based on both administration of alternative antimicrobials and development of advanced delivery systems, are extremely desirable. y Among natural molecules, polyphenols have gained increasing interest in the last few years as potential candidates for Candida treatment [8,9]. Polyphenols are phytochemicals that can be principally found in cereals, grains, legumes, fruits, vegetables and in beverages, e.g., tea, coﬀee, fruit juice and cocoa [10]. Based on their chemical structure, polyphenols can be categorized into four main groups: ﬂavonoids, stilbenes, lignans and phenolic acids. In humans, besides the well-established antioxidant activity, polyphenols display many other biological eﬀects, being able to act as anti-inﬂammatory, antidiabetic, cardioprotective, antiaging [11] and antimicrobial agents [12,13]. 1. Introduction In the present work, we have, for the ﬁrst time, utilized quercetin (Q) and gallic acid (GA) as polyphenols of interest in topical therapy of vaginal Candida. Quercetin (3,3′,4′,5,6- pentahydroxyﬂavone) is the most potent antioxidant among polyphenols that has been proposed for the treatment of a wide spectrum of pathologies, including diabetes, circulatory dysfunctions and cancers [14]. The therapeutic eﬀects of quercetin mainly depend on its abilities to scavenge reactive oxygen radicals such as O2−and ONOO−, protect from lipid peroxidation and chelate metal ions. In addition, its inhibitory eﬀects on cytokine production (i.e., TNF-α and IL-8) and histamine release [15] lead to a reduction of the inﬂammation state. Moreover, several groups [16,17] have also suggested that quercetin may be an antinociceptive in animal models, thus relieving the pain associated with inﬂammation. Interestingly, Maramaldi et al. [18] demonstrated that phytosomes containing quercetin were able to exert a lenitive and anti-itch eﬀect in a group of volunteers with skin damage. Gallic acid (3,4,5-trihydroxybenzoic acid) shares with quercetin some important biological features, mostly imparted by the hydroxyl group. It is largely studied for its strong antioxidant and anti-inﬂammatory properties, as well as anti-carcinogenesis and anti-arteriosclerosis eﬀects. Furthermore, gallic acid is reported to inhibit microbial bioﬁlm formation, possess bactericidal eﬀects towards both Gram-positive and Gram-negative bacteria and exhibit antiviral activities against a human immunodeﬁciency virus [19]. Several authors proposed that gallic acid is also able to exert an antifungal activity against planktonic cultures and bioﬁlm of C. albicans [12,13,20]. Moreover, Li et al. [21] demonstrated that gallic acid inhibited the growth of diﬀerent clinical isolates of Candida spp. and proposed that the fungicidal outcome was due to the impairment of biosynthesis of ergosterol, a fundamental component of the fungal membrane. Despite the favorable pharmacological properties of polyphenols, their applicability remains limited by their low solubility and bioavailability as well as high susceptibility to environmental conditions, including biological environment. In this regard, the incorporation of polyphenols inside a nanocarrier can be an appropriate approach to avoid the degradation of active molecules and promote their deposition at the site of administration assuring enhanced biological eﬀect [22–24]. In this work, a novel liposomal system for the simultaneous delivery of quercetin and gallic acid to the vaginal cavity was developed. 2.1. Materials Lipoid S 100 from fat-free soybean lecithin, containing not less than 94% phosphatidylcholine (SPC), was a kind gift from Lipoid GmbH (Ludwigshafen, Germany). Quercetin (Q, Mw: 302.2, logP: 2.16, purity ≥95%), gallic acid (GA, Mw: 188.1, logP: 0.7 purity ≥99%), potassium phosphate monobasic, propylene glycol (PG), 2,20-azino bis(3-ethylbenzothiazoline)-6-sulfonic acid diammonium salt (ABTS), 1,1-diphenyl-2-picrylhydrazyl (DPPH), vitamin C (ascorbic acid), vitamin E (α-tocopherol), RPMI 1640 medium, bovine serum albumin (BSA), glutamine and Cell Counting Kit-8 (CCK-8) were purchased from Sigma-Aldrich, Chemie GmbH (Steinheim, Germany). Lipopolysaccharide (LPS; Escherichia coli, 055:B5), sulfanilamide, naphthylethylenediamine dihydrochloride, phosphoric acid and sodium nitrite were acquired from Sigma Life Science (Sigma-Aldrich Norway AS, Oslo, Norway). Murine macrophage RAW 264.7 cell line and Candida albicans strain (ATCC 10231) were supplied by ATCC (Manassas, VA, USA). All solvents were of analytical grade and were provided by VWR International bvba/sprl (Leuven, Belgium). Candida was grown in potato dextrose broth (Sigma-Aldrich Co., St. Louis, MO, USA) with 2% of glucose (PDBGlu) or on agar plates of the same medium (PDAGlu). Buﬀer solution at pH 4.5 simulating vaginal pH (3.5–5.5) was prepared as follows: KH2PO4 0.1 M. The composition of Griess reagent was: sulphanilamide 1%, naphthylethylenediamine 0.1% dihydrochloride, phosphoric acid 2.5%. 2.2. Preparation of Liposomes The Q and GA co-loaded liposomes (LP-Q+GA) were prepared by a ﬁlm hydration method as previously described [25], with some modiﬁcations. Brieﬂy, Q (10 mg) and SPC (200 mg) were dissolved in 10 mL of methanol in a round-bottom ﬂask. The solvent was removed by evaporation (Büchi rotavapor R-124 with vacuum controller B-721, Büchi Vac® V-500, Büchi Labortechnik, Flawil, Switzerland) under vacuum (45 mmHg) for 2 h at 50 ◦C. The resulting dry lipid ﬁlm was rehydrated with a solution of GA in distilled water (1 or 2 mg/mL), obtaining multilamellar vesicles (MLV). Liposomes containing only Q (LP-Q) or GA (LP-GA) were prepared under the same conditions by adding only Q (10, 15 or 20 mg) in the methanol lipid solution or dissolving GA in the aqueous phase (1 or 2 mg/mL), respectively. Plain liposomes (plain-LP) were prepared using only SPC (200 mg). All liposomal dispersions were stored at 4–8 ◦C overnight. 1. Introduction The combination of quercetin and gallic acid was chosen 3 of 21 Pharmaceutics 2020, 12, 9 with the aim of achieving a double eﬀect, namely the immediate and long-lasting alleviation of VVC symptomatology and the eradication of the fungus infection. The purpose was to obtain a formulation able to guarantee a modiﬁed release of both active compounds, thus assuring adequate concentration of polyphenols in the vaginal mucosa. Liposomes are vesicular nanostructures composed of one or more lipid bilayers, and are particularly suitable for the combined delivery of two molecules with diﬀerent physicochemical properties since they are able to accommodate both the lipophilic compounds, such as quercetin, and more hydrophilic substances, such as gallic acid, inside the lipid and aqueous compartments, respectively [24,25]. Liposomal formulations were characterized for their technological (size, zeta potential, entrapment eﬃciency, release behavior and stability over time) and biological properties (antioxidant and anti-inﬂammatory activities and cytotoxicity). Finally, the capability of proposed formulations to exert an antifungal eﬀect against Candida was also investigated. 2.3. Size Reduction of Liposomes Sonication was used to obtain liposomes of the desired size [26]. The sonicator (Ultrasonic processor 500 W, Sigma-Aldrich, St. Louis, MO, USA) was set to 40% amplitude and ultrasonic 4 of 21 Pharmaceutics 2020, 12, 9 irradiation for 4 cycles (30 s on/60 s oﬀ) was applied to liposomal dispersions. An ice bath was used to prevent heating of the dispersions. All sonicated liposomal formulations were stored at 4–8 ◦C. 2.7. Evaluation of Liposomes Storage Stability Samples were stored at 4–8 ◦C and kept out of light by applying aluminum foil. The stability of prepared liposomes was evaluated by measuring their mean diameter, PDI, zeta potential and entrapment eﬃciency for Q and GA after 15, 30, 60 and 90 days. Four diﬀerent batches of each formulation were analyzed in triplicate. 2.5. Zeta Potential Zeta potential analysis was performed using a Malvern Zetasizer Nano-ZS (Malvern, Oxford, UK) [28]. Measurement cells (DTS1060) were rinsed with ethanol and ﬁltered water (0.22 µm syringe ﬁlter) before loading the sample. Liposomal dispersions were diluted 1:20 (v/v) with ﬁltered water to achieve an attenuation value of 6–7. Measurements were made at 25 ◦C with an equilibration time of 180 s and number of runs set to automatic. 2.6. Determination of Polyphenol Entrapment Eﬃciency Since Q and GA display very diﬀerent water solubility (the experimentally determined values were 4.1 ± 0.3 µg/mL and 13.7 ± 0.2 mg/mL, respectively), diﬀerent protocols were applied for the determination of their entrapment eﬃciency. Therefore, the entrapment eﬃciency of quercetin (Q EE %) in LP-Q and LP-Q+GA was determined after mild centrifugation (3000× g, 15 min, room temperature; Biofuge stratos centrifuge, Heraeus instruments GmbH, Hanau, Germany) to precipitate non-encapsulated Q. Then, Q was quantiﬁed by UV–visible spectroscopy (Spark Multimode Microplate Reader, Tecan® Trading AG, Männedorf, Switzerland) at 372 nm both in the supernatants containing quercetin incorporated inside vesicles, and in not-centrifuged liposomes after dilution with methanol. The standard curve of Q in methanol was plotted at a concentration range of 2.5–37.5 µg/mL (R2 = 0.9997). Considering the good solubility of GA in distilled water, to determine its entrapment eﬃciency (GA EE %) in LP-GA and LP-Q+GA, liposomes were separated from supernatant (containing freely unentrapped polyphenol) by ultracentrifugation (92,000× g, 1 h, 10 ◦C; Beckman model L8-70M with SW 60 Ti rotor, Beckman Instruments, Brea, CA, USA). Aliquots of pellet and supernatant, as well as not-centrifuged liposomes, were diluted with methanol and GA was quantiﬁed in each fraction by UV-visible spectroscopy (Spark Multimode Microplate Reader, Tecan, Männendorf, Switzerland) at 272 nm. The standard curve of GA in methanol was set up in the 2.5–37.5 µg/mL range (R2 = 0.9996). 2.4. Determination of Vesicle Size Distribution The particle size and polydispersity index (PDI) of liposomal preparations were determined by photon correlation spectroscopy (submicron particle sizer model 370, Nicomp, Santa Barbara, CA, USA). To prevent interference from dust particles, the preparative procedure was carried out in a laminar airﬂow bench, as previously described [27]. Small aliquots of liposomal dispersions were diluted with ﬁltered distilled water (0.22 µm syringe ﬁlter, VWR International, Leuven, Belgium) to obtain a particle count range of 250–350 kHz. All measurements were run (run time of 10 min) in the vesicle mode and intensity-weight distribution at room temperature (24–25 ◦C). 2.8. In Vitro Polyphenol Release Studies The in vitro polyphenol release studies were performed by using Franz diﬀusion cells (1.77 cm2 diﬀusion area, 12 mL acceptor volume; PermeGear, Bethlehem, PA, USA) [29] equipped with a V6A Stirrer (PermeGear, Bethlehem, PA, USA). The heating circulator (Julabo Laboratechnik, F12-ED, Seelbach, Germany) was set to maintain a temperature of 37 ◦C. To mimic the vaginal pH [30], the 5 of 21 Pharmaceutics 2020, 12, 9 acceptor chambers were ﬁlled with phosphate buﬀer pH 4.5 and ethanol was added at the ﬁnal concentration of 20% (v/v) [31–33] to ensure sink conditions during the study (the solubility of Q and GA in the releasing conditions was found to be 0.226 ± 0.042 mg/mL and 78.7 ± 3.4 mg/mL, respectively). Dialysis membranes (Mw cut-oﬀ: 12,000–14,000 Daltons; Medicell International Ltd., London, UK) pre-soaked in phosphate buﬀer pH 4.5 were ﬁxed between donor and acceptor compartments [34]. LP-Q, LP-GA and LP-Q+GA were added to the donor chamber (500 µL). Solutions of Q (in PG solution 50%, w/v) and GA (in distilled water) at the same concentrations as in liposomes were used as controls. Samples (500 µL) were withdrawn from the receptor chamber every hour for 8 h and immediately replaced by an equal amount of fresh medium. The amount of Q and GA was quantiﬁed spectrophotometrically at 372 nm and 260 nm, respectively. Calibration curves for Q (R2 = 0.9998) and GA (R2 = 0.9993) in releasing medium were obtained within the concentration range 2.5–25 µg/mL. Experiments were performed in triplicate and results are expressed as a cumulative percentage of polyphenols released over time. 2.9.1. ABTS·+ Radical Scavenging ABTS·+ radicals were generated by mixing equal volumes (3 mL) of the stock solutions of ABTS (7.4 µM) and potassium peroxodisulphate (2.6 µM) in distilled water at room temperature [24]. The reaction mixture was allowed to stabilize for 18 h and then diluted with 100 mL of ethanol. A total of 300 µL of ABTS·+ radicals working solution was then added to equal volume of samples, shaken vigorously and kept in the dark for 30 min at room temperature. Reduction of blue-green colored radical solution by hydrogen-donating anti-oxidant was measured spectrophotometrically at 751 nm (Spark Multimode Microplate Reader, Tecan® Trading AG, Männedorf, Switzerland). 2.9. Antioxidative Assays The in vitro antioxidant activity of polyphenols formulated inside liposomes was assessed by employing two diﬀerent methods, namely ABTS and DPPH assay [35]. LP-Q, LP-GA and LP-Q+GA and plain-LP were serially diluted in ethanol and tested at various SPC concentrations (10–600 µg/mL), corresponding to Q and GA concentrations of 0.5–30 µg/mL. The activity levels of vitamin E and C were also tested under the same experimental conditions and used as controls for comparison. 2.10. Cell Culture The murine macrophage RAW 264.7 cell line was used to investigate a possible cytotoxic eﬀect of free and formulated polyphenols and their ability to exert an anti-inﬂammatory activity. Cells were maintained in RPMI 1640 medium supplemented with BSA 10%, streptomycin 100 µg/L, and penicillin 100 IU/mL at 37 ◦C in a 5% CO2 atmosphere. 2.9.2. DPPH Radical Scavenging DPPH radical scavenging activity was determined as reported by Jøraholmen et al. [24]. A total of 300 µL of ethanolic DPPH solution (60 µM) was thoroughly mixed with equal volume of each sample and kept in the dark for 30 min at room temperature. The disappearance of violet color after incubation time indicated a high free radical scavenging activity, spectrophotometrically quantiﬁable at 519 nm (Spark Multimode Microplate Reader, Tecan® Trading AG, Männedorf, Switzerland). Experiments were performed in triplicate and the radical scavenging activity (RSA) was calculated according to Equation (1): (1) RSA (%) = (1 −Asample/Acontrol) × 100 (1) where Asample is the absorbance of the sample and Acontrol is the absorbance of the control samples prepared by mixing the same amount of ethanol and ABTS·+ or DPPH solution. Ethanol was used as a blank sample. The combined eﬀect of Q and GA in liposomes was investigated according to Loewe additivity model [36] and the interaction index (γ) was calculated (Equation (2)) as follows: γ = d1/D1 + d2/D2 (2) (2) γ = d1/D1 + d2/D2 6 of 21 Pharmaceutics 2020, 12, 9 where D1 and D2 are the doses of Q (alone) and GA (alone) that have the RSA value of 50 (EC50) while d1 and d2 are the doses of Q and GA in the mixture that elicit the same eﬀect. where D1 and D2 are the doses of Q (alone) and GA (alone) that have the RSA value of 50 (EC50) while d1 and d2 are the doses of Q and GA in the mixture that elicit the same eﬀect. γ > 1, =1 and <1 mean that the combination eﬀect is antagonism, additive and synergy, respectively. 2 10 Cell Culture 2.12. Anti-Inﬂammatory Activity Determination The capability of liposomes to inhibit nitric oxide (NO) production in LPS-induced macrophages was evaluated as previously described and expressed as anti-inﬂammatory activity [37]. The capability of liposomes to inhibit nitric oxide (NO) production in LPS-induced macrophages was evaluated as previously described and expressed as anti-inﬂammatory activity [37]. Cells were cultured in 24-well plates (5 × 105 cells/mL) for 24 h (Section 2.10). Old medium was then removed and replaced with fresh RPMI 1640 medium (1 mL) containing LPS (1 µg/mL) to induce NO production. Cells were treated with 10 µL of liposomal formulations or solutions at the same concentrations as applied in toxicity assay (Section 2.11), and incubated for 24 h. For a negative control, cells were treated with only LPS. The NO production by macrophages was correlated to nitrite formation in the media which was quantiﬁed by Griess methods. Equal volume of media and Griess reagent (300 µL) were mixed and incubated for 30 min. The absorbance was determined at 550 nm (Agilent Technologies, Santa Clara, CA, USA) and a calibration curve (0.5–20 µM) was constructed by using NaNO2 as standard with Griess reagent. All experiments were performed in triplicate and the anti-inﬂammatory activity was expressed as percentage of NO production’s inhibition calculated with respect to control (untreated cells). 2.11. In Vitro Cell Viability Study To assess the in vitro toxicity, RAW 264.7 cells, grown as described in Section 2.10, were seeded (90 µL) in 96-well ﬂat bottom plates at the density of 5 × 104 cells/well. The plates were pre-incubated for 24 h at 37 ◦C in 5% CO2 to allow the cells to stabilize. The adherent cells were then treated with 10 µL of media only (negative control) or liposomal formulations (LP-Q, LP-GA and LP-Q+GA) and subsequently incubated for 24 h. In particular, liposomes were diluted in the growth medium and tested at three diﬀerent ﬁnal SPC concentrations, namely 1, 10 and 50 µg/mL [28], corresponding to Q and GA concentrations of 0.05, 0.5 and 2.5 µg/mL, respectively. Solutions of Q and GA (alone or in combination) were prepared in PG, diluted in growth medium and tested at the same concentrations as in liposomal formulations. Plain-LP, at the same SPC concentrations as in loaded liposomes, was also evaluated. Distilled water and PG solution (added at the ﬁnal concentration of 0.25% w/v, corresponding to the maximum volume of PG applied to cells) served as control. After incubation, living cells were quantiﬁed using the Cell Counting Kit-8 (CCK-8), following the manufacturing instructions. Brieﬂy, 10 µL of CCK-8 were added to the cells and the absorbance was measured at 450 nm (Spark Multimode Microplate Reader, Tecan® Trading AG, Männedorf, Switzerland) after 4 h at 37 ◦C. Since slight spontaneous absorbance may occur in a culture medium incubated with CCK-8, growth medium without cells was used as a blank. All experiments were performed in triplicate and results were expressed as percentage of living cells with respect to control (untreated cells). 2.14. Statistical Analysis All results were expressed as mean ± standard deviation (SD). For the comparison of two means, Student’s t-test was applied. One-way ANOVA followed by Bonferroni correction was used for multiple comparison. All statistical analyses were performed using GraphPad Prims version 8.0.0 for Windows (GraphPad Software, San Diego, CA, USA, www.graphpad.com) and diﬀerences were considered signiﬁcant for p value < 0.05. 2.13. Anti-Candida Activity Testing The antifungal activity against Candida albicans ATCC 10231 was evaluated by broth microdilution following the method reported by Andersen et al. [26]. C. albicans was grown aerobically in PDBGlu at 37 ◦C. After 24 h, Candida suspension was diluted in PDBGlu to reach a concentration of 4 × 105 cells/mL (determined by counting in a Bürker chamber). The yeast suspension (50 µL) was inoculated 7 of 21 Pharmaceutics 2020, 12, 9 in 96-well plates along with 50 µL of liposomal formulations (LP-Q, LP-GA and LP-Q+GA) or solutions of Q (in DMSO) and GA (in distilled water). Samples were diluted in a two-fold sequence to test concentrations ranging from 250 to 2 µg/mL. Plain-LP, DMSO and distilled water were used as negative, solvent, and growth control, respectively. Blank control, consisting only of growth medium, and sterility controls, containing formulations and sterile medium, were also included. Plates were incubated aerobically without shacking at 37 ◦C for 24 h. Afterwards, the growth inhibition was established through microscope observation (20×; Axiovert 40 Inverted Microscope, Carl Zeiss, Thornwood, NY, USA) and IC50 was deﬁned as the minimal concentration of the sample that inhibits 50% or more of the visible growth, as compared to untreated control. Aliquots of the samples (20 µL) were spotted onto PDAGlu and the minimal lethal dose (MLD) was deﬁned as the concentration at which no growth was observed after 24 h of incubation at 37 ◦C. All experiments were repeated in triplicate. 3.1. Characterization of Liposomes 3.1. Characterization of Liposomes 3. Results and Discussion Vaginal drug administration for localized therapy is an interesting therapeutic choice considering the treatment of sexually transmitted diseases, fungal and bacterial infections, and cancer. The vaginal site is regarded as one of the most highly challenging sites for drug action and diﬀerent approaches have been proposed as superior treatments in topical vaginal therapy. Conventional vaginal pharmaceutical forms are often associated with poor retention, low bioavailability, inability to modulate the release of drug and need for frequent administrations that reduce the patient compliance [6]. Smart nanocarrier-based drug delivery platforms oﬀer possibilities to achieve sustained drug release, eﬃcient cellular targeting, and even intrinsic antimicrobial properties [38,39]. Among nanocarriers, liposomes are particularly suitable for vaginal delivery because they are not interfering with vaginal microbiota and are able to protect active substances against external enzymatic degradation and rapid perturbations that can occur in the vaginal cavity [22]. Moreover, they are biodegradable, biocompatible, weakly immunogenic and non-irritating to vaginal mucosa. Their characteristics, such as composition, size and surface properties, will aﬀect their fate in the vaginal site [23]. This study focused on the feasibility of simple phosphatidilcholine-based liposomes. Indeed, the entrapment of highly lipophilic molecules, such as Q, could be hampered by the presence of additional components in the lipidic bilayer. Considering that antifungal formulations need to act primarily on the vaginal epithelial surface, where Candida infection occurs, conventional liposomes could be a valid choice to obtain not-expensive and simple vesicles able to simultaneously accommodate two diﬀerent active molecules. 3.1.1. Liposomal Size and Surface Properties To optimize the loading of polyphenols, liposomal formulations varying in concentrations of Q and/or GA were prepared and characterized in terms of their size, PDI and zeta potential. Results are summarized in Table 1. Firstly, liposomes containing only Q were investigated. 8 of 21 Pharmaceutics 2020, 12, 9 Table 1. Liposomal characteristics: size, size distribution and zeta potential (mean ± SD, n = 4). Type of Liposomes Vesicle Size(nm) PDI Zeta Potential (mV) plain-LP 166.9 ± 18.0 0.35 ± 0.14 −1.5 ± 0.2 LP-Q (Q 1 mg/mL) 194.4 ± 28.5 0.35 ± 0.05 −4.9 ± 0.7 LP-Q (Q 1.5 mg/mL) 194.6 ± 8.8 0.44 ± 0.01 −5.6 ± 0.2 LP-Q (Q 2 mg/mL) 594.3 ± 72.8 0.90 ± 0.01 −6.4 ± 0.3 LP-GA (GA 1 mg/mL) 180.7 ± 21.1 0.38 ± 0.03 −3.7 ± 0.5 LP-GA (GA 2 mg/mL) 289.4 ± 8.3 0.35 ± 0.08 −5.6 ± 0.2 LP-Q+GA (Q 1 mg/mL; GA 1 mg/mL) 220.4 ± 21.6 0.44 ± 0.04 −7.1 ± 0.4 LP-Q+GA (Q 1 mg/mL; GA 2 mg/mL) 366.5 ± 9.2 0.55 ± 0.02 −7.8 ± 0.7 Table 1. Liposomal characteristics: size, size distribution and zeta potential (mean ± SD, n = 4). Although extrusion is a widely used vesicle size reduction method [24,40], sonication was found more suitable in this work. When forced step-wise through polycarbonate membranes (0.8, 0.4 and 0.2 µm pore size ﬁlters, respectively) LP-Q precipitated, possibly because Q was expelled out of the lipid bilayer during the process, leading to unstable dispersions (mean size: 366.5 ± 9.2; PDI: 0.56 ± 0.01). 0. µ po e s e te s, espect ve y) Q p ec p tated, poss b y because Q was e pe ed out o t e p d bilayer during the process, leading to unstable dispersions (mean size: 366.5 ± 9.2; PDI: 0.56 ± 0.01). The starting concentrations of Q and/or GA are indicated in brackets. Formulations selected for further studies are presented in bold. y g p g p The starting concentrations of Q and/or GA are indicated in brackets. Formulations selected for further studies are presented in bold. Considering the maximum Q load in vesicles, we observed that the size and PDI increased with the increasing starting concentration of Q (Table 1). 3.1.1. Liposomal Size and Surface Properties In particular, when Q was added at the concentration of 2 mg/mL, liposomes were signiﬁcantly bigger and the high PDI (>0.8) indicated the presence of precipitates. This was probably due to Q, being very poorly soluble in water, immediately precipitating when not incorporated in the lipid bilayers [41]. LP-Q prepared with the intermediate concentration (Q 1.5 mg/mL) were also excluded because of the stability issue; although equal in size to liposomes prepared with Q 1 mg/mL, they were not stable after just one week of storage at 4 ◦C (mean size: 504.6 ± 14.1; PDI: 0.564 ± 0.052). On the contrary, liposomes formulated with GA in the aqueous phase at concentration of 2 mg/mL were stable over time, however when co-entrapped with Q 1 mg/mL, the resulting LP-Q+GA formulation precipitated. For these reasons, the liposomal preparations LP-Q (Q 1 mg/mL), LP-GA (GA 1 mg/mL) and LP-Q+GA (Q 1 mg/mL; GA 1 mg/mL) were selected for further studies. They displayed an average diameter of around 200 nm that is estimated to be optimal for delivery to vaginal mucosa [31,42] and for targeting microorganisms able to grow bioﬁlms [43], such as Candida spp. [44]. In particular, mean size of liposomes containing either only Q or GA did not diﬀer signiﬁcantly from plain-LP, while liposomes comprising both polyphenols (LP-Q+GA) were slightly bigger (p = 0.0012) as a consequence of the simultaneous incorporation of both polyphenols. Although sonication led to the formation of dispersions less homogeneous than those obtained by extrusion technique [40], results were reproducible and all PDI values were below 0.7 (0.348–0.438), considered acceptable for liposomal dispersions [45]. As previously observed [28,46], plain-LP exhibited a zeta potential close to neutral (Table 1) because phosphatidylcholine is a zwitter ionic lipid that can acquire slightly negative values in distilled water, as a result of the orientation of lipid headgroups and formation of a hydration layer around vesicles [47]. Liposomes incorporating Q and/or GA were more negative than plain-LP (p < 0.05), coherently with the fact that both Q and GA were negatively charged in water (−18.5 ± 0.6 mV and −14.3 ± 0.7 mV, respectively). 3.1.2. Polyphenols Entrapment Eﬃciency Despite the wide spectrum of beneﬁcial properties exerted by Q [14], the low solubility of this molecule limits its wider therapeutic use. Moreover, chemical stability of Q is highly compromised in the presence of oxygen [48] and metal ions [49], as well as in alkaline conditions [50]. Among polyphenolic acids, GA exhibits several biological relevant activities [19,51]. However, its therapeutic Pharmaceutics 2020, 12, 9 9 of 21 potential is also hampered by low bioavailability [52] and susceptibility to environmental factors [53], such as the tendency to easily oxidize at pH above 7 [54] giving rise to metabolites, namely 4-O-methyl gallic acid and pyrogallol, that possess lower antioxidant activity compared to GA. Due to their structure composed of phospholipid bilayers, liposomes can straightforwardly entrap both hydrophilic and lipophilic substances. In this regard, encapsulation in liposomal vesicles is expected to increase the physicochemical stability and local accumulation of active substances at the site of administration, and therefore the therapeutic eﬃciency of both polyphenols [55]. Co-encapsulation of active molecules in the same nanocarrier is particularly attractive because it allows the simultaneous delivery of molecules to their target, thus simplifying and improving the therapy. To date, Q was successfully co-encapsulated in liposomes with resveratrol [56] and epigallocatechin-3-gallate [57], whereas GA was co-loaded with resveratrol [58]. These formulations were mainly focused on the treatment of skin pathologies related to microbial infections or oxidative stress. To the best of our knowledge, no formulations comprising Q and GA have been proposed for localized vaginal treatment. g In Table 2 are reported the EE % of Q and GA formulated either as single polyphenol or in combination after 120 s of sonication (see Section 2.3). Table 2. Entrapment eﬃciency of quercetin (Q EE %) and gallic acid (GA EE %) in the ﬁnal formulations (liposomes sonicated for 120 s) (mean ± SD, n = 4). Liposomes Q EE % GA EE % LP-Q 85.1 ± 4.6 - LP-GA - 25.4 ± 0.9 LP-Q+GA 86.0 ± 7.0 30.2 ± 1.7 Table 2. Entrapment eﬃciency of quercetin (Q EE %) and gallic acid (GA EE %) in the ﬁnal formulations (liposomes sonicated for 120 s) (mean ± SD, n = 4). Q was eﬃciently entrapped both in LP-Q and LP-Q+GA (EE % >85%), as a consequence of its lipophilic nature that favors its incorporation inside the lipid bilayer. 3.1.2. Polyphenols Entrapment Eﬃciency This is in agreement with the reported high EE % for Q in liposomes, ranging from 67% to 89%, [56,59–61]. Recently, Riva et al. [41] also demonstrated that the employment of phytosomes can increase the solubility of this polyphenol, thus promoting the entrapment of Q into the vesicles. The EE %, as well as size and PDI, of GA inside liposomes before and after diﬀerent sonication times are reported in Table 3. The EE % of GA inside MLV was found to be 50.6 ± 0.6% and 47.5 ± 1.0% for LP-GA and LP-Q+GA, respectively. After sonication, the EE % decreased with increasing sonication time. Two minutes of sonication were required to obtain vesicles with the desired size; indeed, mean diameter of LP-GA was 600.1 ± 19.0 nm after 60 s of sonication and 314.8 ± 4.5 nm after 90 s, respectively, whereas LP-Q+GA had a size of 897.7 ± 5.4 nm after 60 s of sonication and 369.7 ± 4.7 nm after 90 s of sonication. Following 2 min of sonication the EE % was 25% for liposomes containing only GA and, interestingly, was higher for LP-Q+GA (30%, p = 0.0005) (Table 2). Our entrapment was higher than reported by Vitonyte et al. [58] for GA and resveratrol co-loaded liposomes. Considering its hydrophilic nature, GA was probably located inside aqueous core [57]. During ultrasound treatment, MLV undergo breakage and rearrangements that induce instability and formation of smaller vesicles. Our results suggest that the presence of Q in the surrounding lipid bilayer partially prevented leaks of GA from liposomes due to sonication process. 10 of 21 Pharmaceutics 2020, 12, 9 Table 3. Characteristics of liposomes containing GA before (multilamellar vesicles, MLV) and after diﬀerent sonication times (60 s and 90 s): size, polydispersity index (PDI) and GA EE % (mean ± SD, n = 4). Liposomes Vesicle Size (nm) PDI GA EE % LP-GA MLV >1 µm >0.9 50.6 ± 0.6 LP-GA sonicated-60 s 600.1 ± 19.0 nm 0.58 ± 0.09 36.8 ± 0.9 LP-GA sonicated-90 s 314.8 ± 4.5 0.44 ± 0.02 31.0 ± 0.7 LP-Q+GA MLV >1 µm >0.9 47.5 ± 1.0 LP-Q+GA sonicated-60 s 897.7 ± 5.4 0.64 ± 0.02 39.9 ± 1.5 LP-Q+GA sonicated-90 s 369.7 ± 4.7 0.56 ± 0.01 35.4 ± 0.7 3.1.3. Stability of Liposomal Preparations 3.1.3. Stability of Liposomal Preparations Storage stability of liposomal dispersions is an important aspect to predict the quality of formulations since the leakage of active molecules as well as aggregation should be avoided [62]. We evaluated stability by using entrapment eﬃciency (Figure 1a), particle size (Figure 1b) and zeta potential (Figure 1c) as parameters indicating changes from original liposomal characteristics. No relevant changes in respect to size and PDI were detected for all formulations; moreover, polyphenol-containing liposomes appeared more stable than plain-LP (Figure 1b). As reported earlier by other authors [56,59,60], Q was eﬀectively retained inside liposomes over the course of 3 months, with no signiﬁcant loss and no signs of degradation. On the contrary, the EE % of GA formulated alone tended to slowly decrease after more than one month of storage indicating leakage from liposomes. After 90 days the EE % of LP-GA fell to 19.6 ± 0.6%, which corresponds to a decline of 22.8%, similarly to what was described by Manosroi et al. for GA encapsulated inside noisome formulations [55]. Notably, this phenomenon was not observed for liposomes comprising both Q and GA; those liposomes were able to prevent leakage of both polyphenols. Zeta potential of LP-GA and LP-Q+GA became more negative during storage while the surface charge of LP-Q was extremely stable. This evidence could be attributed to the tendency of GA to move and accumulate at the lipid bilayer–water interface. However, signiﬁcant (p < 0.0001) decrease in zeta potential occurred after 30 days for LP-GA and after 60 days for LP-Q+GA. 11 of 21 11 of 21 Pharmaceutics 2020, 12, 9 Pharmaceutics 2019, 11, x F maceutics 2020, 12, 9 11 o Day 0 Day 15 Day 30 Day 60 Day 90 0 100 200 300 400 500 0.0 0.1 0.2 0.3 0.4 0.5 Figure 1. Stability of liposomes: changes in (a) Q and GA EE %; (b) liposomes size (bars, left Y axis) and PDI (symbols, right Y axis) and (c) zeta potential over a storage period of 90 days at 4 °C (mean ± SD, n = 4). The statistical significance was calculated with respect to day 0; * p < 0.05. 4. Release of Polyphenols from Liposomes Figure 1. 3.1.3. Stability of Liposomal Preparations Stability of liposomes: changes in (a) Q and GA EE %; (b) liposomes size (bars, left Y axis) and PDI (symbols, right Y axis) and (c) zeta potential over a storage period of 90 days at 4 ◦C (mean ± SD, n = 4). The statistical signiﬁcance was calculated with respect to day 0; * p < 0.05. 4. Release of Polyphenols from Liposomes The in vitro release studies of polyphenols were performed by using Franz diﬀusion cells a h b ﬀ H 4 5 i i h i i h idi f l i l i Day 0 Day 15 Day 30 Day 60 Day 90 0 100 200 300 400 500 0.0 0.1 0.2 0.3 0.4 0.5 Day 0 Day 15 Day 30 Day 60 Day 90 0 100 200 300 400 500 0.0 0.1 0.2 0.3 0.4 0.5 Day 0 Day 15 Day 30 Day 60 Day 90 0 100 200 300 400 500 0.0 0.1 0.2 0.3 0.4 0.5 Figure 1. Stability of liposomes: changes in (a) Q and GA EE %; (b) liposomes size (bars, left Y axis) and PDI (symbols, right Y axis) and (c) zeta potential over a storage period of 90 days at 4 °C (mean ± Figure 1. Stability of liposomes: changes in (a) Q and GA EE %; (b) liposomes size (bars, left Y axis) and PDI (symbols, right Y axis) and (c) zeta potential over a storage period of 90 days at 4 ◦C (mean ± SD, n = 4). The statistical signiﬁcance was calculated with respect to day 0; * p < 0.05. SD, n = 4). The statistical significance was calcul 3.1.4. Release of Polyphenols from Liposomes SD, n = 4). The statistical significance was calcul 3.1.4. Release of Polyphenols from Liposomes 3.1.4. Release of Polyphenols from Liposomes The in vitro release studies of polyphenols were performed by using Franz diﬀusion cells and phosphate buﬀer at pH 4.5 as receiving phase to mimic the acidity of a normal vaginal environment. 12 of 21 ronment. d GA in Pharmaceutics 2020, 12, 9 phosphate buffer a The release behavi The release behaviors of Q and GA from liposomes are depicted in Figure 2. Free Q and GA in solutions were used as controls for comparison. For all liposomal formulations, the cumulative release rates of polyphenols from liposomes were much slower compared to the respective controls (p < 0.001). p p release rates of polyphenols from liposomes were much slower compared to the respective controls (p < 0.001). The release behaviors of Q and GA from liposomes are depicted in Figure 2. Free Q and GA in solutions were used as controls for comparison. For all liposomal formulations, the cumulative release rates of polyphenols from liposomes were much slower compared to the respective controls (p < 0.001). p p release rates of polyphenols from liposomes were much slower compared to the respective controls (p < 0.001). Q released (%) GA released (%) (a) (b) Figure 2. In vitro polyphenol release expressed as cumulative percentages of (a) Q and (b) GA released over time from liposomes containing only Q (LP-Q), liposomes containing only GA (LP-GA) and liposomes containing both polyphenols (LP-Q+GA) compared to free Q and free GA (mean ± SD, n = 3). Figure 2. In vitro polyphenol release expressed as cumulative percentages of (a) Q and (b) GA released over time from liposomes containing only Q (LP-Q), liposomes containing only GA (LP-GA) and liposomes containing both polyphenols (LP-Q+GA) compared to free Q and free GA (mean ± SD, n = 3). Q released (%) (b) Figure 2. In vitro polyphenol release expressed as cumulative percentages of (a) Q and (b) GA released over time from liposomes containing only Q (LP-Q), liposomes containing only GA (LP-GA) and liposomes containing both polyphenols (LP-Q+GA) compared to free Q and free GA (mean ± SD, n = 3) Figure 2. In vitro polyphenol release expressed as cumulative percentages of (a) Q and (b) GA released over time from liposomes containing only Q (LP-Q), liposomes containing only GA (LP-GA) and liposomes containing both polyphenols (LP-Q+GA) compared to free Q and free GA (mean ± SD, n = 3). SD, n = 4). The statistical significance was calcul 3.1.4. Release of Polyphenols from Liposomes The release proﬁles of GA (Figure 2b) from both LP-GA and LP-Q+GA were biphasic, with an initial burst release in the ﬁrst hour followed by a sustained release. Interestingly, signiﬁcant diﬀerences in GA release rates were detected between LP-GA and LP-Q+GA at the earlier time points up to 4 h (p < 0.002). In particular, after 1 h, ~50% of GA was released from LP-GA and only ~34% from 13 of 21 Pharmaceutics 2020, 12, 9 LP-Q+GA. After 8 h, both formulations released more than 80% of GA (p > 0.05) and release was complete within 24 h. The lipophilic nature of Q allowed a more evident sustained release of this polyphenol from liposomes (Figure 2a) [60]. About 47% of Q was released after 8 h from LP-Q. Notably, the co-presence of GA inside liposomes favored the release of Q (p < 0.03), and reached ~58% from LP-Q+GA after 8 h. Thus, the combination of two polyphenols not only extended the release of GA, guaranteeing a long-lasting activity at the site of administration, but also improved the release of Q required to elicit a stronger biological eﬀect. 3.2. Biological Characterization VVC is associated with increased levels of nitric oxide which leads to inﬂammatory response and consequent vulvar pain that can heavily impair the quality of life of many women worldwide [63]. Fidel et al. [5] also demonstrated that symptomatic, but not asymptomatic, women displayed high levels of polymorphonuclear leukocytes (PMN) responsible for release of pro-inﬂammatory mediators and free radicals in the vaginal lumen. The authors highlighted that recruitment of PMN was not only ineﬀective in protecting from fungus infections, but actually exacerbated VVC symptoms, including itching, burning and redness at the vulva and vaginal mucosa. In this regard, polyphenols are attractive natural substances able to exert antioxidative and anti-inﬂammatory activities [10], minimizing the tissue damages and preventing the onset of chronic infections. 3.2.1. Antioxidant Activity of Liposomal Polyphenols 3.2.1. Antioxidant Activity of Liposomal Polyphenols EC50 was deﬁned as eﬀective concentrations required for the 50% decrease of radicals in ABTS DPPH assays (mean ± SD, n = 3). 3.2.1. Antioxidant Activity of Liposomal Polyphenols Antioxidant eﬀect of polyphenols was evaluated after the complete dissolution of liposomal formulations in ethanol, by two colorimetric methods employed. Figure 3 reports the results at four diﬀerent concentrations of Q and/or GA (1, 2.5, 5 and10 µg/mL) obtained for ABTS (Figure 3a) and DPPH assays (Figure 3b). With respect to well-known antioxidants (vitamin E and C), the two polyphenols exhibited signiﬁcantly higher concentration-dependent scavenging activity against both free radicals (p < 0.0001), while plain-LP were ineﬀective, as expected. It is evident that Q and GA possessed strong antioxidant potentials; both were active also at very low concentrations (~11–31% at 1 µg/mL) and both ABTS·+ and DPPH radicals were almost completely inhibited at 10 µg/mL (~90–99%). This is in agreement with evidence reported in the literature for free Q [56] and free GA [64]. Considering that antioxidant activity plays a crucial role in the pharmacological eﬀects of Q and GA, it is important to assure that their incorporation inside lipid nanocarriers and the entire preparative method did not compromise the capacity of the molecules to quickly and eﬃciently remove free radicals. It seems that the therapeutic activity of many polyphenols could be potentiated when co-administered with other polyphenols or antibiotics [65]. Recently, Q was reported to act in synergism with epigallocatechin-3-gallate [57] and with curcumin [65], enhancing the antioxidant and anti-inﬂammatory activities, respectively. To investigate the eﬀect of the combination of Q and GA, EC50 values were obtained from linear regression analysis (Table 4). The γ values, calculated according to the Equation (2) for ABTS and DPPH assays, were below 1, indicating that co-delivered polyphenols exerted an antioxidant activity in a synergistic way. EC50 was deﬁned as eﬀective concentrations required for the 50% decrease of radicals in ABTS and DPPH assays (mean ± SD, n = 3). 14 of 21 14 of 21 Pharmaceutics 2020, 12, 9 Pharmaceutics 2019, 11, x F Figure 3. Antioxidant activities of LP-Q, LP-GA and LP-Q+GA, vitamin E and vitamin C expressed as (a) ABTS and (b) DPPH free radicals scavenging activity (mean ± SD, n = 3). The statistical significance with respect to vitamin E and vitamin C (used as comparison) was reported; * p < 0.0001. Statistical differences between liposomes holding both polyphenols (LP-Q+GA) and liposomes holding only one polyphenol (LP-Q/LP-GA) were also calculated; a: p < 0.001. (a) (b) Figure 3. 3.2.1. Antioxidant Activity of Liposomal Polyphenols Antioxidant eﬀect of polyphenol-containing liposomes. Table 4. Antioxidant effect of polyphenol-containing liposomes. Liposomes ABTS Assay DPPH Assay EC50 (μg/mL) γ EC50 (μg/mL) γ LP-Q 1.61 ± 0.02 - 2.92 ± 0.17 - LP-GA 2.49 ± 0.17 - 3.48 ± 0.03 - LP-Q+GA 1.11 ± 0.06 0.79 1.54 ± 0.02 0.88 EC50 was defined as effective concentrations required for the 50% decrease of radicals in ABT and DPPH assays (mean ± SD, n = 3). Table 4. Antioxidant eﬀect of polyphenol-containing liposomes. Liposomes ABTS Assay DPPH Assay EC50 (µg/mL) γ EC50 (µg/mL) γ LP-Q 1.61 ± 0.02 - 2.92 ± 0.17 - LP-GA 2.49 ± 0.17 - 3.48 ± 0.03 - LP-Q+GA 1.11 ± 0.06 0.79 1.54 ± 0.02 0.88 3.2.2. Anti-Inﬂammatory Activity of Free and Liposomal Polyphenols EC50 was defined as effective concentrations required for the 50% and DPPH assays (mean ± SD, n = 3). 3.2.2. Anti-Inﬂammatory Activity of Free and Liposomal Polyphenols EC50 was defined as effective concentrations required for the 50% and DPPH assays (mean ± SD, n = 3). 3.2.2. Anti-Inﬂammatory Activity of Free and Liposomal Polyphenols 3.2.2. Anti-Inflammatory Activity of Free and Liposomal Polyphenols It is well documented that both Q and GA are able to reduce inﬂammatory response in LPS-induced macrophages by acting at diﬀerent levels of the inﬂammation cascade [19,66]. 3.2.2. Anti-Inflammatory Activity of Free and Liposomal Polyphenols It is well documented that both Q and GA are able to reduce inﬂammatory response in LPS-induced macrophages by acting at diﬀerent levels of the inﬂammation cascade [19,66]. It is well documented that both Q and GA are able to reduce inflammatory response in LPS- induced macrophages by acting at different levels of the inflammation cascade [19,66]. In the current work, we evaluated the inhibitory effects of liposomal polyphenols on the production of NO, a signaling molecule that plays a crucial role in the pathogenesis of inflammation d i f ti LPS i d d h t t d f 24 h ith li l f l h l In the current work, we evaluated the inhibitory eﬀects of liposomal polyphenols on the production of NO, a signaling molecule that plays a crucial role in the pathogenesis of inﬂammation and infections. LPS-induced macrophages were treated for 24 h with liposomal or free polyphenols at three diﬀerent Q/GA concentrations (0.05, 0.5 and 2.5 µg/mL) (Figure 4). 3.2.1. Antioxidant Activity of Liposomal Polyphenols Antioxidant activities of LP-Q, LP-GA and LP-Q+GA, vitamin E and vitamin C expressed as (a) ABTS and (b) DPPH free radicals scavenging activity (mean ± SD, n = 3). The statistical signiﬁcance with respect to vitamin E and vitamin C (used as comparison) was reported; * p < 0.0001. Statistical diﬀerences between liposomes holding both polyphenols (LP-Q+GA) and liposomes holding only one polyphenol (LP-Q/LP-GA) were also calculated; a: p < 0.001. ( ) (a) (a) (b) (b) Figure 3 Antioxidant activities of LP-Q LP-GA and LP-Q+GA vitamin E and vitamin C expressed Figure 3. Antioxidant activities of LP-Q, LP-GA and LP-Q+GA, vitamin E and vitamin C expressed as Figure 3. Antioxidant activities of LP-Q, LP-GA and LP-Q+GA, vitamin E and vitamin C expressed as (a) ABTS and (b) DPPH free radicals scavenging activity (mean ± SD, n = 3). The statistical significance with respect to vitamin E and vitamin C (used as comparison) was reported; * p < 0.0001. Statistical differences between liposomes holding both polyphenols (LP-Q+GA) and liposomes holding only one polyphenol (LP-Q/LP-GA) were also calculated; a: p < 0.001. Figure 3. Antioxidant activities of LP-Q, LP-GA and LP-Q+GA, vitamin E and vitamin C expressed as (a) ABTS and (b) DPPH free radicals scavenging activity (mean ± SD, n = 3). The statistical signiﬁcance with respect to vitamin E and vitamin C (used as comparison) was reported; * p < 0.0001. Statistical diﬀerences between liposomes holding both polyphenols (LP-Q+GA) and liposomes holding only one polyphenol (LP-Q/LP-GA) were also calculated; a: p < 0.001. Figure 3. Antioxidant activities of LP-Q, LP-GA and LP-Q+GA, vitamin E and vitamin C expressed as (a) ABTS and (b) DPPH free radicals scavenging activity (mean ± SD, n = 3). The statistical significance with respect to vitamin E and vitamin C (used as comparison) was reported; * p < 0.0001. Statistical differences between liposomes holding both polyphenols (LP-Q+GA) and liposomes holding only one polyphenol (LP-Q/LP-GA) were also calculated; a: p < 0.001. g Q Q p (a) ABTS and (b) DPPH free radicals scavenging activity (mean ± SD, n = 3). The statistical signiﬁcance with respect to vitamin E and vitamin C (used as comparison) was reported; * p < 0.0001. Statistical diﬀerences between liposomes holding both polyphenols (LP-Q+GA) and liposomes holding only one polyphenol (LP-Q/LP-GA) were also calculated; a: p < 0.001. Table 4. Antioxidant effect of polyphenol-containing liposomes. Table 4. 3.2.1. Antioxidant Activity of Liposomal Polyphenols LP-Q seemed to be more potent with respect to LP-GA (inhibition of ~67% and ~43%, respectively, at 2.5 μg/mL) and the anti- inflammatory activity significantly increased (p < 0.0005) when the two molecules were delivered together (LP-Q+GA inhibited ~79% of NO production at 2.5 μg/mL), consistently with what observed for antioxidant activity. Inhibition of NOproduction (%) Figure 4. Inhibitory effect of liposomal and free Q and/or GA on nitric oxide (NO) production in lipopolysaccharide (LPS)-induced macrophages (mean ± SD, n = 3). The statistical significance was calculated with respect to control (untreated cells); * p < 0.0001. Statistical differences between coupled polyphenols (LP-Q+GA or Q+GA) and single polyphenol (LP-Q/LP-GA or Q/GA) were also investigated; a: p < 0.001. Statistical differences between liposomes and corresponding free polyphenols were reported as follows: b’: p < 0.01; b’’: p < 0.001; b’’’: p < 0.0001. Inhibition of NOproduction (%) Figure 4. Inhibitory eﬀect of liposomal and free Q and/or GA on nitric oxide (NO) production in lipopolysaccharide (LPS)-induced macrophages (mean ± SD, n = 3). The statistical signiﬁcance was calculated with respect to control (untreated cells); * p < 0.0001. Statistical diﬀerences between coupled polyphenols (LP-Q+GA or Q+GA) and single polyphenol (LP-Q/LP-GA or Q/GA) were also investigated; a: p < 0.001. Statistical diﬀerences between liposomes and corresponding free polyphenols were reported as follows: b′: p < 0.01; b”: p < 0.001; b′′′: p < 0.0001. Figure 4. Inhibitory effect of liposomal and free Q and/or GA on nitric oxide (NO) production in lipopolysaccharide (LPS)-induced macrophages (mean ± SD, n = 3). The statistical significance was calculated with respect to control (untreated cells); * p < 0.0001. Statistical differences between coupled polyphenols (LP-Q+GA or Q+GA) and single polyphenol (LP-Q/LP-GA or Q/GA) were also investigated; a: p < 0.001. Statistical differences between liposomes and corresponding free polyphenols were reported as follows: b’: p < 0.01; b’’: p < 0.001; b’’’: p < 0.0001. Figure 4. Inhibitory eﬀect of liposomal and free Q and/or GA on nitric oxide (NO) production in lipopolysaccharide (LPS)-induced macrophages (mean ± SD, n = 3). The statistical signiﬁcance was calculated with respect to control (untreated cells); * p < 0.0001. Statistical diﬀerences between coupled polyphenols (LP-Q+GA or Q+GA) and single polyphenol (LP-Q/LP-GA or Q/GA) were also investigated; a: p < 0.001. 3.2.1. Antioxidant Activity of Liposomal Polyphenols Plain-LP were also tested at the same SPC It is well documented that both Q and GA are able to reduce inflammatory response in LPS- induced macrophages by acting at different levels of the inflammation cascade [19,66]. In the current work, we evaluated the inhibitory effects of liposomal polyphenols on the production of NO, a signaling molecule that plays a crucial role in the pathogenesis of inflammation and infections LPS induced macrophages were treated for 24 h with liposomal or free polyphenols In the current work, we evaluated the inhibitory eﬀects of liposomal polyphenols on the production of NO, a signaling molecule that plays a crucial role in the pathogenesis of inﬂammation and infections. LPS-induced macrophages were treated for 24 h with liposomal or free polyphenols at three diﬀerent Q/GA concentrations (0.05, 0.5 and 2.5 µg/mL) (Figure 4). Plain-LP were also tested at the same SPC 15 of 21 o tested Pharmaceutics 2020, 12, 9 at three different Q/G concentration as in liposomes containing polyphenols. No eﬀect on NO production was observed at the lower concentration, but at 0.5 and 2.5 µg/mL polyphenol-containing liposomes displayed a strong concentration-dependent inhibition of NO production, that was signiﬁcantly higher compared to plain-LP (p < 0.0001). A concentration-dependent behavior was also conﬁrmed for free polyphenols. Notably, in all cases, the inhibitory eﬀects on NO production were higher when Q and/or GA were incorporated inside liposomes (p < 0.009), in agreement with previous works on liposomal curcumin [37], resveratrol [46] and epicatechin [24] formulations. LP-Q seemed to be more potent with respect to LP-GA (inhibition of ~67% and ~43%, respectively, at 2.5 µg/mL) and the anti-inﬂammatory activity signiﬁcantly increased (p < 0.0005) when the two molecules were delivered together (LP-Q+GA inhibited ~79% of NO production at 2.5 µg/mL), consistently with what observed for antioxidant activity. was observed at the lower concentration, but at 0.5 and 2.5 μg/mL polyphenol-containing liposomes displayed a strong concentration-dependent inhibition of NO production, that was significantly higher compared to plain-LP (p < 0.0001). A concentration-dependent behavior was also confirmed for free polyphenols. Notably, in all cases, the inhibitory effects on NO production were higher when Q and/or GA were incorporated inside liposomes (p < 0.009), in agreement with previous works on liposomal curcumin [37], resveratrol [46] and epicatechin [24] formulations. 3.2.1. Antioxidant Activity of Liposomal Polyphenols Statistical diﬀerences between liposomes and corresponding free polyphenols were reported as follows: b′: p < 0.01; b”: p < 0.001; b′′′: p < 0.0001. Our results suggest that the choice of an appropriate delivery system improves the beneficial activity of polyphenols. Furthermore, combining Q and GA can be a promising strategy to take advantage of different mechanisms of action, thus heightening the overall anti-inflammatory effect and improving relief of pain associated with inflammation. Our results suggest that the choice of an appropriate delivery system improves the beneﬁcial activity of polyphenols. Furthermore, combining Q and GA can be a promising strategy to take advantage of diﬀerent mechanisms of action, thus heightening the overall anti-inﬂammatory eﬀect and improving relief of pain associated with inﬂammation. 3.2.3. Effect of Free and Liposomal Polyphenols on Cell Viability 3.2.3. Eﬀect of Free and Liposomal Polyphenols on Cell Viability p yp y The effect of liposomal formulations and free Q and/or GA on macrophage viability was also investigated and results are depicted in Figure 5. Cells were exposed to free and liposomal polyphenols for 24 h at the same concentrations tested for the anti-inflammatory activity. No ytoto i effe t e e fou d The eﬀect of liposomal formulations and free Q and/or GA on macrophage viability was also investigated and results are depicted in Figure 5. Cells were exposed to free and liposomal polyphenols for 24 h at the same concentrations tested for the anti-inﬂammatory activity. No cytotoxic eﬀects were found. cytotoxic effects were found. LP-Q and LP-GA expressed weak mitogenic effects at the highest polyphenol concentration tested (2.5 μg/mL), that was also observed for LP-Q+GA at all concentrations. LP-Q and LP-GA expressed weak mitogenic eﬀects at the highest polyphenol concentration tested (2.5 µg/mL), that was also observed for LP-Q+GA at all concentrations. 16 of 21 duction 16 of 21 duction Pharmaceutics 2020, 12, 9 of cytotoxic action o in LPS induced mac Figure 5. Effect of liposomal and free polyphenols on RAW 264.7 cell viability compared to viability of untreated cells (100%) (mean ± SD, n = 3). The statistical significance was calculated with respect to control; * p < 0.01. Figure 5. Eﬀect of liposomal and free polyphenols on RAW 264.7 cell viability compared to viability of untreated cells (100%) (mean ± SD, n = 3). The statistical signiﬁcance was calculated with respect to control; * p < 0.01. Figure 5. Effect of liposomal and free polyphenols on RAW 264.7 cell viability compared to viability of untreated cells (100%) (mean ± SD, n = 3). The statistical significance was calculated with respect to control; * p < 0.01. Figure 5. Eﬀect of liposomal and free polyphenols on RAW 264.7 cell viability compared to viability of untreated cells (100%) (mean ± SD, n = 3). The statistical signiﬁcance was calculated with respect to control; * p < 0.01. 3.3. Antifungal Potential Polyphenols not only possess useful biological properties, but also show few or no side effects and toxicity and thus may be promising natural molecules to challenge the problem of antifungal drug resistance. Candida albicans is the most frequent species isolated from women suffering from candidiasis and accounts for 80%–90% of total vaginal fungal infections [1]. 3.2.3. Effect of Free and Liposomal Polyphenols on Cell Viability 3.2.3. Eﬀect of Free and Liposomal Polyphenols on Cell Viability For this reason, the antifungal activity of polyphenol containing liposomes and free polyphenols was evaluated against These preliminary results indicated that both liposomal and free polyphenols were not toxic to living cells at the polyphenol concentration up to 2.5 µg/mL, corresponding to lipid concentration of 50 µg/mL. Moreover, the results suggest that the decreased NO production (Section 3.2.2) was actually due to the inhibition of inﬂammation and not to the inhibition of cell proliferation. The lack of cytotoxic action of Q on RAW 264.7 cell line and its ability to signiﬁcantly inhibit NO production in LPS-induced macrophages was previously observed also by Mu et al. [67]. C. albicans ATCC 10231 d l i LP d 3.3. Antifungal Potential expected, plain-LP made only from SPC had no effect [26]. Although Q is known to possess antibacterial activity [14], neither free Q nor LP-Q were active in inhibiting Candida growth. On the contrary, the two liposomal formulations containing GA considerably hampered Candida growth in the concentration range between 31 and 63 μg/mL after 24 h of treatment (Figure S1 Supplementary Materials). Free GA displayed a similar behavior, indicating that liposomes effectively retained the antifungal activity of this phenolic acid. To better clarify the antifungal effect of liposomes, aliquots of samples exhibiting less than 50% growth were spotted onto agar plates and MLD determined after 24 h of incubation. Free GA, as well as liposomes comprising GA, showed fungicidal effects at concentrations higher than 125 μg/mL (Table 5) demonstrating that LP-GA and Polyphenols not only possess useful biological properties, but also show few or no side eﬀects and toxicity and thus may be promising natural molecules to challenge the problem of antifungal drug resistance. Candida albicans is the most frequent species isolated from women suﬀering from candidiasis and accounts for 80–90% of total vaginal fungal infections [1]. For this reason, the antifungal activity of polyphenol-containing liposomes and free polyphenols was evaluated against C. albicans ATCC 10231. Firstly, IC50 values were microscopically determined (Table 5) and, as expected, plain-LP made only from SPC had no eﬀect [26]. Although Q is known to possess antibacterial activity [14], neither free Q nor LP-Q were active in inhibiting Candida growth. -Q+GA were efficient in counteracting Candida proliferation. Our findings were consistent w evious results obtained by Gehrke et al. [12] and Ozcelik et al. [13] on the GA activity against icans ATCC 10231. The authors utilized either pure GA isolated from Schinus lentiscifolius extr standard GA purchased from Sigma-Aldrich, respectively. Table 5. Anti-Candida activity of polyphenols-containing liposomes reported as IC50 and minimal lethal dose (MLD) (mean ± SD, n = 3). Liposomes IC50 (µg/mL) MLD (µg/mL) LP-Q no inhibition LP-GA 31–63 125 LP-Q+GA 31–63 125 plain-LP no inhibition free Q no inhibition free GA 31–63 125 Q+GA were efficient in counteracting Candida proliferation. Our findings were consistent w vious results obtained by Gehrke et al. [12] and Ozcelik et al. [13] on the GA activity against ATCC 10231 Th th tili d ith GA i l t d f S hi l ti if li t Table 5. C. albicans ATCC 10231 d l i LP d 3.3. Antifungal Potential Anti-Candida activity of polyphenols-containing liposomes reported as IC50 and minimal lethal dose (MLD) (mean ± SD, n = 3). The authors utilized either pure GA isolated from Schinu ased from Sigma-Aldrich, respectively. Liposomes IC50 (µg/mL) MLD (µg/mL) LP-Q no inhibition LP-GA 31–63 125 LP-Q+GA 31–63 125 plain-LP no inhibition free Q no inhibition free GA 31–63 125 17 of 21 Pharmaceutics 2020, 12, 9 On the contrary, the two liposomal formulations containing GA considerably hampered Candida growth in the concentration range between 31 and 63 µg/mL after 24 h of treatment (Figure S1 Supplementary Materials). Free GA displayed a similar behavior, indicating that liposomes eﬀectively retained the antifungal activity of this phenolic acid. To better clarify the antifungal eﬀect of liposomes, aliquots of samples exhibiting less than 50% growth were spotted onto agar plates and MLD determined after 24 h of incubation. Free GA, as well as liposomes comprising GA, showed fungicidal eﬀects at concentrations higher than 125 µg/mL (Table 5), demonstrating that LP-GA and LP-Q+GA were eﬃcient in counteracting Candida proliferation. Our ﬁndings were consistent with previous results obtained by Gehrke et al. [12] and Ozcelik et al. [13] on the GA activity against C. albicans ATCC 10231. The authors utilized either pure GA isolated from Schinus lentiscifolius extract or standard GA purchased from Sigma-Aldrich, respectively. These promising ﬁndings could be further explored for localized treatment of infections and/or inﬂammation. By combining two or more polyphenols within a single nanocarrier, it would be possible to achieve eﬃcient multitargeted treatment. 4. Conclusions We have successfully developed innovative multi active-substance loaded liposomal formulations containing Q and GA, either as a single substance or in combination, for the treatment of vaginal diseases, in particular VVC. Liposomes containing both Q and GA had a desired size of about 200 nm and were able to encapsulate a high level of Q as well as favor the entrapment of GA. Moreover, they remained stable over time and provided a controlled release of both Q and GA. LP-Q+GA displayed a strong antioxidant activity due to the synergism between the two entrapped polyphenols. Q and GA delivered together also exhibited the potent anti-inﬂammatory proﬁle while remaining nontoxic. Finally, LP-Q+GA eﬃciently inhibited the growth of C. albicans, the most frequent etiological agent of VVC. The eﬀect could be attributed to the presence of GA and its fungicidal activity. Supplementary Materials: The following are available online at http://www.mdpi.com/1999-4923/12/1/9/s1, Figure S1: Microscopically determination of anti-Candida activity. Supplementary Materials: The following are available online at http://www.mdpi.com/1999-4923/12/1/ Figure S1: Microscopically determination of anti-Candida activity. Author Contributions: N.Š.-B. and B.L. designed the study and supervised the research fellow. B.G., P.B. and E.M. performed the experiments and statistical analysis. All authors contributed in data interpretation and discussions. B.G., N.Š.-B. and B.L. contributed to manuscript generation. All authors have read and agreed to the published version of the manuscript. Funding: B. Giordani was supported by the Marco Polo mobility fellowship of University of Bologna. Acknowledgments: The authors are grateful to Lipoid GmbH, Germany, for providing the lipids. Conﬂicts of Interest: The authors declare no conﬂict of interest. Abbreviations ABTS 2,2-azino bis(3-ethylbenzothiazoline)-6-sulfonic acid diammonium salt; DPPH 1,1-diphenyl-2-picrylhydrazyl; GA gallic acid; LP liposome; LPS lipopolysaccharide; NO nitric oxide; PDI polydispersity index; Q quercetin; SPC phosphatidylcholine; VVC vulvovaginal candidiasis 18 of 21 18 of 21 Pharmaceutics 2020, 12, 9 References 1. Sobel, J.D. Vulvovaginal candidosis. Lancet 2007, 369, 1961–1971. [CrossRef] 2. Soong, D.; Einarson, A. Vaginal yeast infections during pregnancy. Can. Fam. Physician 2009, 55, 255–256. [PubMed] 3. Ahangari, F.; Farshbaf-khalili, A.; Javadzadeh, Y.; Adibpour, M.; Oskouei, B.S. Comparing the eﬀectiveness of Salvia oﬃcinalis, clotrimazole and their combination on vulvovaginal candidiasis: A randomized, controlled clinical trial. J. Obstet. Gynaecol. Res. 2019, 45, 897–907. [CrossRef] [PubMed] 4. Richardson, J.P.; Willems, H.M.E.; Moyes, D.L.; Shoaie, S.; Barker, K.S.; Tan, S.L.; Palmer, G.E.; Hube, B.; Naglik, J.R.; Peters, M. Candidalysin drives epithelial signaling, neutrophil recruitment, and immunopathology at the vaginal mucosa. Infect. Immun. 2018, 86, e00645-17. [CrossRef] 5. Fidel, P.L.J.; Barousse, M.; Espinosa, T.; Ficarra, M.; Sturtevant, J.; Martin, D.H.; Quayle, A.J.; Dunlap, K. An intravaginal live Candida challenge in humans leads to new hypotheses for the immunopathogenesis of vulvovaginal candidiasis. Infect. Immun. 2004, 72, 2939–2946. [CrossRef] [PubMed] 6. Johal, H.S.; Garg, T.; Rath, G.; Goyal, A.K. Advanced topical drug delivery system for the management of vaginal candidiasis. Drug Deliv. 2016, 23, 550–563. [CrossRef] [PubMed] 7. Ksiezopolska, E.; Gabald, T. Evolutionary emergence of drug resistance in Candida opportunistic pathogens. Genes 2018, 9, 461. [CrossRef] 8. Martins, N.; Barros, L.; Henriques, M.; Silva, S.; Ferreira, I.C.F.R. In Vivo anti-candida activity of phenolic extracts and compounds: Future perspectives focusing on eﬀective clinical interventions. BioMed Res. Int. 2015, 2015, 1–14. . Seleem, D.; Pardi, V.; Murata, M.R. Review of ﬂavonoids: A diverse group of natural compounds w anti-Candida albicans activity in vitro. Arch. Oral Biol. 2017, 76, 76–83. [CrossRef] 0. Ganesan, K.; Xu, B. A critical review on polyphenols and health beneﬁts of black soybeans. Nutrients 201 455. [CrossRef] 11. Rasines-Perea, Z.; Teissedre, P. Grape polyphenols’ eﬀects in human cardiovascular diseases and diabetes. Molecules 2017, 22, 68. [CrossRef] [PubMed] 12. Gehrke, I.T.S.; Neto, A.T.; Pedroso, M.; Mostardeiro, C.P.; Da Cruz, I.B.M.; Silva, U.F.; Ilha, V.; Dalcol, I.I.; Morel, A.F. Antimicrobial activity of Schinus lentiscifolius (Anacardiaceae). J. Ethnopharmacol. 2013, 148, 486–491. [CrossRef] [PubMed] 13. Özçelik, B.; Kartal, M.; Orhan, I. Cytotoxicity, antiviral and antimicrobial activities of alkaloids, ﬂavonoids, and phenolic acids C. Pharm. Biol. 2011, 49, 396–402. [CrossRef] [PubMed] 14. D’Andrea, G. Quercetin: A ﬂavonol with multifaceted therapeutic applications? Fitoterapia 2015, 106, 256–271. [CrossRef] 14. D Andrea, G. Quercetin: A ﬂavonol with multifaceted therapeutic applications? Fitoterapia 2015, 106, 256 271. References [CrossRef] 15 Mlcek J ; Jurikova T ; Skrovankova S ; Sochor J Quercetin and its anti-allergic immune response Molecules [CrossRef] 15. Mlcek, J.; Jurikova, T.; Skrovankova, S.; Sochor, J. Quercetin and its anti-allergic immune response. Molecules Mlcek, J.; Jurikova, T.; Skrovankova, S.; Sochor, J. Quercetin and its anti-allergic immune response. Molecules 2016, 21, 623. [CrossRef] 16. Britti, D.; Crupi, R.; Impellizzeri, D.; Gugliandolo, E.; Fusco, R.; Schievano, C.; Morittu, V.M.; Evangelista, M.; Di Paola, R.; Cuzzocrea, S. A novel composite formulation of palmitoylethanolamide and quercetin decreases inﬂammation and relieves pain in inﬂammatory and osteoarthritic pain models. BMC Vet. Res. 2017, 13, 299. [CrossRef] 17. Valério, D.A.; Georgetti, S.R.; Magro, D.A.; Casagrande, R.; Cunha, T.M.; Vicentini, F.T.M.C.; Vieira, S.M.; Fonseca, M.J.V.; Ferreira, S.H.; Cunha, F.Q.; et al. Quercetin reduces inﬂammatory pain: Inhibition of oxidative stress and cytokine production. J. Nat. Prod. 2009, 72, 1975–1979. [CrossRef] 18. Maramaldi, G.; Togni, S.; Pagin, I.; Giacomelli, L.; Cattaneo, R.; Eggenhöﬀner, R.; Burastero, S.E. Soothing and anti-itch eﬀect of quercetin phytosome in human subjects: A single-blind study. Clin. Cosmet. Investig. Dermatol. 2016, 9, 55–62. [CrossRef] 19. Choubey, S.; Goyal, S.; Varughese, L.R.; Kumar, V.; Sharma, A.K.; Beniwal, V. Probing gallic acid for its broad spectrum applications. Mini Rev. Med. Chem. 2018, 8, 1283–1293. [CrossRef] 20. Teodoro, G.R.; Ellepola, K.; Seneviratne, C.J.; Koga-Ito, C.Y. Potential use of phenolic acids as anti-candida Agents: A review. Front. Microbiol. 2015, 6, 1420. [CrossRef] 21. Li, Z.; Liu, M.; Dawuti, G.; Dou, Q.; Ma, Y.; Liu, H.; Aibai, S. Antifungal activity of gallic acid in vitro and in vivo. Phyther. Res. 2017, 31, 1039–1045. [CrossRef] [PubMed] 19 of 21 19 of 21 Pharmaceutics 2020, 12, 9 22. Brako, F.; Mahalingam, S.; Rami-abraham, B.; Craig, D.Q.M.; Edirisinghe, M. Application of nanotechnology for the development of microbicides. Nanotechnology 2017, 28, 052001. [CrossRef] [PubMed] 23. Vani´c, Ž.; Škalko-Basnet, N. Nanopharmaceuticals for improved topical vaginal therapy: Can they deliver? Eur. J. Pharm. Sci. 2013, 50, 29–41. [CrossRef] [PubMed] 24. Jøraholmen, M.W.; Basnet, P.; Tostrup, M.J.; Moueﬀaq, S.; Škalko-Basnet, N. Localized therapy of vaginal infections and inﬂammation: Liposomes-in-hydrogel delivery system for polyphenols. Pharmaceutics 2019, 11, 53. [CrossRef] [PubMed] 25. Voltan, A.R.; Quindós, G.; Alarcón, K.P.M.; Fusco-almeida, A.M.; Mendes-Giannini, M.J.S.; Chorilli, M. Fungal diseases: Could nanostructured drug delivery systems be a novel paradigm for therapy? Int. J. Nanomed. 2016, 11, 3715–3730. [CrossRef] [PubMed] 26. References Andersen, T.; Mishchenko, E.; Flaten, G.E.; Sollid, J.U.E.; Mattsson, S.; Tho, I.; Škalko-Basnet, N. Chitosan-based nanomedicine to ﬁght genital candida infections: Chitosomes. Mar. Drugs 2017, 15, 64. [CrossRef] 27. Jøraholmen, M.W.; Vani´c, Ž.; Tho, I.; Škalko-Basnet, N. Chitosan-coated liposomes for topical vaginal therapy: Assuring localized drug eﬀect. Int. J. Pharm. 2014, 472, 94–101. [CrossRef] 28. Ternullo, S.; Basnet, P.; Holsæter, A.M.; Flaten, G.E.; de Weerd, L.; Škalko-Basnet, N. Deformable liposomes for skin therapy with human epidermal growth factor: The eﬀect of liposomal surface charge. Eur. J. Pharm. Sci. 2018, 125, 163–171. [CrossRef] 29. Jøraholmen, M.W.; Basnet, P.; Acharya, G.; Škalko-Basnet, N. PEGylated liposomes for topical vaginal therapy improve delivery of interferon alpha. Eur. J. Pharm. Biopharm. 2017, 113, 132–139. [CrossRef] 30. Ravel, J.; Gajer, P.; Abdo, Z.; Schneider, G.M.; Koenig, S.S.K.; Mcculle, S.L.; Karlebachd, S.; Gorlee, R.; Russellf, J.; Tacketf, C.O.; et al. Vaginal microbiome of reproductive-age women. Proc. Natl. Acad. Sci. USA 2011, 108, 4680–4687. [CrossRef] 31. Abruzzo, A.; Giordani, B.; Parolin, C.; Vitali, B.; Protti, M.; Mercolini, L.; Cappelletti, M.; Fedi, S.; Bigucci, F.; Cerchiara, T.; et al. Novel mixed vesicles containing lactobacilli biosurfactant for vaginal delivery of an anti-Candida agent. Eur. J. Pharm. Sci. 2018, 112, 95–101. [CrossRef] [PubMed] 32. Giordani, B.; Abruzzo, A.; Musazzi, U.M.; Cilurzo, F.; Nicoletta, F.P.; Dalena, F.; Parolin, C.; Vitali, B.; Cerchiara, T.; Luppi, B.; et al. Freeze-dried matrices based on polyanion polymers for chlorhexidine local release in the buccal and vaginal cavities. J. Pharm. Sci. 2019, 108, 2447–2457. [CrossRef] [PubMed] 33. Park, S.N.; Lee, M.H.; Kim, S.J.; Yu, E.R. Preparation of quercetin and rutin-loaded ceramide liposomes and drug-releasing eﬀect in liposome-in-hydrogel complex system. Biochem. Biophys. Res. Commun. 2013, 435, 361–366. [CrossRef] [PubMed] 34. Bose, S.; Michniak-Kohn, B. Preparation and characterization of lipid based nanosystems for topical delivery of quercetin. Eur. J. Pharm. Sci. 2013, 48, 442–452. [CrossRef] 35. Liang, T.; Guan, R.; Wang, Z.; Shen, H.; Xia, Q.; Liu, M. Comparison of anticancer activity and antioxidant activity between cyanidin-3-O-glucoside liposomes and cyanidin-3-O-glucoside in Caco-2 cells in vitro. RSC Adv. 2017, 7, 37359–37368. [CrossRef] 36. Lee, J.J.; Kong, M.; Ayers, G.D.; Lotan, R. Interaction index and diﬀerent methods for determining drug interaction in combination therapy. J. Biopharm. Stat. 2007, 17, 461–480. [CrossRef] 37. Basnet, P.; Hussain, H.; Tho, I.; Škalko-Basnet, N. Liposomal delivery system enhances anti-inﬂammatory properties of curcumin. J. Pharm. Sci. 2012, 101, 598–609. [CrossRef] 38. References Ensign, L.M.; Cone, R.; Hanes, J. Nanoparticle-based drug delivery to the vagina: A review. J. Control. Release 2014, 190, 500–514. [CrossRef] 39. Leyva-Gómez, G.; Piñón-Segundo, E.; Mendoza-Muñoz, N.; Zambrano-Zaragoza, M.L.; Mendoza-Elvira, S.; Quintanar-Guerrero, D. Approaches in polymeric nanoparticles for vaginal drug delivery: A Review of the state of the art. Int. J. Mol. Sci. 2018, 19, 1549. [CrossRef] 40. Ong, S.G.M.; Chitneni, M.; Lee, K.S.; Ming, L.C.; Yuen, K.H. Evaluation of extrusion technique for nanosizing liposomes. Pharmaceutics 2016, 8, 36. [CrossRef] 41. Riva, A.; Ronchi, M.; Petrangolini, G.; Bosisio, S.; Allegrini, P. Improved oral absorption of quercetin from quercetin phytosome, a new delivery system based on food grade lecithin. Eur. J. Drug Metab. Pharmacokinet. 2019, 44, 169–177. [CrossRef] [PubMed] 42. Das Neves, J.; Amiji, M.; Sarmento, B. Mucoadhesive nanosystems for vaginal microbicide development: Friend or foe? Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2011, 3, 389–399. [CrossRef] [PubMed] 20 of 21 20 of 21 Pharmaceutics 2020, 12, 9 43. Rukavina, Z.; Vani´c, Ž. Current trends in development of liposomes for targeting bacterial bioﬁlms. Pharmaceutics 2016, 8, 18. [CrossRef] [PubMed] 44. Cavalheiro, M.; Teixeira, M.C. Candida bioﬁlms: Threats, challenges, and promising strategies. Front. Med. 2018, 5, 28. [CrossRef] 45. Refai, H.; Hassan, D.; Abdelmonem, R. Development and characterization of polymer-coated liposomes for vaginal delivery of sildenaﬁl citrate. Drug Deliv. 2017, 24, 278–288. [CrossRef] Refai, H.; Hassan, D.; Abdelmonem, R. Development and characterization of polymer-coated liposomes for vaginal delivery of sildenaﬁl citrate Drug Deliv 2017 24 278 288 [CrossRef] 46. Jøraholmen, M.W.; Škalko-Basnet, N.; Acharya, G.; Basnet, P. Resveratrol-loaded liposomes for topical treatment of the vaginal inﬂammation and infections. Eur. J. Pharm. Sci. 2015, 79, 112–121. [CrossRef] 47. Garcia-Manyes, S.; Oncins, G.; Sanz, F. Eﬀect of ion-binding and chemical phospholipid structure on the nanomechanics of lipid bilayers studied by force spectroscopy. Biophys. J. 2005, 89, 1812–1826. [CrossRef] 47. Garcia-Manyes, S.; Oncins, G.; Sanz, F. Eﬀect of ion-binding and chemical phospholipid structure on the nanomechanics of lipid bilayers studied by force spectroscopy. Biophys. J. 2005, 89, 1812–1826. [CrossRef] 48. Buchner, N.; Krumbein, A.; Rohn, S.; Kroh, L.W. Eﬀect of thermal processing on the ﬂavonols rutin and 48. Buchner, N.; Krumbein, A.; Rohn, S.; Kroh, L.W. Eﬀect of thermal processing on the ﬂavonols rutin and quercetin. RAPID Commun. MASS Spectrom. 2006, 20, 3229–3235. [CrossRef] 49. Chen, W.; Sun, S.; Liang, Y.; Song, J. Antioxidant property of quercetin-Cr(III) complex: The role of Cr(III) ion. J. Mol. Struct. References 2009, 918, 194–197. [CrossRef] 50. Moon, Y.J.; Wang, L.; DiCenzo, R.; Morris, M.E. Quercetin pharmacokinetics in Humans. Biopharm Dispos. 2008, 29, 205–217. [CrossRef] 50. Moon, Y.J.; Wang, L.; DiCenzo, R.; Morris, M.E. Quercetin pharmacokinetics in Humans. Biopharm. Drug Dispos. 2008, 29, 205–217. [CrossRef] 51. Choubey, S.; Varughese, L.R.; Kumar, V.; Beniwal, V. Medicinal importance of gallic acid and its ester 51. Choubey, S.; Varughese, L.R.; Kumar, V.; Beniwal, V. Medicinal importance of gallic acid and i derivatives: A patent review. Pharm. Pat. Anal. 2015, 4, 305–315. [CrossRef] [PubMed] 52. Bhattacharyya, S.; Ahammed, S.M.; Saha, B.P.; Mukherjee, P.K. The gallic acid—Phospholipid complex improved the antioxidant potential of gallic acid by enhancing its bioavailability. AAPS PharmSciTech 2013, 14, 1025–1033. [CrossRef] [PubMed] 53. Acevedo, F.; Hermosilla, J.; Sanhueza, C.; Mora-lagos, B.; Fuentes, I.; Rubilar, M.; Concheiro, A.; Alvarez-lorenzo, C. Gallic acid loaded PEO-core/zein-shell nano ﬁbers for chemopreventive action on gallbladder cancer cells. Eur. J. Pharm. Sci. 2018, 119, 49–61. [CrossRef] [PubMed] 54. Eslami, A.C.; Pasanphan, W.; Wagner, B.A.; Buettner, G.R. Free radicals produced by the oxidation of gallic acid: An electron paramagnetic resonance study. Chem. Cent. J. 2010, 4, 15. [CrossRef] [PubMed] 55. Manosroi, A.; Jantrawut, P.; Akazawa, H.; Akihisa, T.; Manosroi, W.; Manosroi, J. Transdermal absorption enhancement of gel containing elastic niosomes loaded with gallic acid from Terminalia chebula galls. Pharm. Biol. 2011, 49, 553–562. [CrossRef] [PubMed] , , [ ] [ ] 56. Caddeo, C.; Nacher, A.; Vassallo, A.; Francesca, M.; Pons, R.; Fernàndez-busquets, X.; Carbone, C.; Valenti, D.; Maria, A.; Manconi, M. Eﬀect of quercetin and resveratrol co-incorporated in liposomes against inﬂammatory/oxidative response associated with skin cancer. Int. J. Pharm. 2016, 513, 153–163. [CrossRef] 57. Chen, W.; Zou, M.; Ma, X.; Lv, R.; Ding, T.; Liu, D. Co-encapsulation of EGCG and quercetin in liposomes for optimum antioxidant activity J Food Sci 2019 84 111 120 [CrossRef] 56. Caddeo, C.; Nacher, A.; Vassallo, A.; Francesca, M.; Pons, R.; Fernàndez-busquets, X.; Carbone, C.; Valenti, D.; Maria, A.; Manconi, M. Eﬀect of quercetin and resveratrol co-incorporated in liposomes against inﬂammatory/oxidative response associated with skin cancer. Int. J. Pharm. 2016, 513, 153–163. [CrossRef] Valenti, D.; Maria, A.; Manconi, M. Eﬀect of quercetin and resveratrol co incorporated in liposomes against inﬂammatory/oxidative response associated with skin cancer. Int. J. Pharm. 2016, 513, 153–163. [CrossRef] 57. Chen, W.; Zou, M.; Ma, X.; Lv, R.; Ding, T.; Liu, D. References Co-encapsulation of EGCG and quercetin in liposomes for optimum antioxidant activity. J. Food Sci. 2019, 84, 111–120. [CrossRef] 57. Chen, W.; Zou, M.; Ma, X.; Lv, R.; Ding, T.; Liu, D. Co-encapsulation of EGCG and quercetin in liposomes for optimum antioxidant activity. J. Food Sci. 2019, 84, 111–120. [CrossRef] 58. Vitonyte, J.; Manca, M.L.; Caddeo, C.; Valenti, D.; Peris, J.; Usach, I.; Nacher, A.; Matos, M.; Gutiérrez, G.; Orrù, G.; et al. Bifunctional viscous nanovesicles co-loaded with resveratrol and gallic acid for skin protection against microbial and oxidative injuries. Eur. J. Pharm. Biopharm. 2019, 114, 278–287. [CrossRef] 59. Zhou, X.; Liu, H.Y.; Zhao, H.; Wang, T. RGD-modiﬁed nanoliposomes containing quercetin for lung cancer targeted treatment. OncoTargets Ther. 2018, 11, 5397–5405. [CrossRef] 60. Jangde, R.; Singh, D. Preparation and optimization of quercetin-loaded liposomes for wound healing, using response surface methodology. Artif. Cells Nanomed. Biotechnol. 2016, 44, 635–641. [CrossRef] 61. Castangia, I.; Nácher, A.; Caddeo, C.; Valenti, D.; Fadda, A.M.; Díez-Sales, O.; Ruiz-Saurí, A.; Manconi, M. Fabrication of quercetin and curcumin bionanovesicles for the prevention and rapid regeneration of full-thickness skin defects on mice. Acta Biomater. 2014, 10, 1292–1300. [CrossRef] [PubMed] 62. Liang, T.; Guan, R.; Shen, H.; Xia, Q.; Liu, M. Optimization of Conditions for Cyanidin-3-OGlucoside (C3G) Nanoliposome Production by Response Surface Methodology and Cellular Uptake Studies in Caco-2 Cells. Molecules 2017, 22, 457. [CrossRef] [PubMed] 63. Alvendal, C.; Ehrstrom, S.; Brauner, A.; Lundberg, J.O.; Bohm-starke, N. Elevated nitric oxide in recurrent vulvovaginal candidiasis—Association with clinical ﬁndings. Acta Obstet. Gynecol. Scand. 2017, 96, 295–301. [CrossRef] [PubMed] 64. De Cristo Soares Alves, A.; Mainardes, R.M.; Khalil, M.N. Nanoencapsulation of gallic acid and evaluation of its cytotoxicity and antioxidant activity. Mater. Sci. Eng. C 2019, 60, 126–134. [CrossRef] [PubMed] 21 of 21 21 of 21 Pharmaceutics 2020, 12, 9 65. Güran, M.; ¸Sanlıtürk, G.; Kerküklü, N.R.; Altunda˘g, E.M.; Yalçın, A.S. Combined eﬀects of quercetin and curcumin on anti-inﬂammatory and antimicrobial parameters in vitro. Eur. J. Pharmacol. 2019, 859, 172486. [CrossRef] [PubMed] 6. Li, Y.; Yao, J.; Han, C.; Yang, J.; Chaudhry, M.T.; Wang, S.; Liu, H.; Yin, Y. Quercetin, inﬂammation immunity. Nutrients 2016, 8, 167. [CrossRef] [PubMed] 67. Mu, M.M.; Chakravortty, D.; Sugiyama, T.; Koide, N.; Takahashi, K.; Mori, I.; Yoshida, T.; Yokochi, T. The inhibitory action of quercetin on lipopolysaccharide-induced nitric oxide production in RAW 264.7 macrophage cells. J. Endotoxin Res. 2001, 7, 431–438. [CrossRef] [PubMed] © 2019 by the authors. References 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/W4239162690	https://www.researchsquare.com/article/rs-602220/v1.pdf?c=1631899548000	English	null	The Water Incident Database (WAID) 2012 to 2019: A Systematic Evaluation of the Documenting of UK Drownings	Research Square (Research Square)	2,021	cc-by	9,452	Background Globally, drowning accounted for 372,000 lives lost in 2014 and is a leading cause of accidental death in most countries (1). This death toll is almost two thirds that of malnutrition and over half that of malaria (1). Drowning is also amongst the ten leading causes of death in children and young people in every region of the World (1), is a primary cause of occupational death and injury, and is a particular problem in middle- and low-income countries (2). This global figure is thought to represent an underestimation by four or five times due to very limited mechanisms for documenting drowning deaths in many countries (2, 3). Even in developed countries such as the United Kingdom (UK), drowning is a leading cause of accidental and intentional death. Indeed, fatal drowning accounts for 400 to 750 deaths, including suicides, each year (4). Non-fatal water-based incidents are between 20-50 times that of the drowning rate although these data are rarely reported (5). Data from comparable developed countries such as Australia suggest approximately $188 million AUD (~£105.9 million GBP) is spent annually on responding to drowning and water-based incidents, indicating a substantial economic cost to society in addition to the emotional burden associated with each case (6). In many cases such death and injury can be avoided with evidence-based safety advice and planning to support intervention (1, 7). Since 2009, the Water Incident Database (WAID) has been used to document UK fatal and non-fatal water-related incidents (referred to as drownings here on). The database was conceived and constructed by the National Water Safety Forum (NWSF) with the support of the Royal Society for the Prevention of Accidents (RoSPA), who host the database. WAID brings together water-related incident data from a wide range of sources within the UK search and rescue region. In doing so, the database could meet the key initiative of the World Health Organization (WHO) to improve the quality of data describing drowning frequency (1) and may potentially underpin effective drowning prevention interventions (8). The key aims of WAID are to i) provide insights into levels of risk (including risk acceptability), enabling meaningful comparisons with activities outside the water sector; ii) to supersede the uncoordinated efforts of organisations trying to establish national trends based on limited data of uncertain quality; iii) to produce much higher quality evidence; and iv) to maximise value and minimise aggregate cost of data collection (9). The Water Incident Database (WAID) 2012 to 2019: A Systematic Evaluation of the Documenting of UK Drownings Samuel Hills Bournemouth University Matthew Hobbs University of Canterbury Michael Tipton University of Portsmouth Martin Barwood (  M.Barwood@leedstrinity.ac.uk ) Leeds Trinity University Keywords: Immersion, cold water, accident, suicide, injury, environmental risk factor License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Page 1/14 Abstract Background. Death by drowning is a leading cause of accidental death in the United Kingdom (UK) and worldwide. The World Health Organization (WHO) states that effective documentation of drowning is required to describe drowning frequency and to underpin effective drowning prevention intervention, thus improving the quality of data describing drowning frequency represents a key initiative. The water incident database (WAID) has been used to document UK fatal and non-fatal water-based incidents since 2009. WAID has not undergone a systematic evaluation of its data or data collection procedures to establish if the database meets the WHO requirements. The present study investigated the characteristics of UK fatal drowning incidents and audited current WAID data capture procedures. Methods. Data for the fatal drowning cases recorded between 2012 and 2019 were reviewed. Summary statistics were produced to describe the prevalence of UK drownings and a two-phase blind audit was conducted to establish a) the completeness of each of the 22 WAID fields (i.e., columns) and b) the reliability of data entry processes by reviewing written data sources originally used to populate WAID. Results. A total of 5,501 fatalities were recorded between 2012-2019. Drowning was most frequent amongst males aged 35 to 60 years (n=1,346), whilst suspected accidents and suicides accounted for 44% and 35% of fatalitie. Suicide by drowning was at a peak in the most recent year of data analysed (i.e., 2019; 279 cases) highlighting an urgent need for targeted intervention. Audit phase one indicated that 16% of all fields were incomplete, thus indicating potential redundancy, duplication, or the need for onward review. Phase two indicated high levels of agreement (80±12%) between audited cases and the ‘true’ WAID entries. Conclusions. This study confirms WAID as a rigorous, transparent and effective means of documenting UK drownings thereby meeting WHO requirements for data quality. Such findings allow researchers and policy makers to use WAID to further investigate UK drowning with a view to improving public safety measures and drowning prevention interventions; our work these data is ongoing. Observations alongside several expert recommendations proposed in this manuscript have informed a revised version of the WAID database. Background Data relating to each incident are entered into WAID by water safety agencies (primary data; e.g., Royal National Lifeboat Institution, Her Majesty’s Maritime Coastguard Agency, the National Fire Chiefs Council, the Royal Lifesaving Society & RoSPA) in accordance with pre-defined fields and taxonomies. Presently WAID includes many thousands of entries from several sources relating to each drowning or water-based incident, including inquests. Drowning is a multifactorial phenomenon (10) and it is therefore important that data relating to each water-based incident distinguish between the outcomes and, if possible, the contributory factors that are specific to a given country or event (8). Establishing these distinctions and contributory factors may enable targeted prevention strategies. For example, it is likely that at least one causal factor in many fatal and non-fatal water-based incidents relates to persons entering the water accidently (4). Further complications can partly be attributed to the low average annual water temperature around the UK which is between 11 and 13 ºC (11). Such water temperatures are known to evoke the life-threatening cold shock response (12, 13) on immersion which increases the chance of aspirating water to the lung causing asphyxiation precipitated by a loss of respiratory control (12, 13). Cold water is also one of the attributed reasons that persons often drown within 3 metres of the safe refuge of land as a result of swim-failure (14). Recorded data must also distinguish those circumstances where persons voluntarily enter the water for reasons of recreation (e.g., 15) where natural causes (such as an underlying health condition; e.g., 16) and suicide are potential outcomes, the risk of which may be compounded by drug or alcohol intoxication (e.g., 17). There are also plausible scenarios where a victim forcibly enters the water as part of criminal related activity (18). Clearly an effective database must comprise of a transparent and unbiased mechanism for documenting and distinguishing between these eventualities. To date WAID processes and data have not been independently assessed to verify the rigour and quality of the collection procedures and outputs. This study aimed to describe the characteristics of UK fatal drowning incidents, including water related fatalities, that occurred between 2012 and 2019. Thereafter we aimed to audit the procedures of data capture relating to a sub-set of drowning cases to examine the agreement between the extant data and Page 2/14 Page 2/14 the data generated by a blind audit. WAID Data Set Specific project approval was granted by the Leeds Trinity University School of Health and Social Sciences research ethics committee (SHSS-2020-03), and a formal data agreement was signed between the authors and RoSPA (the data owner) prior to any data being exchanged. This written agreement explicitly defined the scope of data use, secure storage, and dissemination of related research findings. This study focused only on data relating to fatal incidents recorded by RoSPA over a pre-defined time period. Accordingly, anonymised WAID data were received for drowning incidents occurring between 1st January 2012 and 31st December 2019, inclusive. After removing any confirmed non-fatal and duplicate cases, the final dataset for this research consisted of 5,051 fatalities by drowning that were recorded in WAID. Background This study forms the first step in a programme of work planning to undertake epidemiological research with this UK drowning data set. Scientific convention in such onward studies indicates that the data must first be audited (19, 20) to ensure the data quality is fit to address the proposed research questions (21). In doing so our project represents the first step in scrutinising the database against the WHO’s drowning prevention implementation guide relating to data collection and quality (22). Also in line with this implementation strategy, we sought to make onward developmental recommendations for inclusions and revisions to WAID to improve the data captured in relation to each fatal and non-fatal drowning event. These recommendations were to inform a second iteration of the WAID database planned for future launch; WAIDv2. the data generated by a blind audit. This study forms the first step in a programme of work planning to undertake epidemiological research with this UK drowning data set. Scientific convention in such onward studies indicates that the data must first be audited (19, 20) to ensure the data quality is fit to address the proposed research questions (21). In doing so our project represents the first step in scrutinising the database against the WHO’s drowning prevention implementation guide relating to data collection and quality (22). Also in line with this implementation strategy, we sought to make onward developmental recommendations for inclusions and revisions to WAID to improve the data captured in relation to each fatal and non-fatal drowning event. These recommendations were to inform a second iteration of the WAID database planned for future launch; WAIDv2. Current WAID Data Entry Procedures Having established a date and time for the incident and a stakeholder reference number (i.e., a reference to the safety, rescue or public organisation responsible for entering the data into the main database), each onward WAID data entry phase requires one or more tick box or narrative entries as part of expanded sub-sections. Table 1 provides operational definitions for the 22 WAID field taxonomies comprising this process. Figure 1 provides a schematic overview of the sequence followed for incident entered into WAID including mandatory fields. Figure 2 provides examples of the sub-sets of the choices available for the fields of age category, sex, (suspected) intoxication and the mandatory field describing the suspected outcome. The relevant information for each incident is then converted in summarised form, with each row representing a distinct water-based incident and each column representing a WAID field taxonomy. Data audit A two-phase blind audit approach was taken to the WAID data audit. Phase one involved examining the computerised database itself to quantify the completeness of each field and producing summary statistics describing the incidence and characteristics of fatal drowning in the UK. Phase two assessed the reliability of the process of data entry into WAID by reviewing the stored written data (e.g., in the form of media articles, coroners’ reports, accident reports, court proceedings, etc.) that had been originally used to populate the main database. Phase one: database completeness audit and summary For phase one, the available WAID data were imported into R Studio (V 3.6.1; Vienna, Austria) before each of the 22 primary fields were assessed for completeness. To indicate any potential redundancy or duplication of field information in the present iteration of WAID and allow recommendations to be made for subsequent editions of the database, the number of incomplete or missing entries for each field was determined as an absolute value and then as a percentage of the 5,051 total cases. Thereafter, the prevalence of recorded UK drowning cases was quantified according to different person-, date- (i.e., year), and outcome-related categories available within WAID. The number of WAID entries were therefore calculated specifically in relation to each drowning outcome category listed within the available taxonomies (i.e., ‘accident suspected’, ‘not recorded’, ‘suicide suspected’, ‘natural causes suspected’, ‘crime suspected’ ; see figure 2), the age of the victim (i.e., ‘0 to 2 years’, ‘3 to 5 years’, ‘6 to 12 years’, ’13 to 18 years’, ’19 to 35 years’, ’36 to 60 years’, ‘over 60 years’, ‘not recorded’), the sex of the victim (i.e., ‘male’, ‘female’, ‘not recorded’), and whether the victim was suspected to have been intoxicated (i.e., ‘alcohol’, ‘drugs’, ‘alcohol and drugs’, ‘none’). Data were further examined by calendar year from 2012 to 2019, inclusive. Phase one - database completeness audit and data summary Table 2 shows the completeness of each of the 22 WAID fields examined in phase one of the audit. Several fields were populated for all or most of the 5,051 cases such as latitude and longitude, whereas some fields such as “visibility” and “water depth” were largely incomplete. From a total of 111,122 potential field entries (i.e., 22 fields multiplied by 5,051 recorded fatal cases), 84% were complete. A greater prevalence of drowning was observed in males (a total of 3,722 of the 5,051 fatalities; ~74%) compared with females (1,021 cases; ~20%), whilst there were 308 (~6%) incidents for which the casualty’s sex was “not recorded”. The raw count distribution of drowning cases across the age categories outlined in WAID is displayed in Table 3, whereby the most deaths (36%) occurred in 36- to 60-year-olds. When data were expressed relative to the number of years that each age category spanned (e.g., age category 13 to 18 spanned six years; 189 drowning deaths divided by six years = 32 deaths per year in age range), the highest drowning rate remained within the 36 to 60 age category (67 cases per year; p/y). This calculation was not conducted for the over 60 years age category or where age category was “not recorded” because it was not possible to ascertain the number of years that these age categories spanned. The number of drowning deaths expressed by sex and age group are displayed in Table 4. The most numerous age category for both males and females was 36 to 60 years old indicating a common age for drowning prevalence. When the data were expressed per year included in each age category, the greatest prevalence of drowning remained 36 to 60 years in females whereas the highest per year value for males was observed in the 19 to 35 years category. Drug and/or alcohol intoxication was implicated in a total of 820 deaths (16% of total cases), with 604 (74%) of this 820 associated with intoxication by alcohol alone, 104 (13%) cases associated with drugs alone, and 112 (14%) drownings associated with a combination thereof. The distribution of the 5,051 cases across each drowning outcome category is expressed in Figure 3a, whilst Figure 3b shows the number of WAID recorded cases for each year of the study period combined with the distribution of these cases across the five downing outcomes. Phase two: audit of data entry processes Phase two: audit of data entry processes Prior to phase two, data from WAID age categories ‘0 to 2 years’, ‘3 to 5 years’, ‘6 to 12 years’, and ‘13 to 18 years’ were removed from the dataset. Removal of such data was a condition of the ethical approval that was granted to minimise any onward emotional trauma to those undertaking the audit potentially caused by reading written evidence relating to child and youth drowning incidents. The remaining cases were grouped according to the drowning outcomes of ‘accident suspected’, ‘crime suspected’, ‘natural causes’, ‘not recorded’ and ‘suicide suspected’ before 50 cases, 10 relating to each outcome, were selected for auditing based on their unique identifier (i.e., “Waidised ID”) using a random sampling function in R Studio. The written reports (e.g., newspaper articles, coroner reports, other media, etc.) that had originally been used to populate WAID for each of these cases were printed and redacted by a member of RoSPA staff. This person was independent to the research team and oversaw the removal of personal identifier information such as the names, dates of birth, and addresses of any victims or witnesses. Redacted hard copy documents were then reviewed by two members of the research team (MB & SH) to establish the reliability of WAID data entry processes. It is worthy to note that more than one fatality can be associated with a “Waidised ID” which relate to cases rather than individual persons. Prior to commencing the main audit, four randomly selected cases were pilot tested by both auditors (MB & SH) to verify and familiarise themselves with the method of data entry into WAID that would be replicated during the main audit thereafter. This pilot testing involved each auditor in a blinded manner (i.e., Page 3/14 Page 3/14 without consultation between auditors and without knowing the ‘true’ WAID entries for each case) separately reviewing the stored written details for each case, recording the information that they would have entered into WAID for each field taxonomy based upon the written evidence available to them. Auditors then compared their compiled would-be entries between each other and with reference to the ‘true’ WAID record for the relevant case, cross-checking and discussing in detail any inconsistencies. Agreement levels between auditors were ~80% at the pilot stage and ~77% (grouped mean across auditors) between the auditor entry and the ‘true’ entry latterly confirmed on WAID. Phase one - database completeness audit and data summary Approximately 44% and 35% of incidents recorded over the study period were suspected to be results of accidents and suicides, respectively. “Outcome not recorded” has increased since 2014. The most total cases were recorded in 2013, while the least fatalities occurred in 2018. Main Phase Two Audit The 50 case records randomly selected for the phase two audit described a total of 58 fatalities (as some records documented multiple drownings within a single incident). Following assignment of agreement values (i.e., ‘0’, ‘0.5’, or ‘1’) and subsequent analysis, the overall level of agreement between auditor dummy entries and the true WAID entries was 80 ± 12%. These values were 79 ± 13% and 81 ± 11% respectively for each of the two auditors. Table 5 shows the level of agreement between auditor-entered data and actual WAID records for each WAID field assessed, highlighting that ‘injury’, ‘water depth’, and ‘wind’ showed the greatest agreement (>95%). Conversely, ‘Ordinance Survey reference’ and ‘what happened’ had agreement values <60%. The auditors’ subjective experiences of undertaking the audit suggested that establishing the identifier characteristics of the casualty (e.g., age and sex) was typically deemed to be relatively straightforward, whereas specific details about the incident such as “what happened” were often more difficult to ascertain. Frequently, a lack of evidence to suggest other possibilities meant that “body recovery” had to be concluded for the ‘what happened’ field. Main phase two audit To reflect the typical mode of data entry into WAID, the auditors used the redacted hard copy documents supplied by RoSPA to populate an electronic “dummy” version of WAID fields and taxonomies using the NWSF test site. This site is functionally the same as the main WAID data entry site but populates a separate database used for training. Once all 50 randomly selected cases had been individually considered and entered into the dummy database, auditors’ entries were compared to the true WAID record. The extent of agreement in each case was quantified by assigning each field a value of ‘0’, ‘0.5’, or ‘1’ when entries did not agree, partially agreed (i.e., determined by consensus amongst the research team), or were identical to those listed in WAID. For the fields of latitude and longitude, data were considered to be identical (i.e., a value of 1 was assigned) when dummy and WAID entries were consistent to an accuracy of ≥4 decimal places. This represented a threshold of 11 m leeway within which a value of 1 would be assigned. The agreement between auditor- entered data and WAID data was then calculated in percentage terms (i.e., 100% agreement would indicate that values of 1 had been assigned for all entries) for each field and overall. In addition, both auditors documented their subjective experiences of the processes of data entry, agreement and analysis using a notebook. Such information was recorded to assist in providing recommendations for potential improvement in the design and administration of subsequent editions of WAID. The phase two audit did not consider the fields of “Waidised ID” and “stakeholder reference” as these fields pertain purely to pseudo- anonymised case identifier information. Results Phase one - database completeness audit and data summary Phase two: audit of data entry processes Further consultation with RoSPA clarified any outstanding questions as to the appropriate interpretation of the WAID taxonomies and the processes of data entry into the main database. Based on the outcome of the initial pilot of four cases and detailed consultation with RoSPA, the auditors progressed to the main phase of auditing. without consultation between auditors and without knowing the ‘true’ WAID entries for each case) separately reviewing the stored written details for each case, recording the information that they would have entered into WAID for each field taxonomy based upon the written evidence available to them. Auditors then compared their compiled would-be entries between each other and with reference to the ‘true’ WAID record for the relevant case, cross-checking and discussing in detail any inconsistencies. Agreement levels between auditors were ~80% at the pilot stage and ~77% (grouped mean across auditors) between the auditor entry and the ‘true’ entry latterly confirmed on WAID. Further consultation with RoSPA clarified any outstanding questions as to the appropriate interpretation of the WAID taxonomies and the processes of data entry into the main database. Based on the outcome of the initial pilot of four cases and detailed consultation with RoSPA, the auditors progressed to the main phase of auditing. Main phase two audit Discussion Page 4/14 Page 4/14 Page 4/14 The WHO data set and drowning data from other countries have tended not to focus on intentional drownings (e.g., 3) because no reporting sub-categories existed until recently in the International Statistical Classification of Diseases and Related Health Problems [ICD] codes for suicide by drowning (3) to additionally document intentional drowning. The eleventh edition of the ICD has recognised and addressed this imbalance (25). The case for the reporting of such data must be balanced against the role this could play in highlighting to vulnerable people the potential for drowning as a mechanism for intentional death (26). Observations from the present data set indicate that, acknowledging that sex was not recorded for a further 188 suspected suicides, a higher proportion of female drownings were recorded as “suicide suspected” (44% of female drownings and 30% of male drownings between 2012 and 2019). Whilst not directly comparable, publicly available UK data for 2018 report 4.4% of suicide deaths were by drowning in females compared to 3.8% in males (27). Collectively, it is of significant concern that suicide by drowning was at a peak in the most recent year of data analysed in the present data set (i.e. 279 cases) highlighting an urgent need for targeted intervention. As part of this study we sought to use our procedures to make onward developmental recommendations for inclusions and revisions to WAID to improve the data captured in relation to each fatal and non-fatal drowning event. The recommendations were reached in consultation with RoSPA, on the basis of the auditors’ shared experiences of undertaking the audit, alongside published evidence and the experience of the research team. The recommendations relate primarily to documenting additional factors relating to drowning events and improving the clarity of existing factors considered as part of WAID that may substantially influence an individual’s likelihood of drowning and subsequently inform prevention strategies. These observations are being made openly available in order to inform the practices of other researchers and drowning databases. To avoid potential redundancy of the suggested fields, the value of including these variables must be considered in light of their feasibility and viability for documentation. Recommendations for potential inclusion within WAID were as follows: i. Estimated water temperature and water conditions. It is known that low water temperatures are linked to the magnitude of the life-threatening cold shock response (12) with lower temperatures linked to increased likelihood of drowning (13, 28). Page 4/14 The WHO data set and drowning data from other countries have tended not to focus on intentional drownings (e.g., 3) because no reporting sub-categories existed until recently in the International Statistical Classification of Diseases and Related Health Problems [ICD] codes for suicide by drowning (3) to additionally document intentional drowning. The eleventh edition of the ICD has recognised and addressed this imbalance (25). The case for the reporting of such data must be balanced against the role this could play in highlighting to vulnerable people the potential for drowning as a mechanism for intentional death (26). Observations from the present data set indicate that, acknowledging that sex was not recorded for a further 188 suspected suicides, a higher proportion of female drownings were recorded as “suicide suspected” (44% of female drownings and 30% of male drownings between 2012 and 2019). Whilst not directly comparable, publicly available UK data for 2018 report 4.4% of suicide deaths were by drowning in females compared to 3.8% in males (27). Collectively, it is of significant concern that suicide by drowning was at a peak in the most recent year of data analysed in the present data set (i.e. 279 cases) highlighting an urgent need for targeted intervention. Drowning in persons up to the age of 18 accounted for around 5% (262) of total deaths considered in the present study; these national data are comparable to historic, regional data which show drowning as a leading cause of accidental death in children and adolescents (23). Persons aged 36 to 60 saw the highest frequency of drownings and these comprised of mostly males when sub-divided by sex. Males tend to take greater risks on or around water and intoxication by alcohol or drugs is known to exacerbate this risk taking (24). Indeed, alcohol was suspected to be implicated, whether solely or in combination with drugs, in a total of 16% of the fatal cases reviewed in the present study. Approximately 44% of the cases reviewed in the phase one audit were documented as ‘accident suspected’. The next most numerous outcome category after ‘accident suspected’ was ‘suicide suspected’ (35% of cases). Page 4/14 This study sought to describe the characteristics of UK fatal drowning incidents that occurred between 2012 and 2019 and to establish the completeness of WAID across the fields currently included in the database. Thereafter we aimed to audit the procedures of data capture relating to a sub-set of drowning cases in WAID and to examine the agreement between the extant data and the data generated by a blind audit. In doing so, this study has been able to establish whether the database could meet the key initiative of the WHO to improve data quality describing drowning frequency (1). Our findings show that the number of documented UK drownings remained between 585 (in 2018) and 669 (in 2013) deaths per year for each year from 2012-2019, inclusive. Moreover, the fact that males most frequently drowned during this time period (i.e., male deaths comprising approximately 74% of the recorded fatalities, acknowledging that the victim’s sex was not recorded for a further 6% of the sample), is a finding broadly consistent with other developed countries; Australia 78%; Canada 81%; New Zealand 82% (8). During the audit phase of the study we were able to establish a high level of completeness of most fields in WAID indicating the effectiveness of the database in capturing key characteristics relating to water related fatalities and drowning. We were also able to establish that the stored written evidence associated with each case enabled a high level of agreement to be achieved between auditor and true WAID entry by reading the case details and following the procedures of data entry. This study therefore confirms WAID as a rigorous, transparent and effective means of documenting UK drownings. Very few studies have undertaken an independent audit of drowning database entry procedures and reported the findings in the open scientific literature. Collectively, these findings now legitimise our intentions to undertake a programme of research using this UK drowning dataset. This study sought to describe the characteristics of UK fatal drowning incidents that occurred between 2012 and 2019 and to establish the completeness of WAID across the fields currently included in the database. Thereafter we aimed to audit the procedures of data capture relating to a sub-set of drowning cases in WAID and to examine the agreement between the extant data and the data generated by a blind audit. Page 4/14 In doing so, this study has been able to establish whether the database could meet the key initiative of the WHO to improve data quality describing drowning frequency (1). Our findings show that the number of documented UK drownings remained between 585 (in 2018) and 669 (in 2013) deaths per year for each year from 2012-2019, inclusive. Moreover, the fact that males most frequently drowned during this time period (i.e., male deaths comprising approximately 74% of the recorded fatalities, acknowledging that the victim’s sex was not recorded for a further 6% of the sample), is a finding broadly consistent with other developed countries; Australia 78%; Canada 81%; New Zealand 82% (8). During the audit phase of the study we were able to establish a high level of completeness of most fields in WAID indicating the effectiveness of the database in capturing key characteristics relating to water related fatalities and drowning. We were also able to establish that the stored written evidence associated with each case enabled a high level of agreement to be achieved between auditor and true WAID entry by reading the case details and following the procedures of data entry. This study therefore confirms WAID as a rigorous, transparent and effective means of documenting UK drownings. Very few studies have undertaken an independent audit of drowning database entry procedures and reported the findings in the open scientific literature. Collectively, these findings now legitimise our intentions to undertake a programme of research using this UK drowning dataset. Drowning in persons up to the age of 18 accounted for around 5% (262) of total deaths considered in the present study; these national data are comparable to historic, regional data which show drowning as a leading cause of accidental death in children and adolescents (23). Persons aged 36 to 60 saw the highest frequency of drownings and these comprised of mostly males when sub-divided by sex. Males tend to take greater risks on or around water and intoxication by alcohol or drugs is known to exacerbate this risk taking (24). Indeed, alcohol was suspected to be implicated, whether solely or in combination with drugs, in a total of 16% of the fatal cases reviewed in the present study. Approximately 44% of the cases reviewed in the phase one audit were documented as ‘accident suspected’. The next most numerous outcome category after ‘accident suspected’ was ‘suicide suspected’ (35% of cases). Page 4/14 Including simple measures such as the drowning victim’s postcode (where available) may enable targeted Page 5/14 Page 5/14 intervention according to expected behaviours. Moreover, this variable can also enable the victim's social demographic to be considered as an influential factor similar to other health conditions (20). iv. Ethnicity. Presently we have only a limited indication of drowning prevalence in minority ethnic groups in the UK. It is plausible that drowning risk or incidence may be higher for social or cultural reasons and targeted intervention based on examining this possibility may be valuable. For example data from Canada, another country where cold water is an influential factor in drowning, indicate the age-standardized drowning rate is significantly higher among men of Asian, African, or Hispanic ethnicity compared to men of Greater European ethnicity and for women of Asian, African, or Hispanic ethnicity compared to women of Greater European ethnicity (33). Manual entry of this variable currently exists in WAID. v. Clothing worn. We have previously suggested that air trapped between clothing layers in normal seasonal clothing assists short term buoyancy and therefore flotation (4). Moreover, protecting the skin from rapid skin cooling reduces the extent of the cod shock response (12) and it has been reported that 50% of persons who enter water unintentionally are clothed and drown within 3 metres of safe refuge (4). Knowledge of the extent of clothing worn in cases of drowning may establish the differing role that clothing plays in drowning case outcomes. vi. Establishing a flotation factor. Our laboratory studies have shown variability in flotation capability between individuals (4). We suggested the ability to float may impact upon the likelihood of drowning (4). Yet, factors such as body density, lung volume, clothing and water temperature have all been suggested to impact upon flotation capability (34) and could be derived from entries on WAID. Identifying flotation capability as an influential factor in drowning may underpin the educational basis for focussing on the skillset to achieve floating as part of learn to swim and survival skills. vii. Separate data for drowning event and recovery. With items i) to iii) in mind, where possible future recorded data should try to distinguish between data available at the time of water entry (if known) and that derived from the time of body recovery. Conclusions In summary, this study sought to describe the characteristics of UK fatal drowning incidents that occurred between 2012 and 2019 and to audit the processes and data stored in WAID. In doing so we are able to conclude that the database meets the key initiative of the WHO to improve data quality describing drowning frequency (1). However, revisions and potential improvements to WAID are feasible including considering including variables that describe pre-drowning behaviour and distinguishing between the environmental conditions at the time of water entry from the time of body recovery. The data in WAID now appear suitable to be considered for onward study to address the often-neglected public health issue of drowning (3). This programme of work can now progress to step two of the WHO’s drowning prevention implementation guide with the aim of identifying drowning risk factors (22). Page 4/14 Many of the WAID phase two audited cases included information from the time of recovery where some of the key variables (e.g., air temperature) may have changed significantly between the time of water entry and body recovery. viii. Point of entry to emergency care. In cases where the immersed victim is rescued alive, the proximity of a hospital or the required emergency care may be critical in determining the outcome. Documenting the details around the point of entry to emergency care (if completed) or the proximity of the nearest emergency care to the water location may help contextualise the importance of the transition time to receiving medical support; this inclusion would also raise the possibility of Utstein style reporting on the drowning case should it include resuscitation attempts (35). This study is not without limitations. Whereas the procedures underpinning the phase two data audit were standardised between auditors and pilot testing was conducted before the main audit, the persons involved were not experienced users of WAID. It is therefore plausible that some of the audited entries were subject to judgement error on the part of the auditor. However, as further standardised training (similar to the in-house training provided by RoSPA to those individuals responsible for populating WAID) is likely to improve agreement with the true WAID case, it is reasonable to suggest that the present study probably represents a conservative estimate of the agreement. Secondly, we were only able to audit 50 cases (58 fatalities) in phase two, a sample comprising approximately 1% of the database and excluding child and adolescent drownings. It is plausible that the effects we report in this fraction of the data do not reflect the wider picture or that seen in the non-fatal cases and that such observations are restricted to adult drownings only. Nevertheless, the randomly selected sample of cases examined a complete range of all of the recorded WAID drowning outcomes, was completed independently of RoSPA, and required a significant outlay of time, manpower and resource. Future studies including data collected after 2019 may seek to verify our observations. List Of Abbreviations ICD: International Statistical Classification of Diseases and Related Health Problems NWSF: National Water Safety Forum RoSPA: The Royal Society for the Prevention of Accidents UK: United Kingdom WAID: Water Incident Database WHO: World Health Organisation NWSF: National Water Safety Forum WHO: World Health Organisation Page 4/14 Accordingly, it would be valuable to include a default (rather than optional) estimate of water temperature and water conditions to examine the role these variables play in future WAID cases. Moreover, a water temperature of 6 °C or below has been linked to extended underwater survival time whilst submerged which is also linked to search and rescue (SAR) duration (29). Therefore, knowledge of water temperature in drowning and non-fatal water accident cases may enable decision-making models to be verified or refined that underpin SAR. A plausible option for populating the database may be to triangulate “live” measured data or data derived from seasonal estimates per given body of water particularly in countries where cold water is a seasonal threat. ii. Estimated air temperature. In circumstances where persons voluntarily enter the water the extant air temperature may partially underpin the decision to do so (30). Data from Canada indicate ambient air temperatures exceeding 30 °C increased the likelihood of outdoor drowning by 69%. Fralick et al. (30) also showed that drowning risk in all age groups and genders increased with increasing ambient temperature but to the greatest extent in males. Given that coastal and inland water temperatures tend to be lowest during the Spring (11) and high ambient temperatures are plausible at this time, knowledge of the extant air temperature at the approximate date and time of water entry may enable proactive drowning prevention interventions based on weather forecasting. A default entry to WAIDv2 with Triangulation of “live” or recorded data may be a viable option. iii. Behavioural factors in drowning. Drowning is a multifactorial event that may include a significant behavioural component (31). The inclusion of measures that document behavioural factors may provide valuable information about the behaviours that precede drowning such as visitation of water sites for recreation (32). For example, it is plausible that drowned victims may travel short or long distances for reasons of leisure to access the water environment. In the case of the latter, a transition may take place from a low drowning risk and warning environment (e.g., a city) into one that carries far greater risk (e.g., coastal or river environments). If proven, proactive drowning prevention interventions could target the point of origin of the potential victim if the behavioural pattern is established. Declarations Page 6/14 Page 6/14 Ethics approval and consent to participate Specific project approval was granted by the Leeds Trinity University School of Health and Social Sciences research ethics committee (SHSS-2020-03), and a formal data agreement was signed between the authors and RoSPA (the data owner) prior to any data being exchanged. This written agreement explicitly defined the scope of data use, secure storage, and dissemination of related research findings. Participant consent was not obtained because this study focused only on data relating to fatal incidents recorded by RoSPA over a pre-defined time period. Consent for publication Not applicable. The formal data agreement included RoSPA consent for publication of the data included within this manuscript. Not applicable. The formal data agreement included RoSPA consent for publication of the data included within this manuscript. A il bili f d d i l Not applicable. The formal data agreement included RoSPA consent for publication of the data included within this manuscript. Availability of data and materials The dataset describing the individual drowning cases upon which this manuscript is based cannot be made available. The sensitive nature of the data meant that non-disclosure was a condition of the signed data agreement and granting of ethical approval. An anonymised annual report of WAID data at the drowning cohort level is available from the National Water Safety Forum https://nationalwatersafety.org.uk/waid/annual-reports-and-data/ Competing interests Competing interests The authors declare that they have no competing interests This work was funded by quality related (QR) research funding through Research England. MB and MH developed the idea for the study. MB, SH and MH applied for funding and ethical approval. SH, MH, and MB analysed the data. All authors interpreted the data, drafted and revised the manuscript. Acknowledgements The authors would like to thank Mr David Walker (RoSPA) and Mrs Rachael Brogan (RoSPA) for their guidance throughout the audit. The authors would also like to thank the members of the National Water Safety Forum. References 1. WHO. Global report on drowning: preventing a leading killer. World Health Organization 2014. Available from: https://www.who.int/publications/i/item/global-report-on-drowning-preventing-a-leading-killer Accessed: 09/09/2020 1. WHO. Global report on drowning: preventing a leading killer. World Health Organization 2014. Available from: https://www.who.int/publications/i/item/global-report-on-drowning-preventing-a-leading-killer Accessed: 09/09/2020 2. Bierens JJ, Lunetta P, Tipton MJ, Warner DS. Physiology of drowning: a review. Physiol. 2016;31:147-66. 3. Lu T-H, Lunetta P, Walker S. Quality of cause-of-death reporting using ICD-10 drowning codes: a descriptive study of 69 countries. BMC Med Res Methodol. 2010;10(30):1-6. 3. Lu T-H, Lunetta P, Walker S. Quality of cause-of-death reporting using ICD-10 drowning codes: a descriptive study of 69 countries. BMC Med Res Methodol. 2010;10(30):1-6. 4. Barwood MJ, Bates V, Long GM, Tipton MJ. “Float First”: trapped air between clothing layers significantly improves buoyancy on water immersion in adults, adolescents and children. Int J Aquat Res Ed. 2011;5(2):147-63. 4. Barwood MJ, Bates V, Long GM, Tipton MJ. “Float First”: trapped air between clothing layers significantly improves buoyancy on water immersion in adults, adolescents and children. Int J Aquat Res Ed. 2011;5(2):147-63. 5. Onyekwelu E. Drowning and near drowning. Internet J Health. 2008;8(8):1-4. 6. A 13-year national study of non-fatal drowning in Australia. Royal Life Saving Society Australia 2017. Available from: https://www royallifesaving com au/ data/assets/pdf file/0003/19938/3985 v4 RLS NonFatalSymposium ReportHR PROO 6. A 13-year national study of non-fatal drowning in Australia. Royal Life Saving Society Australia 2017. Available from: https://www.royallifesaving.com.au/__data/assets/pdf_file/0003/19938/3985_v4_RLS_NonFatalSymposium_ReportHR_PROOF_LR.pdf. Accessed: 09/09/2020 from: https://www.royallifesaving.com.au/__data/assets/pdf_file/0003/19938/3985_v4_RLS_NonFatalSymposium_ReportHR_PROOF_LR.pdf. Accessed: 09/09/2020 7. Pearn J, Nixon JW, Franklin RC, Wallis B. Safety legislation, public health policy and drowning prevention. Int J Inj Contr Saf Promot. 2008;15(2):122- 3. 7. Pearn J, Nixon JW, Franklin RC, Wallis B. Safety legislation, public health policy and drowning prevention. Int J Inj Contr Saf Promot. 2008;15(2):122- 3. 8. Peden AE, Franklin RC, Clemens T. Exploring the burden of fatal drowning and data characteristics in three high income countries: Australia, Canada and New Zealand. BMC Public Health. 2019;19(1):794. 8. Peden AE, Franklin RC, Clemens T. Exploring the burden of fatal drowning and data characteristics in three high income countries: Australia, Canada and New Zealand. BMC Public Health. 2019;19(1):794. 8. Peden AE, Franklin RC, Clemens T. Exploring the burden of fatal drowning and data characteristics in three high income countries: Australia, Canada and New Zealand. BMC Public Health. 2019;19(1):794. 9. NWSF. National Water Safety Forum coordinating group terms of reference. National Water Safety Forum. 2019. Available from: https://www.nationalwatersafety.org.uk/media/1195/rospa-nwsf-paper-oct19-final.pdf. Accessed: 10/09/2020 10. References Examining longitudinal associations between the recreational physical activity environment, change in body mass index, and obesity by age in 8864 Yorkshire Health Study participants. Soc Sci Med. 2019;227:76-83. 19. Hobbs M, Griffiths C, Green M, Christensen A, McKenna J. Examining longitudinal associations between the recreational physical activity environment, change in body mass index, and obesity by age in 8864 Yorkshire Health Study participants. Soc Sci Med. 2019;227:76-83. 20. Hobbs M, Griffiths C, Green M, Jordan H, Saunders J, McKenna J. Neighbourhood typologies and associations with body mass index and obesity: A cross-sectional study. Prev Med. 2018;111:351-7. 20. Hobbs M, Griffiths C, Green M, Jordan H, Saunders J, McKenna J. Neighbourhood typologies and associations with body mass index and obesity: A cross-sectional study. Prev Med. 2018;111:351-7. 21. Vogel C, Zwolinsky S, Griffiths C, Hobbs M, Henderson E, Wilkins E. A Delphi study to build consensus on the definition and use of big data in obesity research. Int J Obes. 2019;43(12):2573-86. 22. WHO. Preventing drowning: an implementation guide. World Health Organization 2017. Available 22. WHO. Preventing drowning: an implementation guide. World Health Organization 2017. Available from: https://www.who.int/publications/i/item/preventing-drowning-an-implementation-guide. Accessed: 29/09/2020 22. WHO. Preventing drowning: an implementation guide. World Health Organization 2017. Available f htt // h i t/ bli ti /i/it / ti d i i l t ti id A d 29/09/2020 from: https://www.who.int/publications/i/item/preventing-drowning-an-implementation-guide. Accessed: 29/0 23. Pearson G. Why children die: a pilot study 2006, England (South West, North East and West midlands), Wales and Northern Ireland. London. 2008. Available from: https://www.publichealth.hscni.net/sites/default/files/Why%20Children%20Die%20-%20a%20pilot%20study%202006.pdf. Accessed: 09/09/2020 24. Morgan D, Ozanne‐Smith J, Triggs TJA. Self‐reported water and drowning risk exposure at surf beaches 24. Morgan D, Ozanne‐Smith J, Triggs TJA. Self‐reported water and drowning risk exposure at surf beaches. Aust NZ J Public Health. 2009;33(2):180-8. 25. WHO. International statistical cassification of diseases and related health problems, eleventh revision. World Health Organization 2018. Available from: https://icd.who.int/en. Accessed: 09/09/2020 25. WHO. International statistical cassification of diseases and related health problems, eleventh revision. World Health Organization 2018. Available from: https://icd.who.int/en. Accessed: 09/09/2020 26. Agerbo E, Gunnell D, Bonde JP, Mortensen PB, Nordentoft M. Suicide and occupation: the impact of socio-economic, demographic and psychiatric differences. Psychol Med. 2007;37(8):1131-40. 26. Agerbo E, Gunnell D, Bonde JP, Mortensen PB, Nordentoft M. Suicide and occupation: the impact of socio-economic, demographic and psychiatric differences. Psychol Med. 2007;37(8):1131-40. 27. Manders B, Kaur J. Suicides in the UK: 2017 registrations. 2018. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/suicidesintheunitedkingdom/2018registrations. References Stoop J. Maritime accident investigations. In: Bierens JJ, editor. Handbook on drowning: prevention, rescue, treatment Berlin Heidelberg, Germany: Springer 2006. Page 7/14 Page 7/14 11. Holliday NP, Kennedy J, Kent EC, Marsh R, Hughes SL, Sherwin T, et al. Marine Climate Change Impacts Partnership Scientific Review – Sea Temperature 2007-2008. Available from: http://www.mccip.org.uk/media/3371/sea-temperature.pdf. Accessed: 90/09/2020 11. Holliday NP, Kennedy J, Kent EC, Marsh R, Hughes SL, Sherwin T, et al. Marine Climate Change Impacts Partnership Scientific Review – Sea Temperature 2007-2008. Available from: http://www.mccip.org.uk/media/3371/sea-temperature.pdf. Accessed: 90/09/2020 12. Tipton M. The initial responses to cold-water immersion in man. Clin Sci. 1989;77(6):581-8. 12. Tipton M. The initial responses to cold-water immersion in man. Clin Sci. 1989;77(6):581-8. 12. Tipton M. The initial responses to cold-water immersion in man. Clin Sci. 1989;77(6):581-8. 13. Tipton MJ. Cold water immersion: sudden death and prolonged survival. Lancet. 2003;362:s12-s3. 13. Tipton MJ. Cold water immersion: sudden death and prolonged survival. Lancet. 2003;362:s12-s3. 14. Office H. Report of the working party on water safety. London, United Kingdom; 1977. 14. Office H. Report of the working party on water safety. London, United Kingdom; 1977. 14. Office H. Report of the working party on water safety. London, United Kingdom; 1977. 15. Franklin RC, Scarr JP, Pearn JH. Reducing drowning deaths: the continued challenge of immersion fatalities in Australia. Med J Aust. 2010;192(3):123-6. 15. Franklin RC, Scarr JP, Pearn JH. Reducing drowning deaths: the continued challenge of immersion fatalities in Australia. Med J Aust. 2010;192(3):123-6. 16. Mahony AJ, Peden AE, Franklin RC, Pearn JH, Scarr J. Fatal, unintentional drowning in older people: an assessment of the role of preexisting medical conditions. Healthy Aging Res. 2017;6:e7. 16. Mahony AJ, Peden AE, Franklin RC, Pearn JH, Scarr J. Fatal, unintentional drowning in older people: an assessment of the role of preexisting medical conditions. Healthy Aging Res. 2017;6:e7. 17. Hamilton K, Keech JJ, Peden AE, Hagger MS. Alcohol use, aquatic injury, and unintentional drowning: a systematic literature review. Drug and Alcohol Rev. 2018;37(6):752-73. 17. Hamilton K, Keech JJ, Peden AE, Hagger MS. Alcohol use, aquatic injury, and unintentional drowning: a systematic literature review. Drug and Alcohol Rev. 2018;37(6):752-73. 18. Quan L, Bennett EE, Branche CM. Interventions to prevent drowning. In: Doll LS, Bonzo SE, Mercy JA, Sleet DA, editors. Handbook of injury and violence prevention. New York: Springer; 2008. p. 81-96. 19. Hobbs M, Griffiths C, Green M, Christensen A, McKenna J. References Accessed: 09/09/2020 27. Manders B, Kaur J. Suicides in the UK: 2017 registrations. 2018. Available from: https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/suicidesintheunitedkingdom/2018registrations. Accessed: 09/09/2020 28. Tipton MJ, Eglin C, Gennser M, Golden F. Immersion deaths and deterioration in swimming performance 28. Tipton MJ, Eglin C, Gennser M, Golden F. Immersion deaths and deterioration in swimming performance in cold water. Lancet. 1999;354(9179):626-9. 29. Tipton MJ, Golden F. A proposed decision-making guide for the search, rescue and resuscitation of submersion (head under) victims based on expert opinion Resus 2011;82(7):819-24 28. Tipton MJ, Eglin C, Gennser M, Golden F. Immersion deaths and deterioration in swimming performance in cold water. Lancet. 1999;354(9179):626-9. 29 Tipton MJ Golden F A proposed decision making guide for the search rescue and resuscitation of submersion (head under) victims based on expert 28. Tipton MJ, Eglin C, Gennser M, Golden F. Immersion deaths and deterioration in swimming performance in cold water. Lancet. 1999;354(9179):626-9. ( ) 29. Tipton MJ, Golden F. A proposed decision-making guide for the search, rescue and resuscitation of submersion (head under) victims based on expert opinion. Resus. 2011;82(7):819-24. 29. Tipton MJ, Golden F. A proposed decision-making guide for the search, rescue and resuscitation of submersion (head under) victims based on expert opinion. Resus. 2011;82(7):819-24. 30. Fralick M, Denny CJ, Redelmeier DA. Drowning and the influence of hot weather. PloS one. 2013;8(8):e71689. 30. Fralick M, Denny CJ, Redelmeier DA. Drowning and the influence of hot weather. PloS one. 2013;8(8):e71689. 30. Fralick M, Denny CJ, Redelmeier DA. Drowning and the influence of hot weather. PloS one. 2013;8(8):e7 31. Schmidt A, Hawkins S, Quan L. Drowning is never dry. Expert Rev Respir Med. 2019;13(4):313-5. 32. Peden AE, Franklin RC, Leggat PA. Exploring visitation at rivers to understand drowning risk. Inj Prev. 20 33. Gallinger ZR, Fralick M, Hwang SW. Ethnic differences in drowning rates in Ontario, Canada. J Immigr Minor Health. 2015;17(5):1436-43. 34. Barwood M, Massey HC, Tipton M. “Float First”:: evidence-base for the Royal National Lifeboat Institution's “Respect the Water” campaign. Portsmouth, United Kingdom: University of Portsmouth; 2018. 35. Idris A, Bierens J, Perkins G, Wenzel V, Nadkarni V, Morley P. Revised Utstein-style recommended guidelines for uniform reporting of data from drowning-related resuscitation: an ILCOR advisory statement. Resus. 2015;118:147-58. Page 8/14 Page 8/14 Page 8/14 Tables Field number Field name Operational definition 1 Waidised ID Unique identifier number assigned to each case by the Royal Society for the Prevention of Accidents 2 Stakeholder reference Identifier number associated with the specific water safety agency entering the case 3 What happened Classification of how a person came into difficulty. Tables Alternatively, ‘body recovered’ can be entered 4 Date Date and time that an incident or body recovery occurred 5 Activity Classification of the action or type activity being undertaken by the victim prior to drowning 6 Postcode Postcode at which the incident or body recovery occurred 7 Latitude Latitude at which the incident or body recovery occurred 8 Longitude Longitude at which the incident or body recovery occurred 9 Ordinance Survey reference Ordinance Survey coordinates at which the incident or body recovery occurred 10 Location name Name of the location at which the incident or body recovery occurred 11 Location type Type of location in which the incident or body recovery occurred 12 Location feature Type of feature at or near which the incident or body recovery occurred 13 Wind Progressive numerical scale used to indicate the wind conditions at the time of an incident occurring 14 Visibility Visibility at the time of an incident occurring 15 Water depth Water depth in which the incident or body recovery occurred 16 Age (years) Age of the victim in years 17 Age category Age category to which the victim belonged 18 Injury Classification of the fate of a victim 19 Sex The sex of the victim 20 Coroner report Whether or not the outcome of a coroner’s report is currently pending 21 Narrative A free-text field allowing a brief description of the incident to be entered 22 Intoxication Whether or not intoxication by alcohol and/or drugs was suspected or confirmed Page 9/14 Page 9/14 Field number Field name Total missing cases Percentage missing cases (%) 1 Waidised ID 0 0.0 2 Stakeholder reference 14 0.5 3 What happened 24 0.5 4 Date 0 0.0 5 Activity 6 0.1 6 Postcode 544 10.8 7 Latitude 0 0.0 8 Longitude 0 0.0 9 Ordinance Survey reference 492 9.7 10 Location name 28 0.6 11 Location type 24 0.5 12 Location feature 41 0.8 13 Wind 4,606 91.2 14 Visibility 4,492 88.9 15 Water depth 4,917 97.3 16 Age (years) 734 14.5 17 Age category 637 12.6 18 Injury 0 0.0 19 Sex 308 6.1 20 Coroner report 0 0.0 21 Narrative 0 0.0 22 Intoxication 0 0.0 Total missing cases Percentage missing cases (%) Number of Water Incident Database (WAID) recorded drowning cases sub-divided by age category (n=5,051) Table 3. Table 5. Percentage agreement between auditor-entered dummy cases and true Water Incident Database (WAID) entries for each field taxonomy (n=50 ) Tables Number of Water Incident Database (WAID) recorded drowning cases sub-divided by age category (n=5,051) Age category 0 to 2 yrs 3 to 5 yrs 6 to 12 yrs 13 to 18 yrs 19 to 35 yrs 36 to 60 yrs Over 60 yrs Not reco-rded Total cases (cases p/y) 39 (13) 26 (9) 34 (5) 189 (32) 1,138 (67) 1,807 (72) 1,181(NA) 637 (NA) Values in brackets indicate drowning rate relative to the number of years (per year; p/y) in each WAID age category; NA - Not applicable. Table 3. Number of Water Incident Database (WAID) recorded drowning cases sub-divided by age categ Values in brackets indicate drowning rate relative to the number of years (per year; p/y) in each WAID age category; NA - Not applicable. Table 4. Number of Water Incident Database (WAID) recorded drowning cases sub-divided by sex and age category (n=5,051) Table 4. Number of Water Incident Database (WAID) recorded drowning cases sub-divided by sex and age category (n=5,051) Page 10/14 Page 10/14 Age category (category range) Sex Total cases (cases p/y) 0 to 2 years (3 years) Female 11 (4) Male 27 (9) Sex not recorded 1 (0) 3 to 5 years (3 years) Female 6 (2) Male 19 (6) Sex not recorded 1 (0) 6 to 12 years (7 years) Female 13 (2) Male 21 (3) Sex not recorded 0 (0) 19 to 35 years (17 years) Female 167 (10) Male 949 (56) Sex not recorded 22 (1) 36 to 60 years (25 years) Female 414 (17) Male 1,346 (54) Sex not recorded 47 (2) Over 60 years Female 301 (NA) Male 854 (NA) Sex not recorded 26 (NA) Age not recorded Female 83 (NA) Male 346 (NA) Sex not recorded 208 (NA) ( ) h l bl Values in brackets indicate drowning rate per year (p/y) in each WAID age category. NA - Not applicable Table 5. Percentage agreement between auditor-entered dummy cases and true Water Incident Database (WAID) entries for each field taxonomy (n=50 cases). Table 5. Tables Percentage agreement between auditor-entered dummy cases and true Water Incident Database (WAID) entries for each field taxonomy (n=50 cases) Page 11/14 Page 11/14 Field number Field name Percent agreement (%) 1 What happened 59 2 Date 69 3 Activity 71 4 Postcode 71 5 Latitude 76 6 Longitude 65 7 Ordinance Survey reference 55 8 Location name 87 9 Location type 82 10 Location feature 66 11 Wind 96 12 Visibility 91 13 Water depth 96 14 Age (years) 88 15 Age category 96 16 Injury 100 17 Sex 91 18 Coroner report 64 19 Outcome 80 20 Intoxication 94 Field number Field name Figures Page 12/14 Page 12/14 Figure 1 Schematic overview of WAID data entry sequence for each fatal and non-fatal case. Figure 2 Figure 1 Schematic overview of WAID data entry sequence for each fatal and non-fatal case. chematic overview of WAID data entry sequence for each fatal and non-fatal case. Figure 2 Figure 2 Examples of the sub-sets of the choices available in WAID for the fields of age category, sex, (suspected) intoxication and the mandatory field describing the suspected outcome. Examples of the sub-sets of the choices available in WAID for the fields of age category, sex, (suspected) intoxication and the mandatory field describing the suspected outcome Page 13/14 Figure 3 Number of Water Incident Database (WAID) recorded drowning cases sub-divided by drowning outcome (panel 3a.) and calendar year (panel 3b; n=5,051). Figure 3 Number of Water Incident Database (WAID) recorded drowning cases sub-divided by drowning outcome (panel 3a.) and calendar year (panel 3b; n=5,051). Number of Water Incident Database (WAID) recorded drowning cases sub-divided by drowning outcome (panel 3a.) and calendar year (panel 3b; n=5,051). Page 14/14 Page 14/14 Page 14/14
https://openalex.org/W3018455884	https://eprints.whiterose.ac.uk/170144/1/cmb.2020.0445.pdf	English	null	Reconstructing Genotypes in Private Genomic Databases from Genetic Risk Scores	Journal of computational biology	2,021	cc-by	13,625	Article: Article: Paige, Brooks, Bell, James, Bellet, Aurelien et al. (2 more authors) (2021) Reconstructing Genotypes in Private Genomic Databases from Genetic Risk Scores. Journal of Computational Biology. pp. 435-451. ISSN 1066-5277 Paige, Brooks, Bell, James, Bellet, Aurelien et al. (2 more authors) (2021) Reconstructing Genotypes in Private Genomic Databases from Genetic Risk Scores. Journal of Computational Biology. pp. 435-451. ISSN 1066-5277 https://doi.org/10.1089/cmb.2020.0445 https://doi.org/10.1089/cmb.2020.0445 Reuse This article is distributed under the terms of the Creative Commons Attribution (CC BY) licence. This licence allows you to distribute, remix, tweak, and build upon the work, even commercially, as long as you credit the authors for the original work. More information and the full terms of the licence here: https://creativecommons.org/licenses/ Version: Published Version Article: Paige, Brooks, Bell, James, Bellet, Aurelien et al. (2 more authors) (2021) Reconstructing Genotypes in Private Genomic Databases from Genetic Risk Scores. Journal of Computational Biology. pp. 435-451. ISSN 1066-5277 https://doi.org/10.1089/cmb.2020.0445 This is a repository copy of Reconstructing Genotypes in Private Genomic Databases from Genetic Risk Scores. White Rose Research Online URL for this paper: https://eprints.whiterose.ac.uk/170144/ Version: Published Version ABSTRACT Some organizations such as 23andMe and the UK Biobank have large genomic databases that they re-use for multiple different genome-wide association studies. Even research studies that compile smaller genomic databases often utilize these databases to investigate many related traits. It is common for the study to report a genetic risk score (GRS) model for each trait within the publication. Here, we show that under some circumstances, these GRS models can be used to recover the genetic variants of individuals in these genomic databases—a recon- struction attack. In particular, if two GRS models are trained by using a largely overlapping set of participants, it is often possible to determine the genotype for each of the individuals who were used to train one GRS model, but not the other. We demonstrate this theoretically and experimentally by analyzing the Cornell Dog Genome database. The accuracy of our recon- struction attack depends on how accurately we can estimate the rate of co-occurrence of pairs of single nucleotide polymorphisms within the private database, so if this aggregate information is ever released, it would drastically reduce the security of a private genomic database. Caution should be applied when using the same database for multiple analysis, especially when a small number of individuals are included or excluded from one part of the study. Keywords: genetic risk scores, genomic privacy, GWAS, long-term privacy, reconstruction attack. Reconstructing Genotypes in Private Genomic Databases from Genetic Risk Scores BROOKS PAIGE,1,2 JAMES BELL,1 AURE´LIEN BELLET,3 ADRIA` GASCO´ N,1,4 and DAPHNE EZER1,4,5 Takedown If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing eprints@whiterose.ac.uk including the URL of the record and the reason for the withdrawal request. eprints@whiterose.ac.uk https://eprints.whiterose.ac.uk/ eprints@whiterose.ac.uk https://eprints.whiterose.ac.uk/ eprints@whiterose.ac.uk https://eprints.whiterose.ac.uk/ RECOMB 2020 JOURNAL OF COMPUTATIONAL BIOLOGY Volume 28, Number 0, 2021 Mary Ann Liebert, Inc. Pp. 1–17 DOI: 10.1089/cmb.2020.0445 JOURNAL OF COMPUTATIONAL BIOLOGY Volume 28, Number 0, 2021 Mary Ann Liebert, Inc. Pp. 1–17 DOI: 10.1089/cmb.2020.0445 1The Alan Turing Institute, London, United Kingdom. 2Department of Computer Science, University College London, London, United Kingdom. 3Inria, Parc Scientiﬁque de la Haute Borne Park Plaza, Villeneuve d’Ascq, France. 4University of Warwick, Coventry, United Kingdom. 5Department of Biology, University of York, York, United Kingdom. Brooks Paige, et al., 2021. Published by Mary Ann Liebert, Inc. This Open Access article is distributed under the terms of the Creative Commons 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. Inria, Parc Scientiﬁque de la Haute Borne Park Plaza, Villeneuve d Ascq, France. 4University of Warwick, Coventry, United Kingdom. 5Department of Biology, University of York, York, United Kingdom. Brooks Paige, et al., 2021. Published by Mary Ann Liebert, Inc. This Open Access article is distributed under the terms of the Creative Commons 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. Brooks Paige, et al., 2021. Published by Mary Ann Liebert, Inc. This Open Access article is distributed under the terms of the Creative Commons 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. Keywords: genetic risk scores, genomic privacy, GWAS, long-term privacy, reconstructio 1The Alan Turing Institute, London, United Kingdom. 2Department of Computer Science, University College London, London, United Kingdom. 3Inria, Parc Scientiﬁque de la Haute Borne Park Plaza, Villeneuve d’Ascq, France. 4University of Warwick, Coventry, United Kingdom. 5Department of Biology, University of York, York, United Kingdom. Brooks Paige et al 2021 Published by Mary Ann Liebert Inc This Open Access article is distributed under th PAIGE ET AL. PAIGE ET AL. 2 research focus on long-term privacy of genomic databases rests on the longevity of the encryption scheme (Huang et al., 2015), it is also important to remember that these genomic databases are not just sitting on a server somewhere, but are also being continually utilized for making new scientiﬁc discoveries. Each time these databases are accessed and the scientiﬁc results are published, there is a risk that information will be leaked and that eventually this would enable an attacker to reconstruct private information held in the database. Genomic researchers are already aware that some forms of aggregate data from their databases should not be released publicly, because there is a risk that an attacker may be able to determine whether a particular individual is a member of the database (a membership inference attack). For instance, such attacks have already been developedforsummarystatisticsabout the frequencyofsingle nucleotidepolymorphisms(SNPs;Cai etal.,2015; Dwork et al., 2015; Simmons and Berger, 2015). Membership inference attacks have also been developed for the case where a person is allowed to repeatedly query a database to learn whether at least one individual contains a particular SNP(ShringarpureandBustamante,2015; Raisaroetal.,2017; vonThenenetal.,2018).Thesekindsof aggregate statistics about the frequency or presence/absence of a particular SNP in a database might be useful to release to the broader research community, but it is not an essential output of the research process. However, the main research ﬁndings—that is, the SNPs associated with the trait of interest and their strength of association—are essential to publish since the entire purpose of these genomic research projects is to uncover the relationship between genetic variants and phenotypic traits. Moreover, knowledge of these SNPs can lead to new diagnosis procedures or new potential drug targets, so their release is important for the public interest (Visscher et al., 2017). However, even this information can potentially leak private information about individuals in the database. For instance, Im et al. (2012) found that information about individuals in a genomic database is leaked when studies publish whether each SNP is correlated or anti- correlated to the trait of interest. It is important to quantify how much information is leaked by publishing these research ﬁndings, so that scientists can make informed decisions about when to publish their results and whether it is worth risking the privacy of the participants. PAIGE ET AL. g p y p p In this study, we demonstrate that the kind of research output that is published from genome-wide association studies (GWAS) has the potential to leak enough information to recover the SNPs of individuals in the database (a reconstruction attack), under speciﬁc circumstances. In particular, we focus on the release of genetic risk scores (GRS), a common research output for ﬁnding genetic associations with continuous traits (Qi et al., 2011; Belsky et al., 2013; Zhao et al., 2014; Chouraki et al., 2016; Day et al., 2017; Knowles and Ashley, 2018). We also focus on cases where a database is repeatedly used to perform a GWAS analysis, but not all the individuals are part of all the analyses. This could be the case because some individuals drop out of the study or skip speciﬁc survey questions. Alternatively, some databases, such as 23andMe, may grow in size over time and allow several GWAS to be performed within a short period. Under these circumstances, we demonstrate that it is possible to completely reconstruct the SNPs of an individual by using a custom expectation–maximization (EM) algorithm. We also provide suggestions for avoiding this kind of attack. To be clear, this article focuses on the simpler case in which the exact same trait is investigated in multiple GWAS studies; however, we expect that some version of this attack may be developed in the near future for the case of multiple highly correlated traits. 1. INTRODUCTION I n a survey of genomic privacy experts, the long-term privacy of genomic information was deemed both the most important and the most challenging problem to overcome (Mittos et al., 2019). If an individual’s password or ID number gets leaked, it is always possible to change it. However, it is impossible for a person to change their genetic code and they will pass part of it onto their children, so any information leaks can have long-term impacts on both the individual and their descendants. Although much of the I 1 2 1.1. Overview of scenarios that will be investigated We demonstrate a series of reconstruction attacks that enable us to infer the genotypes of individuals in private genomic databases, based on publicly released GRS. These attacks will initially be deployed in a very favorable scenario, but the scope of the attack will be subsequently expanded, building up to the scenario shown in Figure 1. It is worth noting that the reconstruction attacks that we will describe do not depend on (1) how the SNPs were initially ﬁltered or (2) how strongly they associate with the trait of interest. We will begin by investigating a simple scenario: Two GWAS studies are performed to identify SNPs associated with the same trait, and the two studies use the same set of participants, except that the second study includes one extra individual. In addition, we will assume that we know the frequencies of each SNP and the frequencies that pairs of SNPs co-occur in the same individual. We assume that both studies publish the coefﬁcients associated with the GRS models that they infer as part of the analysis. Next, we will consider the case in which the second study includes more than one additional participant and we demonstrate that in many circumstances this still allows us to easily reconstruct the individual genotypes of all the individuals that are found in the second study but not the ﬁrst (see Section 3.2). RECONSTRUCTION ATTACKS FROM GRS RECONSTRUCTION ATTACKS FROM GRS 3 FIG. 1. We investigate the case where two GWAS studies are performed on two datasets that mostly contain the same individuals. We reconstruct the genotype of those individuals added to the second study, using the GRS from each study and an estimate of SNP frequencies. GRS, genetic risk score; GWAS, genome-wide association studies; SNP, single nucleotide polymorphism. FIG. 1. We investigate the case where two GWAS studies are performed on two datasets that mostly contain the same individuals. We reconstruct the genotype of those individuals added to the second study, using the GRS from each study and an estimate of SNP frequencies. GRS, genetic risk score; GWAS, genome-wide association studies; SNP, single nucleotide polymorphism. Afterward, we will demonstrate that we do not need to know the precise frequencies of SNPs and frequencies of co-occurring SNPs, as long as we have a reasonable estimate of these values from public databases (see Section 3.3). 1.1. Overview of scenarios that will be investigated We also brieﬂy discuss how loosening additional restrictions would impact our ability to predict individual genotypes. In particular, we analyze the case where the two sets of SNPs that are used by the two studies are not identical. These results imply that if two sets of GRS are released on two genetic datasets with largely overlapping populations, it may be possible to reconstruct the genotypes of those individuals who participated in one study but not the other (Fig. 1). PAIGE ET AL. 4 where we have deﬁned the symmetric (N + 1) · (N + 1) matrix K as K = 1 M F> MFM: (4) K = 1 M F> MFM: (4) Now, suppose a second study is run, targeting the same phenotype, which adds a single extra individual with SNPs represented by the N length vector x0. This corresponds to adding the row /> 0 = [ x> 0 1 ] to the design matrix, and extending y with the additional phenotypic value y0 for the new individual. The updated estimator (i.e., the GRS values for the second study) is given by ^bM + 1 = (F> MFM + /0/> 0 ) - 1(F> MyM + y0/0): (5) (5) We assume that both GRS models ^bM and ^bM + 1 are released publicly. An attacker aims at using this knowledge to reconstruct /0 (the genotype of the added individual). Through algebraic re-arrangement (see Section 5.2), we ﬁnd that: /0 = 1 C K(^bM + 1 - ^bM) (6) (6) where C is a scalar, speciﬁcally C = 1 M (y0 - /> 0 ^bM + 1). Eq. (6) means that /0 is a scalar multiple of K(^bM + 1 - ^bM). proach, thus, centers on the use of the vector that we deﬁne as d1, Our approach, thus, centers on the use of the vector that we deﬁne as d1, d1 = D K(^bM + 1 - ^bM) = C/0‚ (7) (7) corresponding to a rescaled copy of the input SNP data in the design matrix /0, which can be easily computed from the two parameter vectors if the matrix K is known. As seen in Section 3.1, we can use d1 to exactly reconstruct the added individual with 100% accuracy. corresponding to a rescaled copy of the input SNP data in the design matrix /0, which can be easily computed from the two parameter vectors if the matrix K is known. As seen in Section 3.1, we can use d1 to exactly reconstruct the added individual with 100% accuracy. We additionally consider the case where m additional individuals have been included in the second study, yielding a new GRS model ^bM + m including these M + m participants. PAIGE ET AL. The extra rows of the design matrix now form a matrix Fm of size m · (N + 1), where each row is an individual that was added to the second study and each column is an SNP (and the last column contains only 1). The corresponding analog to Eq. (7) for multiple individuals, which we derive in Section 5.2, is dm = D K(^bM + m - ^bM) = F> mCm‚ (8) (8) where Cm is a vector of length m. For sufﬁciently small m (relative to N), exact reconstruction of all m added individual genomes is also possible in this setting, following the algorithm we will introduce in Section 3.2. where Cm is a vector of length m. For sufﬁciently small m (relative to N), exact reconstruction of all m added individual genomes is also possible in this setting, following the algorithm we will introduce in Section 3.2. The previous examples have focused on cases in which the participants in the ﬁrst study are a subset of the individuals in the second study. In Section 5.7, we consider the case in which the ﬁrst study has some participants that are not found in the second study and vice versa. We show that the same strategies for reconstructing the genome can be used as in the previous scenario that we discussed, in which multiple participants are added to the second study. 2. METHODS GRS models describe the relationship between a particular phenotype of interest and particular SNPs. These models are ﬁt in a two-stage process: First, a reduced set of SNPs is selected from a potentially very large pool of candidates; then, this reduced set is used as the independent variables in a linear regression analysis. The set of SNPs is selected by ﬁrst ﬁltering for those that signiﬁcantly correlate to the trait of interest, after controlling for other covariates. These SNPs are then further ﬁltered to ensure that they are far apart from one another, to decrease the correlation between them. In this setting, we suppose that M individuals have taken part in a study, and N SNPs have passed the ﬁltration steps to be used in a linear model. Let yM be the vector of M real-valued phenotypes, and XM be an M · N binary matrix, where XM[i‚ j] = 1 if individual i has SNP j. To include an intercept term in the linear model, we deﬁne the design matrix FM to be the M · (N + 1) matrix FM = XM 1M ½ : (1) (1) The GRS model parameter bM is just the coefﬁcient vector of the linear model yM = FMbM + e‚ (2) yM = FMbM + e‚ (2) where e is independent Gaussian noise. Given FM and phenotypes yM, the maximum likelihood estimate of this parameter has a closed form where e is independent Gaussian noise. Given FM and phenotypes yM, the maximum likelihood estimate of this parameter has a closed form ^bM = D 1 M K - 1(F> MyM)‚ (3) (3) 4 – For i = 1‚ . . . ‚ N: Kii estimates the probability that SNP i has value 1 (i.e., the frequency of the SNP in the population). – For i = 1‚ . . . ‚ N - 1 and j > i: Kij = Kji estimates the probability that both SNP i and SNP j are 1 simultaneously (i.e., the frequency of SNP i and SNP j co-occurring in the same individual). – For i = 1‚ . . . ‚ N and j = N + 1: Kij = Kji also estimates the probability that SNP i has value 1, that is, Ki‚ N + 1 = KN + 1‚ i = Kii. – Finally, KN + 1‚ N + 1 = 1. or i = 1‚ . . . ‚ N and j = N + 1: Kij = Kji also estimates the probability that SNP i has value 1, that i + 1 = KN + 1‚ i = Kii. 3.1. Complete reconstruction of one individual’s genotype when SNP frequency information is known 3.1. Complete reconstruction of one individual’s genotype when SNP frequency inform is known The ﬁrst, most straightforward case is when only one participant is added between the ﬁrst and second studies, that is, where ^bM is the GRS for the ﬁrst study (containing M participants), and ^bM + 1 is the GRS for the second study as described in Eqs. (3) and (5). Both of these are vectors of length N + 1, where the ﬁrst N indices correspond to the relationship between each SNP and the trait and the last element is the intercept of the linear model. For now, we also assume we are in the setting where the matrix K is known, for example, because the SNP frequency information has been publicly released. ^ ^ Given K, ^bM + 1, and ^bM, we can use d1 (a vector of length N + 1) to precisely determine the genotype of the individual who was added to the database. For each i = 1‚ . . . ‚ N, the ith entry of d1 is either equal to 0 if /0 contains a 0 (i.e., the individual does not have the SNP at that index) or to C if /0 contains a 1 (i.e., the individual has the SNP at that index). In other words, it is possible to exactly read off the SNPs of the added individual in this setting. Indeed, we tested this strategy on the Cornell Dog Database and found that we were able to reconstruct the genotype of the dog that was added to the second study with 100% accuracy, on both common and uncommon SNPs (Fig. 2A). A B C FIG. 2. (A) We have perfect accuracy in reconstructing the genotype when K is known (using 200 random SNPs to estimate average breed weight in the Cornell Dog Database). (B) We can reconstruct all the genotypes of multiple dogs that are added to the second study and (C) this works in practice by using the data from the Cornell Dog Database, as in (A). B C A C C B A FIG. 2. (A) We have perfect accuracy in reconstructing the genotype when K is known (using 200 random SNPs to estimate average breed weight in the Cornell Dog Database). RECONSTRUCTION ATTACKS FROM GRS 5 Thus, knowledge of SNP frequencies and pairwise co-frequencies from the original study are all that is required to compute K. In the following Sections 3.1 and 3.2, we consider adding one and multiple individuals at once, respectively, in the setting where this matrix K can be estimated exactly. However, although ^bM, ^bM + 1 and M are likely to be published along with the study, an attacker would often need to estimate K from other publicly available data. Most studies will report some information about the study population (such as whether the study focused on individuals from a speciﬁc continent), which can help with estimating K. From this information, we can estimate the value of K in similar populations as those used in the study using publicly available data, for example, from the HapMap project. Our additional experiments in Section 3.3 use a custom EM algorithm to ﬁnd maximum likelihood estimates of /0 when the matrix ^K K is estimated from independent public data. The derivation of this EM algorithm is given in Section 5.4.3, and a formal analysis of the reconstruction error of /0 given the error in ^K is found in Section 5.2.1. 2.1. Estimation of K As it turns out, the entries of matrix K correspond to simple population-level statistics of the SNPs, which could either be inadvertently released (under the assumption they would be safe to share), or could be estimated from another sample from the same population. In fact, the entries of K depend only on the SNP frequencies and SNP co-occurrence frequencies in the dataset: – For i = 1‚ . . . ‚ N: Kii estimates the probability that SNP i has value 1 (i.e., the frequency of the SNP in the population). – For i = 1‚ . . . ‚ N - 1 and j > i: Kij = Kji estimates the probability that both SNP i and SNP j are 1 simultaneously (i.e., the frequency of SNP i and SNP j co-occurring in the same individual). – For i = 1‚ . . . ‚ N and j = N + 1: Kij = Kji also estimates the probability that SNP i has value 1, that is, Ki N + 1 = KN + 1 i = Kii. – Finally, KN + 1‚ N + 1 = 1. 3. RESULTS The key observation from the previous section is that the vectors d1 and dm, derived from the change in parameter vectors ^b from a ﬁrst study to a second study, take only a ﬁnite number of values thanks to the fact that the design matrices F contain only zeros and ones. In particular, when m new individuals are added to the second study, each entry of the vector dm can only take at most 2m values, and a zero value corresponds to the setting where all individuals have the most common variant for that SNP. This section describes algorithmically how these vectors can be used to recover the genomes of the additional individuals, as well as empirical tests that use the Cornell Dog Genome dataset as a case study (Hayward et al., 2016). More details on the experimental setup can be found in Section 5.1. PAIGE ET AL. 6 3.2. Complete reconstruction of multiple individuals’ genotype when SNP frequency information is known We now consider the case where m additional individuals have been included in the second study, yielding a new GRS model ^bM + m, including these M + m participants. Consider again Eq. (8) described earlier. The ith row of Fm is a binary vector that represents the com- bination of the m individuals who have SNP i. This means that, for a ﬁxed value of Cm, the value of the vector dm at index i is uniquely determined by the combination of individuals who have SNP i (Fig. 2B). In other words, there will be at most 2m unique values taken by entries of dm, each corresponding to a combination of the values in vector Cm (Fig. 2C). If we were to learn which values of dm are also found in Cm, then we could infer the complete genotypes of all the m individuals added to the second study. We would be able to reconstruct m complete genotype vectors, although it would be impossible to know which of the genotypes corresponded to which of the m individuals. In fact, in many cases it is extremely straightforward to determine which values in dm cor- respond to values in Cm. Here, we describe a simple algorithm for ﬁnding Cm when there are exactly 2m unique values in dm. If this is not the case, please refer to the more complete algorithm in Section 5.3. 1. First, extract all unique, nonzero values from dm. Find the sum of all pairs of values in this vector p 3. Find all values that are in Eq. (1), but not in Eq. (2). The values of Cm appear in this list. There is no way to know which value of Cm corresponds to which index, so for simplicity we can randomly assign them indices. 3. Find all values that are in Eq. (1), but not in Eq. (2). The values of Cm appear in this list. There is no way to know which value of Cm corresponds to which index, so for simplicity we can randomly assign them indices. 4. Each value in Cm corresponds to a speciﬁc individual who was added to the second study. Each value in dm can be described as a sum of a unique combination of values in Cm. PAIGE ET AL. For instance, if dm[i] = Cm[j] + Cm[k], this means that the SNP at position i is found in individual j and k, but no one else. 4. Each value in Cm corresponds to a speciﬁc individual who was added to the second study. Each value in dm can be described as a sum of a unique combination of values in Cm. For instance, if dm[i] = Cm[j] + Cm[k], this means that the SNP at position i is found in individual j and k, but no one else. We tested this approach by using the Cornell Dog Database, in a test scenario where the second study added three different dogs. We were able to uniquely identify the genotypes of all three dogs with 100% accuracy, with both common and uncommon SNPs (Fig. 2C). 3.1. Complete reconstruction of one individual’s genotype when SNP frequency information is known (B) We can reconstruct all the genotypes of multiple dogs that are added to the second study and (C) this works in practice by using the data from the Cornell Dog Database, as in (A). PAIGE ET AL. 3.3. Accurate estimation of an individual’s genotype when SNP frequency information is estimated from a public database Previously, we assumed that the attacker had access to the matrix K, which consists of population-level statistics on frequencies and co-occurrence frequencies of SNPs. Although this could be released volun- tarily by organizations that are not aware of the risk, we now consider the case where K is not directly available to the attacker but is instead estimated from a separate public database assumed to correspond to individuals from the same population. We simulated this scenario by using the Cornell Dog Database by taking one random set of dogs for building the GRS model, and a second non-overlapping set of dogs for estimating ^K. We compared the value of ^d1 = ^K(^bM + 1 - ^bM) with the known value of /0. We observe that ^d1 has signiﬁcantly different values at indices where /0[i] = 0 and /0[i] = 1; examples for the cases where one and three dogs are added can be seen in Figure 3. ^ Themainchallengeisthatthevector ^d1 nowincludesadditionalnoise,sowecannotsimplyuseitsentryatindex N + 1 to estimate C, nor do the entries i with /0[i] = 0 also correspond directly to ^d1[i] = 0. Instead, we develop a custom EM algorithm to ﬁnd a maximum likelihood estimate of the constant C and recover /0, that is, to determine the probability that each /0[i] = 0 or /0[i] = 1, based on the value of ^d1 (see Section 5.4.3 for details). We ﬁnd that this method can successfully reconstruct the correct value of /0[i] much better than a baseline that uses the public dataset to independently estimate the most common variant for each of the SNP (Fig. 4). Crucially, we show that our approach is able to reconstruct, with relatively high accuracy, the genotypes of dogs even when they differ signiﬁcantly from those in the public dataset (Fig. 5). This shows that our attack is able to extract information about the particular individuals that differ across the two studies, not merely about the general population as in the most-common-variant baseline. By deﬁnition, dogs that have genotypes that differ signiﬁcantly from the general population have a higher proportion of uncommon SNPs, and the ability to recover these uncommon SNPs is particularly important from a privacy per- spective. Indeed, uncommon SNPs can be used to identify a particular individual and are also more likely to be associated with disease phenotypes, which is sensitive information. In general, we ﬁnd that the larger the FIG. 3. 4. DISCUSSION In this study, we demonstrate that private information is leaked when GRS models are published, speciﬁcally in the case where two sets of largely overlapping individuals are used for multiple studies. In particular, we show that we can recover SNPs from an individual in a private database—a reconstruction attack. Even though we would not have a name associated with this genotype, it may be possible to identify the individual once the genotypic data are available to the attacker. For instance, the attacker may have FIG. 4. Accuracy at reconstruction of genomes x0 using EM estimation and a noisy estimate ^K, as compared with a natural baseline that always predicts the most common variant at each SNP locus. We use this as a baseline, because without any additional information about bM and bM + 1, the most accurate prediction of the dog’s genotype would be to predict the most common variant at each locus. Here, we deﬁne accuracy as the proportion of SNPs that are correctly identiﬁed in the dog that was found in the second GWAS study, but not the ﬁrst. Each distribution is constructed from 500 experimental test points, in which we (1) took 10 random splits of the full dog dataset, assigning dogs to either the public or private dataset; (2) for each split, we tested the reconstruction 50 times, each time adding a different randomly sampled dog to the second GWAS study. The private dataset always has 1000 individuals; the public test dataset is of increasing size, improving performance. EM, expectation–maximization. FIG. 4. Accuracy at reconstruction of genomes x0 using EM estimation and a noisy estimate ^K, as compared with a natural baseline that always predicts the most common variant at each SNP locus. We use this as a baseline, because without any additional information about bM and bM + 1, the most accurate prediction of the dog’s genotype would be to predict the most common variant at each locus. Here, we deﬁne accuracy as the proportion of SNPs that are correctly identiﬁed in the dog that was found in the second GWAS study, but not the ﬁrst. 3.4. Accurate estimation of an individuals’ genotype when different SNPs are used in each study When GRS models are constructed, the ﬁrst step is to ﬁlter the set of SNPs down to a small set of SNPs that are (1) signiﬁcantly correlated to the trait after covariates are considered and (2) far apart from one another along the genome. If the two studies use two different sets of SNPs to construct the GRS model, it is still possible to recover whether or not each of the SNPs in the overlap is present in the new individual. This process is highly analogous to the previous cases and is detailed in Section 5.6. 3.3. Accurate estimation of an individual’s genotype when SNP frequency information is estimated from a public database Example values taken by the noisy vector ^d, given the true value of the corresponding SNP in the genome. (Left) adding one new participant; (right) adding three new participants. These ﬁgures are analogous to those in Figure 2, although in the case where K is not known and instead estimated from an independent public database. RECONSTRUCTION ATTACKS FROM GRS 7 RECONSTRUCTION ATTACKS FROM GRS 7 FIG. 3. Example values taken by the noisy vector ^d, given the true value of the corresponding SNP in the genome. (Left) adding one new participant; (right) adding three new participants. These ﬁgures are analogous to those in Figure 2, although in the case where K is not known and instead estimated from an independent public database. public dataset available, and the more similar the dataset is to the unknown private dataset, the better we are able to reconstruct the genome of the added individual. Full details and description of the experimental setting are given in Section 5.1. We also derive theoretical error bounds for our estimate of /0 based on the error in ^K in Section 5.4.1. public dataset available, and the more similar the dataset is to the unknown private dataset, the better we are able to reconstruct the genome of the added individual. Full details and description of the experimental setting are given in Section 5.1. We also derive theoretical error bounds for our estimate of /0 based on the error in ^K in Section 5.4.1. This task becomes more challenging when multiple individuals are added simultaneously and K is unknown; an algorithm for estimating Fm for m > 1, along with additional empirical results, is given in Section 5.4. 4. DISCUSSION Each distribution is constructed from 500 experimental test points, in which we (1) took 10 random splits of the full dog dataset, assigning dogs to either the public or private dataset; (2) for each split, we tested the reconstruction 50 times, each time adding a different randomly sampled dog to the second GWAS study. The private dataset always has 1000 individuals; the public test dataset is of increasing size, improving performance. EM, expectation–maximization. 8 PAIGE ET AL. FIG. 5. Results of Figure 4 broken down by individual dogs. Here, each point represents a dog and we deﬁne atypicality as the proportion of uncommon variants that the dog has compared with the public database—for instance, if 51% or more of dogs in the public database have a G in a speciﬁc locus, but this dog has a T, then this would count toward the dog’s atypicality. In other words, dogs further to the right are less and less similar to average dogs present in the public dataset (measured by percentage of different variants). In contrast to the most-common-variant baseline, our method generalizes well even to dogs that are highly dissimilar to those in the public dataset. Larger public databases (right) provide more accurate population estimates of ^K, leading to more accurate reconstructions overall. FIG. 5. Results of Figure 4 broken down by individual dogs. Here, each point represents a dog and we deﬁne atypicality as the proportion of uncommon variants that the dog has compared with the public database—for instance, if 51% or more of dogs in the public database have a G in a speciﬁc locus, but this dog has a T, then this would count toward the dog’s atypicality. In other words, dogs further to the right are less and less similar to average dogs present in the public dataset (measured by percentage of different variants). In contrast to the most-common-variant baseline, our method generalizes well even to dogs that are highly dissimilar to those in the public dataset. Larger public databases (right) provide more accurate population estimates of ^K, leading to more accurate reconstructions overall. access to partial genotypic information of the individual and then be able to identify them. Alternatively, they could use the genotype information to predict ethnicity and other phenotypic traits that could then be used to uniquely identify the individual. 4.1. Suggestions for good practice We provide a number of simple suggestions for good practice that would help limit this attack. We provide a number of simple suggestions for good practice that would help limit this attack. 1. Aggregate statistics about the frequency of SNPs in the database or the frequency of co-occurrence of SNPs should never be released. We have shown that this information, combined with GRS, allows to precisely reconstruct individual genomes in various settings. It may be possible to release noisy versions of SNP frequency data, but this would be equivalent to releasing ^K (our estimated K from the public database). With our EM algorithm, we have demonstrated that it is still possible to do some genotypic reconstruction with a noisy ^K, but this becomes harder as the noise in ^K increases. On the other hand, providing a very noisy ^K may be of limited utility to the scientiﬁc community. 2. If a genetic dataset is intended to serve for multiple complementary analyses, it is important that all study participants are used in every analysis performed. If there are missing phenotypic data from a few individuals, they should not be included in any of the analyses performed, or their privacy may be compromised. 3. When multiple individuals are added in between two studies, then the ability to reconstruct the genomes dependsonthenumberofSNPs,beinglargerelativetothenumberofindividuals.Inparticular,ifmnewdogs are added, exact reconstruction is only possible by using the approach in Section 3.2 if the number of SNPs N> 2m. Thus, we suggest to avoid releasing multiple studies that differ by fewer than log2 N individuals. 4. DISCUSSION We also note that even an incomplete reconstruction attack (in which only a proportion of the SNPs are correctly identiﬁed) is likely to be sufﬁcient to perform a membership inference attack. Investigating the relationship between the reconstruction attack and the membership inference attack will be a subject of future research. Importantly, if the attackers were unable to link the genomic data with a particular individual, the reconstruction attack would still be a breach in privacy that could have serious conse- quences. For instance, the patient may have only consented to have their genomic data used in particular kinds of research studies, whereas the attacker may use the reconstructed genomic data for a different (potentially unethical) purpose. 4.2. Extensions and future work Although we have analyzed the case where the genome is represented by binary values of 0 or 1, often studies instead count the number of times each allele is present, which would lead to a design matrix F containing values 0‚ 1‚ or 2. In this scenario, K no longer contains the frequencies of SNPs and their co- occurrences, but it is something slightly more complicated that we describe in Section 5.8. This does not dramatically change the approach in this study, except in that the vector dm can take 3m possible values, 9 RECONSTRUCTION ATTACKS FROM GRS rather than 2m. In practice, then, studies that use allele counts are somewhat more robust to attacks; the multiple dog reconstruction attack would likely be ambiguous if 3m > N, rather than 2m > N. p g y g A possible countermeasure to our reconstruction attack could consist of randomly perturbing the GRS models before releasing them, as done in differentially private linear regression (Wang, 2018). However, a naive application of this strategy could destroy the utility of the models. A formal and empirical analysis of the effectiveness of such protection against reconstruction attacks, as well as of the usefulness of the resulting GRS models to genomic researchers, is beyond the scope of this article and left for future work. resulting GRS models to genomic researchers, is beyond the scope of this article and left for future work. Another countermeasure is to refrain from releasing precise information about the population structure of the study population to prevent the attacker from estimating K effectively. This would, however, limit the utility of the research study, because the researchers would not know to what populations the research applies. Our work has a number of limitations. For instance, we only test our EM algorithm on dog data. Dog populations may have different population structures than human populations due to selective breeding Another countermeasure is to refrain from releasing precise information about the population structure of the study population to prevent the attacker from estimating K effectively. This would, however, limit the utility of the research study, because the researchers would not know to what populations the research applies. Our work has a number of limitations. For instance, we only test our EM algorithm on dog data. 5.1. Experimental details 5.1.1. Cornell dog database. To experimentally test the reconstruction attacks, we used data from the Cornell Dog Genome Database, which contains data about SNPs from a wide range of dog breeds and a number of associated phenotypic traits. The two traits we focused on were average breed weight and average breed height, because these two phenotypes had the fewest number of missing values. For the initial investigation, we binarized the genotype matrix—considering all heterogenous alleles to have a value of 1. (We also repeated the analysis with the original genotype matrix.) Only common SNPs (i.e., SNPs that were found in 25%–75% of the dogs) were used, leaving 23,497 SNPs. For each linear model built, M = 1000 dogs were randomly sampled as the ‘‘private’’ dataset and N = 200 SNPs were randomly selected. To ensure that the SNPs that were sampled were spatially distributed, the SNPs were randomly sampled in a stratiﬁed way, so one SNP was selected in every 23‚ 497 200 -sized bin. 5.1.2. Experiment with imprecise K. First, two linear models were constructed to predict average breed weights: one with the M = 1000 randomly sampled dogs and another that contained one additional randomly sampled dog. This gives ^bM and ^bM + 1. To mimic the process of estimating K from a public database, we randomly sampled an additional 200, 400, or 800 dogs that were not included as part of the original set and used this to estimate K, which we denote by ^K. Now, we could calculate ^K(^bM + 1 - ^bM) and compare this with the known /0 for the additional dog from the second study. These additional dogs are taken from a third ‘‘test’’ dataset, disjoint from both the public and private data. The plots in Figures 4 and 5 are produced by re-running the algorithms across 10 random public/private/test splits, where the ‘‘test’’ dataset has 50 dogs that are each individually considered as candidates for the (M + 1)th dog added to the private dataset. 4.2. Extensions and future work Dog populations may have different population structures than human populations due to selective breeding, so in the future we aim at investigating how properties of population structure will impact our ability to estimate K and the accuracy of our reconstruction attack. It may seem unlikely on the surface that two GWAS analyses will include nearly the same participants. One potentially common setting where this could arise is when a single study collects both genotype and phenotype data from a single set of participants, and it releases multiple models to predict multiple traits. In this case, there may be a small number of individuals who are used in one analysis, but not the other; for instance, there may be a small subset of participants who skip a particular survey question that was used to collect phenotype information, and this is, indeed, evident in a recent study (Jiang et al., 2019). In such settings, it could be very possible for multiple released GRS models to be computed on sets of individuals that differ by only a few participants. In future work, we aim at extending our analysis and attack to settings where multiple GRS models are released, each predicting different but highly correlated traits. 5.3. Algorithm for identifying unique genotypes of multiple dogs when K is known Although the simple approach described in the main article will work in many cases, there are a few special circumstances where a more complex algorithm may be required. In particular, it would not work if there are combinations of SNPs that are not observed among the individuals added to the database. For instance, if there is not an SNP location where the ﬁrst individual has an SNP variant and the others do not, then we would miss the corresponding value in Cm. However, it is still possible to identify all the values in Cm through a more complex algorithm: 1. First, extract all unique, nonzero values from dm. 2. Find the sum of all pairs of values in Eq. (1). 3. Find all values that are in Eq. (1), but not in Eq. (2). 4. If there are exactly m values in Eq. (3) and the sum of these values equal the last value of dm (corresponding to the intercept term), then we have found the correct values of Cm. p g p 5. Otherwise, this suggests that there are one or more elements of Cm that are missing from Eq. (3) and possibly a few values in Eq. (3) that are not in Cm. 5. Otherwise, this suggests that there are one or more elements of Cm that are missing from Eq. (3) and possibly a few values in Eq. (3) that are not in Cm. 6. Begin by subtracting every pair of values in Eq. (3). These are now also potential values of C 7 S h f f l f E (3) d (6) h h l l f d Th 7. Search for a set of m values from Eqs. (3) and (6) that sum to the last element of dm. There may be more than one set of values for which this is true. 8. If this search is unsuccessful, repeat steps 6–7. Eventually, a set of m values summing to dm should be found. 8. If this search is unsuccessful, repeat steps 6–7. Eventually, a set of m values summing to dm should be found. 9. If more than one possible set of values is found for Cm in Eq. (7), it is still possible to compare these sets and identify which is the most likely to contain the true values of Cm. 5.3. Algorithm for identifying unique genotypes of multiple dogs when K is known For each possible Cm vector, a set of genotypes can be constructed for the m additional individuals. Using the frequencies of each SNP, it is possible to calculate the probability of observing each genotype. The set of values that produces the most likely genotypes for the m individuals is most likely to be the correct one. In addition, this algorithm depends on the fact that it is extremely unlikely that if someone were to sample three random continuous numbers i, j, and k, it would just so happen that i + j = k. There is an extremely small chance that a value of Cm would be un-discoverable because of a coincidence of this nature. 5.2. Adding multiple dogs Here, we explain Eqs. (7) and (8). Note that the former is a special case of the latter so we will only explain the latter in detail. First note that, by deﬁnition, ^bM = (F> MFM) - 1F> MyM = (MKM) - 1FT MyM‚ ^bM + m = (F> M + mFM + m) - 1F> M + myM + m = (MKM + mKm) - 1F> M + myM + m: ^bM = (F> MFM) - 1F> MyM = (MKM) - 1FT MyM‚ 10 PAIGE ET AL. Substituting these into the left hand side of the following equation gives the right hand side: Substituting these into the left hand side of the following equation gives the right hand side: (MKM + mKm)^bM + m - MKM ^bM = FT mym: (9) (9) (MKM + mKm)^bM + m - MKM ^bM = FT mym: (9) (MKM + mKm)^bM + m - MKM ^bM = FT mym: This equation can be rearranged to give KM(^bM + m - ^bM) = 1 M FT mym - m M Km ^bM + m = 1 M FT m(ym - Fm ^bM + m): Deﬁning the length m vector Cm = 1 M (ym - Fm ^bM + m) yields the form used in Eq. (8). For the special case of m = 1, Cm is a scalar and we recover Eq. (7). 5.3. Algorithm for identifying unique genotypes of multiple dogs when K is known RECONSTRUCTION ATTACKS FROM GRS 11 5.4.1. Analytic bound on k /0 - ^/0 k. For convenience, we only consider the case of adding a single individual, though the generalization is quite straightforward. If ^K is substituted for K in our reconstruction Eq. (7), we get an approximation of /0 that we denote ^/0. We would like to bound the (relative) error between /0 and ^/0. Later, we ignore the constant factors C and ^C for simplicity, noting that these scaling factors are estimated from the resulting /0 or ^/0 anyway. We, thus, consider u0 = K(^bM + 1 - ^bM) and ^u0 = ^K(^bM + 1 - ^bM). Using k k on vectors, and also on matrices, we denote the corresponding operator norm. The relative error between u0 and ^u0 is given by: k u0 - ^u0 k k u0 k = k ( ^K ^K - 1 - ^KK - 1)u0 k k u0 k k ^K ^K - 1 - ^KK - 1 k = k ^K( ^K - 1 - K - 1) k : Note that ^K - 1 - K - 1 = ^K - 1(K - ^K)K - 1 and hence k ^u0 - u0 k k u0 k k K - 1 kk K - ^K k ‚ (10) (10) This means that we can bound the error by two quantities. The term k K - 1 k is bounded earlier by 1= min (eig(K)), which is ﬁnite as soon as K is nonsingular. This is not a strong requirement, as in the case of linear regression it is required for ^bM + 1 and ^bM to exist. Note that in the case of L2-regularized linear regression (i.e., ridge regression), K is replaced by K + kI, where k is the regularization parameter, and we can directly bound this term by k. The key term in Eq. (10) is k K - ^K k, the error in estimating K by ^K. Let us assume that the public database used to obtain ^K follows the same distribution as the private database used to ﬁt the GRS models. Denote by ^M the number of individuals used to estimate ^K. Then, under classic boundedness assumptions and leveraging matrix concentration inequalities such as matrix Bernstein Tropp (2015), we can show that E[ k K - ^K k ] = O(1= ﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃﬃ min ( ^M‚ M) p ). RECONSTRUCTION ATTACKS FROM GRS This shows that the error in estimating K is small as long as the private and public databases are large enough. 5.4.2. Modeling the error in ^K. In this section, we deﬁne a model to capture the error in ^K, which leads to the EM algorithm for estimating /0, which is used in the experiments. As our estimated ^K drifts from the true K, this expression ^K(^bM + 1 - ^bM) would produce a wider range of values than just 0 and C. Let eijN (0‚ r2) be independent noise, which we assume corrupts each element of Kij; that is, given the estimated matrix ^K suppose 5.4.2. Modeling the error in ^K. In this section, we deﬁne a model to capture the error in ^K, which leads to the EM algorithm for estimating /0, which is used in the experiments. As our estimated ^K drifts from the true K, this expression ^K(^bM + 1 - ^bM) would produce a wider range of values than just 0 and C. leads to the EM algorithm for estimating /0, which is used in the experiments. As our estimated K drifts from the true K, this expression ^K(^bM + 1 - ^bM) would produce a wider range of values than just 0 and C. Let eijN (0‚ r2) be independent noise, which we assume corrupts each element of Kij; that is, given the estimated matrix ^K, suppose Let eijN (0‚ r2) be independent noise, which we assume corrupts each element of Kij; that is, given the estimated matrix ^K, suppose KijN ( ^Kij‚ r2)‚ (11) (11) for some small r2. This is clearly an oversimpliﬁcation (as we know K is e.g., bounded and symmetric), but it is a useful starting point that allows derivation of a simple estimation algorithm. For notational brevity, in this and the following section we deﬁne the vector (12) D = ^bM + m - ^bM (12) which corresponds to the difference between the two GRS model parameter vectors when m additional dogs are added. Given the true value of K, the system of equations F> mcm = KD relates the known quantity D and the Gaussian-distributed K with the matrix Fm and the unknown values in the vector cm 2 Rm. 5.4. Estimating K If the true matrix K is unknown, it can be estimated with public data. We denote this estimator by ^K. For ^K to be an accurate estimate, the data that it is generated from must be drawn from the same (or a sufﬁciently similar) population as that used in the private study. We will model this assuming no dis- crepancy between population distributions; however, when we discuss how to evaluate whether the esti- mate is good, that assessment should account for this systematic error as well. Later, we will be primarily concerned with the error due to the subsampling in both the private and public datasets. Also of note, the same analysis given next applies to the scenario in which the researchers do not release K, but rather release a ‘‘noisy’’ version of K, where the noise is drawn from a normal distribution. They might consider doing this if they feel that releasing information about SNP frequencies is important for the research community, but they do not wish to release the real K because this would allow for an exact reconstruction of genotype. This noisy K could still be used in a reconstruction attack in the same way as an estimate of K from a public database is used. RECONSTRUCTION ATTACKS FROM GRS PAIGE ET AL. With some algebraic re-arrangement, and since for the true underlying value of K we have KD = F> mcm, we can write this as With some algebraic re-arrangement, and since for the true underlying value of K we have KD = F> mcm, we can write this as ^KDN ( X m j = 1 Cj/j‚ r2D>DI) (13) (13) where C1‚ . . . ‚ Cm and r are parameters we need to estimate. The vector ^KD is observed ‘‘data,’’ computed from the public SNP database and the two released parameter vectors. We can model each of the entries of Fm, which are zeros and ones, as Bernoulli distributions, whose prior probabilities correspond to the public dataset estimated frequencies. This suggests a model for ^KD that is akin to a constrained mixture of Gaussians. where C1‚ . . . ‚ Cm and r are parameters we need to estimate. The vector ^KD is observed ‘‘data,’’ computed from the public SNP database and the two released parameter vectors. We can model each of the entries of Fm, which are zeros and ones, as Bernoulli distributions, whose prior probabilities correspond to the public dataset estimated frequencies. This suggests a model for ^KD that is akin to a constrained mixture of Gaussians. For the special case of m = 1, with only a single scalar C and vector /0, this reduces to ^KDN (C/0‚ r2D>DI): (14) (14) 5.4.3. Parameter estimation with EM. We now can use this model to derive EM algorithms for ﬁnding maximum likelihood estimates of all parameters, and estimate the posterior distribution over SNP variants for the added individuals between the two studies. For notational convenience in this section, denote the entries of the m new individuals Fm 2 f0‚ 1gN + 1‚ m as zi‚ j, for i = 1‚ . . . ‚ N + 1 and j = 1‚ . . . ‚ m, and let zj denote the column vector z1‚ j‚ . . . ‚ zN + 1‚ j. Denote the prior probabilities for each i as a1‚ . . . ‚ aN + 1, where a1‚ . . . ‚ aN are the (public) population frequencies for each SNP, and aN + 1 = 1. Let x1‚ . . . PAIGE ET AL. ‚ xN + 1 denote the entries of the ﬁxed (observed) vector x = ^KD, which in this simpliﬁed notation is distributed as p(xjcm‚ Fm‚ r2) = N (xj X m j = 1 Cjzj‚ r2D>DI): p(xjcm‚ Fm‚ r2) = N (xj X m j = 1 Cjzj‚ r2D>DI): posing we know values of C1‚ . . . ‚ Cm‚ r2, to estimate the entries of Fm we want to ﬁnd p(zjx‚ C‚ r2 Supposing we know values of C1‚ . . . ‚ Cm‚ r2, to estimate the entries of Fm we want to ﬁnd p(zjx‚ C‚ r2), p(zjx‚ cm‚ r2) / p(xjcm‚ Fm‚ r2)p(z): p(zjx‚ cm‚ r2) / p(xjcm‚ Fm‚ r2)p(z): An EM algorithm to estimate cm‚ r2 would proceed by alternately: An EM algorithm to estimate cm‚ r2 would proceed by alternately: 1. Given cm‚ r2, estimate the posterior distribution p = p(zjx‚ cm‚ r2); 1. Given cm‚ r2, estimate the posterior distribution p = p(zjx‚ cm‚ r2); 2. Given the posterior p, maximize L = Ep[ log p(xjcm‚ Fm‚ r2)] with respect to cm and r2. 1. Given cm‚ r2, estimate the posterior distribution p = p(zjx‚ cm‚ r2); 2. Given the posterior p, maximize L = Ep[ log p(xjcm‚ Fm‚ r2)] with respect to cm and r2. For each zi‚ j, we can analytically compute the distribution For each zi‚ j, we can analytically compute the distribution p(zi‚ j = 1jx‚ cm‚ zk6¼j‚ r2) p(zi‚ j = 1jx‚ cm‚ zk6¼j‚ r2) (15) = aiN (xijCj + P k6¼j Ckzi‚ k‚ r2D>D) aiN (xijCj + P k6¼j Ckzi‚ k‚ r2D>D) + (1 - ai)N (xij P k6¼j Ckzi‚ k‚ r2D>D) ‚ (15) the conditional probability of each particular entry taking a value of 1, rather than 0, for each zj given the values of the other zk, k 6¼ j. Note that each SNP location i can be treated independently; however, each of the individuals j = 1‚ . . . ‚ m individuals must be considered jointly. the conditional probability of each particular entry taking a value of 1, rather than 0, for each zj given the values of the other zk, k 6¼ j. Note that each SNP location i can be treated independently; however, each of the individuals j = 1‚ . . . ‚ m individuals must be considered jointly. 5.4.4. Exact EM algorithm when 1 individual is added. RECONSTRUCTION ATTACKS FROM GRS This breaks down into a sum across the entries in cm, with KD = X m j = 1 Cj/j: KD = X m j = 1 Cj/j: We need to estimate all m constants Cj‚ j = 1‚ . . . ‚ m. j If K is Gaussian [following Eq. (11)], then the linear transformation KD is Gaussian as well. We denote each of the rows of K as a vector ki, i = 1‚ . . . ‚ N; then, for each row, the scalar value k> i D*N (^k> i D‚ r2D>D)‚ meaning overall the vector KD is distributed N ( ^KD‚ r2D>DI). 12 PAIGE ET AL. RECONSTRUCTION ATTACKS FROM GRS 13 This yields @L @C = X i pi r2DTD (xi - C)‚ @L @r2 = X N i = 1 pi @ @r2 log N (xijC‚ r2D>D) + (1 - pi) @ @r2 log N (xij0‚ r2D>D)‚ which we set equal to zero and solve to ﬁnd which we set equal to zero and solve to ﬁnd ^C = P i pixi P i pi ‚ (17) ^C = P i pixi P i pi ‚ (17) ^r2 = 1 ND>D X N i = 1 pi(xi - C)2 + (1 - pi)x2 i : (18) (18) These updates taken together can be used to deﬁne an EM algorithm that optimizes the values of C and r2, despite the fact that the entries of /0 are unknown; once C and r2 are then known, the vector p will give probability estimates for each entry of /0. p y y /0 The overall EM algorithm can be summarized by the following iterative updates: The overall EM algorithm can be summarized by the following iterative updates: 1. pi p(zi = 1jx‚ ^C‚ ^r2) = aiN (xij ^C‚ ^r2D>D) aiN (xij ^C‚ ^r2D>D) + (1 - ai)N (xij0‚ ^r2D>D), 2. ^C) P i pixi P i pi , 3. ^r2) 1 ND>D PN i = 1 pi(xi - ^C)2 + (1 - pi)x2 i : 1. pi p(zi = 1jx‚ ^C‚ ^r2) = aiN (xij ^C‚ ^r2D>D) aiN (xij ^C‚ ^r2D>D) + (1 - ai)N (xij0‚ ^r2D>D), P 1. pi p(zi = 1jx‚ ^C‚ ^r2) = aiN (xij ^C‚ ^r2D>D) aiN (xij ^C‚ ^r2D>D) + (1 - ai)N (xij0‚ ^r2D>D), 2. ^C) P i pixi P i pi , 3. ^r2) 1 ND>D PN i = 1 pi(xi - ^C)2 + (1 - pi)x2 i : 1. pi p(zi = 1jx‚ ^C‚ ^r2) = aiN (xij ^C‚ ^r2D>D) aiN (xij ^C‚ ^r2D>D) + (1 - ai)N (xij0‚ ^r2D>D), P i 3. RECONSTRUCTION ATTACKS FROM GRS ^r2) 1 ND>D PN i = 1 pi(xi - ^C)2 + (1 - pi)x2 i : To initialize the algorithm, we can set pi to some initial probabilities, and ﬁnd initial values for ^C‚ ^r2; we experimented with setting both to the prior probabilities per SNP estimated from the public data and to the vector of all zeros (corresponding to a ‘‘hard’’ initialization at the value of the baseline estimate), and we found no qualitative difference in performance. 5.4.5. Stochastic EM (SEM) for multiple individuals. For m > 1, the exact posterior depends on all individuals and does not have a compact form. However, we can easily approximate the posterior by Gibbs sampling using Eq. (15), which describes the full conditional distribution p(zi‚ j = 1jx‚ cm‚ zk6¼j‚ r2), iteratively drawing samples for each individual j. We can use this for parameter estimation of r2 and each C1‚ . . . ‚ Cm by using the stochastic EM algorithm Celeux and Diebolt (1985), which differs from a standard EM algorithm in that the expectation step (evaluating the posterior) is replaced by Monte Carlo sampling. In this algorithm, we alternately 1. draw approximate posterior samples of zi‚ j by one or more sweeps of Gibbs sampling, following Eq. (15); 1. draw approximate posterior samples of zi‚ j by one or more sweeps of Gibbs sampling, following Eq. (15); 2 2. conditioned on the current sampled values zi‚ j, ﬁnd values of r2 and C1‚ . . . ‚ Cm which maximize the likelihood N (xij Pm j = 1 Cjzi‚ j‚ r2D>DI). 2. conditioned on the current sampled values zi‚ j, ﬁnd values of r2 and C1‚ . . . ‚ Cm which maximize the likelihood N (xij Pm j = 1 Cjzi‚ j‚ r2D>DI). Although this does not converge to an exact parameter value, under suitable conditions the algorithm converges in distribution to a Gaussian centered on the maximum likelihood estimate of the parameter. A point estimate can be extracted by averaging across many iterations after convergence. In contrast to the EM updates, the updates for values of Cj and r2 given actual sampled values of zj are straightforward and do not scale combinatorially in m. PAIGE ET AL. For the special case of m = 1, this yields a tractable exact EM algorithm. Since there are no other individuals, Eq. (15) reduces to p(zjx‚ C‚ r2), with 5.4.4. Exact EM algorithm when 1 individual is added. For the special case of m = 1, this yields a tractable exact EM algorithm. Since there are no other individuals, Eq. (15) reduces to p(zjx‚ C‚ r2), with pi = p(zi = 1jx‚ C‚ r2) = aiN (xijC‚ r2D>D) aiN (xijC‚ r2D>D) + (1 - ai)N (xij0‚ r2D>D) (16) (16) the posterior probability of each particular entry taking a value of 1, rather than 0. To maximize L = Ep[ log p(xjC‚ /0‚ r2)] with respect to C and r2, we ﬁrst compute the derivatives of L = X i X zi p(zij . . . ) log p(xijC‚ zi‚ r2) = X N i = 1 pi log N (xijC‚ r2D>D) + (1 - pi) log N (xij0‚ r2D>D): RECONSTRUCTION ATTACKS FROM GRS 5.5. Scaling of EM algorithm with size of private dataset Figure 7 demonstrates the change in accuracy of the EM algorithm over a range of different private database sizes. For this test, a synthetic dataset with 100 SNPs and 1,000,000 individuals is generated; 10,000 are held out as a public database, and 30 individuals are taken as a ﬁxed test dataset of new dogs to add and are used to estimate EM algorithm accuracy, across increasingly large private database sizes. The algorithm has stable performance for increasingly large private databases. PAIGE ET AL. PAIGE ET AL. The maximum likelihood estimate of r2 given this estimated ^cm is simply the mean squared error The maximum likelihood estimate of r2 given this estimated ^cm is simply the mean squared error ^r2 = 1 N + 1 X N i = 1 (xi - ^c> mzi)2: (20) (20) To address permutation invariance in the entries 1‚ . . . ‚ m, we enforce an ordering on the estimated values of Cj, with C1 C2 . . . Cm. This breaks the symmetry across the indices of the m new indi- viduals added in the second study, and it is handled by a projection operation at each iteration, in which the estimated values are sorted in ascending order after each maximization step. g p Empirical results quantifying the performance of this algorithm are shown in Figure 6, in an experi- mental setup similar to that for evaluating EM when a single dog is added to a dataset in the main article, with unknown K. A private dataset is assumed to contain 1000 individuals, whereas a separate public dataset of 800 is available; m = 3 new individuals are added to the private dataset to produce two parameter vectors bM and bM + m. On average, the SEM algorithm predicts the correct SNP 75.5% of the time, relative to 71.5% for the ‘‘most common variant’’ baseline, a moderate improvement. RECONSTRUCTION ATTACKS FROM GRS Optimizing cm corresponds to solving a least-squares problem, that is, min C1‚ ...‚ Cm X N + 1 i = 1 (xi - X j = 1 Cjzi‚ j)2 = min cm k x - Zcm k2 2 ‚ using the vector notation cm = [C1‚ . . . ‚ Cm]> 2 Rm, x = [x1‚ . . . ‚ xN + 1]> 2 RN + 1, and Z 2 [0‚ 1]N + 1‚ m, has the solution using the vector notation cm = [C1‚ . . . ‚ Cm]> 2 Rm, x = [x1‚ . . . ‚ xN + 1]> 2 RN + 1, and Z 2 [0‚ 1]N + 1‚ m, has the solution using the vector notation cm = [C1‚ . . . ‚ Cm]> 2 Rm, x = [x1‚ . . . ‚ xN + 1]> 2 RN + 1, and Z 2 [0‚ 1]N + 1‚ m, has the solution ^cm = (Z>Z) - 1Z>x: (19) ^cm = (Z>Z) - 1Z>x: (19) 14 RECONSTRUCTION ATTACKS FROM GRS 15 FIG. 7. Accuracy at reconstruction of the genome of one additional individual, using EM estimation and a noisy estimate ^K, measured as the size of the initial private database increases. For very small private databases, accuracy is very high, as changes in entries of b are clearly attributable to the new individual. Beyond a certain threshold, overall accuracy is quite stable. Error bars show mean and two standard deviations. FIG. 7. Accuracy at reconstruction of the genome of one additional individual, using EM estimation and a noisy estimate ^K, measured as the size of the initial private database increases. For very small private databases, accuracy is very high, as changes in entries of b are clearly attributable to the new individual. Beyond a certain threshold, overall accuracy is quite stable. Error bars show mean and two standard deviations. and now, from the second experiment, we have and now, from the second experiment, we have and now, from the second experiment, we have and now, from the second experiment, we have (K0 + /0T 0/0 0)^b0 = F0Ty0: (22) (22) Taking the difference between these expressions, as earlier, gives Taking the difference between these expressions, as earlier, gives K0 ^b0 - K ^b = F0Ty0 - FTy - /0T 0/0 0 ^b0: (23) (23) Restricting to the overlapping set gives that Restricting to the overlapping set gives that [K0 ^b0] - [K ^b] = [F0Ty0 - FTy - /0T 0/0 0 ^b0]: (24) Noting that [K] = [K0] and that [F0Ty0] - [FTy] = [/T 0y0] we get that [K]([^b0] - [^b]) = [/T 0](y0 0 - /0 ^b0): (25) (24) (25) Analogous to the previous cases, (y00 - /0 ^b0) is a scalar that we can label C and we get Analogous to the previous cases, (y00 - /0 ^b0) is a scalar that we can label C and we get [/T 0] = 1 C [K]([^b0] - [^b]): (26) (26) Thus, if K is known, it can be used to deduce whether the additional individual has each of the SNPs in the overlapping set. If K is not known exactly, it can be estimated from public data just as in the same SNP case. 5.6. Estimating /0 with different SNP sets Here, we analyze what can still be said in the event that the two studies do not use exactly the same set of SNPs. We will still assume that the sets of SNPs are considered to have a signiﬁcant overlap. For this purpose we will need a greater variety of notation. A primed variable denotes that it corresponds to the second set of SNPs, for example, K0 is the co-occurrence matrix from the original M users for the second experiment. If a vector or matrix is surrounded by square brackets, this denotes the same object but with rows and columns corresponding to SNPs not in the overlap removed, for example, [K] denotes the co-occurrence matrix from the ﬁrst experiment restricted to the overlapping SNPs. As described earlier, from the ﬁrst experiment, we have As described earlier, from the ﬁrst experiment, we have K ^b = FTy (21) (21) K ^b = FTy FIG. 6. Results for running the stochastic EM algorithm when estimating SNPs for three additional dogs simulta- neously. This experimental setup replicates the experiment for one additional dog, across 5 public/private/test dataset splits, with 20 different test sets of three additional dogs for each. (Left) Accuracy at predicting SNP presence relative to the ‘‘most common variant’’ baseline. On average, the SEM algorithm predicts the correct SNP 75.5% of the time, relative to 71.5% for the baseline. (Right) As in the one-dog example, we see relative improvement in the performance of our algorithm when considering more atypical dogs. SEM, stochastic EM. FIG. 6. Results for running the stochastic EM algorithm when estimating SNPs for three additional dogs simulta- neously. This experimental setup replicates the experiment for one additional dog, across 5 public/private/test dataset splits, with 20 different test sets of three additional dogs for each. (Left) Accuracy at predicting SNP presence relative to the ‘‘most common variant’’ baseline. On average, the SEM algorithm predicts the correct SNP 75.5% of the time, relative to 71.5% for the baseline. (Right) As in the one-dog example, we see relative improvement in the performance of our algorithm when considering more atypical dogs. SEM, stochastic EM. RECONSTRUCTION ATTACKS FROM GRS AUTHOR DISCLOSURE STATEMENT The authors declare they have no competing ﬁnancial interests. 5.8. Description of K when the genotypes are non-binary yg – For i = 1‚ . . . ‚ N and j = N + 1: Kij = Kji = pAa + 2pAA. – For i = 1‚ . . . ‚ N and j = N + 1: Kij = Kji = pAa + 2pAA. – For i = 1‚ . . . ‚ N and j = N + 1: Kij = Kji = pAa + 2pAA. – Finally, KN + 1‚ N + 1 = 1. PAIGE ET AL. 16 and the following a + b length vector: ra + b = (ya - Fa ^bM + a)‚ - (yb - Fb ^bM + b) h i (28) (28) Then, this gives us: KM(^bM + a - ^bM + b) = 1 M Fa + bra + b (29) (29) This means that having two nonoverlapping participant sets is equivalent to the setting in which the ﬁrst study is a subset of the second (only m is now a + b). FUNDING INFORMATION This project was funded by the Alan Turing Institute Research Fellowship under EPSRC Research grant (TU/A/000017); EPSRC/BBSRC Innovation Fellowship (EP/S001360/1), and under the EPSRC grant EP/N510129/1. It was also partly funded by a grant from CPER Nord-Pas de Calais/FEDER DATA Advanced data science and technologies 2015–2020. 5.8. Description of K when the genotypes are non-binary 5.8. Description of K when the genotypes are non-binary In many cases, GRS are calculated on genotype matrices that are non-binary. In particular, they may take on three discrete values 0, 1, and 2, where 0 indicates that the most common variant is homozygous, 1 indicates that the individual is heterozygous for the uncommon variant, and 2 indicates that the individual is homozygous for the uncommon variant. If this is the case, the description of K will change. However, it is still the case that the entries of K depend only on the SNP frequencies and SNP co-occurrence frequencies in the dataset, and that knowledge of SNP frequencies and pairwise co-frequencies from the original study, are all that is required to compute K. – For i = 1‚ . . . ‚ N: Kii = pAa + 4pAA where paa is the frequency of individuals being heterozygou uncommon variant and pAA is the frequency of individuals being homozygous for the uncommon – For i = 1‚ . . . ‚ N: Kii = pAa + 4pAA where paa is the frequency of individuals being heterozygous for the uncommon variant and pAA is the frequency of individuals being homozygous for the uncommon variant. uncommon variant and pAA is the frequency of individuals being homozygous for the uncommon variant. – For i = 1‚ . . . ‚ N - 1 and j > i: Kij = Kji = pAa=Bb + 2pAA=Bb + 4pAA=BB where pAa=Bb is the frequency that both SNPs are simultaneously heterogygous, pAA=Bb is the frequency that one SNP is homozygous for the rare variant and the other is heterogygous simultaneously, and pAA=BB is the frequency that that uncommon variants are found to be homozygous simultaneously. pAA q y g yg – For i = 1‚ . . . ‚ N - 1 and j > i: Kij = Kji = pAa=Bb + 2pAA=Bb + 4pAA=BB where pAa=Bb is the frequency that both SNPs are simultaneously heterogygous, pAA=Bb is the frequency that one SNP is homozygous for the rare variant and the other is heterogygous simultaneously, and pAA=BB is the frequency that that uncommon variants are found to be homozygous simultaneously. yg y – For i = 1‚ . . . ‚ N and j = N + 1: Kij = Kji = pAa + 2pAA. – Finally, KN + 1‚ N + 1 = 1. 5.7. Case in which each GWAS study adds two new sets of participants 5.7. Case in which each GWAS study adds two new sets of participants This article mostly explores the case in which one study’s participants are a subset of the other study’s participants. Here, we demonstrate that this is equivalent to the case where each of the two studies contains a small number of participants that are not found in the other study. In particular, let us say that the ﬁrst study has M + a participants and the second study has M + b participants, where the ﬁrst M participants are shared between the studies, but there are a participants that are found in the ﬁrst study but not the second, and b participants that are found in the second study but not in the ﬁrst. Following on from Eq. (9), we see that: KM(^bM + a - ^bM + b) = KM(^bM + a - ^bM) - KM(^bM + b - ^bM) = 1 M FT a(ya - Fa ^bM + a) - FT b(yb - Fb ^bM + b) h i : Let us deﬁne the following (N + 1) · (a + b) matrix obtained by concatenating the two genotype matrices: ﬁne the following (N + 1) · (a + b) matrix obtained by concatenating the two genotype matrices: Fa + b = Fa‚ Fb ½ (27) Fa + b = Fa‚ Fb ½ (27) PAIGE ET AL. PAIGE ET AL. REFERENCES Belsky, D.W., Mofﬁtt, T.E., Sugden, K., et al. 2013. Development and evaluation of a genetic risk score for obesity. Biodemography Soc. Biol. 59, 85–100. g p y Cai, R., Hao, Z., Winslett, M., et al. 2015. Deterministic identiﬁcation of speciﬁc individuals from GWAS results. Bioinformatics 31, 1701–1707. f Celeux, G., and Diebolt, J. 1985. The SEM algorithm: A probabilistic teacher algorithm derived from the EM algorithm for the mixture problem. Comput. Stat. Q. 2, 73–82. Chouraki, V., Reitz, C., Maury, F., et al. 2016. Evaluation of a genetic risk score to improve risk prediction for Alzheimer’s disease. J. Alzheimers Dis. 53, 921–932. Day, F.R., Thompson, D.J., Helgason, H., et al. 2017. Genomic analyses identify hundreds of variants associated with age at menarche and support a role for puberty timing in cancer risk. Nat. Genet. 49, 834–841. Dwork, C., Smith, A., Steinke, T., et al. 2015. Robust traceability from trace amounts. 2015 IEEE 56th Annual Symposium on Foundations of Computer Science. Berkeley, California. pp. 650–669. Dwork, C., Smith, A., Steinke, T., et al. 2015. Robust traceability from trace amounts. Symposium on Foundations of Computer Science. Berkeley, California. pp. 650–669. RECONSTRUCTION ATTACKS FROM GRS 17 Hayward, J.J., Castelhano, M.G., Oliveira, K.C., et al. 2016. Complex disease and phenotype mapping in th dog. Nat. Commun. 7, 10460. g Huang, Z., Ayday, E., Fellay, J., et al. 2015. Genoguard: Protecting genomic data against brute-force attacks. 2015 IEEE Symposium on Security and Privacy. San Jose, California. pp. 447–462. Im, H., Gamazon, E., Nicolae, D., et al. 2012. On sharing quantitative trait gwas results in an era of multiple-omics data and the limits of genomic privacy. Am. J. Hum. Genet. 90, 591–598. Jiang, L., Zheng, Z., Qi, T., et al. 2019. A resource-efﬁcient tool for mixed model association analysis of large-scale data. Nat. Genet. 51, 1749–1755. Knowles, J.W., and Ashley, E.A. 2018. Cardiovascular disease: The rise of the genetic risk score. PLoS Med. 15, e1002546. Mittos, A., Malin, B., and Cristofaro, E.D. 2019. Systematizing genome privacy research: A privacy-enhancing technologies perspective. Proc. Privacy Enhanc. Technol. 2019, 87–107. Qi, L., Ma, J., Qi, Q., et al. 2011. Genetic risk score and risk of myocardial infarction in hispanics. Circu 374–380. Raisaro, J.L., Trame`r, F., Ji, Z., et al. 2017. Addressing beacon re-identiﬁcation attacks: Quantiﬁcation and mitigation of privacy risks. J. Am. Med. Inform. Assoc. 24, 799–805. p y f Shringarpure, S., and Bustamante, C. 2015. Privacy risks from genomic data-sharing beacons. Am. J. Hum. Genet. 97, 631–646. Simmons, S., and Berger, B. 2015. One size doesn’t ﬁt all: Measuring individual privacy in aggregate genomic data. 2015 IEEE Security and Privacy Workshops. San Jose, California. pp. 41–49. ropp, J.A. 2015. An introduction to matrix concentration inequalities. Found. Trends Mach. Learn. 8, 1–230 Visscher, P.M., Wray, N.R., Zhang, Q., et al. 2017. 10 years of gwas discovery: Biology, function, and translation. Am. J. Hum. Genet. 101, 5–22. von Thenen, N., Ayday, E., and Cicek, A.E. 2018. Re-identiﬁcation of individuals in genomic data-sharing beacons via allele inference. Bioinformatics 35, 365–371. Wang, Y.X. 2018. Revisiting differentially private linear regression: Optimal and adaptive prediction & estimation in unbounded domain. Proceedings of the Conference on Uncertainty in Artiﬁcial Intelligence (UAI). Monterey, California. pp. 93–103. Zhao, X., Xi, B., Shen, Y., et al. 2014. An obesity genetic risk score is associated with metabolic syndrome in Chinese children. Gene 535, 299–302. E-mail: daphne.ezer@york.ac.uk
https://openalex.org/W4288060670	https://kula.uvic.ca/index.php/kula/article/download/171/473	English	null	Leveraging Wikidata to Build Scholarly Profiles as Service	KULA	2,022	cc-by	9,320	Lemus-Rojas, Mairelys, Jere Odell, Lucille Frances Brys, and Mirian Ramirez Rojas. 2022. Leveraging Wikidata to Build Scholarly Profiles as Service. KULA: Knowledge Creation, Dissemination, and Preservation Studies 6(3). https://doi.org/10.18357/kula.171 Lemus-Rojas, Mairelys, Jere Odell, Lucille Frances Brys, and Mirian Ramirez Rojas. 2022. Leveraging Wikidata to Build Scholarly Profiles as Service. KULA: Knowledge Creation, Dissemination, and Preservation Studies 6(3). https://doi.org/10.18357/kula.171 Lemus-Rojas, Mairelys, Jere Odell, Lucille Frances Brys, and Mirian Ramirez Rojas. 2022. Leveraging Wikidata to Build Scholarly Profiles as Service. KULA: Knowledge Creation, Dissemination, and Preservation Studies 6(3). https://doi.org/10.18357/kula.171 RESEARCH ARTICLE Mairelys Lemus-Rojas Brown University Jere Odell IUPUI University Library Lucille Frances Brys IUPUI University Library Mirian Ramirez Rojas Ruth Lilly Medical Library, Indiana University School of Medicine Mirian Ramirez Rojas Ruth Lilly Medical Library, Indiana University School of Medicine Mirian Ramirez Rojas In this article, the authors share the different methods and tools utilized for supporting the Scholarly Profiles as Service (SPaS) model at Indiana University–Purdue University Indianapolis (IUPUI). Leveraging Wikidata to build a scholarly profile service aligns with interests in supporting open knowledge and provides opportunities to address information inequities. The article accounts for the authors' decision to focus first on profiles for women scholars at the university and provides a detailed case study of how these profiles are created. By describing the processes of delivering the service, the authors hope to inspire other academic libraries to work toward establishing stronger open data connections between academic institutions, their scholars, and their scholars' publications. Keywords: scholarly profiles; Wikidata; Scholia; linked data Keywords: scholarly profiles; Wikidata; Scholia; linked data Infrastructure and Tools for Scholarly Profiles y Academic librarians who support faculty and researchers in tracking and showing scholarly productivity data rely on different approaches. For example, one approach is to populate a university content management system with lists of publications and provide links to the curriculum vitae files of the affiliated academics. But these lists are outdated in most cases, and their structure does not allow enough flexibility for the data to be reused. Another approach is to use faculty directories, researcher and author profiles (e.g., researchers’ ORCID profiles and Google Scholar profiles), personal websites, online bibliographic databases (e.g., Scopus, PubMed, and Web of Science), and open access institutional repositories to retrieve and display the researchers’ scholarly outputs. However, these sources are seldom perfectly accurate or complete. y p p y p Other tools specifically designed for scholarly profiles may provide a more reliable alternative. These sys tems allow any user to register, collect, and evaluate the productivity of academics at the institutional level, tracking their works and the relationships among scholars with common interests as well as managing scholars’ identities. Academic institutions often rely on proprietary software and services to maintain researchers’ profiles and research funding (e.g., Pure by Elsevier, Pivot by ProQuest, and Activity Insight by Digital Measures) (“Comparison of Research Networking Tools and Research Profiling Systems” 2021). However, open-source tools to support this work are available and openly accessible, and their use aligns with UL’s commitment to advancing openness. These tools include ReCiter (“Wcmc-Its/ReCiter” 2021), VIVO, DSpace-CRIS, and Scholia. ReCiter, developed by the Weill Cornell Medical College, is an author disam biguation system that uses a machine learning process and available identity data to generate bibliogra phies. It maintains publication lists of scholars by extracting data from PubMed or Scopus (Johnson et al. 2014; Albert et al. 2021). VIVO and DSpace-CRIS are tools for displaying researcher profiles (Obeid et al. 2014; Mornati 2019). In contrast, Scholia, a free web-based application, builds researcher profiles from data retrieved from Wikidata (Nielsen, Mietchen, and Willighagen 2017). The rendering of these profiles is dynamic because the application queries the Wikidata Query Service and directly displays the most up-to-date information available for a particular subject. p p j Scholarly profile tools implemented at IUPUI include Scholars@IU and ORCID. Scholars@IU—a ProQuest Pivot product offered by the university’s Office for Research—provides public profiles for 10,440 faculty and academics. Introduction Contributing to projects that offer a community-driven solution to sharing knowledge freely and openly is at the core of the mission of many academic libraries. The University Library (UL) of the Indiana University– Purdue University Indianapolis (IUPUI) campus has been an early adopter of open-source platforms and efforts to advance open knowledge and is committed to making scholarship openly accessible to a broader community of users (IUPUI University Library 2019) through, in part, its commitment to open knowledge projects (IUPUI University Library Center for Digital Scholarship 2021). IUPUI faculty likewise demonstrated their commitment to open knowledge when the faculty council adopted an open access policy (OAP) in 2014 (IUPUI University Library n.d.). As a result of this policy, articles written by institution-affiliated scholars that fit the criteria for the open access policy are archived in ScholarWorks, the institutional repository. Since the implementation of the OAP, the library has been managing a yearly data collection of more than three thousand articles authored by university scholars; of these, roughly 70 percent are deposited into ScholarWorks (IUPUI University Library Center for Digital Scholarship 2020). The OAP work is fundamental in that it provides free access to more than thirteen thousand articles that might otherwise be behind a paywall while also increasing the visibility of their authors. However, the infrastructure the library uses for sharing these resources has its limitations. The current version of the software used for the institutional repository (DSpace 5.6) is not configured to provide linked data profiles for authors. Therefore, the connections among authors, works, and institutions must be estab lished in other platforms. Wikidata has the infrastructure in place to facilitate these connections and more. Using Wikidata—the structured linked data repository that serves all Wikimedia projects—and the tools it enables, such as Scholia—a web-based service that feeds from Wikidata—will enhance findability while also page 2 of 14 Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service contributing to the dissemination of scholarship. Wikidata offers an opportunity for extending the reach of scholars’ works and connecting them with other research while also making the data available in a format that is easily accessible and reusable. The library leverages the power of Wikidata to build profiles for scholars that may be less visible or systematically underserved in other systems. Introduction This article shares the methods, approach, and priorities that inform a Scholarly Profiles as Service (SPaS) model at Indiana University–Purdue University Indianapolis (IUPUI), a campus that includes 3,693 people with academic positions. The article describes the background for the development of the service, explains a decision to focus on gender equity, and provides a description of the process of delivering the service for all women faculty affiliated with IUPUI. Infrastructure and Tools for Scholarly Profiles This profile system lists faculty and researchers by campus, school, department, division, or center and by specific expertise. It is also intended to be a tool for identifying potential collaborators and mentors throughout the Indiana University system (Indiana University n.d.). In addition, IUPUI recently became an institutional ORCID member (ORCID n.d.). The libraries and the Offices for Research, Graduate Studies, and Academic Affairs use ORCID in several systems and programs—including annual reviews, Scholars@IU, Wikidata, journal publishing, and consultations for scholarly profile management. Thus, ORCID is a crucial tool facilitating the customization, integration, and connection of IUPUI affiliates’ identi fiers with other research networking tools and research profiling systems already implemented at the insti tution. However, ORCID adoption depends in part on the participation of individual authors. At the same time, some systems, like Scholars@IU, offer few editing opportunities to authors and the librarians that serve them. As an openly edited database, Wikidata provides an opportunity to consolidate disparate data elements and to supplement data profiles with information that is missing from proprietary systems. Wikidata for Scholarly Profiles Wikidata, a structured linked data knowledge base, is part of the Wikimedia ecosystem of free and openly accessible projects. It was conceived as a central repository to support all Wikimedia-related projects but has since grown beyond the limits of its original conception. Many galleries, libraries, archives, and museums (GLAMs) have been working on building capacity through community engagement within their institutions and beyond to advance open knowledge (Lemus-Rojas 2019; Allison-Cassin et al. 2019). The contributions Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 3 of 14 made in Wikidata by these communities play a significant role in the pursuit of knowledge equity. A factor that has attracted users from around the world is the multilingual nature of Wikidata; while Wikipedia currently has 310 active language versions, there is only one instance of Wikidata, where all languages coexist in one central location (“List of Wikipedias” 2021). The multilingual capacity of the knowledge base has made it an ideal environment for collaboration among the growing global community of users—almost twenty- seven thousand so far (“Wikidata:Statistics” 2021). Wikidata users are able to create, edit, maintain, and use the linked open data while also taking part in the development of a growing ontology to accommodate the needs of the community. The knowledge base can even accommodate conflicting information, which can be further defined with qualifying statements and references to support the assertions being made. made in Wikidata by these communities play a significant role in the pursuit of knowledge equity. A factor that has attracted users from around the world is the multilingual nature of Wikidata; while Wikipedia currently has 310 active language versions, there is only one instance of Wikidata, where all languages coexist in one central location (“List of Wikipedias” 2021). The multilingual capacity of the knowledge base has made it an ideal environment for collaboration among the growing global community of users—almost twenty- seven thousand so far (“Wikidata:Statistics” 2021). Wikidata users are able to create, edit, maintain, and use the linked open data while also taking part in the development of a growing ontology to accommodate the needs of the community. The knowledge base can even accommodate conflicting information, which can be further defined with qualifying statements and references to support the assertions being made. Wikidata for Scholarly Profiles q y g pp g Wikidata’s data model is matched to Resource Description Framework (RDF) triples, facilitating interop erability between Wikidata and external data sources (“Wikidata:Relation Between Properties in RDF and in Wikidata” 2017). These triples contain a subject, a predicate, and an object. By forming these statements or assertions to describe a particular concept, new relationships are made, which contribute to the expansion of the knowledge graph. Wikidata entities—the content of an entry—may refer to an item, a property, or a lexeme. These entities have their corresponding namespaces: items in the main namespace, properties in the property namespace, and lexemes in the lexeme namespace (“Help:Namespaces” 2022). The system also assigns unique identifiers composed of a letter (Q for items and P for properties) and numbers to all the entities in these namespaces. The QID and PID allow machines to easily read and understand the data, while the values linked to them are written in a form more easily understood by humans. For example, in Table 1, the item Q63470490 represents the work “Conscientious Women: The Dispositional Conditions of Institutional Treatment on Civic Involvement”; this item is linked to the item Q56486841, which represents the scholar Amanda Friesen, through the use of the P50 property, which is used to store authors’ names. In other words, Q63470490 (“Conscientious Women: The Dispositional Conditions of Institutional Treatment on Civic Involvement”) is a work by P50 (an author) known as Q56486841 (Amanda Friesen). Table 1: Example of RDF triple in Wikidata RDF triple Machine friendly Human friendly Subject (Item) Q63470490 “Conscientious Women: The Dispositional Conditions of Institutional Treatment on Civic Involvement” Predicate (Property) P50 author Object (Value) Q56486841 Amanda Friesen Table 1: Example of RDF triple in Wikidata Table 1: Example of RDF triple in Wikidata While Wikidata stores data on human knowledge broadly, the community has taken an interest in increas ing the representation of bibliographic data with a focus on scholarly articles. Roughly, publications repre sent 43 percent of items (“Statistics” n.d.) in Wikidata and the majority of these are scholarly articles representing 31 percent of the items stored in the knowledge base (“Wikidata:Statistics” 2021). While there is a growing number of users contributing article data to the knowledge base either by using external tools or by manual editing, most of the article contributions are being made through bots—tools used for making automated contributions without the need for human intervention. Wikidata for Scholarly Profiles Many of these contributions are the efforts of WikiCite, an initiative and a community that aims at building an open citation database in Wikidata. Wikidata currently surpasses 37 million entries for scholarly articles and 240 million citation links to these articles and other publications (Scholia n.d). Challenges and Issues: Gender Equity in Wikimedia Projects The gender inequities in both the content and the culture of Wikimedia sites have been the focus of journalism, scholarship, and specific efforts of the Wikimedia Foundation. In all cases, most of the attention has focused on gender inequities in Wikipedia. Recent stories have focused on gaps in coverage, a culture of harassment among Wikipedians, and programmatic efforts to address the problems. For example, The Atlantic was one of several news outlets to cover the prior omission of Donna Strickland, a scientist who won the Nobel Prize for Physics (Koren 2018). The following year, The Guardian reported on the efforts of the Wikimedia Foundation to increase the number of women editors of Wikipedia (Balch 2019). In that same year, The New York Times covered the harassment of cisgendered women and transgender editors of Wikipedia (Jacobs 2019). More recently, The Washington Post published a commentary on a Wikipedian’s efforts to address gender bias in the site’s coverage of political science topics. The author observed, among other biases, the infrequency of women authors of works cited by Wikipedia’s political science entries (Baltz 2021). page 4 of 14 Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service Scholarly studies of gender inequities have also focused on Wikipedia. These studies explore the gender distribution of Wikipedia contributors, biases in the character and the quantity of content about women, and the structural features of Wikipedia that contribute to these inequities (Bear and Collier 2016; Ford and Wajcman 2017; Graells-Garrido, Lalmas, and Menczer 2015; Lir 2019; Sun and Peng 2021; Wagner et al. 2015; Wagner et al. 2016). A 2018 Wikimedia Foundation-supported investigation of Wikimedia editors found that 90 percent of gender self-reporting contributors identified as male and only 9 percent identified as female (“Community Insights/2018 Report/Contributors” 2019). By 2020, this imbalance eased, but only by six percentage points, with women contributors making up 15 percent of all Wikimedians (“Community Insights/Community Insights 2021 Report” 2021). Although these reports rely on opt-in surveys and may, therefore, underestimate the number of women contributors, efforts to adjust the estimates for survey-re sponse bias improved the gender balance by less than 5 percentage points (Hill and Shaw 2013). This gender imbalance in editing means that the interests of male editors are more likely to result in new contributions. Challenges and Issues: Gender Equity in Wikimedia Projects Because predominantly men contribute to and edit Wikimedia content, their interests begin to form links between subjects in such a way that women—when included—are cast as tangential to men (Wagner et al. 2015; Wagner et al. 2016; Graells-Garrido, Lalmas, and Menczer 2015). g ) Research addressing gender inequities in Wikidata is less common, but recent work shows that the gender disparities in Wikipedia are also found in Wikidata. To this point, Wikidata itself can be used as a tool to assess the gender imbalance of its content (Pellissier Tanon and Suchanek 2019). For example, by querying Wikidata for instances of “human” with known-gender records, Klein and Konieczny (2015) found that instances were “84.4% male, 15.6% female, and ≈ 0.0001% nonbinary.” This research formed the foundation for an ongoing dashboard of gender representation in Wikimedia projects, Humaniki. Based on a Wikidata search completed on June 14, 2021, Humaniki (n.d.) reports that Wikidata instances for “human” with the property sex or gender (P21) remain overwhelming imbalanced, with 81.9 percent of records “male,” 18.1 percent “female,” and .05 percent “other genders.” Along with other content, Wikidata now includes records for more than forty million journal articles; about half of these articles have been linked to author records. Of the records for authors, only 15 percent have statements indicating the author’s sex or gender (P21) (Cobb 2020). Although many of the records for scholarly authors have yet to be described by the Wikidata property for sex or gender (P21), the overall trend on the site skews disproportionately “male,” much like in Wikipedia. Case Study: Scholarly Profiles as Service (SPaS) Context Wikidata stores structured linked data, making them interoperable, and releases these data under a CC0 license, which enables their reuse by external tools and applications. One of these web-based applications is Scholia. Among other things, Scholia was developed to facilitate the exploration of scholarship contained in Wikidata (Nielsen, Mietchen, and Willighagen 2017). It is a freely accessible and open solution for generating scholarly profiles and facilitating integration with other web services supporting open infrastructure. By using and sharing open data, the Scholia service is in compliance with FAIR (Findable, Accessible, Interoperable, Reusable) principles. Scholia makes live SPARQL queries (SPARQL Protocol and RDF Query Language) against Wikidata to generate the profiles, which means that they always present Wikidata’s most up-to-date information. While other scholarly profile services may necessitate curation by individual scholars/researchers, in Scholia the data presented are curated by the global community of Wikidata users. When the efforts of the global community are supplemented by targeted institutional approaches (like those described here), Scholia profiles can have comparably complete coverage. The Scholia service not only generates the profiles to make it easier for users to access and analyze Wikidata’s data, but it also provides links to external tools that can be used to enhance the data of a particular section. For instance, the co-authors graph section in a researcher’s profile page can include a link for missing co-author items that may need curation (see Figure 1). Using these links to identify the missing pieces and making these enhancements in Wikidata will improve the rendering of the co-authors graph the next time the query is run. While displaying publicly accessible data about scholars and their works, the Scholia web service does not collect any private information from users. The concept for the development and implementation of the Scholarly Profiles as Service (SPaS) model at IUPUI emerged after a pilot project conducted by UL in the summer of 2017 (“Wikidata:WikiProject IUPUI University Library” 2021). This pilot afforded us the opportunity to explore the potential of Wikidata and Scholia as free and open solutions for scholarly profiles, for which the Lilly Family School of Philanthropy was chosen as a use case. Context page 6 of 14 Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service Furthermore, these tenured and tenure-track male faculty historically have had and currently have an out sized influence in the direction of the university’s research and creative focus. IUPUI includes seventeen degree-offering schools and, with the exception of Nursing, most schools have been led by male deans (Gregory H. Mobley, Archives Specialist, email correspondence to authors, May 5, 2020). Currently, only six degree-offering schools are led by female deans. The exact local impact of these gender inequities at IUPUI is beyond the scope of our efforts on this project and beyond the scope of this article. However, these inequities are not unique to IUPUI and, in fact, gender inequity is a widely discussed challenge that has troubled Wikimedia projects for many years. With these inequities and the biases that they perpetuate in mind, we decided to put IUPUI’s women scholars at the forefront of our efforts. We have done so by creating Wikidata entries for all of the women scholars in a school or department while saving the male scholars for later. This may seem, at the outset, a strange approach for a project that aims to eventually provide a full open data profile of IUPUI’s authors and their scholarly products. However, it is a first step in our efforts to center the work of women in the intellectual history of the campus. By creating entries for women scholars first, we are taking a small step to reverse the typical relationships and linkages between male and female scholars. If we were to start with senior male scholars, our efforts would replicate gender dynamics that have centered men in citation networks and Wikimedia projects. While this work to enhance the visibility of IUPUI women scholars is happening in Wikidata, the reach goes beyond the ecosystem of Wikimedia projects. Wikidata is used by a number of organizations to support their products and research. For this reason, ensuring an accurate representation of women authors will ultimately contribute to having information about them and their works more readily available across different platforms and utilities. For instance, Wikidata’s data is being used by AI technologies, integrated into digital assistant tools such as Siri and Alexa (Simonite 2019; Kinsella 2019; Abellán 2019), and used by Google to generate knowledge graphs. Process At IUPUI, we have taken several approaches to gather, curate, and contribute scholarly profile data to Wikidata. These approaches have been modified and adapted to facilitate the work based on the individual needs for the dataset; in our case, while we have worked toward ensuring the representation of IUPUI women scholars in Wikidata and linking scholars to their works, the creation of new scholar entries has proven to be more impactful. Creating entries for scholars in the knowledge base enables other Wikimedia contributors to create content and link articles and other data to these entries. Context As discussed elsewhere, our prior work with Scholia and the resulting SPaS model have been key to demonstrating the value of contributing to open platforms (Lemus-Rojas and Odell 2018). This work adds a new layer to the campus’s support for the dissemination of scholarship that crosses the boundaries of library systems and helps improve general information resources on various subjects across disciplines. Context The goal of the pilot was to have represented in Wikidata the core faculty from the selected school, their co-authors—regardless of institutional affiliations—and some of their publications Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 5 of 14 Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 5 of 14 page 5 of 14 Figure 1: Co-author graph generated for an IUPUI-affiliated women scholar showing a link to the curation page for the disambiguation and/or creation of co-author entries. Figure 1: Co-author graph generated for an IUPUI-affiliated women scholar showing a link to the curation page for the disambiguation and/or creation of co-author entries. Figure 1: Co-author graph generated for an IUPUI-affiliated women scholar showing a link to the curation page fo the disambiguation and/or creation of co-author entries. (Lemus-Rojas and Odell 2018). Connecting works to authors is vital for expanding the open citation graph, which is why this task was also explored during the pilot phase. This required looking at the reference sec tions of the faculty publications to get the necessary information to create entries for the works and connect them to their authors. This pilot project provided the foundation for the SPaS model, discussed in the rest of this paper, which aims to provide an accurate representation of IUPUI-affiliated scholars and their publica tions in Wikidata. Given the fact that women are largely underrepresented across all Wikimedia projects, the first phase of the SPaS work has focused on helping bridge this prevalent gender divide in the knowledge base by prioritizing the creation of entities for IUPUI women scholars while also enhancing the citation graph for their works. g p At IUPUI, academic positions include executive administration, tenured or tenure-track faculty and librar ians, clinical faculty, lecturers, and research faculty. In 2020, the headcount of full-time academic positions at IUPUI totalled 3,693 people. In survey data reporting on self-identification, the Office of Institutional Research and Decision Support found that 1,626 of these 3,693 people identified as “female,” 44 percent of the faculty population. However, of the 1,338 tenured or tenure-track faculty and librarians, only 479 (36 percent) identified as “female” (IUPUI Institutional Research and Decision Support n.d.). In other words, on this campus, men are more likely to have job security and salaries that reflect promotion on the tenure track. Representing Women Scholars Until early 2021, the SPaS work had been carried out by two librarians at UL who not only were making direct contributions to Wikidata to represent women scholars and their publications, but were also organizing and facilitating Wikidata edit-a-thons for library employees to participate in this effort. Building capacity to get others in the library involved with Wikidata-related projects has been at the forefront of our efforts to contribute to open projects. Our campus was one of the participating institutions in the Program for Cooperative Cataloging (PCC) Wikidata Pilot led by the Library of Congress. This pilot provided us with an opportunity to put out a call for volunteers and formalize a SPaS Working Group that attracted library employees from IUPUI University Library, Ruth Lilly Medical Library, and Ruth Lilly Law Library. The composition of the group was a mix of six librarians and two staff members from public services and other more technical areas. At this time, all current IUPUI women scholars have been either added to the knowledge base or, in cases where they were already represented, the entries about them have been enhanced with additional data and references. To ensure the Wikidata representation of all IUPUI women scholars in assistant, associate, or full professor positions, the SPaS Working Group has taken a school-by-school approach, adding or revising entries for all tenured and tenure-track women scholars on campus one school at a time. We begin the SPaS work by accessing the institutional web pages for women scholars by school. To gather the data, we use functions in Google Sheets to scrape web pages or, in some cases, we collect the data manually. At a minimum, from the scholars’ institutional web pages we can retrieve their names, affiliations, and website links. We also gather their education history and any links available for their professional networking and social media websites. Once the data have been compiled in Google Sheets, we use OpenRefine to clean up the data and run the Wikidata reconciliation service (“Wikidata Reconciliation for OpenRefine” n.d.). This reconciliation service is a quick way to determine if an author is already represented in Wikidata. In such cases, we can avoid duplication Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 7 of 14 but still have an opportunity to use the data gathered to enhance the existing entries. Representing Women Scholars While it is not common to find identifiers on scholars’ web pages, some scholars do include links to their Twitter and LinkedIn profiles. If these links are provided, we record the information in Wikidata in the Twitter username (P2002) and LinkedIn personal profile ID (P6634) properties respectively. We also look for identifiers in other external web services to link them to the scholars’ entries—for instance, Google Scholar author ID (P1960), ORCID iD (P496), and Scopus author ID (P1153). Assertions in Wikidata can be supported by the inclusion of references. We use the property reference URL (P854) and the date the URL was retrieved (P813) to support our assertions. Having these data points set up in the schema allows us to either upload edits directly to Wikidata or export the data to QuickStatements—a tool used to make batch edits to Wikidata. Once the items are in Wikidata, we often enhance the entries with additional information from a variety of sources. For instance, we look for education information in the scholar’s curriculum vitae (CV), if available through their institutional web page, as well as in their LinkedIn and ORCID profiles whenever possible. Any education data found for the scholar is recorded using the property educated at (P69) with the appropriate qualifiers to specify, for instance, the start time (P580), end time (P582), academic degree (P512), and aca demic major (P812). When CVs are available, it is often possible to add the employment information for the scholar prior to their time at IUPUI and include qualifiers to indicate the dates of employment. For instance, prior employment information is included under the existing employer (P108) statement added when creat ing or enhancing the scholar’s entry in Wikidata (see Figure 3). The new values to represent other employers can include qualifiers such as position held (P39), start time (P580), and end time (P582). Figure 3: Example of an employer (P108) statement for an IUPUI-affiliated woman scholar containing prior employment information. Figure 3: Example of an employer (P108) statement for an IUPUI-affiliated woman scholar containing prior employment information. Typically, we used the scholar’s institutional web page as the main source of reference to support the statements. Representing Women Scholars In addition to reconciling the data in OpenRefine, we create a schema, which is essentially a template where the various edits needed for a particular Wikidata statement can be specified. For instance, the scholars’ names are used to label their Wikidata entity, the property affiliation (P1416) is used to connect the name of the school and/or unit they are affiliated with, and official website (P586) is used for the web page containing their scholarly data. but still have an opportunity to use the data gathered to enhance the existing entries. In addition to reconciling the data in OpenRefine, we create a schema, which is essentially a template where the various edits needed for a particular Wikidata statement can be specified. For instance, the scholars’ names are used to label their Wikidata entity, the property affiliation (P1416) is used to connect the name of the school and/or unit they are affiliated with, and official website (P586) is used for the web page containing their scholarly data. As part of the schema creation process, we also include a number of core properties and constant values for all entries (see Figure 2). These include the properties instance of (P31) to indicate that the entry is for a “human”; sex or gender (P21) with the value “female” because we are focusing on women scholars; languages spoken, written, or signed (P1412) with a default value of “English”; occupation (P106), which for faculty includes both values “university teacher” and “researcher”; employer (P108) to indicate they are employed by “Indiana University – Purdue University Indianapolis”; and work location (P937) with a default value of “Indianapolis.” Figure 2: Schema created in OpenRefine containing core properties and constant values for the creation of entries to represent IUPUI scholars. Figure 2: Schema created in OpenRefine containing core properties and constant values for the creation of entries to represent IUPUI scholars. Figure 2: Schema created in OpenRefine containing core properties and constant values for the creation of entries to represent IUPUI scholars. page 8 of 14 Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service An important feature of Wikidata is that entities can be linked to external sources through the use of identifiers. This is not only beneficial in enriching the data in Wikidata but in making connections with other data sources. Representing Women Scholars However, as we made progress in creating new Wikidata items to represent IUPUI women schol ars and/or enhance existing entries, we were conscious of the fact that every year scholars retire or move to other institutions and their web pages are taken down. This meant that the information in a particular statement would no longer be verified through the reference we had used in Wikidata. Therefore, in an effort to ensure that users can still access inactive websites to verify the information in Wikidata, we initiated a project in the summer of 2020 to archive the official websites for all IUPUI women scholars using either the Internet Archive Wayback Machine or archive.today services. That way, the reference for the statement could include the archive URL (P1065) for the source and archive date (P2960) in addition to the reference URL (P854) and retrieved (P813) date (see Figure 4). Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 9 of 14 Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 9 of 14 Figure 4: Example of an occupation (P106) statement for an IUPUI-affiliated woman scholar containing a supporting reference URL, retrieved date, archived URL, and archived date. Figure 4: Example of an occupation (P106) statement for an IUPUI-affiliated woman scholar containing a supportin reference URL, retrieved date, archived URL, and archived date. Table 2: Sample of Wikidata-related events hosted at UL focused on building capacity and contributing to the enhancement of entries representing IUPUI women scholars Program/event name Date Number of editors Number of women editors Total edits Bringing IUPUI Female Faculty Members to Wikidata November 9, 2017 5 3 744 Wikidata for “Women Creating Excellence at IUPUI” March 7, 2018 7 6 1,050 Wiki Learning Event: Wikidata January 3, 2019 14 8 131 Building Faculty Scholarly Profiles using Wikidata January 14, 2019 7 6 69 able 2: Sample of Wikidata-related events hosted at UL focused on building capacity and contributing o the enhancement of entries representing IUPUI women scholars While SPaS is a more recent effort, UL has been engaging library personnel by organizing and hosting numerous Wikidata edit-a-thons to build capacity and enhance entries for IUPUI women scholars for several years (see Table 2). These events are reflective of UL’s academic open knowledge efforts and commitment to supporting Wikimedia campaigns and other projects that contribute to the collection of knowledge. Representing Women Scholars The development and delivery of hands-on training sessions on how to use Wikidata has been a key element for the success of these initiatives. We endeavored to ensure that both the data that we were contributing and the participants in these events focused on addressing gender inequities. In 2017, the library organized its first Wikidata edit-a-thon, “Bringing IUPUI Female Faculty Members to Wikidata,” with five active editors enrolled who completed nearly 750 edits. In 2018, to honor International Women’s Day, the library organ ized another Wikidata edit-a-thon, “Wikidata for ‘Women Creating Excellence at IUPUI,’” which produced over a thousand edits. During the “Wiki Learning Event: Wikidata” program in January 2019, fourteen edi tors enrolled and recorded 131 edits to enhance existing entries for IUPUI-affiliated scholars. The success of the initial events hosted by UL inspired greater collaboration between UL and other campus libraries such as the Ruth Lilly Medical Library. For example, a one-hour Wikidata workshop hosted by UL later in January 2019, “Building Faculty Scholarly Profiles using Wikidata,” covered the basics of Wikidata and provided par ticipants with the necessary skills to create and edit Wikidata entries for IUPUI women scholars with a focus on the Department of Obstetrics & Gynecology (“IUPUI University Library WikiProject Programs” n.d.; “Wikipedia:GLAM/IUPUI University Library/Events” 2021). Linking Scholars to Their Articles Knowing that there are bots making contributions of article entries to Wikidata, we have prioritized finding Wikidata article entries for works written by IUPUI women scholars and establishing connections between the works and author entries rather than adding new articles. Making these connections is critical when it comes to enhancing the citation graphs of IUPUI-affiliated scholars. The fact that this approach does not require making edits directly in Wikidata, but rather leans on the functionality of the Author Disambiguator tool, has afforded us the opportunity to increase participation from library personnel in support of the SPaS work. Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 10 of 14 page 10 of 14 Table 3: Sample of Wikidata-related events hosted at UL focused on linking IUPUI women scholars to their articles Program/event name Date Number of editors Number of women editors Articles linked Wikidata Editing Competition January 29, 2020 4 1 588 Wikidata Editing Competition 2 February 6, 2020 4 4 367 Linking IUPUI Women Faculty to Their Works in Wikidata March 5, 2020 13 7 2,254 Women Faculty Articles April 10, 2020 5 2 1,665 Table 3: Sample of Wikidata-related events hosted at UL focused on linking IUPUI women scholars to their articles ple of Wikidata-related events hosted at UL focused on linking IUPUI women scholars to Women Faculty Articles April 10, 2020 One way in which this has been accomplished at UL is by continuing to organize and host edit-a-thons and editing competitions. In the first quarter of 2020, four events (see Table 3) were hosted at UL: “Wikidata Editing Competition,” “Wikidata Editing Competition 2,” “Linking IUPUI Women Faculty to Their Works in Wikidata,” and “Women Faculty Articles” (“IUPUI University Library WikiProject Programs” n.d.; “Wikipedia:GLAM/IUPUI University Library/Events” 2021). The objective of these events was to link existing profile entries of IUPUI- affiliated women scholars to their corresponding scholarly article entries already present in Wikidata. Of the various efforts to increase library participation, editing competitions have been the most produc tive. Participants of these events worked from a Google Sheet which provided a list of all of the IUPUI women scholars that were already present in Wikidata at the time of the event. Participants claimed the entry they wanted to work on by adding their Wiki username in a column next to the scholar’s name. Linking Scholars to Their Articles Then, using the Author Disambiguator tool, they searched for the scholar’s name by copying a name from the spreadsheet and pasting it into the “author name” input area of the tool. Clicking on the “Look for author” button retrieved a list of potential publications for the scholar. Next, participants checked each publication to see if it had been authored by the IUPUI scholar or if the publication belonged to someone else. Those that did belong to an IUPUI-affiliated scholar were then linked using the “Link selected works to author” button. Once this process was complete, participants returned to the shared Google Sheet, where they recorded the number of articles that they were able to link to the chosen scholar. Turning this task into a competition by keeping track of who was able to link the most articles to the most scholars helped to turn what might have been seen as a tedious task into a fun and engaging one by providing bragging rights to the winner and a sense of accomplishment to all participants. Conclusion and Future Work In this article, we have shared the different methods used to advance the Scholarly Profiles as Service (SPaS) at IUPUI. This service not only increases the visibility of the campus scholars in an open environment but may also contribute to raising the reputation of our institution. As profiles for our scholars and the works that they created are added to Wikidata, the collective work of the university’s scholars becomes easier to discover and cite. While we see the value of engaging in this type of work, we also understand that integrating it into exist ing, often more traditional, library services can take time. However, following the successful contribution of works produced by three campus schools in 2019, we are enthusiastic about the potential for the SPaS work to include all campus-affiliated schools prospectively. The IUPUI open access policy workflow manages a dataset of more than three thousand articles and other campus publications every year. These data, derived from institutional reports and other sources, are often incomplete (e.g., lacking a DOI) or incorrect (e.g., typos in titles or sources). To date, UL has cleaned the metadata to the point that it can retrieve accurate DOIs, de-duplicate, identify open access status, and (if needed) notify an author for participation in the pol icy. We aim to find ways to build on this work (without overly increasing the labor involved) to systematically and annually add all articles from the IUPUI open access policy workflow to Wikidata. Effectively, this would create a complete collection of scholarly articles by IUPUI authors within a specific year. Libraries have an opportunity to place themselves at the forefront of the open knowledge movement by contributing data to open infrastructures, but convincing library administrators of the importance of taking action now may prove challenging. This may be due, in part, to the nature of the Wikidata environment, where all users have access to contribute, edit, and curate the data. Many libraries may be more comfortable with the controlled environments that they have traditionally used for the curation of knowledge. Library- based Wikidata services, therefore, need consistent and ongoing internal and external outreach. Other SPaS Efforts The potential undercoverage of social science and humanities scholarship in Wikidata is beyond the scope of this article, but the comparative gaps that we found may inform what data we prioritize for future Wikidata contributions. School of Education’s 2019 publications, only one of the forty-five works had been previously added by a bot to Wikidata—an article that was indexed by PubMed. The potential undercoverage of social science and humanities scholarship in Wikidata is beyond the scope of this article, but the comparative gaps that we found may inform what data we prioritize for future Wikidata contributions. Other SPaS Efforts Although we have prioritized creating author entries for IUPUI women scholars and linking those entries to existing entries for articles that those scholars have written, we have also completed work on specific topics and on all authors from selected schools. Specifically, we have made an effort to add or enhance Wikidata entries for all COVID-related works by IUPUI authors, ensure that the scholars were also represented, and link the works to the authors. In addition, we have contributed or enhanced records for all works authored in 2019 by all scholars from three campus schools: Education, Philanthropy, and Public Health. As a feature of IUPUI’s open access policy workflow, the UL manages a complete dataset of all articles authored by campus authors in any given year. If these works are scholarly articles, they meet the criteria for inclusion under the terms of the policy. All articles that are already open access or can be made open access in the institutional repository are deposited in the institutional repository, IUPUI ScholarWorks. The remaining articles are flagged and the IUPUI authors receive an email notification requesting that they send the library a version (typically the accepted manuscript) that can be openly archived. Because this work requires the library to maintain a complete and relatively clean metadata collection for works authored by IUPUI scholars, we have the opportunity to reuse the data in our SPaS effort. In completing this work for the three schools, we contributed or enhanced entries for 198 articles using external tools and utilities to automate the work (e.g., SourceMD, Zotero Wikidata Translator) and linked these articles to their IUPUI authors (either manually or using the Author Disambiguator tool). In doing this work, we learned that some disciplines are more likely to benefit from bots that regularly contribute data from open databases such as PubMed. Thus, for the School of Public Health 2019 articles, more than 90 percent (101 of 111) were already in Wikidata. In contrast, of the forty-two works published in 2019 by the School of Philanthropy authors, none had been previously added to Wikidata. Similarly, for the Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 11 of 14 page 11 of 14 School of Education’s 2019 publications, only one of the forty-five works had been previously added by a bot to Wikidata—an article that was indexed by PubMed. Conclusion and Future Work In addition to this ongoing outreach to stakeholders, one idea we hope to move forward as we continue the SPaS work is the creation of an online submission form to gather consent from scholars to contribute their images to Wikimedia Commons—the media repository for all Wikimedia projects—and for collecting their CVs. These images could either be the ones displayed on the scholar’s institutional web page or another image they have the rights to. For the CVs, we anticipate having to test and adopt PDF data extraction utili ties to be able to more programmatically access specific data from the submitted files. Due to the current focus of the SPaS work on increasing visibility of IUPUI women scholars and their works, we are also in the planning stages for revisiting all entries created to represent authors to ensure that they all include the sex or gender (P21) statement with proper references (see Figure 5). To this end, we have created a model for Figure 5: Example of a reference added to support the sex or gender (P21) statement for an IUPUI-affiliated woman scholar. Figure 5: Example of a reference added to support the sex or gender (P21) statement for an IUPUI-affiliated woman scholar. gure 5: Example of a reference added to support the sex or gender (P21) statement for an IUPUI-affiliated woman scholar. page 12 of 14 Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service how to reference this statement, which includes the reference URL (P854) for the scholar’s institutional web page, the retrieved date (P813), the archive URL (P1065), the archive date (P2960), and a quotation (P1683) to insert a quotation from the scholar’s institutional web page to capture the pronouns they used to describe themselves. Whenever a quotation is added, we include the property based on heuristic (P887) with value inferred from grammatical gender used in text (Q94997488) and a second value, as applicable, to indicate that it was also inferred from given name (Q69652498). By adding references to gender statements on Wikidata, we aim to avoid misgendering. At the same time, we believe that our work on Wikidata entries can be responsive to the expressed gender identities of authors at our university in a way that closed, proprietary systems cannot be. Conclusion and Future Work y We hope that the methods described in this article inspire other academic institutions to take on pilot projects that support ongoing efforts to bridge the gender divide in Wikidata while also contributing to strengthening the connections between academic institutions, their scholars, and their scholarly output. Competing Interests The authors have no competing interests to declare. References Abellán, Jorge. 2019. “Wikidata le da alas al conocimiento abierto.” In The Internet Health Report 2019. https://internethealthreport.org/2019/wikidata-le-da-alas-al-conocimiento-abierto/?lang=es. Archived at https://archive.ph/CuAYF. Abellán, Jorge. 2019. “Wikidata le da alas al conocimiento abierto.” In The Internet Health Report 2019. https://internethealthreport.org/2019/wikidata-le-da-alas-al-conocimiento-abierto/?lang=es. Archived at https://archive.ph/CuAYF. Albert, Paul J., Sarbajit Dutta, Jie Lin, Zimeng Zhu, Michael Bales, Stephen B. Johnson, Mohammad Mansour, Drew Wright, Terrie R. Wheeler, and Curtis L. Cole. 2021. “ReCiter: An Open Source, Identity-Driven, Authorship Prediction Algorithm Optimized for Academic Institutions.” PLoS ONE 16 (4): e0244641. https://doi.org/10.1371/journal.pone.0244641. Allison-Cassin, Stacy, Alison Armstrong, Phoebe Ayers, Tom Cramer, Mark Custer, Mairelys Lemus-Rojas, Sally McCallum, Merrilee Proffitt, Mark A. Puente, Judy Ruttenberg, and Alex Stinson. 2019. “ARL White Paper on Wikidata: Opportunities and Recommendations.” ScholarWorks. https://scholarworks.iupui.edu/ handle/1805/18902. Balch, Oliver. 2019. “Making the Edit: Why We Need More Women in Wikipedia.” The Guardian, November 28, 2019. http://www.theguardian.com/careers/2019/nov/28/making-the-edit-why-we-need-more- women-in-wikipedia. Archived at: https://archive.vn/3L713. Baltz, Samuel. 2021. “Wikipedia’s Political Science Coverage Is Biased. I Tried to Fix It.” The Washington Post, February 24, 2021. https://www.washingtonpost.com/politics/2021/02/24/wikipedias-political-sci ence-coverage-is-biased-i-tried-fix-it/. Archived at: https://archive.ph/PAjhe. g p p j Bear, Julia B., and Benjamin Collier. 2016. “Where Are the Women in Wikipedia? Understanding the Different Psychological Experiences of Men and Women in Wikipedia.” Sex Roles 74 (5): 254–65. https://doi. org/10.1007/s11199-015-0573-y. Cobb, Simon. 2020. “Author Items in Wikidata.” WikiCite Virtual Conference 2020. https://commons.wikimedia. org/wiki/File:Author_items_in_Wikidata.pdf. Archived at: https://web.archive.org/web/20220601053916/ https://upload.wikimedia.org/wikipedia/commons/7/79/Author_items_in_Wikidata.pdf p p g p p “Community Insights/2018 Report/Contributors.” 2019. Meta. https://meta.wikimedia.org/wiki/ Community_Insights/2018_Report/Contributors. Archived at: https://archive.ph/vGKxz. “Community Insights/Community Insights 2021 Report.” 2021. Meta. https://meta.wikimedia.org/wiki/ Community_Insights/Community_Insights_2021_Report. Archived at: https://archive.ph/T8V8B. “Comparison of Research Networking Tools and Research Profiling Systems.” 2021. Wikipedia. https:// en.wikipedia.org/w/index.php?title=Comparison_of_research_networking_tools_and_research_profil ing_systems&oldid=1023327450. Archived at: https://archive.vn/9Jm1R. g y p Ford, Heather, and Judy Wajcman. 2017. “‘Anyone Can Edit’, Not Everyone Does: Wikipedia’s Infrastructure and the Gender Gap.” Social Studies of Science 47 (4): 511–27. https://doi.org/10.1177/0306312717692172. , , J y j y , y p and the Gender Gap.” Social Studies of Science 47 (4): 511–27. https://doi.org/10.1177/0306312717692172. Graells-Garrido, Eduardo, Mounia Lalmas, and Filippo Menczer. 2015. “First Women, Second Sex: Gender Bias in Wikipedia.” In HT ’15: Proceedings of the 26th ACM Conference on Hypertext & Social Media, 165–74. New York: Association for Computing Machinery. https://doi.org/10.1145/2700171.2791036. “Help:Namespaces ” 2022 Wikidata https://wwwwikidata org/wiki/Help:Namespaces Archived at: Graells-Garrido, Eduardo, Mounia Lalmas, and Filippo Menczer. 2015. “First Women, Second Sex: Gender Bias in Wikipedia.” In HT ’15: Proceedings of the 26th ACM Conference on Hypertext & Social Media, 165–74. New York: Association for Computing Machinery. https://doi.org/10.1145/2700171.2791036. References p g y p g “Help:Namespaces.” 2022. Wikidata. https://www.wikidata.org/wiki/Help:Namespaces. Archived at: https://archive.vn/Dr24I. Hill, Benjamin Mako, and Aaron Shaw. 2013. “The Wikipedia Gender Gap Revisited: Characterizing Survey Response Bias with Propensity Score Estimation.” PLoS ONE 8 (6): e65782. https://doi.org/10.1371/ journal.pone.0065782. mus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 13 of 1 page 13 of 14 page 13 of 14 Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service umaniki. n.d. “Gender Gap by Language Editions in Wikimedia Projects.” Accessed June 24, 2021. https:// humaniki.wmcloud.org/gender-by-language. g g y g g diana University. n.d. “Scholars@IU.” Accessed June 3, 2021. https://scholars.proquest.com/gallery/indi ana. Archived at: https://archive.vn/kgHtk. IUPUI Institutional Research and Decision Support. n.d. “Employee Headcount Calculator.” Accessed May 5, 2021. https://tableau.bi.iu.edu/t/prd/views/FacultyStaffandStudentEmployees/EmployeeHeadcountCalculator?: iid=6&:isGuestRedirectFromVizportal=y&:embed=y. Archived at: https://archive.vn/62nqL. pp p y y https://tableau.bi.iu.edu/t/prd/views/FacultyStaffandStudentEmployees/EmployeeHeadcountCalculator?: iid=6&:isGuestRedirectFromVizportal=y&:embed=y. Archived at: https://archive.vn/62nqL. IUPUI University Library. 2019. “Open Values.” https://cds.ulib.iupui.edu/openvalues. Archived at: http archive.vn/hrrvB. PUI University Library. n.d. “Open Access Policy, IUPUI Faculty Council (October 7, 2014).” Accessed June 8, 2021. https://openaccess.iupui.edu/policy. Archived at: https://archive.vn/dgNsP. IUPUI University Library Center for Digital Scholarship. 2020. “IUPUI Open Access Policy: Annual Report for 2019.” https://scholarworks.iupui.edu/handle/1805/24383. IUPUI University Library Center for Digital Scholarship. 2021. “IUPUI University Library Commitment to Open Knowledge.” https://cds.ulib.iupui.edu/digitalscholarship/openknowledge. Archived at: https:// archive.vn/TVcAJ. “IUPUI University Library WikiProject Programs.” n.d. Accessed May 30, 2021. https://outreachdashboard. wmflabs.org/campaigns/iupui_university_library_wikiproject/programs. Archived at: https://archive. vn/QjfhW. j Jacobs, Julia. 2019. “Wikipedia Isn’t Officially a Social Network. But the Harassment Can Get Ugly.” The New York Times, April 8, 2019. https://www.nytimes.com/2019/04/08/us/wikipedia-harassment-wikime dia-foundation.html. Archived at: https://archive.ph/Eye1A. p p y Johnson, Stephen B., Michael E. Bales, Daniel Dine, Suzanne Bakken, Paul J. Albert, and Chunhua Weng. 2014. “Automatic Generation of Investigator Bibliographies for Institutional Research Networking Systems.” Journal of Biomedical Informatics 51 (October): 8–14. https://doi.org/10.1016/j.jbi.2014.03.013. Kinsella, Bret. 2019. “Voice Assistants Alexa, Bixby, Google Assistant and Siri Rely on Wikipedia and Yelp to Answer Many Common Questions about Brands.” Voicebot.Ai. July 11, 2019. http://voicebot. ai/2019/07/11/voice-assistants-alexa-bixby-google-assistant-and-siri-rely-on-wikipedia-and-yelp-to-an swer-many-common-questions-about-brands/. Archived at: https://archive.ph/fGww2. Klein, Maximilian, and Piotr Konieczny. 2015. “Gender Gap Through Time and Space: A Journey Through Wikipedia Biographies and the ‘WIGI’ Index.” ArXiv:1502.03086 [cs.CY]. https://doi.org/10.48550/ arXiv.1502.03086. Koren, Marina. 2018. “One Wikipedia Page Is a Metaphor for the Nobel Prize’s Record With Women.” The Atlantic, October 2, 2018. https://www.theatlantic.com/science/archive/2018/10/nobel-prize-physics- donna-strickland-gerard-mourou-arthur-ashkin/571909/. Archived at: https://archive.vn/Vxyg2. g p yg mus-Rojas, Mairelys. 2019. “Open Knowledge Report (2017-2018).” ScholarWorks. https://scholarworks iupui.edu/handle/1805/18586. Lemus-Rojas, Mairelys, and Jere D. Odell. 2018. References “Creating Structured Linked Data to Generate Scholarly Profiles: A Pilot Project Using Wikidata and Scholia.” Journal of Librarianship and Scholarly Communication 6 (1): p.eP2272. https://doi.org/10.7710/2162-3309.2272. ( ) p p g Lir, Shlomit Aharoni. 2019. “Strangers in a Seemingly Open-to-All Website: The Gender Bias in Wikipedia.” Equality, Diversity and Inclusion: An International Journal 40 (7): 801–18. https://doi.org/10.1108/EDI-10-2018-0198. “List of Wikipedias.” 2021. Wikimedia. Accessed May 5, 2021. https://meta.wikimedia.org/wiki/List_of_ Wikipedias Archived at: https://archive ph/7RWzu ( ) p p g Lir, Shlomit Aharoni. 2019. “Strangers in a Seemingly Open-to-All Website: The Gender Bias in Wikipedia.” Equality, Diversity and Inclusion: An International Journal 40 (7): 801–18. https://doi.org/10.1108/EDI-10-2018-0198. “Li f Wiki di ” 2021 Wiki di A d M 5 2021 h // iki di / iki/Li f , g g y p p q Diversity and Inclusion: An International Journal 40 (7): 801–18. https://doi.org/10.1108/EDI-10-2018-019 “List of Wikipedias.” 2021. Wikimedia. Accessed May 5, 2021. https://meta.wikimedia.org/wiki/List_ Wikipedias. Archived at: https://archive.ph/7RWzu. ist of Wikipedias.” 2021. Wikimedia. Accessed May 5, 2021. https://meta.wikimedia.org/wiki/List_of_ Wikipedias. Archived at: https://archive.ph/7RWzu. p p p Mobley, Greg H. Email correspondence. May 5, 2020. Archives Specialist, IUPUI University Library, Ruth Lilly Special Collections and Archives. Mornati, Susanna. 2019. “DSpace-CRIS: The Free Open-Source CRIS – What’s New in 2019.” euroCRIS. https://dspacecris.eurocris.org/handle/11366/979. p p g Nielsen, Finn Årup, Daniel Mietchen, and Egon Willighagen. 2017. “Scholia and Scientometrics with Wikidata.” ArXiv:1703.04222 [cs.DL]. https://doi.org/10.48550/arXiv.1703.04222. Obeid, Jihad S., Layne M. Johnson, Sarah Stallings, and David Eichmann. 2014. “Research Networking Systems: The State of Adoption at Institutions Aiming to Augment Translational Research Infrastructure.” Journal of Translational Medicine & Epidemiology 2 (2): 1026. https://www.ncbi.nlm.nih.gov/pmc/arti cles/PMC4610407/. ORCID. n.d. “IUPUI.” Accessed June 3, 2021. https://orcid.org/members/0010f00002Jfe4XAAR-iupui. Archived at: https://archive.vn/X35ap. Pellissier Tanon, Thomas, and Fabian Suchanek. 2019. “Querying the Edit History of Wikidata.” The Semantic Web: ESWC 2019 Satellite Events, 161–66. https://doi.org/10.1007/978-3-030-32327-1_32. Lemus-Rojas et al.: Leveraging Wikidata to Build Scholarly Profiles as Service page 14 of 14 Scholars@IU. n.d. Accessed June 3, 2021. https://scholars.proquest.com/gallery/indiana. Archived a https://archive.vn/kgHtk. Scholars@IU. n.d. Accessed June 3, 2021. https://scholars.proquest.com/gallery/indiana. Archived at: https://archive.vn/kgHtk. Scholia n d Acc d May 5 2021 http :// cholia toolforg org/ Archiv d at: http ://archiv ph/vwG1O https://archive.vn/kgHtk. Scholia n d Accessed May 5 2021 https://scholia toolforge org/ Archived at: https://archive ph/vwG1O p g Scholia. n.d. Accessed May 5, 2021. https://scholia.toolforge.org/. Archived at: https://archive.ph/vwG1O Scholia. n.d. Accessed May 5, 2021. https://scholia.toolforge.org/. Archived at: https://archive.ph/vwG1O. Simonite, Tom. 2019. “Inside the Alexa-Friendly World of Wikidata.” WIRED. February 18, 2019. kidata:Statistics.” 2021. Wikidata. https://www.wikidata.org/wiki/Wikidata:Statistics. Archived at: https://web.archive.org/web/20210401012913/https://www.wikidata.org/wiki/Wikidata:Statistics. References https:// www.wired.com/story/inside-the-alexa-friendly-world-of-wikidata/. Archived at: https://archive. ph/6JMiM. p “Statistics.” n.d. WikiCite. Accessed January 4, 2022. http://wikicite.org/statistics.html. Archived at: https:// archive.vn/kdCaR. Sun, Jiao, and Nanyun Peng. 2021. “Men Are Elected, Women Are Married: Events Gender Bias on Wikipedia.” arXiv:2106.01601 [cs.CL]. https://doi.org/10.48550/arXiv.2106.01601. Wagner, Claudia, David Garcia, Mohsen Jadidi, and Markus Strohmaier. 2015. “It’s a Man’s Wikipedia? Assessing Gender Inequality in an Online Encyclopedia.” Proceedings of the International AAAI Conference on Web and Social Media 9 (1). https://ojs.aaai.org/index.php/ICWSM/article/view/14628. Wagner, Claudia, Eduardo Graells-Garrido, David Garcia, and Filippo Menczer. 2016. “Women through the Glass Ceiling: Gender Asymmetries in Wikipedia.” EPJ Data Science 5: 5. https://doi.org/10.1140/epjds/ s13688-016-0066-4. “Wcmc-Its/ReCiter.” 2021. GitHub. https://github.com/wcmc-its/ReCiter. “Wikidata Reconciliation for OpenRefine.” n.d. Wikidata. Accessed December 15, 2021. https://wikidata. reconci.link/. “Wikidata:Relation between properties in RDF and in Wikidata.” 2017. Wikidata. https://www.wikidata.org/ wiki/Wikidata:Relation_between_properties_in_RDF_and_in_Wikidata. Archived at: https://archive. ph/lZLub. p “Wikidata:Statistics.” 2021. Wikidata. https://www.wikidata.org/wiki/Wikidata:Statistics. Archived at: https://web.archive.org/web/20210401012913/https://www.wikidata.org/wiki/Wikidata:Statistics. “Wikidata:WikiProject IUPUI University Library.” 2021. Wikidata. https://www.wikidata.org/wiki/ Wikidata:WikiProject_IUPUI_University_Library. Archived at: https://archive.ph/WHCpQ. j y y p p p “Wikipedia:GLAM/IUPUI University Library/Events.” 2021. Wikipedia. https://en.wikipedia.org/wiki/ Wikipedia:GLAM/IUPUI_University_Library/Events. Archived at: https://archive.vn/xPNC1. How to cite this article: Lemus-Rojas, Mairelys, Jere Odell, Lucille Frances Brys, and Mirian Ramirez Rojas. 2022. Leveraging Wikidata to Build Scholarly Profiles as Service. KULA: Knowledge Creation, Dissemination, and Preservation Studies 6(3). https://doi.org/10.18357/kula.171 How to cite this article: Lemus-Rojas, Mairelys, Jere Odell, Lucille Frances Brys, and Mirian Ramirez Rojas. 2022. Leveraging Wikidata to Build Scholarly Profiles as Service. KULA: Knowledge Creation, Dissemination, and Preservation Studies 6(3). https://doi.org/10.18357/kula.171 Submitted: 25 June 2021 Accepted: 28 April 2022 Published: 27 July 2022 kidata:WikiProject IUPUI University Library.” 2021. Wikidata. https://www.wikidata.org/wiki/ Wikidata:WikiProject_IUPUI_University_Library. Archived at: https://archive.ph/WHCpQ. Submitted: 25 June 2021 Accepted: 28 April 2022 Published: 27 July 2022 Submitted: 25 June 2021 Accepted: 28 April 2022 Published: 27 July 2022 Copyright: @ 2022 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/. Copyright: @ 2022 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/ KULA: Knowledge Creation, Dissemination, and Preservation Studies is a peer-reviewed open access journal published by University of Victoria Libraries.
W3191378658.txt	https://journals.iaepan.pl/khkm/article/download/2791/2673	pl		„By umarłą widzieć… okno się musiało w trunnie wyrząć”. Rozważania o początkach portretu trumiennego	Kwartalnik Historii Kultury Materialnej	2,021	cc-by	8,400	KWARTALNIK HISTORII KULTURY MATERIALNEJ 69 (1), 2021 PL ISSN 0023-5881 www.iaepan.edu.pl the CC BY 4.0 license (https://creativecommons.org/licenses/by/4.0/) DOI: 10.23858/KHKM69.2021.1.004 Aleksander Jankowski „By umarłą widzieć… okno się musiało w trunnie wyrząć”. Rozważania o początkach portretu trumiennego1 Abstrakt: Artykuł dotyczy portretu trumiennego, czyli malowanego na blasze wizerunku nieboszczyka, który przytwierdzany był do trumny na czas egzekwiów. Autor analizuje ten artystyczny fenomen sztuki i kultury śmierci doby staropolskiej oraz związaną z nim problematykę badawczą. Wyjaśnia zagadnienie jego genezy, metryki oraz funkcji w sarmackiej obyczajowości w okresie nowożytnym. Abstract: The article concerns the coffin portrait, i.e. a portrait of the deceased person painted on a tin plate, attached to the coffin for the exequies. The author addresses this phenomenon, characteristic of the art and culture of death in the Old-Polish era, and related research questions, explaining its origin, chronology and functions in the Sarmatian customs of the early modern period. Słowa kluczowe: kultura staropolska, portret trumienny, castrum doloris, egzekwia, uroczystość pogrzebowa, pompa funebris, archimimus, panegiryk pogrzebowy Key words: Old-Polish culture, coffin portrait, castrum doloris, exequies, funeral ceremony, pompa funebris, archimimus, funerary panegyric Portret trumienny, artystyczny fenomen sztuki staropolskiej, uznany za „sól ziemi sarmackiego malarstwa”2, po blisko 100 latach naukowego zainteresowania nadal pobudza do naukowych dociekań, ukierunkowanych zwłaszcza ku mgliście rysującym się początkom zjawiska. Przetrwało blisko 800 tej kategorii zabytków, z czego 350 gruntownie przebadano3. Brakuje jednak obiektów wczesnych, które pozwoliłyby prześledzić proces formowania się portretu trumiennego; nie ma też staropolskich źródeł pisanych bezpośrednio odnoszących się do konterfektów trumiennych i ich funkcji w kulturze funeralnej4. Z kolei dość liczne relacje z pogrzebów i rachunki kosztów egzekwiów dotyczą uroczystości magnackich z drugiej połowy XVII stulecia i młodszych, a więc odbywających się w czasach, gdy w tym środowisku malowany portret trumienny ustępował miejsca portretowi „reprezentacyjnemu”, eksponowanemu w zwieńczeniu castri doloris5. 1 Problem funkcji portretu trumiennego po egzekwiach zostanie podjęty przez autora w kolejnym artykule, który ukaże się na łamach „Kwartalnika Historii Kultury Materialnej” w roku 2022. 2 Ryszkiewicz A. 1974, s. 38. 3 Są to przede wszystkim zbiory muzealne. Stan badań nad portretem trumiennym szerzej omówiony w: Dziubkowa J. 1996, s. 16–19. 4 Portrety trumienne bardzo rzadko wzmiankowane są w protokołach wizytacji biskupich. Nieliczne i bardzo lakoniczne informacje w tych źródłach odnaleźli: Łukaszewicz J. 1858, s. 257 (w Ceradzu Kościelnym); Szołdrski W. 1929–1931, s. 70–72 (kościół dominikanów w Toruniu). Materiały ikonograficzne to zaledwie kilka rycin i rysunkowych projektów katafalków. 5 Chrościcki J.A. 1974, s. 70. 58 ALEKSANDER JANKOWSKI Malowany na blasze popiersiowy portret osoby zmarłej — powszechny w dobie staropolskiej element „pogrzebnej apperencyi”6 szlachetnie urodzonych — zaistniał w naukowych opracowaniach dopiero w roku 1933, przy okazji wystawy zabytków z czasów Stefana Batorego i Jana III Sobieskiego, zorganizowanej w Muzeum Wojska w Warszawie (przy udziale Muzeum Narodowego). W przygotowanym na tę okoliczność katalogu, w dziale tzw. „portretów różnych” wymieniono kilka wizerunków namalowanych na blachach o kształcie owalnym, nazywając je „nagrobnymi” oraz pięć konterfektów namalowanych na blaszanym podobraziu o polu wielobocznym, i te określono „trumiennymi”7, niejako parafrazując przyjęty dla nich trzy lata wcześniej przez Alfreda Brosiga termin „tabliczki z trumien”8. Brosig pojęciem tym konstatował swe pionierskie wówczas (w 1930 r.) spostrzeżenia: „Owe charakterystyczne sześciokątne, na miedzi, wzgl. na blasze cynowej malowane portrety […] pojawiają się w Polsce z początkiem XVII wieku. Stanowią zupełnie odrębną grupę wśród zabytków naszego malarstwa. Są to bowiem wizerunki nieboszczyków, pochodzące z trumien, do których wiek pierwotnie były przymocowane; stąd też ich kształt [...]. Wielka szkoda, że nikt nie zajmował się dotąd kwestią powstania i pochodzenia tego bądź co bądź oryginalnego zwyczaju i nie zbierał wiadomości o tych tablicach portretowych, które porozwieszane w wielkiej nieraz ilości w naszych kościołach lub porozrzucane po zakrystjach i zakamarkach, giną bezpowrotnie”9. Podejmowane z czasem próby bardziej syntetycznego ujęcia zagadnienia portretu trumiennego dotyczyły głównie kwestii jego genezy. Do dziś jednak pozostała ona otwarta. W dociekaniu źródeł przywoływano bowiem różne czynniki (społeczne i twórcze), o różnej metryce. W jednej optyce rozważano początki obrazowania nieboszczyka, konwencję wizerunku, blaszane podobrazie, kształt obrazowego pola itp. Co więcej, od początku, czyli od ustaleń wspomnianego Alfreda Brosiga i autorów Katalogu wystawy jubileuszowej zabytków z czasów króla Stefana i Jana III Sobieskiego… z 1933 r.10, portret trumienny niejako z definicji związano wyłącznie z liturgią pogrzebową i drewnianą trumną o wielobocznym przekroju, do której rzekomo dostosowywano kształt blachy. A przecież trumny o takim kształcie stosowano częściej dopiero od połowy XVII w. Tym samym „wieloboczny kształt blachy” jako jedno z podstawowych kryteriów portretu trumiennego nie uwzględnia wcześniejszych form trumien, gdy skrzynie miały z reguły przekrój prostokąta zbliżonego do kwadratu, ewentualnie „zdecydowanego trapezu”11. Wielobok większości blach raczej nie odwzorowuje ani przekroju, ani czoła trumny, a szczególnie blachy o kształcie regularnego ośmioboku. W literaturze przedmiotu i upowszechnianych w niej cechach portretu trumiennego marginalizowano dość licznie zachowane „dwustronne” blachy. Portret z rewersem zawierającym 6 Pojęcia „pogrzebna apperencyia” na określenie portretu trumiennego i pozostałych elementów mocowanych na trumnie (form heraldycznych, inskrypcyjnych i dekoracyjnych) użył kaznodzieja karmelitów bosych o imieniu zakonnym Anioł od św. Teresy, w kazaniu na pogrzebie Jana Mikołaja Daniłowicza, Anioł od św. Teresy. 1651, bez paginacji; Nowicka-Struska A. 2006, s. 79. 7 Katalog wystawy jubileuszowej zabytków. 1933, s. 88–90. 8 Brosig A. 1930, s. 327–334. 9 Brosig A. 1930, s. 327. 10 Katalog wystawy jubileuszowej zabytków. 1933, s. 88–90. 11 Drewniane trumny używane były w polskiej kulturze funeralnej częściej od XIII–XIV w. Upowszechniały się w późnym średniowieczu i w dobie nowożytnej, na długo jednak (przynajmniej do połowy XVIII w.) pozostały wyrazem statusu społecznego, a nawet świadectwem pewnej zamożności. W środowiskach wiejskich i mniej majętnych ciała grzebano tradycyjnie — owinięte „całunem”, ułożone w wydrążonej kłodzie, albo na desce. Ten ostatni zwyczaj utrwaliło przysłowie: „do grobowej deski” (por. Labudda A. 1983, s. 210). W kulturze sepulkralnej bogatszych warstw społecznych jakość trumny (zastosowany materiał i dekoracje) była środkiem wyrazu świeckiej reprezentacji. Trumny, w których składano zmarłych przedstawicieli patrycjatu, bywały bardzo ozdobne i konieczne było wydawanie specjalnych rozporządzeń, by pohamować przepych w ich zdobnictwie (zob. Cieślak K. 1992, s. 90). ROZWAŻANIA O POCZĄTKACH PORTRETU TRUMIENNEGO 59 istotne dopełnienia identyfikujące wizerunek i charakteryzujące status zmarłego (inskrypcyjne i heraldyczne) nie był zgodny z przyjętymi założeniami integralnego związku konterfektu z czołem trumny. Tym bardziej że portrety trumienne z semantycznym rewersem mają zaczepy ewidentnie służące do ich zawieszania, tak by możliwy był „dwustronny” ogląd wizerunku. Studia nad portretem trumiennym sporadycznie odnosiły się również do tych konterfektów, które były mocowane do trumien metalowych (tj. cynowych, miedzianych, ołowianych), nazywanych w literaturze sarkofagami. Te także nie odpowiadały przyjętej definicji, gdyż bywały malowane bezpośrednio na czole skrzyni, a ich pole miało kształt czworoboczny, tak jak w przypadku portretu trumiennego Jana Karola Opalińskiego (zm. w 1695 r.)12 (ryc. 1). Ryc. 1. Trumna cynowa i portret trumienny Jana Karola Opalińskiego, z ok. 1695 r., w Muzeum „Zamek Opalińskich” w Sierakowie, źródło: https://pl.wikipedia.org/wiki/ Jan_Karol_Opaliński#/media/Plik:JanKarolOpalinski_sarcophagus.jpg (dostęp 25.06.2020) Fig. 1. The tin coffin and coffin portrait of Jan Karol Opaliński, c. 1695, in the Opaliński Castle Museum in Sieraków, source: https://pl.wikipedia.org/wiki/Jan_Karol_Opaliński#/media/ Plik:JanKarolOpalinski_sarcophagus.jpg (accessed 25.06.2020) Wątpliwa jest też ugruntowana w literaturze przedmiotu metryka portretu trumiennego lokowana gdzieś w połowie XVII w. w Wielkopolsce13. Za podstawowy argument takiej proweniencji portretu trumiennego uznano dominację zabytków z tego regionu. Ignorowano wnioski wynikające z innego materiału źródłowego — ze skąpych, ale istniejących archiwaliów. Dowodzą one, że praktyka malowania post mortem portretu na blasze przeznaczonego na ceremonię pogrzebową miała miejsce w kręgu malarzy krakowskich już w XVI w., a dotyczyła 12 O takich rozwiązaniach wzmiankują też kontrakty; wiadomo np. że w połowie XVII w. rodzina wojewody inowrocławskiego Hieronima Radomickiego zleciła konwisarzowi, Maciejowi Gerethowi, wykonanie metalowej trumny, na której znaleźć się miał portret zmarłego, Bystroń J. 1994, s. 100–101. 13 Dziubkowa J. 1996, s. 24; Mrozowski P. 2000, s. 77. 60 ALEKSANDER JANKOWSKI pochówków spoza Wielkopolski. W rachunkach z pogrzebu Krystyny z Radziwiłłów Zamoyskiej, który odbył się w Warszawie w 1580 r., wśród kosztów wymieniono kwotę dla malarza, który „conterfektował nieboszczkę Panią po śmierci”14. Wizerunek wykonał malarz sprowadzony specjalnie z Krakowa („któremu się posłało na drogę z Krakowa i na odiezdnim […], i któremu zapłacono za puzdro na obraz”)15. Z zestawienia wydatków poniesionych w związku z tą uroczystością wynika, że na jednym z czół trumny znajdowała się „tabliczka złotha z napisem, która na cziele z łańcuszka srebrnego wisi”, na wieku zaś „przibita druga mosiądzowa”16. Wymieniony „conterfekt nieboszczki” Zamoyskiej przytwierdzono najpewniej do drugiego czoła (wezgłowia) trumny. Zamówienie na portret zmarłej skierowane z Warszawy do Krakowa świadczy, że już wówczas praktykowano użycie podczas ceremonii pogrzebowej wizerunku osoby zmarłej namalowanego na blasze, choć nie wszędzie jeszcze wtedy takie wykonywano. Teza o „wielkopolskiej kolebce” portretu trumiennego nie wytrzymuje krytyki także w zestawieniu ze źródłami ikonograficznymi. Wystarczy przywołać pochodzący z ok. połowy XVII w. szkic katafalku wykonany przez Giovanniego Battistę Gisleniego, z uczytelnionym na czole trumny wizerunkiem zmarłego17. Architekt pracujący dla warszawskiego dworu Wazów, chętnie zatrudniany też przez możnych w Krakowie, Wilnie i we Lwowie (nigdy w Wielkopolsce), nie wykreował jakiejś pionierskiej, artystycznej wizji trumny z wyobrażeniem nieboszczyka. Po prostu odwzorował zwyczajowy element scenografii ceremonii pogrzebowej. Podobnie potraktował portret trumienny Tylman z Gameren w szkicu castrum doloris Gryzeldy z Zamoyskich Wiśniowieckiej, zaznaczając na przedniej ścianie trumny wizerunek zmarłej (ryc. 2). Tożsamość staropolskiej kultury „urodzonych” i ideologii sarmatyzmu zainspirowała badaczy do cofnięcia się w poszukiwaniach źródeł portretu trumiennego aż do starożytności. Jako pierwszy sugestię o takiej genezie zwyczaju prezentacji malowanego oblicza nieboszczyka w egzekwiach dał wyżej wspomniany Alfred Brosig (w publikacji z 1930 r.): „Nie wiem, skąd w Polsce wziął się ten zwyczaj umieszczania na wiekach trumien podobizn umarłych. Przypomina on podobny zwyczaj Egipcjan, a mianowicie owe znane egipskie, grecko-egipskie i hellenistyczne portrety mumij, malowane na drewnianych deseczkach enkaustyką, tj. gorącemi farbami woskowemi. Czy odgrywała tu rolę jakaś stara tradycja ludów słowiańskich czy raczej zachodzą w naszym wypadku wpływy południowe, nie śmiem rozstrzygnąć […]”18. To luźne i pełne powątpienia skojarzenie trumiennego konterfektu z mumiami, powróciło ostatnio (w 2015 r.) w teorii egiptologa Piotra Otto Scholza. Analizując pamiętnik Mikołaja Krzysztofa Radziwiłła „Sierotki” z podróży do Ziemi Świętej, Syrii i Egiptu (w latach 1582–1584)19 badacz dostrzegł w staropolskich wizerunkach zmarłych symptom hellenistycznego ducha czasu20. Wychodząc z neoplatońskiej koncepcji bytu Karla Jaspersa (osi czasu)21 Scholz uznał, że portret trumienny jest pochodną portretu fajumskiego, a dokładnie, że „portret fajumski jest protoplastą trumiennego”22. Ta konstatacja całkowicie jednak pozbawia trumienny konterfekt kulturowego kontekstu, określanego przez mentalność szlachty i specyficzną sarmacką pobożność23. Ponadto 14 Rachunki z pogrzebu Krystyny z Radziwiłłów Zamoyskiej Kanclerzyny Koronnej, wyprawionego w kościele św. Anny w Warszawie w 1580 r., dokument opublikowany w: Chrościcki J.A. 1974, Aneks 1, s. 267–271. 15 Chrościcki J.A. 1974, s. 267. 16 Chrościcki J.A. 1974, s. 267. 17 Rysunek piórkiem z tzw. Notatnika Gisleniego, reprodukcja w: Wiliński S. 1958, il. 3. 18 Brosig A. 1930, s. 327. 19 Radziwiłł M.K. 1925. 20 Scholz P. O. 2015, s. 73–81. 21 Scholz P. O. 2015, s. 74. 22 Scholz P. O. 2015, s. 75. 23 Problematykę sarmackiej pobożności porusza m.in.: Tazbir J. 1970, s. 7–37; Nowicka-Jeżowa A. 1992, s. 121–137; Łukarska B. 2018. ROZWAŻANIA O POCZĄTKACH PORTRETU TRUMIENNEGO 61 sam Radziwiłł widział w mumiach tylko „maszkary na twarzach malowane i pozłociste”24. Dwie, jakie zakupił („mężczyzny i białej głowy […], obie za dwa cekiny”)25, a które zdaniem Scholza stały się wzorcem dla portretu trumiennego, autor Peregrynacji osobiście wyrzucił za burtę podczas sztormu, uznając, że to właśnie te „ciała pogańskie […] jako że bez pohyby są w opatrzności i straży szatańskiej”26 nawałnicę sprowadziły. W antycznych funeraliach genezę portretu trumiennego dostrzegł także Tadeusz Dobrzeniecki27. Za jego „odległy prototyp” przyjął imago clipeata, reliefowy, popiersiowy wizerunek na sarkofagu, mający uobecniać zmarłego hic et nunc28, a zarazem portret fajumski, ze względu na zawartą w nim sugestywną iluzję ożywionego oblicza śledzącego wzrokiem widza. Znajomość tych wzorów była, zdaniem Dobrzenieckiego, naturalnym rezultatem zainteresowania starożytnością i ukształtowaniem się mitu o sarmackiej genealogii „urodzonych”29. Tendencja do przypisywania Ryc. 2. Szkic projektu castrum doloris Gryzeldy Konstancji Wiśniowieckiej do kościoła Świętego Krzyża w Warszawie, szlachcie wyznania katolickiego Tylman z Gameren, 1672 r., ze zbiorów Gabinetu Rycin fascynacji przedchrześcijańskim Biblioteki Uniwersyteckiej w Warszawie, Archiwum Tylmana, dziedzictwem antyku, skutkująnr inw. GR BUW, Inw.G.R.6425 (AT874) cej (i to w czasie kontrreformaFig. 2. The design of Gryzelda Konstancja Wiśniowiecka’s cji) swobodną adaptacją zwyczacastrum doloris for the Holy Cross church in Warsaw, jów i artystycznych dokonań by Tylman van Gameren, 1672, from the collection of the tamtej epoki na potrzeby miejUniversity Library in Warsaw, the Tylman Archive, scowych, nowożytnych praktyk inventory no. GR BUW, Inw.G.R.6425 (AT874) 24 Radziwiłł M.K. 1925, s. 112. Radziwiłł M.K. 1925, s. 113. 26 Radziwiłł M.K. 1925, s. 131. 27 Dobrzeniecki T. 1990, s. 86–87. 28 Dobrzeniecki T. 1990, s. 82, 85. 29 Zaledwie 20% portretów trumiennych odpowiada schematowi kompozycyjnemu wizerunków rzymskich, Dziubkowa J. 1996, s. 18. 25 62 ALEKSANDER JANKOWSKI religijno-funeralnych, nie wytrzymuje krytyki. Ani ideologia sarmatyzmu z mitem o starożytnych korzeniach szlachty polskiej, ani jaspersowska oś czasu, nie uprawniają do konstatacji odrywających portret trumienny od lokalnej tradycji funeralnej, potrydenckiej pobożności (osadzonej w duchowości średniowiecza) bądź mentalno-kulturowych skutków nowej formuły państwa (monarchii elekcyjnej), a później „potopu” szwedzkiego. Nie można też analizować zjawiska portretu trumiennego bez obrzędowości pogrzebowej, kazań i oracji, bez stawiania pytań o jego funkcje w różnych fazach egzekwiów oraz losy i ideowe znaczenie po takiej ceremonii. Pierwsze przejawy zwyczaju przytwierdzania do trumny malowanego na blasze oblicza zmarłego poświadczone są, o czym już wspomniano, w XVI stuleciu. Zbiegły się z początkami sztalugowego portretu w sztuce polskiej. Oba zjawiska wiążą się z krakowską działalnością Marcina Kobera, nadwornego malarza króla Stefana Batorego. W warsztacie tego artysty powstał m.in. niewielki popiersiowy portret monarchy30 namalowany farbami olejnymi na miedzianej blasze (w owalnym, elipsoidalnym polu, o wymiarach 13,2 × 18,2 cm), a następnie przytwierdzony do trumny zmarłego króla podczas uroczystości pogrzebowej. Konterfekt, znaleziony w 1877 r. w odkrytej przypadkowo krypcie kaplicy mariackiej katedry wawelskiej, jest najstarszym zachowanym dowodem wykonania malunku ukazującego oblicze zmarłego złożonego w zamkniętej trumnie. Tę niedostępną przez stulecia kryptę, „nader szczupłą, do której otwór zaginął w czasie budowania późniejszych grobów królewskich”31, lustrował jako pierwszy profesor Józef Łepkowski. W protokole zanotował: „Trumna cynowa rozpadła. Druga drewniana, ta na której była osadzona metalowa, spróchniała na drzazgi rozpadła. Trzecia drewniana, trzymająca się całości, chociaż spróchniała — oblana żywicą, odziana aksamitem — z antabami żelaznemi bez ozdób[…]”32. Z krypty, spomiędzy tych szczątków wydobyto portret króla Stefana. Podczas zrealizowanej nakładem hrabiny Katarzyny Potockiej „nieumiejętnej i powierzchownej renowacji trumny metalowej”33 portret przytwierdzono do cynowego wieka. Znajdujący się w tym miejscu konterfekt widział Ignacy Józef Kraszewski, który w wydanych w 1888 r. Wizerunkach książąt i królów polskich, pisał: „Na wieku cynowej trumny zwierzchniej Chrystus Pan z N. Panną u krzyża; a na srebrnej blasze epigraf. Obok niego wizerunek króla Stefana pod szkłem, olejno na cynie malowany, podobny do portretu u kks. Missyonarzy w Krakowie”34. W roku 1908 na wniosek Leonarda Lepszego portret złożono w skarbcu katedralnym, a podczas kolejnej renowacji sarkofagu (w 1930 r.), w miejscu oryginału przytwierdzono kopię35. Cynową trumnę króla Stefana Batorego wykonano w gdańskiej pracowni konwisarskiej Daniela Giselera I. W roku 1588 dostarczono ją do Łobzowa. Tu, „[…] w zameczku przed miastem, bo tam ciało królewskie z Grodna przywiezione w Sobotę złożone było, które na wóz w trumnie cynowej do tego przygotowanej włożywszy tym porządkiem do miasta [Krakowa — A.J.] prowadzone było”36. Procesję dokładnie zrelacjonował anonimowy świadek wydarzenia37. Zanotował on: „Wieziono ciało króla Stefana na wozie wielkim, dwunastą koni pod czarnem aksamitem […]”38. Dalej podał informację posłyszaną bądź zaczerpnięta z jakiegoś 30 Konterfekt przypuszczalnie dublował miniaturowy wizerunek króla Stefana pędzla Marcina Kobera, powstały jako dar dla arcyksięcia Ferdynanda (do jego kolekcji portretów monarchów europejskich), Kopera F. 1926, s. 156–157. 31 Kruszyński T. 1932, s. II. 32 Fragment protokołu prof. Józefa Łepkowskiego przytoczony w: Kruszyński T. 1932, s. II. 33 Tak krytyczną ocenę wyrażono w 1930 r., przy okazji kolejnej renowacji metalowej trumny Stefana Batorego podjętej z inicjatywy i nakładem Uniwersytetu Stefana Batorego w Wilnie, Kruszyński T. 1932, s. II. 34 Kraszewski J.I. 1888, s. 305. 35 Kruszyński T. 1932, s. III. 36 Pogrzeb Stefana Batorego w Krakowie. 1860, s. 457. 37 Kronika mieszczanina krakowskiego. 1930, s. 60–64. 38 Kronika mieszczanina krakowskiego. 1930, s. 63. ROZWAŻANIA O POCZĄTKACH PORTRETU TRUMIENNEGO 63 źródła: „Trunnę dębową w trunnę cynową z ciałem włożyli, która bardzo osobliwą robotą była we Gdańsku urobiona, cudnemi kunszty odlewana”39. Samą trumnę zapewne widział, bo szczegółowo opisał: „Była też w tej trunnie po boku uczyniona tabliczka pod szkłem malowana twarz króla Stefana własna, jako żywa co ją mógł dobyć i zasie wetknąć”40. W ceremoniach odbywających się w katedrze ów „krakowski mieszczanin” nie uczestniczył (co sam zaznaczył), ale posiłkując się relacjami świadków odnotował: „pochowano ciało za wielkim ołtarzem, na zamku w kościele w kaplicy i nowy grób z gruntu murowany, w którem leży. W tejże trunnie cynowej Batory Stefan Pierwszy, król Polski”41. Z zapisu wynika, że w trakcie uroczystości eksponowano trumnę cynową (mieszczącą trumnę dębową z ciałem), a na metalowej burcie przytwierdzono („pod szkłem”) portret króla. Niewykluczone, że umiejscowienie konterfektu „po boku” trumny a nie na czole było w XVI w. typowe. Prawdopodobnie najpierw wizerunek umocowano na trumnie z drewna, do której złożono ciało króla w Grodnie. Gdy dwa lata później (w maju 1588 r.) trumnę drewnianą zdeponowano w cynowej, portret zapewne przełożono na burtę tej drugiej, i to w tym miejscu widział wizerunek autor kroniki. Owalna „tabliczka pod szkłem malowana z twarzą króla Stefana własną, jako żywą”, czyli namalowana na blasze, przytwierdzona do bocznej ściany trumny i tak włączona do egzekwiów, to portret trumienny. I jest nim niezależnie od kształtu pola obrazu42. Alternatywnym, stosowanym jeszcze w XVII w. zwyczajem, było ukazywanie w egzekwiach oblicza zmarłego przez otwór wycięty w trumnie. Świadczą o tym opisy funeraliów magnackich. W 1641 r. Jan Kmita, relacjonując pogrzeb Krzysztofa II Radziwiłła, odnotował: „Na tym Catafalku stało Ciało w Trunnie Axamitem Karmazynowym y ćwiekami wielkimi Srebrnymi po szyrokim pasamunie srebrnym bogato obitey. Nad okienkiem [podkreślenie — A.J.] Tablica Srebrna z napisem w głowach. […] Na Tablicy nad okienkiem napis […]”43. W tym samym czasie explicite „okno w trumnie” odnotowuje Stanisław Oświęcim: „gdy kochana siostra [Anna — A.J.] […] przed południem ducha swego panieńskiego […] Panu Bogu oddała, a nas pozostałych nienagrodzonego nigdy nabawieła żalu a niemniej wszytkę okolicą i sąsiadów, którzy jako za żywota przystojność jej i urodę kochali, tak i po śmierci bez przestanku ciało nawiedzali, życząc ją jeszcze i umarłą widzieć, dla czego i okno się musiało w trunnie wyrząć”44 [podkreślenie — A.J.]. Śladem tej praktyki jest metaforyczne określenie trumny jako „domku drewnianego” używane jeszcze w końcu XVII w., np. w legendzie epitafijnej portretu Anny Mielęckiej (zm. w 1694 r.)45. „Wyrzęte w trunnie okno” szklono. W roku 1629 książę Krzysztof II Radziwiłł, organizując pogrzeb siostry stryjecznej Katarzyny Gorajskiej (z Radziwiłłów), zadysponował: „[…] trzeba posłać do Wilna dla zrobienia ćwieczków posrebrzanych, żeby się nimi chędogo truna obiła […]. Także na szkło francuskie do truny”46. Takie przeszklone „okno” dawało przypuszczalnie możliwość oglądania twarzy zmarłej. Chronologia przytoczonych zapisów pozwala na domniemanie, że w pierwszej połowie XVII w. stosowano równocześnie dwa sposoby ukazywania oblicza nieboszczyka w trumnie: przez „wyrzęte okno” (przeszklone) oraz „malowaną twarzą pod szkłem”. Ponieważ „pod 39 Kronika mieszczanina krakowskiego. 1930, s. 63 Kronika mieszczanina krakowskiego. 1930, s. 63. 41 Kronika mieszczanina krakowskiego. 1930, s. 63. 42 Np. Joanna Dziubkowa, odnosząc się do portretu z sarkofagu króla Stefana Batorego, stwierdziła, że nie jest on portretem trumiennym, gdyż nie spełnia kryteriów „typowego portretu trumiennego w sensie roli, jaką mu wyznaczono w sarmackiej ceremonii pogrzebowej”. Przyjmuje, że konterfekt jest „incydentalnym zabytkiem”, będącym zapowiedzią zwyczaju kształtującego się dopiero przed połową XVII w., Dziubkowa J. 1996, s. 21. 43 Kmita J. 1641, bez paginacji. 44 Oświęcim S. 1907, s. 193. 45 Dziubkowa J. 1974, s. 112. 46 Cytat za: Jarczykowa M. 2012, s. 86. 40 64 ALEKSANDER JANKOWSKI Ryc. 3. Portret trumienny Adama Sędziwoja Czarnkowskiego, przed rokiem 1627, na przedniej ścianie trumny cynowej, źródło: https://sarmatyzm.files.wordpress.com/ 2013/11/img_97021.jpg (dostęp 25.06.2020) Fig. 3. The coffin portrait of Adam Sędziwój Czarnkowski, before 1627, on the front board of a tin coffin, source: https://sarmatyzm.files.wordpress.com/ 2013/11/img_97021.jpg (accessed 25.06.2020) szkłem” był także wspomniany konterfekt Krystyny z Radziwiłłów Zamoyskiej (w rachunkach za trumnę uwzględniono też pozycję „skło”)47, można sądzić, że w XVI stuleciu malowany wizerunek zmarłego przytwierdzany do trumny, czyli wczesny portret trumienny, również był szklony. Zarazem źródła sugerują, że zwyczaj umieszczania na trumnie malowanego portretu wyprzedzał praktykę wykonywania w niej „okien”, udokumentowaną dopiero w zapisach siedemnastowiecznych. Jednak wobec skąpego i niejednoznacznego w wymowie materiału archiwalnego nie można wykluczyć, że realistycznie malowany wizerunek „pod szkłem” jest chronologicznie młodszym, elitarnym odpowiednikiem otworu wyciętego w trumnie i oszklonego. Przeszklony konterfekt, tworząc iluzję „okna”, przez które widać zmarłego z „twarzą […] własną, jako żywą”, przede wszystkim gwarantował niezmienność oblicza przedstawionej osoby, co miało szczególne znaczenie, gdy organizacja pogrzebu trwała wiele tygodni, a bywało, że i miesięcy. Dowodem bezpośrednich skojarzeń portretu trumiennego z „oknem” jest konterfekt Adama Sędziwoja Czarnkowskiego (zm. w 1627 r.) z czoła prostopadłościennej skrzyni trumny cynowej (ryc. 3). Ten malowany wizerunek wpasowano w „arkadowe” pole o dekoracyjnej (jakby architektonicznej) oprawie, zespolone z parą skrzydeł budzących jednoznaczne skojarzenia z okiennicami. Zamknięte skrzydła ukazywały herby Czarnkowskich, otwarte — ujmowały portret alegoriami Wiary i Roztropności48. 47 Chrościcki J.A. 1974, s. 267. Cynowa trumna jest dziełem poznańskiego warsztatu konwisarskiego Jakuba Kanadeja. „Okiennice” nie zachowały się. Pisze o nich m.in.: Łukaszewicz J. 1858, s. 187; Wiliński S. 1958, s. 39. Portret trumienny Adama Sędziwoja Czarnkowskiego — skrzydła („okiennice”) i widok cynowej trumny, reprodukcja w: Dziubkowa J. 1996, s. 46–47. 48 ROZWAŻANIA O POCZĄTKACH PORTRETU TRUMIENNEGO 65 Przekazy wzmiankujące trumny z przeszklonymi „oknami” oraz przytwierdzane do trumien „portrety pod szkłem” wymagają analiz porównawczych, tak by, w pierwszej kolejności usystematyzować ich chronologię. Potrzebne są także dalsze kwerendy archiwalne, które przyczynią się do zwiększenia bazy źródłowej. Przy obecnym stanie wiedzy można wstępnie założyć, że kres tradycji „wyrzynania okna” nastąpił wraz z upowszechnieniem się zwyczaju mocowania konterfektu na czole trumny, czego najstarszym zachowanym przykładem jest wyżej wymieniony portret Adama Sędziwoja Czarnkowskiego. Portret przytwierdzony w tym miejscu był dobrze widoczny, co pozwalało na eksponowanie twarzy zmarłego w czasie egzekwiów, zapewniając jej dobrą widoczność in gremio. Przede wszystkim jednak umieszczenie na trumnie oblicza nieboszczyka ‒ obrazu ukazującego „twarz […] własną, jako żywą”, wymownie oddawało ideowo-religijną istotę liturgii pogrzebowej, ceremonialnego pożegnania ciała złożonego w trumnie, a zarazem „uwolnienia” i pożegnania nieśmiertelnej duszy, odchodzącej w zaświaty na spotkanie z Bogiem „twarzą w twarz”49. Realistyczne oblicze ożywione szeroko otwartymi oczami reprezentowało wieczną kwintesencję (istotę) konkretnej osoby ludzkiej, którą św. Augustyn określił jako homo interior (człowiek wewnętrzny) i utożsamił z duszą; duszą — zwróconą w stronę tego, co ponadczasowe i niezmienne, duszą — poszukującą swego Stwórcy, stającą przed nim „twarzą w twarz”50. Uczestnicy ceremonii pogrzebowej nie zawsze byli świadomi owej symboliki — portretu trumiennego jako substytutu homo interior osoby zmarłej, wyrażającego nieśmiertelną duszę. Wierzono, że taki konterfekt zapewnia fizyczną obecność zmarłego w liturgii pogrzebowej. Tym religijnym przekonaniem kierował się hetman wielki litewski Michał Kazimierz Pac, pisząc o swoim portrecie trumiennym: „obraz osobę moią representuiący pod czas depositiy Ciała mojego”51. Portret trumienny był częścią szerszej, teatralnej mistyfikacji udziału nieboszczyka w uroczystości pogrzebowej. Pełny scenariusz tej ceremonii, właściwy kulturotwórczym elitom doby staropolskiej, zakładał również uobecnienie osoby nieżyjącej hic et nunc — żywym sobowtórem, tzw. archimimusem. I w tym przypadku wzorem była ceremonia królewska, tj. średniowieczna tradycja królewskich uroczystości pogrzebowych (zaczerpnięta z antycznych jeszcze ceremoniałów cesarskich)52, w XIV stuleciu przyjęta w Królestwie Polskim53 i kontynuowana w pochówkach królów elekcyjnych. Ryt królewskiego pogrzebu z udziałem archimimusa, czyli sobowtóra odgrywającego króla (ipsius regi mortui personam representans), wprowadzono w oficjalnej (drugiej) ceremonii pochówku Kazimierza III Wielkiego, zorganizowanej przez Ludwika I Węgierskiego 19 listopada 1370 r.54 Przebieg ceremonii, zrelacjonowany przez Janka z Czarnkowa55, nawiązał do egzekwiów zmarłego w 1342 r. króla Węgier, Karola Roberta (ojca Ludwika). W kondukcie pogrzebowym króla Kazimierza „jechał rycerz, odziany w złocistą szatę królewską, na pięknym stępaku królewskim, purpurą pokryty, osobę zmarłego króla wyobrażającą”; w świątyni „pod49 Por. 1 Kor 13,12. Wątek „człowieka wewnętrznego” (homo interior) pojawia się w rozważaniach biskupa Hippony o mocy wiary i odkrywaniu Boga w Trójcy Św. (Augustyn Św. 1962). Interpretację augustiańskiego homo interior m.in. w: Terka M. 2014, s. 410–414. 51 Fragment zapisu testamentowego z 1675 r., Testament Michała Paca. 1840, s. 66; nowsza edycja testamentu por.: Testament Michała Paca. 2006, s. 132–133. 52 Związki średniowiecznego i antycznego ceremoniału pogrzebowego monarchów nie zostały dotychczas jednoznacznie rozstrzygnięte. Jest bardzo prawdopodobne, że wyraźne podobieństwo scenariuszy uroczystego pożegnania zmarłego i honorowania jego godności jest wynikiem swego rodzaju konwergencji, niezależnej od czasu i miejsca, por. Śnieżyńska-Stolot E. 1975, s. 92. 53 Labudda A. 1983, s. 205. 54 Śnieżyńska-Stolot E. 1975, s. 90. 55 Kronika Jana z Czarnkowa. 1996, s. 36–38. 50 66 ALEKSANDER JANKOWSKI koniuszy przyprowadził zbrojnego rycerza w królewskie szaty odzianego. Ów rycerz osobę „zmarłego króla wyobrażał”56. W tym przypadku tylko jedna osoba uosabiała dignitas monarchy, co różniło krakowską ceremonię od tej zorganizowanej w Budzie, gdy Karola Roberta odgrywało trzech archimimusów, występujących in persona et spiritu eiusdem domini regis57. Obrzęd kontynuowano w trakcie pochówków Jagiellonów. Zygmunt August przyjął za obowiązujący w pochówku królewskim wariant ceremonii pogrzebowej Zygmunta Starego (zaplanowany przez biskupa krakowskiego Samuela Maciejowskiego). Ordo pompae funebris serenissimi Sigismundi Regis Poloniae zastosowano w pogrzebach Zygmunta Augusta i Anny Jagiellonki, a później pierwszych królów elekcyjnych — Stefana Batorego i Zygmunta III Wazy. Porządek uroczystości przewidywał m.in.: mszę z wystawionym ciałem zmarłego (praesente corpore), katafalk z trumną (castrum doloris) oraz udział sobowtóra nieboszczyka w rycerskim rynsztunku — archimimusa, wjeżdżającego konno do świątyni (podczas Pater noster), później z konia spadającego, zrzucającego rycerskie insygnia (miecz, hełm i tarczę) oraz kruszącego narzędzia władzy. W ceremonii pogrzebowej Stefana Batorego po raz pierwszy zastosowano nowy wariant uobecnienia zmarłego — malowany portret przytwierdzony do trumny, a sakralny majestat króla reprezentował „pan podstoli Srzeniawa (Śreniawa) […] osoba w szacie królewskiej”58. Władzę ziemską i dowództwo wojskowe uosabiał „Ferens Wesselin, miłośnik króla Stefana mając miecz goły w ręku […] w kirysie hecowanym czarnym a białem, także i na koniu z rzędem i deką żelazną, bardzo śliczną robotą urobioną”59. W trzecim dniu ceremonii, podczas mszy żałobnej w katedrze wawelskiej, „gdy Pater noster śpiewano, kiryśnik, we zbroi, pan Wesselin, z proporcem wjechał z drzewem, które mu z skarbu dano”60. Później „[…] kiedy Agnus Dei śpiewać zaczęto […] kiryśnik we zbroi do kóru wjechał, poskoczywszy na koniu z gromem z konia spadł”61. Gdy scenariusz ceremonii monarszej skopiowano w egzekwiach magnackich, archimimus był przede wszystkim „kiryśnikiem” uosabiającym wojskowe przywództwo. Jego obecność odnotowano już podczas pogrzebu zmarłego w 1561 r. hetmana Jana Amora Tarnowskiego: „po Kazaniu u ołtarza podawano znaki, tak rycerskie, jako i chrześcijańskie tego pana umarłego […] Brzostowski Kiryśnik z konia spadł z srogim y ogromnym trzaskiem”62. Wraz z dojrzewaniem sarmackiej mentalności podczas pochówków upowszechniały się parateatralne formy świeckich rytuałów typowe dla monarszych ceremonii funeralnych. Okazały pogrzeb szybko stał się wyrazem pozycji społecznej zmarłego i jego rodu. W połowie XVII stulecia pochówki magnackie z udziałem archimimusa zyskiwały już iście paradne, barokowe inscenizacje. Jan Kmita, relacjonując uroczystości z okazji pogrzebu Krzysztofa II Radziwiłła, 56 Kronika Jana z Czarnkowa. 1996, s. 36, 37. Śnieżyńska-Stolot E. 1975, s. 91. Ani na Węgrzech, ani w Polsce nie przyjął się zwyczaj niesienia w królewskiej procesji pogrzebowej kukły wyobrażającej zmarłego władcę (tzw. effigies), wykonanej z drewna, skóry bądź wosku. 58 Kronika mieszczanina krakowskiego. 1930, s. 62. Podstolego koronnego Piotra Śreniawę (Szreniawę) występującego w roli archimimusa, reprezentującego królewski majestat, wymienia także autor źródła: Pogrzeb Stefana Batorego w Krakowie. 1860, s. 458. Natomiast w węgierskiej relacji z pogrzebu Stefana Batorego jako archimimus wzmiankowany jest Zramalla Castelon, Janowski P.J. 2015, s. 43. 59 Kronika mieszczanina krakowskiego. 1930, s. 62. 60 Pogrzeb Stefana Batorego w Krakowie. 1860, s. 461. 61 Pogrzeb Stefana Batorego w Krakowie. 1860, s. 461. 62 Żywot y smierc Jana Tarnowskiego. 1773, s. 118. Zdarzały się też bardziej bezpośrednie nawiązania do egzekwiów monarszych, poprzez wprowadzenie dwóch archimimusów. W roku 1591 podczas pogrzebu kasztelana żarnowskiego Jana Myszkowskiego, który odbył się w Krakowie, w procesji „jachał kiryśnik i giermek z kopiją, z tarczą”, za nimi zaś „osoba pańska w szatach nieboszczykowskich, na koniu bardzo cudnym”, Kronika mieszczanina krakowskiego. 1930, s. 90. 57 ROZWAŻANIA O POCZĄTKACH PORTRETU TRUMIENNEGO 67 wspomina o łożu „w nogach którego siedział na Krześle Kirys bogaty pod Kitą: z Szablą Buławą y Kałkanem złotym z Kamieńmi reprezentujący Osobę Nieboszczykowską”63. Później, w kondukcie przed trumną (na wozie ciągniętym przez osiem koni) „między Dardami iechał P. Marcin Oborski w Szatach Nieboszczyka Xcia na Koniu, ktorego on nayczęściey zażywał zwyczaynie ubranym, który tak dobrze et facie et gestu repraesentował Nieboszczyka, że wszytkim był w podziwieniu et lachrymas movebar. […] Za Ciałam zaraz prowadzono Konia Nieboszczykowskiego kochanego w światło Szkarłatney Kapie aż do ziemie. Na którym przerzucona była zbroia Nieboszczykowska y buława leżała a ci co prowadzili, byli w Kapach Atłasowych Karmazynowych”64. Ceremonia pogrzebowa z archimimusem dawała też okazję do przedstawienia genealogii rodu nieboszczyka, explicite wyrażanej nawet w homiliach. We wzorcowym zbiorze takich kazań z 1670 r., autorstwa księdza Aleksandra Lorencewicza (rektora kolegiów jezuickich we Lwowie i Jarosławiu), pojawił się passus: „Tak Rzymianie y Grecy prowadząc ciała żołnierskie, chorągwie, drzewce, oręża ku ziemi obróciwszy, ostatnią posługę oddawali i Hetmanów obrazy z marmuru albo inszego kruszca rysowali, odlewali”65. Karmelita bosy Andrzej Kochanowski w uczestnikach pogrzebu podskarbiego wielkiego koronnego Jana Mikołaja Daniłowicza (zm. w 1650 r.) widział antycznych żałobników. Do wdowy zwrócił się słowami: „Ciebie Jaśnie Wielmożna MMP podskarbina koronna z matronami porównam rzymskiemi”66. W królewskich uroczystościach funeralnych zwyczaj udziału sobowtóra osoby zmarłej wygasł w trzeciej ćwierci XVII stulecia67, natomiast u magnatów konne rajdy zbrojonego jeźdźca po świątyni zdarzały się jeszcze u schyłku XVIII w. Były to jednak ekscesy, w których gest łamania drzewca kopii i kruszenia oznak wojskowych godności zmarłego stracił rangę ideową. Kaznodzieje potępiali ten obyczaj jako przejaw pogaństwa, naruszający powagę przestrzeni sakralnej i chrześcijańskiego pochówku. W roku 1752 ksiądz Jan Poszakowski perorował: „Nie trzeba naśladować pogańskich pogrzebów, na których tu w Polszcze i na Litwie jeszcze pogańskie ceremonie trwają, jako to, że rycerz zbrojny na koniu do kościoła wjeżdża, i kopią o katafalk skruszywszy sam z konia na ziemię się rzuca a szkapę wolno puszcza częstokroć z przestrachem gminu na tę ceremonię nie z nabożeństwa ale z ciekawości cisnącego się”68. W trzeciej ćwierci XVII stulecia w egzekwiach magnackich wprowadzono nowy element teatralizacji ‒ okazjonalną architekturę, aranżowaną doraźnie na czas pogrzebu. Źródłem były ówczesne ceremoniały francuskie i włoskie, których artystyczną oprawę upowszechniały modne podręczniki dekoracji funeralnej69. Docierające z Francji trendy oddziaływały również na malowany konterfekt trumienny, zwyczajowo używany w staropolskich funeraliach przedstawicieli magnaterii. W drugiej połowie tego stulecia nabierał on reprezentacyjnego charakteru. Podczas obrzędów coraz częściej umieszczano go w innym miejscu — na zwieńczeniu architektonicznego „namiotu żalu” (castrum doloris). 63 Kmita J. 1641, bez paginacji. Kmita J. 1641, bez paginacji. 65 Kazania pogrzebne ks. Aleksandra Lorencowica. 1670, s. 39; Chrościcki J.A. 1974, s. 59. 66 Kochanowski A. 1650, k. A4; Nowicka-Struska A. 2006, s. 93. 67 Do ceremoniału pogrzebowego z udziałem archimimusa nawiązano jeszcze w kwietniu 1826 r., podczas warszawskiej uroczystości funeralnej będącej wyrazem hołdu dla zmarłego w 1825 r. cara Aleksandra I. Zwrócił na to uwagę Chrościcki J.A. 1996, s. 31. Wydarzenie szczegółowo relacjonuje: Opis żałobnego obchodu pamięci po Aleksandrze I Cesarzu. 1829; szerzej o politycznych źródłach warszawskiego ceremoniału, zob.: GetkaKenig M. 2016, s. 695–732. 68 Taki wywód ks. Jan Poszakowski zawarł we wstępie do zbioru kazań francuskiego kaznodziei Józefa Lamberta, zob.: Lambert J. 1752, s. 135. 69 Jednym z bardziej znanych podręczników dekoracji pogrzebowych było dzieło Claude’a F. Menestriera (Menestrier C. F. 1683), Chrościcki J.A. 1970, s. 257. 64 68 ALEKSANDER JANKOWSKI Niezależnie od przyjętej konwencji, niezmienny pozostał przepych ceremonii. Odstępstwa od tego zwyczaju wyraźnie zastrzegano w testamentowych zapisach. Uczynił tak hetman Michał Kazimierz Pac w ostatniej woli z 4 czerwca 1675 r.: „Ciało moye włożywszy w Trune prostą czarną s krzyżem farbowaną zadnym suknem daleko więcey bławatem nieobiiaiąc do Kościoła Swiętego Piotra w Wilnie na Antokolu prywatnie bez zadnych Processiey y Assistentiey bez zadnej okazałości y Pompy Swiatowey zaprowadzą y tam na Marach bez Katafalku swiec Jerzęcych wielkich Sześć około Trunę poki się ofiara Swięta za Duszę moie odprawować będzie zapaliwszy postawią, bez kazania tak podczas Pogrzebu iako y przed Pogrzebem bez mowy oratiy Pogrzebowych niekładąc Buławy na Trunę nie strzelając z Dział y Ręczney Strzelby bez kruszenia Kopiy Rumakow y konie nieprowadząc ani nagrobkow inscriptiey Chorągwie Herbow Obrazow osobę moią representuiących pod czas depositiy Ciała mojego y potym niewystawując owo zgoła bez zadney okazałości i splendece na kturey się więc Swiat ten przy takich Ocasiach silic y sadzic zwykł Ten Act Pogrzebowy odprawią, o co Jako naypilniey JchMMPanow Executorow Upraszam Ubogich tylko samych y kapłanow z Domowemi Zaprosiowszy dla nich stypę sprawią […]”70. Często jednak dyspozycje testatora odnośnie do skromnej oprawy uroczystości były lekceważone. Górę brały względy prestiżowe rodu i z „pogrzebnej” pompy nie rezygnowano. Niekiedy, aby uczynić zadość woli zmarłego, ale też nie uchybić reprezentacji, organizowano dwie ceremonie ‒ skromną i tradycyjną (wystawną). Na przykład w roku 1661 chorąży wielki koronny, Jan Sobieski (przyszły król), najpierw pochował matkę skromnie, tak jak prosiła w testamencie, a dzień później w tym samym miejscu (w kościele dominikanów w Żółkwi) zorganizował pompa funebris, stosowną do rangi swego rodu71. Wzorzec elitarnej uroczystości pogrzebowej, nasyconej elementami świeckiego ceremoniału królewskiego (rycerskiego), szybko przenikał również do kultury sepulkralnej szlachty, zarówno tej majętnej, jak i szaraczkowej. Najbardziej okazałe pompa funebris (z wątkiem archimimusa) naśladowano tylko sporadycznie ze względów finansowych i z powodu uwarunkowań kulturowych, gdyż przedstawiciele tej grupy społecznej rzadko należeli do grona wysokiego dowództwa wojskowego. Mniej zamożna szlachta ideę uobecniania zmarłego w ceremonii pogrzebowej sprowadzała tylko do wykorzystania portretu trumiennego, a theatrum in exequiis raczej ograniczała do konduktu, castrum doloris i akcesoriów trumiennych oraz świetlnej oprawy liturgii, bicia w dzwony i panegirycznych oracji. Portret trumienny włączony ok. połowy XVII stulecia do zwyczajowego zestawu funeralnych atrybutów szlachty (a z czasem także bogatszego mieszczaństwa), stał się domeną twórczości cechowej. W jej ramach ostatecznie ukształtowały się charakterystyczne dla tego typu konterfektu cechy formalno-stylistyczne. Dominował w nich dosadny realizm oblicza portretowanego najczęściej en trois quarts, płasko modelowanego, bladego, ujętego zdecydowanym konturem na jednolitym kolorystycznie tle polerowanej blachy (niekiedy srebrzonej lub pokrytej czarną farbą). Przesadnie podkreślane defekty, szeroko otwarte oczy oraz przenikliwe spojrzenie „podążające” za widzem decydowały o wyjątkowej ekspresji wizerunku. Podniosłość chwili potęgowało umiejętne wykorzystanie światła ‒ świec, pochodni i lamp oliwnych, czasami wzmacnianego lustrami72. Na efekty świetlne, często uwzględniane już na etapie projektowania katafalku (ryc. 4) nie szczędzono środków. Zwykle pochłaniały one 70 Testament Michała Paca. 1840, s. 64, 65–66; nowa edycja krytyczna testamentu: Testament Michała Paca. 2006, s. 132–133. Ostatecznie hetmana pochowano tak, jak zdecydował — na Antokolu „[…] w Sklepie w Dziedzincu w samych Drzwiach Kościelnych Za Zywota ieszcze dla mnie wymurowanym”. 71 Podwójny pochówek (najpierw skromy, a następnego dnia wystawny) miał także Aleksander Sobieski, zmarły w 1714 r. w Rzymie, por. Chrościcki J.A. 1974, s. 63. 72 Lustra zastosowano np. dla oświetlenia castrum doloris Tomasza Zamoyskiego w kolegiacie zamojskiej, Ciesielski T. 2001, s. 226. ROZWAŻANIA O POCZĄTKACH PORTRETU TRUMIENNEGO znaczną część wydatków na pogrzeb 73, a fundusze nie raz zabezpieczano testamentem74. Na przykład w czasie mszy „pogrzebnej” Róży z Ogińskich Krasińskiej (zm. w 1724 r.) kościół w Węgrowie „cały gęstymi wewnątrz jaśniał światłami, bo gdzie tylko mogła miejsce wynaleźć, dla świec jarzących dowcipna ciekawość, na najwyższych gzymsach, na filarach, nad chrzcielnicą i amboną, nad zakrystią i skarbcem, nad kapłańskimi stallami, na kratach, lampami je przetykała tak dalece, że dla gorącości lampy się trzaskały, spadały świece”75. Świece związane były też z tradycyjną symboliką liturgii pogrzebowej. Jej sens podnoszono także w kazaniach. W 1593 r. ksiądz Szymon Wysocki, posługując się metaforą wskazywał: „Potym te dwie rzeczy iakoby rozne czarność żałoby, którą wszystkie ściany okryto z iasnością tak wielkiey wielości świec na co wszyscy patrzycie iako się z sobą barzo pieknie szykuią znaczy iż śmierć sprawiedliwych aczkolwiek iest z przyrodzenia swego okrutna jednak dla tego co po niej następuje to jest on szczęśliwy y błogosławiony żywot a też iey wielce pożądaną śliczność przydaie”76. Półtora wieku później (w 1739 r.) ksiądz Protazy Newerani, autor przewodnika po obrzędowości religijnej, świece pogrzebne uznawał za „znak że człowiek z goreiącemi cnot 73 69 Ryc. 4. Studium projektu castrum doloris Gryzeldy Konstancji Wiśniowieckiej do warszawskiej kolegiaty pw. św. Jana Chrzciciela, Tylman z Gameren, 1672 r., ze zbiorów Gabinetu Rycin Biblioteki Uniwersyteckiej w Warszawie, Archiwum Tylmana, nr inw. GR BUW, Inw.G.R.6426 (AT875) Fig. 4. A study design of Gryzelda Konstancja Wiśniowiecka’s castrum doloris for St John’s collegiate church in Warsaw, by Tylman van Gameren, 1672, from the collection of the University Library in Warsaw, the Tylman Archive, inventory no. GR BUW, Inw.G.R.6426 (AT875) Chrościcki J.A. 1974, s. 44. Taki zapis umieścił w testamencie np. pozostający w bardzo złej sytuacji finansowej Kazimierz Strobiszewski, por. Dumanowski J. 2001, s. 319. 75 Giżycki G. 1724, k. D2. 76 Wysocki S. 1593, s. 44. Na to szesnastowieczne źródło wcześniej zwrócił uwagę Juliusz Chrościcki, por.: Chrościcki J.A. 1968, s. 394. 74 70 ALEKSANDER JANKOWSKI Świętych lámpami záyść drogę Oblubieńcowi Duszy swiey Chrystusowi powinien, ieżeli chce bydź puszczony do Krolestwa Niebieskiego”77. Podczas liturgii pogrzebowej, w przestrzeni przesyconej dymem kadzideł, pośród świec i symbolicznej iluminacji, portret trumienny eksponowany w oprawie „majestatu” (castrum doloris) utożsamiał zmarłego, nie tylko jako biernego świadka ceremonii, ale także jako jej aktywnego uczestnika. W budzącym dreszcz grozy theatrum in exequiis sportretowany nie tylko patrzył na żałobników, ale i do nich „mówił” (ustami oratora), a żegnając się wyrażał wdzięczność obecnym: „Żegnam Waszmościów wszystkich […] Żegnam i was wszystkich poddanych […] A na ostatek wszystkim obecnie się tu znajdującym ostatnie a życzliwe dając vale na spokojny odpoczynek te pobożne sprowadzającemu exuvias Waszmościom jako najszczęśliwszej życzę drogi”78. W ten sposób „iluzja naoczności”79 udziału zmarłego w ceremonii zyskiwała na sugestywności. * * * Uroczystości pogrzebowe w dawnej Polsce wiązały pobożność i religijną symbolikę życia wiecznego z efektownym spektaklem manifestacji osobistego i rodowego splendoru. Gdzieś pod koniec trzeciej lub w czwartej ćwierci XVI stulecia w ceremonialno-obrzędowym splocie religijnych i świeckich inspiracji, pod wpływem czyśćcowej pobożności, pojawił się zwyczaj przedstawiania oblicza zmarłego złożonego w zamkniętej trumnie. W tym celu trumna zyskała szklone, „wyrżnięte okno”, przez które widać było twarz nieboszczyka. Niemal równocześnie, z tych samych pobudek, praktykowano rozwiązanie alternatywne — przytwierdzana do trumny (wieka bądź ściany) malowana na blasze niewielka podobizna zmarłego. Ukazaną w ten sposób „twarz własną, jako żywą”, początkowo również przesłaniano „szkłem”, tworząc iluzję prawdziwego oblicza. W ciągu trzeciej ćwierci XVII w. malowany portret całkowicie wyparł „przeszklone okno”. Przytwierdzany teraz do czoła (wezgłowia) trumny stał się nieodzowny w „pogrzebnej apperencyi”. Taki status dawała portretowi trumiennemu jego „podwójna” funkcja ideowa: uobecniająca w liturgii pogrzebowej „człowieka wewnętrznego” w trakcie „przenosin do gabinetu podziemnego”80, oraz po ceremonii pogrzebowej, gdy obraz „osobę nieboszczyka representuiący pod czas depositiy Ciała”, wyeksponowany we wnętrzu świątyni, żywym „pamięć o sobie przywracał”, legitymizując trwałą obecność (duszy) w przestrzeni liturgicznej. Adres Autora: dr hab. Aleksander Jankowski, prof. UKW Wydział Historyczny Uniwersytet Kazimierza Wielkiego w Bydgoszczy ul. Księcia Józefa Poniatowskiego 12 85-667 Bydgoszcz aleksander.jankowski@ukw.edu.pl aleksanderjankowski@wp.pl https://orcid.org/0000-0002-7806-1845 77 Newerani P. 1739, s. 364–365. Fragment mowy pogrzebowej przytoczono za: Bystroń J. 1994, s. 106. 79 Pojęcie „iluzji naoczności” zaproponowano w analizie praktyk kulturowych doby staropolskiej w: Dziechcińska H. 1987, s. 87–89. 80 Boczyłowic J. 1669, s. 518. 78 ROZWAŻANIA O POCZĄTKACH PORTRETU TRUMIENNEGO 71 BIBLIOGRAFIA Źródła i opracowania publikowane Anioł od św. Teresy. 1651. Podskarbi według statutu Ewangielii postępujący w J.W.J.M. Panu P. Janie Mikołaju z Żurowa Daniłowiczu,podskarbim wielkim koronnym... przy egzekwiach w Lublinie w kościele WW.O.O. Karmelitów Bosych […] R.P: 1650 d. 11 kwietnia pokazany, Poznań. Augustyn Św. 1962. Św. Augustyn, O Trójcy Świętej, tłum. M. Stokowska, oprac. J.M. Szymusiak, ks. XII, Poznań. Boczyłowic Jakub. 1669. Oracya pogrzebowa, [w:] J. Boczyłowic , Orator politicus albo wymowny polityk Rożne traktujący Materye, Toruń, s. 517–519, https://jbc.bj.uj.edu.pl/publication/505522 (dostęp 29.03.2018). Brosig Alfred. 1930. Tabliczki z trumien, „Kronika Miasta Poznania”, R. VIII, nr 4, s. 327–334. Bystroń Jan. 1994. Dzieje obyczajów w dawnej Polsce: wiek XVI–XVIII, t. 2, Warszawa. Chrościcki Juliusz A. 1968. „Castris et astris”. Kazania i relacje pogrzebowe jako źródła historii sztuki, „Biuletyn Historii Sztuki”, nr 3, s. 384–395. Chrościcki Juliusz A. 1970. Projektanci i wykonawcy katafalków z 1 połowy XVIII w., [w:] Rokoko. Studia nad sztuką 1 połowy XVIII w. Materiały Sesji Stowarzyszenia Historyków Sztuki zorganizowanej wspólnie z Muzeum Śląskim we Wrocławiu, Wrocław październik 1968, Warszawa, s. 251–274. Chrościcki Juliusz A. 1974. Pompa funebris. Z dziejów kultury staropolskiej, Warszawa. Chrościcki Juliusz A. 1996. Od śmierci do egzekwii, [w:] Vanitas. Portret trumienny na tle sarmackich obyczajów pogrzebowych, red. J. Dziubkowa, Poznań, s. 28–31. Ciesielski Tomasz. 2001. Pogrzeby wojskowe w czasach saskich, [w:] Wesela, chrzciny i pogrzeby w XVI–XVII wieku. Kultura życia i śmierci, red. H. Suchojad, Warszawa, s. 217–233. Cieślak Katarzyna. 1992. Kościół-cmentarzem. Sztuka nagrobna w Gdańsku (XV–XVIII w.). „Długie trwanie” epitafium, Gdańsk. Dobrzeniecki Tadeusz. 1990. Geneza polskiego portretu trumiennego, [w:] Portret. Funkcja–Forma– Symbol. Materiały Sesji Stowarzyszenia Historyków Sztuki, Toruń grudzień 1986, Warszawa, s. 73–88. Dumanowski Jarosław. 2001. Pompa funebris? Z testamentów szlachty wielkopolskiej z XVIII w., [w:] Wesela, chrzciny i pogrzeby w XVI–XVII wieku. Kultura życia i śmierci, red. H. Suchojad, Warszawa, s. 315–322. Dziechcińska Hanna. 1987. Oglądanie i słuchanie w kulturze dawnej Polski, Warszawa. Dziubkowa Joanna. 1996. Stoj pogrzebiencze, a pod glancem tey blachy upatruj prześwietne popioły, [w:] Vanitas. Portret trumienny na tle sarmackich obyczajów pogrzebowych, red. J. Dziubkowa, Poznań, s. 16–19. Getka-Kenig Mikołaj. 2016. Rządowe przedsięwzięcia pomnikowe ku czci Aleksandra I a ideologia „wskrzeszenia” Polski w latach 1815–1830, „Kwartalnik Historyczny”, R. CXXIII, z. 4, s. 695–732. Giżycki Grzegorz. 1724. Delineacya Aktu Załobneg. Przy Pogrzebie Ciała Jaśnie Wielmożney Iey Mci Pani P. Rozy z Oginskich Krasinskiey, Kasztelanowey Płockiey, Starościney Sztumskiey, Prasnyskiey, Nowomieyskiey etc. etc. W Kośćiele Farnym Węgrowskim, Fundacyi Jaśnie WW: Ich Mciow PP: Krasinskich, odprawionego; Dnia 28. Listopada, Roku 1724, [w:] Kwiat Spadaiący, Przećiwnym śmiertelnośći wichrem z wybornego Drzewa Zerwany. Zostawiwszy na nim kosztowne Frukta; To iest Wielka Matka, godnych Synow Iasnie Wielmozna Roza z Oginskich Krasinska, Kasztelanowa Płocka Sztumska, Prasnyska, Nowomieyska etc. etc. Staroscina. W Kośćiele swoim Farnym Węgrowskim Clericorum Saecularium in Comm: Viventium Pogrzebiona. Y Kazaniem Pogrzebowym Przez Xiędza Grzegorza Giżyckiego, Kapłana ejusdem Instituti Obzałowana. Roku Panskiego 1724. Dnia 28. Listopada w Warszawie w Drukarni WW: OO: Scholarum Piarum, Warszawa, k. B2–B2v, https://jbc.bj.uj.edu.pl/dlibra/docmetadata?from=rss&id=647414 (dostęp 29.05.2020). Janowski Piotr J. 2015. Oprawa artystyczno-ideowa pogrzebu Stefana Batorego, „Barok. Historia– –Literatura–Sztuka”, R. XXII/1(43), s. 29–45. 72 ALEKSANDER JANKOWSKI Jarczykowa Mariola. 2012. Przy pogrzebach rzeczy i rytmy. Funeralia Radziwiłłowskie z XVII wieku, Katowice. Katalog wystawy jubileuszowej zabytków. 1933. Katalog wystawy jubileuszowej zabytków z czasów króla Stefana i Jana III w gmachu Muzeum Wojska w czterechsetlecie urodzin Stefana Batorego i dwieściepięćdziesięciolecie odsieczy wiedeńskiej, Warszawa. Kazania pogrzebne ks. Aleksandra Lorencowica. 1670. Kazania pogrzebne miane w różnych żałobnych okazyjach, przez ks. Aleksandra Lorencowica, prowincjała polskiego Societatis Iesu, w Kolegium Kaliskim Societatis Iesu drukowane, Roku Pańskiego 1670, Kalisz. Kopera Feliks. 1926. Malarstwo w Polsce od XVI do XVIII wieku (renesans, barok, rokoko), Dzieje Malarstwa w Polsce, t. II, Kraków. Kmita Jan. 1641. Opisanie Porządku przy Ciele S.Pamięci X. Jegomości Pana Wojewody Wileńskiego Hetmana Wielkiego W.X. Litew[skiego] Administratora Mohilowskiego, Bystrzyckiego, etc. Starosty: Y Wszystkiego Actu Pogrzebowego Który się odprawował w Wiżunach Anno 1641 21. Ianuarij, [w:] Proces pogrzebu S. Pamięci Jaśnie Oświeconego Książęcia Je[g]o Mości na Birżach i Dubinkach P.P. Chrzysztopha Radziwiłła Wojewody Wileńskiego, Hetmana Wielkiego W.X.L. Administratora Mohilewskiego, Bystrzyckiego, etc. etc. Starosty, Lubcz, https://jbc.bj.uj. edu.pl/dlibra/publication/383359/edition/364882/content (dostęp 22.11.2017). Kochanowski Andrzej. 1650. Niebieskie planety górnym oddane sferom, przy pogrzebie Jaśnie Wielmożnego Jego Mości P. Mikołaja z Żurowa Daniłowicza, podskarbiego wielkiego koronnego […] w Lublinie w kościele Panien Karmelitanek Bossych […] Roku Pańskiego 1650 6 Aprilis, Kraków. Kraszewski Józef I. 1888. Wizerunki książąt i królów polskich (z 39 rycinami ks. Pillatiego oraz inicjałami C. Jankowskiego), Warszawa. Kronika Jana z Czarnkowa. 1996. Kronika Jana z Czarnkowa, tłum. J. Zerbiłło, oprac. M.D. Kowalski, Kraków. Kronika mieszczanina krakowskiego. 1930. Kronika mieszczanina krakowskiego z lat 1575–1595, wyd. H. Barycz, Kraków. Kruszyński Tadeusz ks. 1932. Sarkofag Stefana Batorego w Grobach królewskich na Wawelu, „Kurier Literacko-Naukowy”, R. 9, nr 4, s. I–III. Labudda Alfons. 1983. Liturgia pogrzebu w Polsce do wydania Rytuału Piotrkowskiego (1631). Studium historyczno-liturgiczne, Textus et Studia Historiam Theologiae in Polonia Excultae Spectantia, vol. XIV, Warszawa. Lambert Józef. 1752. Kazania na święta, tłum. J. Poszakowski, Wilno. Łukarska Beata. 2018. Religijność sarmacka w przekazie piśmiennictwa polskiego XVII i XVIII wieku. Zarys monograficzny i antologia tekstów źródłowych, Częstochowa. Łukaszewicz Józef. 1858-1859. Krótki opis historyczny kościołów parochialnych, kościółków, kaplic, klasztorów, szkółek parafialnych, szpitali i innych zakładów dobroczynnych w dawnej dyecezyi poznańskiej, t. 1, Poznań. Menestrier Claude F. 1683. Des Décorations funebres, ou il est amplement traité des tentures, des lumieres, des mausolées, catafalques, inscriptions et autres ornements funebres ; avec tout ce qui s’est fait de plus considérable depuis plus d’un siecle, pour les papes, empereurs, rois, reines, cardinaux, princes, prelats, scavans et personnes illustres en naissance, vertu et dignite. Paryż. Mrozowski Przemysław. 2000. „Siste gradam viator” — o upamiętnianiu zmarłych w sztuce dawnej Polski, [w:] „Przeraźliwe echo trąby żałosnej do wieczności wzywającej”. Śmierć w kulturze dawnej polski od średniowiecza do końca XVIII wieku. Katalog wystawy, red. P. Mrozowski, Warszawa, s. 52–84. Newerani Protazy. 1739. Ozdoba Kościoła katolickiego. To iest Cerymonie ktorych Kościoł S. w wszelkich okolicznościach zwykł zażywać. Pismem S. Powagą Oycow SS. dawnością utwierdzone, gruntownym wykładem obiasnione. Naprzod Niemieckim ięzykiem Teraz Na Polski z przydatkiem wielu rzeczy przetłumaczone y do druku podane. Przez X. Protazego Neweraniego Zakonu S.O. Franciszka, Małopolskiego Reformata. Roku 1739, Lwów, https://jbc.bj.uj.edu.pl/ dlibra/publication/491995/edition/466332?language=en (dostęp 28.02.2018). ROZWAŻANIA O POCZĄTKACH PORTRETU TRUMIENNEGO 73 Nowicka-Jeżowa Anna. 1992. Feniks w popiołach. Uwagi o kulturze religijnej drugiej połowy XVII w., [w:] Literatura i kultura polska po „potopie”, red. B. Otwinowska, J. Pelc, współudz. B. Falęcka, Wrocław-Warszawa-Kraków, s. 121–137. Nowicka-Struska Anna. 2006. Elementy teatralne, gest i komunikacja niewerbalna w siedemnastowiecznych kazaniach pogrzebowych, „Annales Universitatis Mariae Curie-Skłodowska. Sectio FF, Philologiae”. Vol. 24, 71–97. Opis żałobnego obchodu pamięci po Aleksandrze I Cesarzu. 1829. Opis żałobnego obchodu po wiekopomnej pamięci Najjaśniejszym Aleksandrze I, Cesarzu Wszech Rosji, Królu Polskim w Warszawie w dniach 7, 9, 10, 11, 12, 13, 17, 19, 23 kwietnia 1826 roku uroczyście odbytego, Warszawa. Oświęcim Stanisław. 1907. Stanislai Oświęcim diarium 1643–1651, wyd. i oprac. W. Czerniak, Scriptores Rerum Polonicarum, t. XIX, Kraków, nr 63. Pisarze polskiego Odrodzenia o sztuce. 1955. Pisarze polskiego Odrodzenia o sztuce, oprac. W. Tomkiewicz, Teksty Źródłowe do Dziejów Teorii Sztuki, t. 4, Wrocław. Pogrzeb Stefana Batorego w Krakowie. 1860. Pogrzeb Stefana Batorego w Krakowie dnia 23go 1588 roku, [w:] Panowanie Henryka Walezyusza i Stefana Batorego Królów Polskich. Z rękopisów Albertrandego podług wydania Ż. Onacewicza z dołączeniem Pamiętników historii Stefana Batorego dotyczących, i listu Jerzego Chiakora sekretarza król. opisującego ostatnie chwile tego monarchy, Kraków, s. 457–461. Radziwiłł Mikołaj Krzysztof. 1925. Mikołaja Krzysztofa Radziwiłła, Peregrynacja do Ziemi Świętej (1582–1584), Archiwum do Dziejów Literatury i Oświaty w Polsce, t. 15, cz. 2, oprac. J. Czubek, Kraków. Ryszkiewicz Andrzej. 1974. Malarstwo około roku 1600, [w:] Sztuka około roku 1600. Materiały z Sesji Stowarzyszenia Historyków Sztuki zorganizowanej przy współpracy Wydziału Kultury Prezydium Wojewódzkiej Rady Narodowej w Lublinie, Lublin, listopad 1972, red. T. Hrankowska, Warszawa, s. 31–49. Scholz Piotr O. 2015. Geneza portretu trumiennego w świetle podróży Mikołaja Krzysztofa Radziwiłła „Sierotki” do Egiptu i Ziemi Świętej (1582–1584). Kilka uwag badawczych, [w:] Cywilizacja prowincji Rzeczpospolitej szlacheckiej, Cykl II: Nie wszystek umrę. Pamięć o zmarłych w kulturze staropolskiej, red. A. Jankowski, A. Klonder, Bydgoszcz, s. 73–81. Szołdrski Władysław. 1929–1931. Z dziejów dominikanów w Toruniu, „Zapiski Towarzystwa Naukowego w Toruniu”, t. 8, s. 43–86. Śnieżyńska-Stolot Ewa. 1975. Dworski ceremoniał pogrzebowy królów polskich w XIV wieku, [w:] Sztuka i ideologia XIV wieku. Materiały sympozjum Komitetu Nauk o Sztuce Polskiej Akademii Nauk, Warszawa, 29 i 30 listopada 1973 r., red. P. Skubiszewski, Warszawa, s. 89–100. Tazbir Janusz. 1970. Sarmatyzacja katolicyzmu w XVII wieku, [w:] Wiek XVII — kontrreformacja–barok. Prace z historii kultury, red. J. Pelc, Wrocław-Warszawa-Kraków, s. 7–37. Terka Mariusz. 2014. Wiara jako poszukiwanie Boga w świetle nauczania św. Augustyna, „Vox Patrum”, R. 34, t. 61, s. 399–426. Testament Michała Paca. 1840. Testament Michała Paca, „Wizerunki i Roztrząsania Naukowe. Poczet Nowy Drugi. Tomik ośmnasty”, Wilno. Testament Michała Paca. 2006. Testament Michała Kazimierza Paca i Aleksandra Hilarego Połubińskiego w przededniu kampanii na Ukrainie przeciwko Turkom i Tatarom w 1675 roku, oprac. K. Bobiatyński, M. Nagielski, Materiały do Historii Wojskowości, t. 3, cz. 1, s. 130–143. Wiliński Stanisław. 1958. U źródeł portretu staropolskiego, Warszawa. Wysocki Szymon. 1593. Kazanie na pogrzebie Jaśnie Oświeconego książęcia Jegomości Pana Pana Jana Symeona Olelkowicza z łaski Bożey Książęcia Słuckiego y Kopylskiego etc. ostatniego meskiey płci zacney staradawney familiey potomka. Miane. Przez X. Symona Wysockiego, Societatis Iesu kapłana. W Kościele teyże Societatis w Lublinie 29. Aprilis. 1593, Wilno, https: //www.wbc.poznan.pl/dlibra/publication/378536?tab=1 (dostęp 15.04.2019). Żywot y smierc Jana Tarnowskiego. 1773. Żywot y smierc Jana Tarnowskiego kasztelana krakowskiego hetmana wielkiego koronnego pisany niegdyś przez Stanisława Orzechowskiego. Teraz z rzadkiego Załuskich Biblioteki rękopisma wyjęty, wydrukowany y przypiskami objaśniony, Warszawa. 74 ALEKSANDER JANKOWSKI “To see the deceased… a window had to be cut in the coffin”. On the origins of the coffin portrait The coffin portrait, i.e. an image of the deceased person painted on metal sheet, attached to the coffin for the exequies, is an artistic phenomenon characteristic of the art and culture of death in the Old-Polish era. Even though it had been researched for almost 100 years, its origin, chronology and functions are still debated. The reason for that is, among others, the set of features defining this genre, especially the assumption that a coffin portrait was attached at the head board of the coffin, determining the shape of the metal plate, which was supposed to match the cross-section of the coffin (two trapeziums sharing the longer parallel sides). Following those criteria, the beginnings of the coffin portrait were sought in the mid-17th century, which excluded portraits of the dead painted on metal plates of different shapes, attached to the lids or sides of coffin of rectangular (almost square) cross-sections already in the 16th c. The oldest known portrait of the departed person fixed at the head board comes from such a casket. Hexagonal coffins began to be used in Poland in the mid-17th c., and accordingly, the portraits fixed to the head board became hexagonal. Since most of the surviving coffin portraits come from Greater Poland, researchers have tended to assume that the custom originated in that region. However, sources indicate that fixing a portrait to the polygonal head board of the coffin was practiced earlier, outside Greater Poland, in the circle of magnates associated with the court of the Vasa kings. During the funeral ceremony, which was conceptualized as a farewell to the soul and body (the Latin pompa funebris), the coffin portrait was an incarnation of the dead person. In this role it was displayed in the “castle of grief” (castrum doloris), completing the scenario of the ceremony, in which the main part was played by a living double of the departed one (archimimus). In the 16th–17th c., an alternative was to show the face of the dead person via a glazed opening cut in the coffin, but this custom was gradually replaced by the coffin portrait to guarantee that the face stayed unchanged, which was important in prolonged ceremonies often taking part many months after the death. Having originated in the circle of Catholic elites, the coffin portrait spread among nobles of all Christian denominations. Despite religious differences and the resulting different visions of the beyond, it remained an important manifestation of religiousness and social status. Translated by Izabela Szymańska
https://openalex.org/W3034677969	http://www.scielo.br/pdf/ref/v27n3/1806-9584-ref-27-03-e58797.pdf	Spanish; Castilian	null	Internet Violence against Chilean Feminists and Other Activists	Revista Estudos Feministas	2,019	cc-by	10,924	Violencia en Internet contra feministas y Violencia en Internet contra feministas y Violencia en Internet contra feministas y Violencia en Internet contra feministas y Violencia en Internet contra feministas y otras activistas chilenas otras activistas chilenas otras activistas chilenas otras activistas chilenas otras activistas chilenas Cecilia Alejandra Ananías Soto1 0000-0002-2331-9837 Karen Denisse Vergara Sánchez1 0000-0002-1917-9608 1ONG Amaranta, Concepción, Biobío, Chile. ong.amaranta@gmail.com Resumen: Resumen: Resumen: Resumen: Resumen: Este trabajo explora la violencia de género en Internet que afecta a feministas y otras activistas por los DDHH en Chile, entregando una aproximación inicial a un tipo de violencia que se ha encontrado invisibilizada en Latinoamérica, y que recientemente ha demostrado lo peligrosa que puede llegar a ser para la vida de quienes la han sufrido. Para conocer esto, se encuestó a 163 mujeres cishétero y transgénero a través de un cuestionario online, dentro de una metodología de muestreo no probabilístico por conveniencia, como una primera exploración de este tema a nivel nacional. De las activistas consultadas, un 73,6% declaró explícitamente haber sido víctima de este tipo de violencia. Ellas principalmente sufrieron ataques verbales (91,7%), acoso (25,8%), amenazas (22%) y publicación de información falsa (15%). También, un 13,3% declaró que le sustrajeron imágenes de las redes sociales y otro 13,3% recibió imágenes y/o videos sexualmente agresivos, evidenciando, además, la nula preparación de las policías y poder judicial para enfrentar estos hechos, incluso a la hora de denunciar. Además, a través de una aproximación al discurso de los atacantes, se determinó que se trata principalmente de hombres jóvenes y adultos con estudios superiores, quienes utilizan Internet para infundir terror psicológico y una sensación de permanente vigilia hacia las víctimas. Se trata de una problemática urgente que necesita ser abordada como sociedad, para que no se continúe amplificando en medio de la impunidad legal y la reticencia a ser abordada como violencia de género digital por parte de autoridades y ciudadanía. g g p p y Palabras clave: Palabras clave: Palabras clave: Palabras clave: Palabras clave: violencia de género; feminismo; discurso de odio; violencia en Internet; acoso. Internet Violence against Chilean Feminists and Other Activists Internet Violence against Chilean Feminists and Other Activists Internet Violence against Chilean Feminists and Other Activists Internet Violence against Chilean Feminists and Other Activists Internet Violence against Chilean Feminists and Other Activists ggggg Summar Summar Summar Summar Summary: y: y: y: y: This work analyses gender-based violence on Internet which affects both feminists and human rights activists in Chile, giving thus a first approach to a kind of violence that has been overshadowed in Latin America and that has proved recently how dangerous it can be for those who have gone through it. To this end, 163 cisgender heterosexual and transgender women were surveyed through an online questionnaire inside a non-probabilistic sampling method through convenience, as the first research of this topic nationwide. Among the activists interviewed, 73.6% explicitly stated to have been victims of this kind of violence. They suffered mainly verbal abuse (91.7%), harassment (25.8%), threats (22%) and false information published about them (15%). Also, 13.3% stated to have had pictures stolen from their social networks, and another 13.3% received sexually aggressive images and/or videos, which also made evident the non-existent preparation of both the police and the judiciary power while facing those issues, even when they are being denounced. Besides, through an approximation to the aggressors’ speech, it was established that most of them were mainly young and adult men with higher education, who use Internet to spread psychological terror and to make the victims feel they are being constantly observed. This is an urgent problematic that the society needs to address, to avoid an increase between the legal impunity and the reluctance of authorities and citizens to consider it as digital gender-based violence. Keywords: Keywords: Keywords: Keywords: Keywords: Gender-based violence; Feminism; Hate speech; Violence on Internet; Harassment. Artigos Artigos Artigos Artigos Artigos Artigos Artigos Artigos Artigos Artigos 1. Introducción 1. Introducción 1. Introducción 1. Introducción 1. Introducción La violencia de género es un problema a nivel global que, de la mano de los Estudios de Género y el activismo feminista, ha comenzado a ser visibilizado, desnormalizado y denunciado. Internet es una plataforma que ha permitido que los movimientos por los derechos de la mujer se organicen, tejan redes y expandan su lucha. Pero, como explican investigadores como Manuel Castells (2001) y Remedios Zafra (2005), la Red no es una plataforma ajena a los comportamientos: más bien, los comportamientos se apropian de la Red y se amplifican, amparados en la distancia física y el anonimato. Esto incluye el sexismo, las desigualdades de género y la violencia. Una prueba de esto es que las mujeres jóvenes, de entre 18 y 35 años, son las que están más expuestas a la violencia basada en tecnología, según un mapeo de Association for Progressive Communications (APC). Esta abarca desde el acoso (individual o grupal), hostigamiento, extorsión, amenazas, robo de identidad, doxing1, alteración y publicación de fotos y videos sin consentimiento. Otro estudio de la Superintendencia de Educación de Chile reveló que el ciberacoso entre y hacia escolares ha aumentado y que el 82% de las denuncias corresponden a situaciones que afectan a mujeres (Claudia MIÑO, 22 de julio de 2018). Por ende, hablamos de un problema de género. j ( j ) p g A esto se suma la lenta o nula reacción de las leyes. En Chile, ni siquiera está normado el acoso como delito penal, a menos que ocurra en el contexto laboral ni mucho menos se ha normado esta figura en Internet. La Brigada de Cibercrimen de la Policía de Investigaciones de Chile, lo más cercano a la fuerza policial en Internet, solo responde ante delitos contra menores en Internet (pedofilia, pornografía infantil, grooming), delitos que afectan el patrimonio (estafas como el phishing y el pharming) y el daño informático. La única forma de denunciar un ataque a una mujer adulta en Internet es a través de antiguas figuras del Código Civil, como la injuria, calumnia y daño al honor, las cuales muchas veces han sido utilizadas por los mismos agresores para desincentivar las denuncias y ‘funas’2; además, los procesos en tribunales civiles suelen ser largos, burocráticos y costosos. Un caso representativo de esto es lo que le ocurrió a la periodista Javiera Tapia y la cantante Daniela González, más conocida como “Dulce y Agraz”. 1 Este término define la recopilación y divulgación de información personal, que está disponible de manera pública, para amedrentar o amenazar a la víctima. Puede tratarse del domicilio, datos financieros, teléfono privado o el nombre de familiares. CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ Das ativistas consultadas, 73,6% declarou explicitamente ter sido vítima deste tipo de violência. Elas sofreram principalmente ataques verbais (91,7%), assédio (25,8%), ameaças (22%) e publicação de informação falsa (15%). Também, 13,3% declarou que lhes roubaram imagens das redes sociais e 13,3% recebeu imagens e/ou vídeos sexualmente agressivos, evidenciando, além disso, a nula preparação dos policiais e do poder judiciário para enfrentar estes feitos, inclusive no momento da denúncia. Além disso, através de uma aproximação ao discurso dos atacantes, determinou-se que se trata principalmente de homens jovens, com estudos superiores, que utilizam a Internet para infundir terror psicológico e uma sensação de permanente vigilância sobre as vítimas. Se trata de uma problemática urgente, que necessita ser abordada como sociedade, para que no se continue amplificando em meio à impunidade legal e a reticência por parte das autoridades e cidadãos em abordá-la como violencia de gênero digital. g g Palavras-chave: Palavras-chave: Palavras-chave: Palavras-chave: Palavras-chave: Violência de gênero; feminismo; discurso de ódio; violência na Internet; assédio Violência na Internet contra feministas e outras ativistas chilenas Violência na Internet contra feministas e outras ativistas chilenas Violência na Internet contra feministas e outras ativistas chilenas Violência na Internet contra feministas e outras ativistas chilenas Violência na Internet contra feministas e outras ativistas chilenas Resumo: Resumo: Resumo: Resumo: Resumo: Este trabalho explora a violência de gênero na Internet que afecta a feministas e outras ativistas por Direitos Humanos no Chile, dando uma aproximação inicial a um tipo de violência que tem estado invisibilizada na América Latinha e que recentemente tem demonstrado o quão perigosa pode chegar a ser para a vida de quem a tem sofrido. Para isto, foi realizado um questionário on-line com 163 mulheres cishétero e transgênero, dentro de una metodología de amostra não-probabilística por conveniência, como uma primeira exploração do tema em nível nacional. 1 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 2 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS Canadá, el pasado 23 de abril de 2018, cuando Alek Minassian, quien se autodenominaba ‘incel’, asesinó a 8 mujeres y dejó a 15 heridas tras arrojarles su vehículo encima. Los antecedentes recabados por la policía arrojaron que en la cuenta de Facebook del asesino había un mensaje machista en el que celebraba la acción que iba a consumar como el inicio de la ‘rebelión incel’. En Estados Unidos, existen estudios del Southern Poverty Law Center (SPLC) de Alabama que indican que, de la mano del discurso sexista y racista del presidente Donald Trump, han aumentado los grupos de odio (EL DESCONCIERTO, 26 de febrero de 2018), incluyendo, por primera vez, los primeros supremacistas masculinos. Se trata de dos agrupaciones: Una Voz Para Hombres y Retorno de Reyes, los cuales abogan por la subyugación total de las mujeres, ya que las consideran una plaga, además de plantearse como enemigos de la equidad y el feminismo. En una conversación con BBC Mundo, Heidi Beirich, directora de Inteligencia del SPLC, explicó que ven a las mujeres como seres inferiores, “como una fuerza maligna en la sociedad [...] para ellos, son malvadas y las llaman ‘zorras’” (Guillermo OLMO, 5 de marzo de 2018) y, por ende, el avance de la conquista de sus derechos es visto como una señal de decadencia. En Chile, durante una marcha a favor del aborto, varias mujeres fueron agredidas y empujadas, se lanzaron vísceras y sangre a las calles y tres activistas fueron apuñaladas. La vocera de Mesa Acción por el Aborto, Macarena Castañeda, relató lo ocurrido en una entrevista al programa Tele Trece Tarde, uno de los noticieros más vistos del país: una turba de encapuchados empezó a armar barricadas en la marcha, a una cuadra de donde estaba el escenario. De repente, un grupo empezó a apuñalar a las que estaban pasando. Algunas supieron que estaban atacando, salieron a proteger y terminaron apuñaladas. Esto es terrorismo, yo no lo quiero llamar de otra manera. Cuando un grupo quiere amedrentar a otro para evitar que pueda expresar sus ideas libremente, sin provocación alguna. Intentaron cortar la marcha para que no pudiéramos pasar. (TELETRECE, 26 de julio de 2018) Muchas activistas sindicaron el ataque al Movimiento Social Patriota, una agrupación de ultraderecha chilena, ya que, ese mismo día, había hecho un llamado a una contra manifestación en sus redes sociales. VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS Tras los ataques, escribieron en Twitter3: “Nuevamente los Social Patriotas hemos salido a contra manifestarnos. Esta vez en la marcha del Aborto Libre. Ya no basta con las RRSS. ¡Es hora de tomar las calles!”. Por esto se hace imperativo dejar de ver la violencia online como algo separado de la violencia offline; ni mucho menos trivializarla, ya que se trata de un continuo de violencia hacia las mujeres. En esa línea, este trabajo busca explorar la violencia relacionada con tecnología que sufren las feministas y otras activistas en Chile. Para esto, se encuestó a 163 feministas de distintas regiones del país, principalmente de Santiago, Concepción y Valparaíso, para conocer si han sufrido este tipo de violencia, cómo se ha manifestado, cómo la han manejado y si la han discutido con su entorno u organizaciones. Este trabajo fue escrito con perspectiva de género y su objetivo no es solo aportar con cifras al estudio de esta temática, sino que, además, como un acto político ante las múltiples voces que intentan silenciar a las mujeres y para posicionar la violencia basada en tecnología en la opinión pública. 2 Nombre dado en Chile a una manifestación de denuncia y repudio público contra una persona o grupo que cometió una acción deshonesta, ilegal o violenta. 3 En Twitter están como @movs_patriota. Cabe destacar que medios como La Tercera les dieron espacio a sus dirigentes para hablar; no así a las víctimas. 1. Introducción 1. Introducción 1. Introducción 1. Introducción 1. Introducción En un reportaje de POTQ Magazine -un medio de cultura y actualidad, titulado Cuando ella habla, escucho la revolución, esta periodista denunció a una serie de músicos involucrados en violencia de género, entre ellos, Jimmy Valenzuela, ex pareja de González. Valenzuela presentó un recurso de protección contra ambas mujeres, mas el tribunal falló a favor tanto de la periodista, como -posteriormente- a favor de la víctima. El fallo determinó que la violencia contra las mujeres era un tema de interés público y no privado (EL DESCONCIERTO, 8 de febrero de 2018). Estos vacíos legales y figuras judiciales antiguas utilizadas para acallar a las denunciantes permiten que se organicen ataques grupales hacia mujeres -principalmente, víctimas de violencia, comunicadoras, mujeres de perfil público y feministas- en foros completamente abiertos, sin miedo a sufrir repercusiones. A nivel internacional, esto ha llevado a que aparezca el fenómeno INCEL, sigla de “celibato involuntario” en inglés; una forma en que miles de hombres se están autodefiniendo para promover discursos de odio contra mujeres y comunidades marginadas. Para la investigadora de la Universidad de Dublín, Debbie Ging, citada en el medio The Huffington Post (7 de mayo del 2018), se trata de un grupo de hombres misóginos que culpa a las mujeres por su nula vida sexual, presentando además un rechazo abierto contra la comunidad LGBTI, inmigrantes y oprimidos. Los ‘incels’ se congregan a través de Internet, en foros y espacios anónimos, en los cuales pueden refugiarse tras un avatar para ocultar su identidad y realizar ataques contra las mujeres de su entorno, llegando a la acción y violencia física; hechos que se materializaron en 2 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 2 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS 2. Género 2. Género 2. Género 2. Género 2. Género, Internet y Ciber , Internet y Ciber , Internet y Ciber , Internet y Ciber , Internet y Ciberfeminismo feminismo feminismo feminismo feminismo Internet partió el año 1969, en Estados Unidos, como una herramienta de uso para el almacenamiento de información y coordinación militar. En sus inicios, el proyecto se denominaba Arpanet y comenzó a expandirse entre centros de investigación, como una herramienta poderosa de conocimiento, almacenamiento e intercambio de información (Gonzalo FERREYRA, 1996). ( ) Quince años después, en 1984, la académica Donna Haraway publica El Manifiesto Cyborg. En él, la autora desmitifica la dominación masculina en la tecnología, proponiendo una nueva identidad política definida como Cyborg, la cual es un híbrido entre cuerpo y máquina, lo que desafía el binarismo occidental hombre-mujer en cuanto al discurso del sistema sexo-género de esa época. En esos años, la tecnología aún no proporcionaba las herramientas para que los usuarios interactuaran con otros a través de un computador, sin embargo, esto no fue impedimento para que comenzaran a surgir teorías relacionadas en cuanto a cuerpo y máquina. En 1992, la académica inglesa Sadie Plant comienza a utilizar el término ciberfeminismo para definir una fusión entre ciberespacio y feminismo, aseverando que la tecnología ha sido cómplice del feminismo, en tanto máquina, computador y red comparten atributos con la mujer, en cuanto a su flexibilidad, fluidez y capacidad de decisión, y que esta puede ser una poderosa herramienta para comunicar y hacer ciencia (Sadie PLANT, 1997). Plant recuerda y homenajea a 2 Nombre dado en Chile a una manifestación de denuncia y repudio público contra una persona o grupo que cometió una acción deshonesta, ilegal o violenta. 3 En Twitter están como @movs_patriota. Cabe destacar que medios como La Tercera les dieron espacio a sus dirigentes para hablar; no así a las víctimas. Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 3 CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ Ada Lovelace, creadora olvidada de la primera Máquina Analítica, la cual es el pasado de lo que conocemos en la actualidad como computadora. Ada Lovelace, creadora olvidada de la primera Máquina Analítica, la cual es el pasado de lo que conocemos en la actualidad como computadora. 2. Género 2. Género 2. Género 2. Género 2. Género, Internet y Ciber , Internet y Ciber , Internet y Ciber , Internet y Ciber , Internet y Ciberfeminismo feminismo feminismo feminismo feminismo Pero, a pesar de que Internet constituye una gran plataforma para la acción política y las redes entre mujeres, como detalla Zafra (2005), la Red no ha resistido el poder patriarcal y sigue reiterando modelos de dominación, amparados en muchos casos por el arrojo que da el anonimato y por los procesos autorregulatorios de aquellos que ven que, las identidades históricamente fuertes y las situaciones de dominación y poder reaccionarias que las mantienen, se desmoronan (web). Para Zafra, las mitologías de género no desaparecen en Internet, sino que se amplifican. La violencia se traslada desde el cuerpo físico al cuerpo virtual de las mujeres; un tipo de violencia que, si bien no deja marcas, puede tener consecuencias igual de nefastas. En la misma línea, Manuel Castells postula que Internet no es una mera tecnología: es la misma sociedad, ya que expresa los procesos, intereses, valores e instituciones sociales. En base a esto, postuló el concepto de Sociedad Red: “una sociedad cuya estructura social está construida en torno a redes de información a partir de tecnología de información microelectrónica estructurada en Internet” (CASTELLS, 2001, p.13). Y como Internet constituye la forma organizativa de nuestras sociedades, esta no modifica los comportamientos, sino que los comportamientos se apropian y amplifican en la Red. Esto incluye el machismo y la desigualdad de género. g g Las investigadoras Paola Bonavitta, Jimena De Garay y Jeli Edith Camancho coinciden que, aunque las redes han permitido la visibilidad de los discursos de las mujeres, una mayor participación y las posibilidades de encuentros y articulaciones, ello no acarrea necesariamente igualdad de acceso, de participación y, mucho menos, reconocimiento y respeto a los derechos humanos. Así, el espacio cibernético no deja de ser un lugar donde el sistema patriarcal ejerce mecanismos de poder, y donde las mujeres por un lado, han logrado una mayor inserción en diversos espacios, sitios virtuales y redes de información, y, por el otro, deben seguir cuestionando y resistiendo las invasiones, censuras y abusos de dicho sistema. (Paola BONAVITTA; Jimena DE GARAY; Jeli Edith CAMANCHO, 2015, p.34) A esto se suma que existen mecanismos de exclusión en la Red, tales como brechas socioeconómicas y de género. El Ministerio de Economía de Chile, junto con la firma F & K Consultores, elaboró en 2016 un estudio que permite conocer la conectividad a Internet en los hogares chilenos. 2. Género 2. Género 2. Género 2. Género 2. Género, Internet y Ciber , Internet y Ciber , Internet y Ciber , Internet y Ciber , Internet y Ciberfeminismo feminismo feminismo feminismo feminismo Posteriormente, la palabra ciberfeminismo fue transformándose en una estrategia política, para promover la inclusión al otorgarles a las mujeres el sitio público y masivo que siempre les fue negado (Remedios ZAFRA, 2008), lo que permitió que los discursos feministas comenzaran a circular públicamente a través de Internet, ignorando barreras geográficas, buscando la apropiación de un espacio que sigue considerándose masculino: el de la tecnología. Justo con esta, nació también el activismo digital: un mecanismo poderoso que, mediante la utilización de Internet como las redes sociales, ha ayudado a dar voz a causas que en los medios de comunicación tradicionales no son muy visibles. y El activismo feminista está utilizando los recursos multimediales que entrega Internet, para constituir espacios que permitan promover su causa y re-conocerse con otros movimientos oprimidos, adquiriendo una voz propia en un territorio que hasta hace veinte años parecía ser solo masculino. Esta concepción se ha ido eliminando gracias a los aportes de teóricas desde la tecnología y también por la resignificación de la red como un entorno que bien podría ser femenino. Se valen para ello de los mismos orígenes de Internet: un sistema que ha sido concebido como una red plana, donde todos los nodos de comunicación son iguales, y donde no existe jerarquía, sino reciprocidad a la hora de dar y recibir información: “La concepción de la red plana hace que sea un espacio en el que las mujeres podemos actuar y relacionarnos de manera más cómoda” (Montserrat BOIX, 2002, p. 6). En otras palabras, las herramientas digitales pueden proporcionar una gran ayuda para fortalecer el proceso de organización y de comunicación desde una perspectiva feminista, cual no había sido considerada en su creación. Para Judith Butler (2017), una de las modalidades de resistencia que las personas afectadas por la precariedad han ido desarrollando, busca romper la relación entre el espacio público, la plaza común y el régimen actual. Esto, para poder apropiarse de los espacios establecidos y permeados por el poder existente. En base a lo dicho por Butler, se podría señalar que Internet es uno de los lugares donde se están planteando estas prácticas de escisión de lo constituido a nivel gubernamental y patriarcal, donde, en este caso, mujeres feministas pueden apropiarse de espacios masculinizados. 4 Revista Estudos Feministas, Florianópolis, 27(3): e58797 4 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 4 , p , ( DOI: 10.1590/1806-9584-2019v27n358797 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología La violencia contra la mujer se considera un delito grave contra los Derechos Humanos y afecta seriamente a nuestro país. A pesar de que los medios de comunicación informan los casos de violencia como algo aislado e incluso, como algo insólito, una de cada tres mujeres chilenas ha sufrido violencia por parte de algún familiar, pareja o expareja. Y de esta cifra, el 74% de ellas la recibióì de parte de su pareja o expareja. A esto hay que sumar que un 64% no denuncia, muchas veces por miedo al agresor (ADIMARK, 2013). Anualmente, según cifras oficiales, mueren entre 40 y 50 mujeres en manos de sus parejas, exparejas o convivientes; otras organizaciones feministas cifran los asesinatos en más de 50 al año. Ante este panorama, las agrupaciones feministas chilenas han denunciado y posicionado la temática a través del activismo digital, un camino que les permita llegar a más mujeres y adolescentes en riesgo de violencia machista. Pero estas plataformas digitales han resultado ser un arma de doble filo, donde, si bien se permite denunciar y formar redes de apoyo, además las mujeres se ven expuestas a la violencia, acoso y revictimización. Por eso, es importante destacar que la violencia de género basada en tecnología es un continuo de la violencia que sufren históricamente las mujeres. Tanto la esfera online como la offline están ligadas y no deben ser vistas de manera separada. Como explica la periodista e investigadora Paz Peña (2018), uno de los principales errores es trivializar la violencia que ocurre en espacios digitales o plataformas tecnológicas, porque se cree que comienza y termina ahí y que es algo efímero. Esto solo permite la impunidad y amplificación del problema. Según el mapeo de APC, los principales ataques se dirigen hacia mujeres, y de estos, un 40% son cometidos por personas conocidas por las sobrevivientes. Este estudio distinguió tres categorías principales de mujeres que sufren violencia de género relacionada con la tecnología: Una mujer en una relación íntima con una pareja que resulta violenta, una profesional con perfil público que participa en espacios de comunicación (por ejemplo, periodistas, investigadoras, activistas y artistas) y una sobreviviente de violencia física o sexual. (APC, 6 de marzo de 2015, web) Las principales plataformas y dispositivos denunciados por los actos de violencia cometidos en ellas son Facebook (26%) y celulares (19%). VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS digital más reducida de Latinoamérica, se evidencia que ser de un estrato socioeconómico bajo y además ser mujer es un doble factor de exclusión social y digital. g A nivel global, como explican Bonavitta, De Garay y Camancho (2015), la participación online de las mujeres es menor en comparación a la de los hombres, pero en el norte global esto no ocurre necesariamente porque tengan menos acceso a esta tecnología, sino porque están más expuestas a la violencia y acoso virtual en todas sus manifestaciones. 2. Género 2. Género 2. Género 2. Género 2. Género, Internet y Ciber , Internet y Ciber , Internet y Ciber , Internet y Ciber , Internet y Ciberfeminismo feminismo feminismo feminismo feminismo El resultado evidencia que el 82% del quintil de mayores ingresos del país tiene acceso a Internet, mientras el extremo opuesto, es decir el quintil más pobre, registra solo un 28% de conectividad. En tanto, la medición dividida por género arroja que la utilización de Internet por parte de hombres alcanza un 56%, versus un 46,3% en el caso de las mujeres. Cifra que va descendiendo en paralelo a medición por acceso en quintiles de recursos. Es decir, a pesar de que nuestro país tiene la brecha 4 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10 1590/1806 9584 2019v27n358797 4 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología El acoso llegó Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 5 CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ al punto de que le crearon una página falsa con su nombre, en la que se “fomentaba” el infanticidio y castración de niños. Además, publicaron sus datos personales, como dirección y el nombre de su madre, invitando a asesinarlas o lincharlas (Leonardo SAKAMOTO, 2015) Algo similar le ocurrió a la periodista especializada en videojuegos, Ana Freitas, quien fue amenazada de ser violada y perseguida hasta su casa por publicar un artículo sobre el machismo en chats y foros. Ella explicó en una entrevista a Florencia Goldsman que desde los ataques: estoy más paranoica, ciertamente. Más atenta en relación a mi privacidad, a los detalles de mi vida que revelo por ahí, tanto en la vida on como offline. En el período más intenso de amenazas, tuve que lidiar con el miedo, inseguridad, ansiedad (Florencia, 2015). Como Ana Freitas (2015) relataba en su artículo original, en su investigación en foros nerds se encontró con miles de jóvenes que trataban a las mujeres como “depósitos de semen”. Ella tenía claro que su experiencia no era aislada y que Internet y el ambiente nerd eran bastante hostiles para las mujeres, tal como ocurre con cualquier otro punto del mundo, pero: “Lo que no me parecía coherente es, ¿por qué chicos tan jóvenes, frecuentemente más informados y ‘cultos’ que la media, a veces tímidos y socialmente desadaptados, olvidan que las mujeres son personas como ellos?” (Ana FREITAS, 2015, web). ( , , ) El caso más reciente, fue la difusión de datos personales de la víctima de una violación grupal durante una fiesta de San Fermín, caso conocido mediáticamente como ‘La Manada’. Alrededor de 200 usuarios de Foro Coches y otros cientos de usuarios de Burbuja. Info lograron dar con la identidad de la víctima cruzando datos de la prensa y los difundieron a través de estas páginas con miles de visitas. El caso fue tan polémico, que se hizo una campaña masiva para que los anunciantes que colocaban publicidad en estos sitios se retiraran. El fundador de Foro Choches, Alex Marín, afirmó que habían bloqueado los perfiles de usuarios involucrados en la difusión de datos y culpó a los medios de comunicación por difundir tanta información de la víctima. 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología Foros como estos –llamado schans y muchos del tipo imageboards- tienen símiles en cada país y se caracterizan por su hostilidad hacia las mujeres, poner en tela de juicio a las víctimas de violencia y atacar a mujeres feministas y otras activistas. De hecho, en este mismo comentado foro existían temas de conversación como “Después de escuchar los audios de La Manada... no hay violación”. En Chile, existen sitios símiles donde, entre otros ataques coordinados, existieron campañas para denigrar y desmentir el testimonio de Nabila Rifo: una mujer que sufrió graves heridas y la pérdida de sus globos oculares a manos de su pareja; se generaron una serie de publicaciones y memes en los que se burlaban de sus heridas o donde la denigraban sexualmente. Y, más recientemente, se denunció un foro que sustraía fotos a mujeres, adolescentes y niñas con el fin de humillarlas, acosarlas y solicitar sus datos, infundiendo el terror. Como resume un artículo de Fundación Karisma, organización colombiana que trabaja en la promoción de los derechos humanos en el mundo digital, los ataques a mujeres en Internet tienden a ser personalistas, con frecuentes referencias a las relaciones personales y familiares; descalificativos en cuanto a la apariencia física y la capacidad intelectual; y sexualizados, en donde el cuerpo es usado como arma y campo de batalla. La intimidación no cae en las ideas o los argumentos, sino, más bien, en el hecho de que es una mujer quien se expresa y opina públicamente. (Amalia TOLEDO, 2016, web) Por lo general, son atacadas por escribir sobre temáticas “de hombres” (política, videojuegos, deportes) o por hablar de los derechos de las mujeres, feminismo, la comunidad LGBT+ o la denuncia del machismo en la sociedad. Como concluye esta fundación colombiana, “escribir sobre estos asuntos provoca agresiones e insultos que buscan deslegitimar el feminismo y reducir el poder de influencia pública de las mujeres a su mínima expresión” (TOLEDO, 2016, web). p p j p ( ) La periodista e investigadora chilena, Karen Vergara (2017), recabó revelar, en una serie de entrevistas, que el principal obstáculo para las organizaciones feministas chilenas es el acoso y virulencia por parte de sujetos masculinos en Internet: Las redes sociales y nuestra web son el acercamiento más rápido que tenemos con las mujeres que nos siguen. 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología Solo el 49% de los casos fueron denunciados a las autoridades y, de estos, menos de la mitad (41%) fueron investigados. La abogada y activista mexicana, Gisela Pérez de Acha (2018), explica que la violencia de género en línea está íntimamente ligada a los estereotipos de género que hemos aprendido desde nuestra infancia. Se excluye a las mujeres de temáticas como deporte, política y tecnología y se les exige que sean ‘señoritas’: buenas, sumisas, recatadas. Como profundiza: A las mujeres que opinamos de esas cosas ‘que no nos incumben’, o quienes disfrutamos de nuestra sexualidad de forma libre a través de la tecnología, se nos responde con violencia. Es un castigo social para mantener a raya a las mujeres disidentes que salen de sus roles (Gisela, 15 de marzo de 2018, web). Como bien resume Paz Peña (2017) en su informe para las Naciones Unidas, las adolescentes y mujeres están más expuestas a esto, al punto de que el 65% de los casos de ciberbullying corresponden a mujeres; además, ellas son las principales víctimas de la difusión de pornografía no consentida, violencia que se ha cuadruplicado en los últimos 4 años. La principal plataforma para promover campañas de odio y difundir contenido sexual es Twitter, mientras que en Facebook es donde más son agredidas las activistas. g Como afirman las periodistas e investigadoras,Florencia Goldsman y Graciela Natansohn (2016), estos ataques en línea tienen consecuencias muy reales en la llamada vida real: dañan la reputación, aíslan a las víctimas, limitan su movilidad tanto en Internet como en la vía pública, generan depresión, miedo, ansiedad, trastornos del sueño, entre otros. El mapeo de APC reveló que un 33% de las víctimas sufren un daño emocional, que impide su plena participación en el mundo offline y online, un daño a la reputación (18%) e invasión a la privacidad (18%). A eso, se suma que en 11% de los casos denunciado hubo daños físicos; por ende, la Internet es utilizada para facilitar las violaciones y violencia en la vida offline. Goldsman y Natansohn recogen en su artículo una serie de casos de violencia digital en Latinoamérica: está el de la bloguera feminista Lola Aronovich. Esta fue amenazada de muerte por haber denunciado a un sitio brasileño masculinista en el que se formentaba el odio. Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS El factor común es evidente: la figura masculina asociada a un sistema patriarcal que resiste cambios. Para la académica Inés Crosas Remón (2016), el único método para combatir a los llamados trolls de Internet es con el contra-discurso. Es decir, desplegar el poder comunicativo y horizontal de la red para llamar la atención y coordinarse contra el ciberacoso que silencia a muchas mujeres. j Se hace preocupante la baja cantidad de denuncias y la poca recepción que tienen estas en las policías y juzgados. Ante esto, la investigadora Paz Peña (2018) recomienda, en su informe para las Naciones Unidas, promover el acceso a la justicia, entrenando a policías, fiscales y jueces en este tipo de violencia y ajustando leyes ya existentes a la contingencia actual. Entre otras recomendaciones, Peña propone a los Estados: reconocer la encriptación, anonimato y la protección de datos como derechos vitales para proteger a mujeres de la violencia vinculada a tecnología; que las víctimas tengan derecho a la reparación y que sea proporcional a la magnitud del ataque; reformular los llamados crímenes de honor, ya que muchas veces sirve de mecanismo para silenciar y desincentivar la crítica; buscar alternativas no criminales y de educación en temáticas de género. En cuanto al sector privado – por ejemplo, compañías como Facebook y Twitter –, Peña recomienda contextualizar el problema a nivel regional, ya que las soluciones que funcionan en el norte global no necesariamente pueden funcionar en países latinoamericanos. Además, promover el acceso a herramientas que reportan los abusos, ya que muchas veces están en inglés o tienen lenguaje muy técnico, y que quienes reciban las denuncias estén calificados para evaluarlas de acuerdo con el contexto regional. 5. Caracterización de las encuestadas 5. Caracterización de las encuestadas 5. Caracterización de las encuestadas 5. Caracterización de las encuestadas 5. Caracterización de las encuestadas La mayoría de las encuestadas se definía como feministas (85,3%); además, un 8% se define como activista por los pueblos y mujeres indígenas y un 3% como activista por los derechos LGBT. Al preguntarles a través de qué actividades ejercían el feminismo o activismo, un 84,7% dijo difundir información sobre feminismo, activismo y/o derechos de la mujer. También, gran parte de ellas lo ejercían a través de la sororidad, defensa y apoyo a otras mujeres (82,8%). Otras respuestas fueron: definiéndose como feministas en la Red o su entorno social (77,9%), participando en protestas y convocatorias (77,9%), a través de la autonomía sobre sus cuerpos o cuerpas (60,7%) y educando a otras sobre feminismos o derechos de la mujer (58,3%). Otras mencionan actividades en particulares a las que se dedican, tales como: trabajar con mujeres indígenas, trabajar en ONG’s, escribiendo e, incluso, a través de las enseñanzas del feminismo antiespecista y comunitario (con una alimentación vegana o comerciando en base a trueque). Unas 50 encuestadas (30,6%) declararon expresamente no pertenecer a ninguna organización. El resto milita o es parte de una gran variedad de colectivos, corporaciones, ONG’s, asociaciones y partidos. Destaca la Red Chilena contra la Violencia a la Mujer, Observatorio Contra el Acoso Callejero Chile, FEM Chile y Colectiva We Monguen. g Un 54% de las mujeres encuestadas tiene entre 19 y 29 años (88 mujeres) y un 25,8% tiene entre 30 y 39. Le siguen mujeres de entre 40 y 49 años (13,5%), mujeres de 50 o más (4,3%) y mujeres de 18 años o menos (2,5%). Cabe destacar que las encuestadas son principalmente de las regiones Metropolitana (38,7%), del Biobío (30,7%), Valparaíso (7,4%) y Araucanía (4,9%), debido a que las autoras del estudio tienen más redes en estas regiones; de paso, estas son, coincidentemente, las regiones con mayor población de Chile. Además, un 4,9% era de Coquimbo, un 1,8%, de Arica y Parinacota y un 1,2% de Atacama. Hubo solo una encuestada en cada una de las regiones de Los Ríos, Los Lagos, Aysén y Magallanes. 4. Metodología 4. Metodología 4. Metodología 4. Metodología 4. Metodología Dado que al momento de comenzar esta investigación no existen cifras sobre violencia vinculada a tecnología en Chile ni mucho menos ciberviolencia contra feministas y otras activistas, este trabajo es exploratorio y busca ser un primer acercamiento a la problemática a nivel nacional. Se encuestó a 163 mujeres chilenas que se definían como feministas o como activistas por los derechos de la mujer, a través de una plataforma online. Esto se enmarca dentro de una investigación de carácter cualitativo, por un lado, con entrevistas a mujeres activistas feministas, y por otro lado cuantitativo, a través de un muestreo no probabilístico por conveniencia, pero que incluyó a mujeres de todo Chile. Como existía el peligro de que la encuesta fuera objeto de ataques digitales, que se viera alterada o que se atacara a quienes la respondían, se le prohibió a las encuestadas difundirla en redes sociales abiertas: solo a través del boca a boca o en mensajes privados. 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología 3. Violencia contra las mujeres basada en tecnología Es ahí donde publicamos las acciones que estamos realizando, y un canal a través del cual la gente sabe que puede escribirnos [...] Por lo mismo, tenemos que lidiar con bastante misoginia, proveniente de hombres y sus privilegios. (entrevista a representante del Observatorio contra el Acoso Callejero, junio de 2017) Nuestro sitio está hecho para que las víctimas puedan entregar sus testimonios en un espacio donde se garantiza su anonimato y que les creemos, pero no faltan los comentarios de hombres diciendo que ‘se lo buscaron’ o que ‘le damos color o mucha importancia a algo común’. Eso como mujeres jóvenes nos duele [...] Nos han pasado cosas fuertes, como que algunos foros machistas de Internet nos busquen y bloqueen nuestras cuentas RUT [una cuenta vista del Banco de Chile] –, o saquen nuestros certificados de nacimiento o domicilio. Más allá de eso no pueden hacer más, pero ya es peligroso ver al nivel que pueden llegar con su odio. (Fragmento de entrevista a C.G., mayo de 2017) Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ veces” o lo califica como “trolleo” o “pesadeces”, intentando bajarle el tono o trivializando lo ocurrido. Por ende, si se suman las respuestas afirmativas, más aquellas que calificaron los ataques como leves, se concluye que un 73,6% de ellas ha sufrido violencia en entornos digitales. Al preguntarles si una mujer estaba más expuesta a violencia cibernética por ser feminista o activista, un 97,5% respondió de forma afirmativa. veces” o lo califica como “trolleo” o “pesadeces”, intentando bajarle el tono o trivializando lo ocurrido. Por ende, si se suman las respuestas afirmativas, más aquellas que calificaron los ataques como leves, se concluye que un 73,6% de ellas ha sufrido violencia en entornos digitales. Al preguntarles si una mujer estaba más expuesta a violencia cibernética por ser feminista o activista, un 97,5% respondió de forma afirmativa. Un 91,7% declaró haber sido atacada verbalmente, a través de burlas, insultos, humillaciones y groserías. Además, un 25,8% fue objeto de acoso y hostigamiento, otro 22% recibió amenazas y un 15% sufrió la publicación de información falsa (injurias y calumnias). También, un 13,3% declaró que le sustrajeron imágenes de las redes sociales y otro 13,3% recibió imágenes y/o videos sexualmente agresivos. g Otras formas de ataques reportados, es la publicación de datos personales en webs y foros (10%), hackeo de redes sociales (8,3%), suplantación de identidad (6,7%) y publicación de imágenes íntimas y eróticas sin consentimiento (1,7%). Además, dos mujeres (1,7%) afirmaron que los ataques pasaron del plano virtual al físico. Algunas mujeres detallaron sus experiencias personales con este tipo de violencia: a una le publicaron las imágenes de sus hijos sin consentimiento y a otra le enviaron las imágenes de una mujer “deformada por los golpes”, insinuando que le ocurriría lo mismo. Las principales medidas que tomaron las encuestadas atacadas fueron: bloquear al atacante en las redes sociales (62,7% lo hace), aumentar la seguridad en sus redes sociales (54,2%), conversar el tema con su círculo de confianza (36,4%) y denunciar lo ocurrido en las mismas redes sociales (28%). Cabe destacar que un 13,6% no hizo nada al respecto y un 3,4% decidió acudir a alguna forma de terapia. También, en cinco casos (4,2%) se acudió a Carabineros (Fuerzas de Orden y Seguridad de Chile), PDI (Policía Civil de Investigaciones de Chile), o juzgados. CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ En 4 de los cinco casos no hubo ningún tipo de constancia o apertura de investigación, un solo caso fue recibido, sin mayores avances posteriores. Otras tres mujeres optaron por cerrar sus redes sociales de manera preventiva (2,5%). En casos particulares, las mujeres declaran haberse defendido respondiendo y contraargumentando; otra mujer logró interponer una querella y otra solicitó asesoría legal a la Fundación Datos Protegidos, organización sin fines de lucro, cuya misión es la promoción, defensa y fortalecimiento de los derechos a la privacidad y protección de los datos personales como derechos fundamentales. En general, existe una ridiculización del feminismo y de las luchas por los derechos de la mujer, por lo que, si una mujer escribe desde este discurso, recibirá burlas agresivas, humillaciones y ataques; la violencia suele ser escondida bajo “bromas”, como enviar a mujeres “a la cocina, donde pertenecen”. Como resume una mujer encuestada: La gente siempre es ofensiva cuando uno tiene un discurso feminista. A las mujeres se nos trata de putas y de fáciles por postear imágenes reveladoras o con un fin artístico. Me molesta que crean que una es un objeto; antes de todos los insultos y clasificación de mi vida, soy una persona que es libre de hacer lo que quiera con su cuerpo. (testimonio anónimo) Debido a los constantes ataques, muchas mujeres optan por la autocensura, por ende, este tipo de violencia afecta su libre circulación en entornos digitales y su libre expresión: Ya no publico tanto con relación al feminismo como antes. Y de hacerlo ignoro o borro comentarios ‘hater’ o muy ignorantes o machistas. Básicamente, al publicar algo me decían en broma ‘feminazi’, se armaban peleas o discusiones eternas entre varios amigos de las redes, lo que me terminaba angustiando. Y al públicamente mostrarme por las redes como feminista, significaba burlas o comentarios mal intencionados en juntas con amigos en la realidad, pelambres o incluso prejuicios sobre mí. (testimonio anónimo) Me hice otro perfil en el que mi foto es la de una madre y sus hijos. En ese perfil opino de igual forma que en mi perfil verdadero, pero el hecho de parecer una madre criando desincentiva las agresiones machistas. (testimonio anónimo) Una de las formas de acoso más utilizadas por los machitrolles chilenos es el envío de imágenes gráficas, morbosas y violentas para amedrentar a la víctima. 6. Resultados y análisis 6. Resultados y análisis 6. Resultados y análisis 6. Resultados y análisis 6. Resultados y análisis De 163 encuestadas, un 71,2% declara expresamente haber sido víctima de violencia cibernética. Solo un 26,4% dijo expresamente que no. El restante 2,4% afirma haberlo sufrido “a Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 7 CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ 8 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS Un usuario me amenazó con ‘violarme hasta la tráquea’. Otro grupo de usuarios sacó fotos de mi Facebook e hicieron memes humillantes de mis fotos y de mi familia. Diariamente recibo insultos en RRSS. (testimonio anónimo) Me mandaron una foto mía amenazando de muerte a mis hijos, mi madre y a mí. Me subieron y bajaron a garabatos, pero no pesqué; me mandaron unos 4 o 5 correos en ese mismo tono y como no respondí, no volvieron a hacerlo. (testimonio anónimo) y baja o a ga aba os, pe o o pesqué; e a da o u os o 5 co eos e ese s o o o y como no respondí, no volvieron a hacerlo. (testimonio anónimo) Las sobrevivientes de violencia sexual también son blanco constante de ataques basados en tecnología, sufriendo revictimización: Un abogado se burló de las personas que hemos sido víctimas de violaciones, a raíz de despenalización de aborto en 3 causales. Al enfrentarlo, trató de corregir sus palabras y luego me mandó al psiquiatra y que me medicara por ‘mi trauma’. Por otra parte, recibí mucho apoyo, [así que] sólo me incentivó más a hablar de lo ocurrido. (testimonio anónimo) Algo que caracteriza a los ataques misóginos en Internet, es que nunca se cree en la palabra de una mujer. Siempre son “exageradas”, “histéricas” o “buscan atención”, motivo por el cual se les ataca y silencia. Existen cuentas en Twitter, Instagram y otras redes sociales que, si bien no atacan directamente, facilitan la agresión y el desprestigio de mujeres: Hombres de una cuenta de Twitter tomaron ciertas publicaciones que hice relacionadas a momentos en que me he sentido acosada. Se mofaron de eso y lo pusieron en duda, instando a otros usuarios a burlarse, agredirme y desearme la muerte. (testimonio anónimo) También son comunes los ataques grupales y las referencias a pornografía y violación: También son comunes los ataques grupales y las referencias a pornografía y violación: Fui atacada por un grupo de amigos que se dedican a denostar mujeres por Twitter, en especial, uno, que fue el más violento, me enviaba por interno imágenes y videos de una actriz porno que, según él, le recordaba a mí […] tengo entendido que sigue acosando a otras niñas, incluso menores de edad. CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ El principal objeto es acallarlas y hacerlas sentir inseguras en espacios digitales y físicos: El episodio más violento fue una imagen que me enviaron a través de Facebook, alguien que no estaba en mi lista de amigos. Era una foto de una mujer sin ojos con un lema que aludía a que yo podía ser la siguiente. (testimonio anónimo) El más grave fue cuando presté apoyo a una chica trans y un grupo de hombres me incluyó en una lista de mensajes directos, donde compartían fotografías explícitas de violaciones y decían que nos iban a hacer eso. (testimonio anónimo) Otra forma de atacar es a través de la sustracción de las mismas imágenes que las mujeres tienen en sus redes sociales de forma pública, incluyendo apariciones en la prensa o fotos familiares. La violencia suele estar dirigida a su intimidad y sus redes más cercanas: 8 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS (testimonio anónimo) Al preguntarles si el tema de la violencia cibernética se discutía en asambleas, colectivos o agrupaciones, un 75,5% declara que este tema no se suele conversar, en contraposición a un 16,6% que sí lo ha hablado. Se intuye que esto ocurre por distintos motivos: miedo, querer dejar lo ocurrido en el pasado, normalización -debido a lo constante y masivo de los ataques-, sensación de que lo que ocurre en Internet es “pasajero” o no equiparable al “mundo real”, por las mismas brechas de género presentes en la tecnología y por la sensación de impunidad. Esta misma situación de incredulidad o idea de que lo que ocurre en Internet no se traspasa al plano físico, es lo que genera mayor debate sobre qué debemos hacer ante ataques recibidos a través de redes sociales. ¿Debemos ignorar, dar la batalla o encontrar nuevas formas de congregarnos en la red? Verónica Engler (2017) traduce lo que Anita Sarkeesian señala en el manifiesto The Feminist Frequency para promover la seguridad digital entre mujeres y comunidad LGBT+ quienes reciben la mayor cantidad de ataques: Desearíamos no tener que escribir esto. Tomar algunas de estas medidas para garantizar tu seguridad online te costará tiempo real y, a veces, dinero. Es una sanción impuesta a las mujeres, la gente de color, queer y transgénero, y otros grupos oprimidos por atrevernos a expresar nuestras opiniones en público. (Verónica ENGLER, 2017, web) Se hace necesario entonces ahondar en investigaciones que sitúen estas discusiones en nuestro espacio cotidiano, en nuestros países y comunas para poder alzar la voz contra el sistema que sigue reproduciendo estos ataques, y que se escuda en el anonimato en la Red para no hacer nada al respecto. 9 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 7. Reflexiones a modo de conclusión 7. Reflexiones a modo de conclusión 7. Reflexiones a modo de conclusión 7. Reflexiones a modo de conclusión 7. Reflexiones a modo de conclusión Este estudio corrobora que la violencia relacionada a tecnología es un problema que constantemente afecta a las mujeres feministas y otras activistas de Chile, sin olvidar que este tipo de violencia también es dirigida a otras personas, ya sea por su religión, etnia, orientación sexual e identidad sexual. Es un problema interseccional y de género, constantemente ignorado y trivializado, incluso por las propias víctimas, ya que se asume que aquello que está en Internet no es real. Esto conlleva una utilización del ciberespacio como un territorio donde se siguen replicando conductas propias de una sociedad patriarcal. Las cifras de acceso a Internet en Chile y el tipo de utilización que se le da al mismo vuelven a reproducir lo que ocurre en nuestro país a nivel cultural y social, en el cual prima la desigualdad, la exclusión y la discriminación, en este caso específico estudiado, en contra de las agrupaciones feministas que se congregan de manera digital. Es importante destacar que cuando se intentó esbozar un perfil típico de los usuarios digitales que atacan colectivos feministas y pro DDHH en la web, se determinó que muchos se auto describían como adultos, ya sea con estudios en ámbitos donde ha calado fuerte la lucha por los derechos Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 9 CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ de la mujer (como las Humanidades y Ciencias Sociales) o de carreras STEAM (como las ingenierías y carreras ligadas a tecnología y matemática); los primeros se sentían amenazados por los avances de las mujeres en estos espacios; mientras que los segundos se encuentran en posiciones de mayor prestigio social, desde donde atacan y miran de manera inferiorizante a las mujeres. Muchos, incluso, declaraban tener hijos, hijas “no feministas”, pareja o esposa. Se hace hincapié en esto, para aclarar que no son casos aislados ni se trata de un problema de “educación”, sino que se trata de un continuo de violencia machista, cometida principalmente por jóvenes y adultos con cierto nivel socioeconómico y de capital cultural -y, por ende, con cierto poder- y que, además, esta estrategia no solo se repite en Chile, sino que el resto del mundo. Quienes cometen estos ataques se escudan en el anonimato, el “rebaño” y la distancia física para manifestar sus verdaderas intenciones, las cuales no manifestarían de manera pública. Su objetivo es infundir temor, angustia y terror. 7. Reflexiones a modo de conclusión 7. Reflexiones a modo de conclusión 7. Reflexiones a modo de conclusión 7. Reflexiones a modo de conclusión 7. Reflexiones a modo de conclusión Muchas veces va de la mano de discursos racistas, clasistas y que tienden a encontrar espacio en grupos de ultraderecha: el Movimiento Social Patriota, que fue acusado por los ataques a las mujeres en la marcha del aborto, eran cercanos a políticos de ultraderecha, como es el caso de José Antonio Kast, quien constantemente dirige su discurso contra las mujeres feministas, los migrantes haitianos o el pueblo mapuche. g Una mujer que sale de la esfera privada y que levanta la voz es vista como un peligro, ya que rompe con lo preestablecido y, por ende, debe ser atacada, humillada, acosada y amenazada. Sus historias, denuncias y saberes son negados, ya que hay una mirada inferiorizante hacia las mujeres. De los testimonios analizados en este trabajo se desprende que los ataques son sumamente personalistas: se dirigen hacia el aspecto, familia, trabajo, datos personales. Las amenazas suelen ir orientadas hacia su integridad física, sexual y psicológica. Llama la atención que al mismo tiempo que los machitrolles atacan y se felicitan públicamente por ser violentos -en foros ya mencionados-, simultáneamente niegan que exista la violencia contra la mujer y la desigualdad de género. Viven en una especie de esquizofrenia social: mientras ellas luchan por la igualdad y los derechos, ellos se resisten y aferran a la sociedad conservadora y patriarcal que conocen, a través de una violencia que fomentan y niegan simultáneamente. Ellas salen a las calles por un mundo mejor; ellos pasan horas y horas en el computador difundiendo odio. p El panorama recogido por este estudio demuestra lo importante que es posicionar este tema en la opinión pública y también educar en torno a la violencia relacionada a tecnología, la violencia de género y el discurso de odio; especialmente, porque este último suele ser confundido con la libertad de expresión y, muchas veces, ha sido el caldo de cultivo para grupos de odios y ataques físicos, como lo ocurrido en la marcha por el aborto libre de Santiago de Chile. Se trata de una problemática urgente que necesita ser abordada como sociedad, para que no se continúe amplificando en medio de la impunidad legal y la reticencia a ser abordada como violencia de género digital por parte de autoridades y ciudadanía. Referencias Referencias Referencias Referencias Referencias ADIMARK. Encuesta Nacional de Victimización por Violencia Intrafamiliar y Delitos Sexuales. Ministerio del Interior y Seguridad Pública. [En línea]. Chile, 2013. Disponible en http:// www.seguridadpublica.gov.cl/filesapp/Presentacion%20VIF_adimark_final.pdf. Acceso el 03/03/16. ADIMARK. Encuesta Nacional de Victimización por Violencia Intrafamiliar y Delitos Sexuales. Ministerio del Interior y Seguridad Pública. [En línea]. Chile, 2013. Disponible en http:// www.seguridadpublica.gov.cl/filesapp/Presentacion%20VIF_adimark_final.pdf. Acceso el 03/03/16. ASSOCIATION FOR PROGRESSIVE COMMUNICATIONS. “Infografía: Mapeo de violencia contra las mujeres relacionada con la tecnología ¡Dominemos la tecnología! 8 datos importantes”. GenderIT [En línea]. España, 6/3/2015. Disponible en https://www.genderit.org/es/resources/infograf-mapeo-de-violencia- contra-las-mujeres-relacionada-con-la-tecnolog-dominemos-la-te. Acceso el 15/05/2018. BONAVITTA, Paola; DE GARAY, Jimena; CAMACHO, Jeli Edith. “Mujeres, feminismos y redes sociales: acceso, censura y potencialización”. Revista Questión, La Plata, v. 1, n. 48, p. 33-44, dic. 2015. BONAVITTA, Paola; DE GARAY, Jimena; CAMACHO, Jeli Edith. “Mujeres, feminismos y redes sociales: acceso, censura y potencialización”. Revista Questión, La Plata, v. 1, n. 48, p. 33-44, dic. 2015. BOIX, Montserrat. “Feminismos, comunicación y tecnologías de la información”. Aportes Andinos, Quito, n. 4, nov. 2002. BOIX, Montserrat. “Feminismos, comunicación y tecnologías de la información”. Aportes Andinos, Quito, n. 4, nov. 2002. BUTLER, Judith. Cuerpos aliados y lucha política: hacia una performatividad de la asamblea. Paidós, Argentina, 2017. BUTLER, Judith. Cuerpos aliados y lucha política: hacia una performatividad de la asamblea. Paidós, Argentina, 2017. CASTELLS, Manuel. Internet y la Sociedad Red. En: LECCIÓN INAUGURAL DEL PROGRAMA DE DOCTORADO SOBRE LA SOCIEDAD DE LA INFORMACIÓN Y EL CONOCIMIENTO, evento sin número, 2001, Barcelona, Universitat Oberta de Catalunya. Sin editorial, documento recuperado por el Sistema Universidad Abierta y Educación a Distancia de la Universidad Nacional Autónoma de México, sin número de página. Disponible en http://fcaenlinea.unam.mx/anexos/1141/ 1141_u5_act1.pdf. Acceso el 15/05/2018. 10 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10 1590/1806 9584 2019v27n358797 10 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 10 0 , p , ( DOI: 10.1590/1806-9584-2019v27n358797 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS CROSAS REMÓN, Inés. Sexismo en la red: análisis de la ciberviolencia en contra del ciberfeminismo en Youtube. 2016. Maestría (Postgrado en Estudios Internacionales sobre Medios, Poder y Diversidad) – Universitat Pompeu Fabra, Barcelona, Provincia de Barcelona, España. EL DESCONCIERTO. “Fallo a favor de periodista demandada por reportaje constituye que violencia contra las mujeres es de interés público y no privado”. El Desconcierto [En línea]. Chile, 8/2/2018. Disponible en http://www.eldesconcierto.cl/2018/02/08/fallo-a-favor-de-periodista-demandada- por-reportaje-constituye-que-violencia-contra-las-mujeres-es-de-interes-publico-y-no-privado/. Acceso el 07/07/2018. EL DESCONCIERTO. “Grupos neonazis se disparan un 22% y aparecen por primera vez colectivos de supremacía masculina con Trump en el poder”. El Desconcierto [En línea]. Chile, 26/2/2018. Disponible en http://www.eldesconcierto.cl/2018/02/26/grupos-neonazis-se-disparan-un-22-y- aparecen-por-primera-vez-colectivos-de-supremacia-masculina-con-trump-en-el-poder/. Acceso el 07/07/2018. ENGLER, Verónica. Antifeminismo online. En Nueva Sociedad, Democracia y Política en América Latina [En línea]. 2017. Disponible en http://nuso.org/articulo/antifeminismo-online/#footnote-17. Acceso el 15/05/18. FERREYRA, Gonzalo. Internet paso a paso: hacia la autopista de la información. Ciudad de México: Alfa Omega, 1996. FREITAS, Ana. “Nerds e machismo: por que mulheres não são bem vindas nos fóruns e chans”. HuffPost Brasil [En línea]. Brasil, 2/2/2015. Disponible en https://www.huffpostbrasil.com/ana-freitas/ nerds-e-machismo-porque-m_b_6598174.html?1422906690. Acceso el 15/05/2018. GOLDSMAN, Florencia. “Una internet sin violencia hacia la mujer solo va suceder en un mundo sin violencia hacia la mujer”. GenderIT [En línea]. España, 11/05/ 2015. Disponible en https:// www.genderit.org/es/feminist-talk/una-internet-sin-violencia-hacia-la-mujer-solo-va-suceder-en-un- mundo-sin-violencia-ha. Acceso el 12/03/18. GOLSDMAN, Florencia; NATANSOHN, Graciela. “Violencia contra las mujeres en red, vigilancia y el derecho a la privacidad”. En: SIMPÓSIO NACIONAL ABCIBER, XI, 2016, São Paulo, Pontificia Universidad Católica de São Paulo. Anales de ABCiber 2016. Sin local de publicación ni número de página. Disponible en https://www.academia.edu/32199540/Violencia_contra_las_mujeres_en _red_vigilancia_y_el_derecho_a_la_privacidad?auto=download. Acceso el 02/04/2018. MIÑO, Claudia. “Aumentan denuncias por ciberbullying en colegios: mujeres son las más afectadas”. BioBioChile [En línea]. Concepción, Chile, 22/7/2018. Disponible en https://www.biobiochile.cl/ noticias/nacional/chile/2018/07/22/aumentan-denuncias-por-ciberbullying-en-colegios-mujeres- son-las-mas-afectadas.shtml. Acceso el 23/07/2018. OLMO, Guillermo. “‘Ven a las mujeres como una plaga’: cómo son los grupos supremacistas masculinos incluidos por primera vez en el mapa del odio de Estados Unidos”. BBC Mundo [En línea]. 5/3/2018. Disponible en https://www.bbc.com/mundo/noticias-internacional-43253925. Acceso el 07/03/2018. PEÑA OCHOA, Paz et. al.. “Reporte de la situación de América Latina sobre la Violencia de Género ejercida por Medios Electrónicos”. Tedic [En línea]. 11/4/2017. Disponible en https://www.tedic.org/ wp-content/uploads/sites/4/2017/11/Latin-American-Report-on-Online-Gender-Violence-final.pdf. Acceso el 10/03/2018. PEÑA OCHOA, Paz. “Recommendationsontechnology-relatedViolenceAgainstWomen (VAW) forthe UN”. Medium [En línea]. 1/03/2018. Disponible en https://medium.com/@pazpena/recommendations-on- technology-related-violence-against-women-vaw-for-the-un-5e27b544e6b2. Acceso el 30/03/2018. PÉREZ DE ACHA, Gisela. “Internet: un espacio político para nosotras”. Derechos Digitales [En línea]. 15/03/2018. Disponible en https://www.derechosdigitales.org/11965/internet-un-espacio-politico- para-nosotras/. Acceso el 20/04/2018. 11 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 CONTRIBU CONTRIBU CONTRIBU CONTRIBU CONTRIBUCIÓN CIÓN CIÓN CIÓN CIÓN DE AUTORÍA DE AUTORÍA DE AUTORÍA DE AUTORÍA DE AUTORÍA La concepción de esta investigación se dio a la par, ya que ambas autoras sufrieron violencia de género en Internet y notaron que era una problemática recurrente en su entorno y que había un vacío legal y estadístico en Chile. El diseño de la encuesta, recolección y análisis de datos la llevó a cabo principalmente Ananías, aunque la redacción y discusión de datos se dio nuevamente a la par. Cabe destacar los aportes desde los Estudios de Género de Vergara, que conforman la base teórica de este estudio. COMO COMO COMO COMO COMO CIT CIT CIT CIT CITAR ESTE ARTÍCUL AR ESTE ARTÍCUL AR ESTE ARTÍCUL AR ESTE ARTÍCUL AR ESTE ARTÍCULO OO OO, DE A , DE A , DE A , DE A , DE ACUERDO CON L CUERDO CON L CUERDO CON L CUERDO CON L CUERDO CON LAS NORMAS DE L AS NORMAS DE L AS NORMAS DE L AS NORMAS DE L AS NORMAS DE LA REVIST A REVIST A REVIST A REVIST A REVISTAAAAA::::: SOTO, Cecilia Alejandra Ananías; SÁNCHEZ, Karen Denisse Vergara. “Título do artigo”. Revista Estudos Feministas, Florianópolis, v. 27, n. 3, e58797, 2019. PLANT, Sadie. Digital Women + the New Technocultures. Nueva York: Doubleday, l997. p. 37 PLANT, Sadie. Digital Women + the New Technocultures. Nueva York: Doubleday, l997. p. 37 SAKAMOTO, Leonardo. “Meu nome é Lola. E estou ameaçada de morte por ser feminista”. Blog Do Sakamoto [En linea]. 8/11/2015. Disponible en https://blogdosakamoto.blogosfera.uol.com.br/2015/ 11/08/meu-nome-e-lola-e-estou-ameacada-de-morte-por-ser-feminista/?cmpid=copiaecola. Acceso el 02/02/2018. 11 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 11 CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ FINANCIACIÓN FINANCIACIÓN FINANCIACIÓN FINANCIACIÓN FINANCIACIÓN Este estudio se llevó a cabo de manera independiente y sin apoyo financiero de ningún tipo. Cabe destacar que en Chile prácticamente no hay forma de solicitar fondos para investigación en torno a Tecnología y Derechos Humanos. CONSENTIMIENTO DE USO DE IMAGEM CONSENTIMIENTO DE USO DE IMAGEM CONSENTIMIENTO DE USO DE IMAGEM CONSENTIMIENTO DE USO DE IMAGEM CONSENTIMIENTO DE USO DE IMAGEM CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ CECILIA ALEJANDRA ANANÍAS SOTO E KAREN DENISSE VERGARA SÁNCHEZ TELETRECE. “Mujer apuñalada en marcha pro aborto: ‘Jamás imaginé lo que podía venir’”. T13 [En línea]. 26/07/2018. Disponible en http://www.t13.cl/noticia/nacional/video-mujer-apunalada- marcha-pro-aborto-jamas-imagine-podia-venir. Acceso el 26/07/2018. THE HUFFINGTON POST. “Los incel, los misóginos que matan porque no consiguen tener sexo”. The Huffington Post [En línea]. 7/05/2018. Disponible en https://www.huffingtonpost.es/2018/05/07/los-incel- los-misoginos-que-matan-porque-no-consiguen-tener-sexo_a_23428957/. Acceso el 07/05/2018. TOLEDO, Amalia. “Misoginia en internet: bombardeo a campo abierto contra las periodistas”. Fundación Karisma [En línea]. Bogotá, 24/02/2016. Disponible en https://karisma.org.co/misoginia- en-internet-bombardeo-a-campo-abierto-contra-las-periodistas/. Acceso el 04/04/2018. VERGARA SÁNCHEZ, Karen Lazos activistas de mujeres chilenas en la red. ¿Puede Internet ser una herramienta al servicio del ciberfeminismo? 2017. Magíster (Programa de Postgrado en Estudios de Género y Cultura, mención Humanidades) - Universidad de Chile, Santiago, Región Metropolitana, Chile. ZAFRA, Remedios. “La escritura invisible, el ojo ciego y otras formas (fragmentadas) del poder y la violencia de género en Internet”. Remedios Zafra blog. [En línea]. España, 2005. Disponible en http://www.remedioszafra.net/carceldeamor/vsc/textos/textorz.html. Acceso el 02/04/2018. ZAFRA, Remedios. “Lo que decimos fue, lo que no quiso ser y lo que queremos del ciberfeminismo”. Mujer y Cultura Visual [En linea]. 2008. Disponible en https://www.remedioszafra.net/mcv/ pensamiento/tx/text_rz08.htm. Acceso el 02/04/2018. ISSN 1697-8218 Cecilia Alejandra Ananías Soto Cecilia Alejandra Ananías Soto Cecilia Alejandra Ananías Soto Cecilia Alejandra Ananías Soto Cecilia Alejandra Ananías Soto (ananiascecilia@gmail.com) es periodista, escritora e investigadora chilena, con maestría en Ciencias de la Comunicación. Es fundadora y Directora General de ONG Amaranta, organización que busca educar en torno a Género e Interculturalidad. Sus líneas de trabajo son: Violencia de Género y Medios de Comunicación y Tecnología y Derechos Humanos. KKKKKaren Denisse V aren Denisse V aren Denisse V aren Denisse V aren Denisse Vergara Sánchez ergara Sánchez ergara Sánchez ergara Sánchez ergara Sánchez (karen.vergara.s@gmail.com) es periodista, Bachiller en Humanidades y candidata a Magíster en Estudios de Género y Cultura en América Latina, mención Humanidades de la Universidad de Chile. Directora de Comunicaciones de ONG Amaranta y colaboradora en temas de comunicación y género en Fundación Datos Protegidos. Sus líneas de investigación abordan Cultura, Violencia de Género, Tecnologia y Medios de Comunicación. 13 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 CONSENTIMIENTO DE USO DE IMAGEM CONSENTIMIENTO DE USO DE IMAGEM CONSENTIMIENTO DE USO DE IMAGEM CONSENTIMIENTO DE USO DE IMAGEM CONSENTIMIENTO DE USO DE IMAGEM 12 Revista Estudos Feministas, Florianópolis, 27(3): e58797 DOI: 10.1590/1806-9584-2019v27n358797 12 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN No aplicable CONFLICTO DE INTERESES CONFLICTO DE INTERESES CONFLICTO DE INTERESES CONFLICTO DE INTERESES CONFLICTO DE INTERESES No aplicable LICENCIA DE USO LICENCIA DE USO LICENCIA DE USO LICENCIA DE USO LICENCIA DE USO Este artículo está licenciado bajo la Licencia Creative Commons CC-BY International. Con esta licencia se puede compartir, adaptar, crear material para cualquier objetivo, siempre que se le atribuya la autoría. HISTORIAL HISTORIAL HISTORIAL HISTORIAL HISTORIAL Recebido em 23/08/2018 Reapresentado em 05/03/2019 Aprovado em 27/03/2019 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN APROBACIÓN DE COMITÉ DE ÉTICA EN INVESTIGACIÓN No aplicable CONFLICTO DE INTERESES CONFLICTO DE INTERESES CONFLICTO DE INTERESES CONFLICTO DE INTERESES CONFLICTO DE INTERESES No aplicable LICENCIA DE USO LICENCIA DE USO LICENCIA DE USO LICENCIA DE USO LICENCIA DE USO Este artículo está licenciado bajo la Licencia Creative Commons CC-BY International. Con esta licencia se puede compartir, adaptar, crear material para cualquier objetivo, siempre que se le atribuya la autoría. HISTORIAL HISTORIAL HISTORIAL HISTORIAL HISTORIAL Recebido em 23/08/2018 Reapresentado em 05/03/2019 Aprovado em 27/03/2019 VIOLENCIA EN INTERNET CONTRA FEMINISTAS Y OTRAS ACTIVISTAS CHILENAS Este artículo está licenciado bajo la Licencia Creative Commons CC-BY International. Con esta licencia se puede compartir, adaptar, crear material para cualquier objetivo, siempre que se le atribuya la autoría. HISTORIAL HISTORIAL HISTORIAL HISTORIAL HISTORIAL Recebido em 23/08/2018 Reapresentado em 05/03/2019 Aprovado em 27/03/2019 Recebido em 23/08/2018 Reapresentado em 05/03/2019 Aprovado em 27/03/2019
https://openalex.org/W2050174527	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0054332&type=printable	English	null	Function of Mouse Embryonic Stem Cell-Derived Supporting Cells in Neural Progenitor Cell Maturation and Long Term Cxpansion	PloS one	2,013	cc-by	11,764	Introduction progenitors in vitro also remains to be addressed. First of all, what cell type is more committed neural progenitor? Or in another word, the critical time when the neural-progenitors are fully competent to undergo neurogenesis and the time of their isolation from other surrounding cells that are not undergoing neurogenesis are yet to be determined. Can these more committed neural progenitors be passaged without losing their potential to differentiate into neurons? The fate and function of cells that do not undergo neurogenesis is yet another interesting question to be answered. Are these cells helpful in the differentiation of NPCs into neurons or are they byproducts of the differentiation? Mouse embryonic stem cells (ES) have the potential to differentiate into many cell types and are thus considered potential cell therapy candidates to treat neurodegenerative diseases [1– 3].To avoid teratoma formation in ES cells and prevent damage to fully differentiated mature neurons during transplantation, ES derived neuronal progenitor cells (NPC) are the preferred cell types in neural degenerative disease research [4–7]. Understand- ing the development of neural progenitor cells becomes important. In mouse, the most frequently used technique to differentiate ES cells to neurons is the 5-step method [8–10], and stromal- derived inducing activity (SDIA) method. In 5-step method, cells in the expanding stage (stage 4) are used as NPCs [6,11–13]. Given SDIA method, ES cells cultured on PA6 or MS5 feeder cells for a specific period are also used as NPCs [14–18]. In both of the methods, the developmental process of neural Mouse embryonic stem cells (ES) have the potential to differentiate into many cell types and are thus considered potential cell therapy candidates to treat neurodegenerative diseases [1– 3].To avoid teratoma formation in ES cells and prevent damage to fully differentiated mature neurons during transplantation, ES derived neuronal progenitor cells (NPC) are the preferred cell types in neural degenerative disease research [4–7]. Understand- ing the development of neural progenitor cells becomes important. In mouse, the most frequently used technique to differentiate ES cells to neurons is the 5-step method [8–10], and stromal- derived inducing activity (SDIA) method. In 5-step method, cells in the expanding stage (stage 4) are used as NPCs [6,11–13]. Given SDIA method, ES cells cultured on PA6 or MS5 feeder cells for a specific period are also used as NPCs [14–18]. ceived June 25, 2012; Accepted December 11, 2012; Published January 14, 2013 an 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: 2013 Guan et al. This is an open-access article distributed under the terms of the Creative Commons Attributi unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: This work was supported by a research grant from the High Level Talent Fund of the Beijing Healthcare System (Grant No: 2009-2-14, 2011-3-093), National Science Foundation of China (Grant No. 81272804; 30940039; 30970939), National Science Foundation of Beijing (Grant No. 7102078), Beijing Nova Program of Science and Technology (Grant No. 2009B22), Youth Foundation of Capital Medical University and Science Foundation of the Ministry of Education in China (Grant No. 11120066) and Key Project of Chinese Ministry of Education (Grant No. 210003). 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: yaz@bjsap.org (YAZ); chlyz34@163.com (LC) ¤ Current address: Institute of Zoology, Chinese Academy of Science, Beijing, P. R. China Function of Mouse Embryonic Stem Cell-Derived Supporting Cells in Neural Progenitor Cell Maturation and Long Term Cxpansion Yunqian Guan1,3, Qing-An Du1,3, Wanwan Zhu1,3¤, Chunlin Zou1,3, Di Wu1,3, Ling Chen2, Yu Alex Zhang1,3 1,3, Qing-An Du1,3, Wanwan Zhu1,3¤, Chunlin Zou1,3, Di Wu1,3, Ling Chen2, Yu Guan1,3, Qing-An Du1,3, Wanwan Zhu1,3¤, Chunlin Zou1,3, Di Wu1,3, Ling Chen2, Y ng1,3* 1 Cell Therapy Center, Beijing Institute of Geriatrics, Xuanwu Hospital, Capital Medical University, Beijing, P. R. China, 2 Department of Neurosurgery, Xuanwu Hospital, Capital Medical University, Beijing, P. R. China, 3 Key Laboratory of Neurodegeneration, Ministry of Education, P. R. China 1 Cell Therapy Center, Beijing Institute of Geriatrics, Xuanwu Hospital, Capital Medical University, Beijing, P. R. China, 2 Depar Capital Medical University, Beijing, P. R. China, 3 Key Laboratory of Neurodegeneration, Ministry of Education, P. R. China January 2013 \| Volume 8 \| Issue 1 \| e54332 Abstract Background: In the differentiation of mouse embryonic stem (ES) cells into neurons using the 5-stage method, cells in stage 4 are in general used as neural progenitors (NPs) because of their ability to give rise to neurons. The choice of stage 4 raises several questions about neural progenitors such as the type of cell types that are specifically considered to be neural progenitors, the exact time when these progenitors become capable of neurogenesis and whether neurogenesis is an independent and autonomous process or the result of an interaction between NP cells and the surrounding cells. Methodology/Principal Findings: In this study, we found that the confluent monolayer cells and neural sphere like cell clusters both appeared in the culture of the first 14 days and the subsequent 6 weeks. However, only the sphere cells are neural progenitors that give rise to neurons and astrocytes. The NP cells require 14 days to mature into neural lineages fully capable of differentiation. We also found that although the confluent monolayer cells do not undergo neurogenesis, they play a crucial role in the growth, differentiation, and apoptosis of the sphere cells, during the first 14 days and long term culture, by secreted factors and direct cell to cell contact. Conclusions/Significance: The sphere cells in stage 4 are more committed to developing into neural progenitors than monolayer cells. Interaction between the monolayer cells and sphere cells is important in the development of stage 4 cell characteristics. Citation: Guan Y, Du Q-A, Zhu W, Zou C, Wu D, et al. (2013) Function of Mouse Embryonic Stem Cell-Derived Supporting Cell Maturation and Long Term Cxpansion. PLoS ONE 8(1): e54332. doi:10.1371/journal.pone.0054332 Citation: Guan Y, Du Q-A, Zhu W, Zou C, Wu D, et al. (2013) Function of Mouse Embryonic Stem Cell-Derived Supporting Cells in Neural Progenitor Cell Maturation and Long Term Cxpansion. PLoS ONE 8(1): e54332. doi:10.1371/journal.pone.0054332 Editor: Irina Kerkis, Instituto Butantan, Brazil Received June 25, 2012; Accepted December 11, 2012; Published January 14, 2013 Received June 25, 2012; Accepted December 11, 2012; Published January 14, 2013 2. Sphere Cells are Neural Progenitor Cells during the First 14 Days and the Subsequent 6 Weeks of Stage 4 Spheres are oval or round with diameters of 100–300 nm and are not completely differentiated even after stage 5. While most sphere cells are nestin positive (Fig. 2B1–B3), undifferentiated Oct- 4 positive cells do exist (Fig. 2A1–A3). We found that spheres are mainly composed of neural progenitors. Stage 4 sphere cells were separated on day 14, cultured for subsequent 6 weeks, and differentiated independently at week 5 and 8 in stage 5 medium for 7 days. The numbers of neurons and astrocytes were counted after Tuj-1 and GFAP staining. Without monolayer cells, the spheres give rise to 35– 40%Tuj-1 positive (Fig. 2C1–C3) neurons and 6–9% GFAP positive astrocytes (Fig. 2D1–D3). In our previous report [29], we observed two cell types on day 14 in the mouse ES cell derived neural progenitor expanding stage, which is stage 4. We reported that one cell type grows like neural spheres and are scattered among the second cell type, which are the confluent monolayer of cells. While the neural sphere cells give rise to neurons and astrocytes, the confluent monolayer cells do not. But we found that the monolayer cells support the growth of spheres by inducing their proliferation, decreasing apoptosis and increasing the overall percentage of neuron formation. However, throughout the 8 week culture of stage 4, monolayer cells lack Oct-4 positive cells (Fig. 3A1–A3), but are positive for nestin (Fig. 3B1–B3). Regardless of the time of testing, these monolayer cells are neither progenitors of neurons (Fig. 3C1–C3) nor astrocytes (Fig. 3D1–D3). In this study, we divided the 8 week long stage 4 into the first 14 days and the subsequent 6 weeks, and explored whether stage 4 cells could be divided into spheres and monolayer cells during the two periods, and more importantly, whether sphere cells had the potential to be neural progenitors. Because NPCs in general require some time to obtain the complete ability to give rise to neurons, we also determined the number of days required for the mouse ES derived NPCs to obtain the ability to undergo neurogenesis. In addition, we studied the effects of monolayer cells on sphere cell differentiation into neurons during the first 14 days and the next 6 weeks, and found that it was achieved by cell- cell contact and (or) secreted factors. Effect of Surrounding Cells on Neural Progenitors sphere cells (Fig. 1C) both of which can be separated and cultured in the respective stage 4 medium. maintained them for 4 weeks with 1,000-fold expansion without significant changes in their phenotype. Similarly, Hayashi et al obtained ‘‘adherent neurospheres’’ with a modified EB formation method and cultured them for 12 weeks [21]. All these results suggest that the NP cells could be cultured for longer duration and harvested in higher quantities. We also found that when stage 4 cells start differentiation to stage 5 on day 14, the percentage of Tuj-1 positive neurons (Fig. 1D, E, F) is 60–70% and GFAP positive astrocytes is 7–12%. In addition, during the subsequent 6 weeks of culture, cells differentiate into stage 5 and on week 5 and 8 have similar percentages of neurons and astrocytes. This indicates that the ability of stage 4 cells to undergo neurogenesis remains unchanged during long culture periods. Other evidences suggest some cells in NPC are more committed to neurons, and the neurogenesis of mES derived neural progenitors is not an autonomous process, but is influenced by surrounding cells. For example, the critical role of the in vitro or in vivo microenvironment in the differentiation of stem cells or NPCs has been studied. Transplantation of the ES cells cultured on MS5 or PA6 for longer time could increase the neural differentiation of graft and decrease the potential of tumor formation risk significantly [15,22–24]. PA6 cell surface activity is required for neural differentiation of hESC, but secreted factors are required for the specific DA neuron-inducing effect [25]. Transplantation of stem cells into different areas of the brain results in a difference in differentiation suggesting that the fate of the graft is influenced by cell-cell contact and secreted factors released either by the graft or the host [26–28]. 3. First 14 Days in Stage 4 is Essential for NPC to Obtain Continuous Ability for Neurogenesis In the experiments about stage 4 cells as a whole and sphere cells alone. To avoid the impact of collagenase disassociation on the characteristics of sphere cells alone, stage 4 cells used in these experiments were recombined sphere cells and monolayer cells. In brief, sphere cells were separated by collagenase, then plated on the monolayer cells, and differentiated in stage 5 medium. To determine the time required for stage 4 cells to obtain the ability to undergo complete neurogenesis, we isolated sphere cells on days 3, 7, 10, and 14 in stage 4 and started stage 5 differentiation. The percentage of Tuj-1 positive neurons after stage 5 differentiation was significantly higher in later than earlier days, and was 0, 12.265.2, 22.467.9, and 39.965.7% (p,0.05) (Fig. 4A, 4C, 4E, 4G). Introduction In both of the methods, the developmental process of neural Cumulating evidences suggest that NPCs can be expanded. Human ES cell derived NPCs maintain the ability to undergo neurogenesis during a long term culture [19]. Chung et al [20] isolated Otx2+ Corin+ NP cells at the end of stage 3 and January 2013 \| Volume 8 \| Issue 1 \| e54332 1 PLOS ONE \| www.plosone.org Effect of Surrounding Cells on Neural Progenitors January 2013 \| Volume 8 \| Issue 1 \| e54332 1. Sphere Cells and Monolayer Cells Exist in the First 14 Days and the Subsequent 6 Week Culture of Stage 4 Cells (C) After collagenase treatment and sorting with 40 mm mesh, the cell aggregates can be disassociated from the surrounding cells and cultured in suspension to form spheres. (D) Phase contrast image of stage 5 cells finished differentiation. (E)Tuj-1staining of the stage 5 cells after 7 days of differentiation. The neurons with Tuj-1 staining are located mainly in the cell clusters. Tuj-1 positive cells could be hardly found in monolayer cells. (F) Merged image of the Tuj-1 staining and DAPI staining of nucleus in the same field. Green, Tuj-1; Red, DAPI. Scale bars, 100 mm. doi:10.1371/journal.pone.0054332.g001 ect o Su ou d g Ce s o eu a oge to s Figure 1. Typical cell types of stage 4 are sphere cells and monolayer cells. (A) Cell aggregates and surrounding monolayer cells at the beginning of stage 4. Within 14 days, the spheres will grow bigger and the monolayer cells will reach 90% confluence. (B)The independently cultured monolayer cells after the removal of cell aggregates by collagenase IV. (C) After collagenase treatment and sorting with 40 mm mesh, the cell aggregates can be disassociated from the surrounding cells and cultured in suspension to form spheres. (D) Phase contrast image of stage 5 cells finished differentiation. (E)Tuj-1staining of the stage 5 cells after 7 days of differentiation. The neurons with Tuj-1 staining are located mainly in the cell clusters. Tuj-1 positive cells could be hardly found in monolayer cells. (F) Merged image of the Tuj-1 staining and DAPI staining of nucleus in the same field. Green, Tuj-1; Red, DAPI. Scale bars, 100 mm. doi:10.1371/journal.pone.0054332.g001 BrdU incorporation rate of stage 4 sphere cells on days 7 and 14 in culture are 40.7611.0 and 50.067.4% and on weeks 5 and 8 are 43.365.0 and 42.366.3% respectively. These values are significantly lower than BrdU incorporation rates observed when sphere and monolayer cells are cultured together, which are 57.064.8, 67.167.6, 66.068.1 and 63.0066.6% respectively. 1. Sphere Cells and Monolayer Cells Exist in the First 14 Days and the Subsequent 6 Week Culture of Stage 4 Cells The re-combined sphere cells and monolayer cells also started to differentiate on days 3, 7, 10, and 14 in stage 4,and the percentage of neurons from these cells significantly increased with the increase in the number of days in culture, from zero to 17.366.4, 48.163.0, and 68.6611.3% (p,0.05) (Fig. 4B, 4D, 4F, 4H). These results demonstrate higher neural yields than sphere cells alone in each time point. This test in addition also served to observe the effect of monolayer cells on the neurogenesis of sphere cells. y g The cell morphology during stage 4 is characterized by cell aggregates scattered amongst monolayer cells in addition to numerous fibers between the ell aggregates (Fig. 1A). In our experiments, when cells in stage 3 are disassociated and plated into stage 4, cell aggregates form on day two or three after plating and can be separated from the monolayer using collagenase. Morphology of these detached cell aggregates are similar to neurospheres isolated from fetal or adult nerve tissue of rodents or humans. Because of this similarity we coin the term ‘‘sphere cells’’ in this study. In contrast, cells surrounding the cell aggregates are termed as ‘‘monolayer cells’’ due to their characteristic morphol- ogy. During the subsequent 6 weeks in culture, stage 4 cells did not require another 14 days in stage 4 medium to obtain the complete potential to differentiate into neurons. Usually they need 5–7 days to reach 80% confluence, and when transferred into stage 5 medium for differentiation, as high as 60–70% neurons were observed. Sphere cells alone, meanwhile, yielded only 40–45% when grown in the subsequent 6 weeks. In general 5–7 days after plating, stage 4 cells attain 80% confluence and are ready for passage. Repeated passing and 8 weeks of continuous culture resulted in monolayer (Fig. 1B) and January 2013 \| Volume 8 \| Issue 1 \| e54332 January 2013 \| Volume 8 \| Issue 1 \| e54332 PLOS ONE \| www.plosone.org 2 Effect of Surrounding Cells on Neural Progenitors Effect of Surrounding Cells on Neural Progenitors Figure 1. Typical cell types of stage 4 are sphere cells and monolayer cells. (A) Cell aggregates and surrounding monolayer cells at the beginning of stage 4. Within 14 days, the spheres will grow bigger and the monolayer cells will reach 90% confluence. (B)The independently cultured monolayer cells after the removal of cell aggregates by collagenase IV. 4. The Effect of Monolayer Cells on the Proliferation of Sphere Cells Whether proliferation, apoptosis and differentiation character- istics of ES derived NPCs are autonomous or influenced by neighboring cells are not clear yet. PLOS ONE \| www.plosone.org January 2013 \| Volume 8 \| Issue 1 \| e54332 3 Effect of Surrounding Cells on Neural Progenitors Figure 2. Characterization of sphere cells. (A) The sphere cells isolated by the end of the long term culture in stage 4 still contain Oct-4 cells. (B) The sphere cells are nestin positive. (C) After 5 stages of differentiation, the sphere cells alone gave rise to 35–40% Tuj-1positive neu After 5 stages of differentiation, GFAP positive astrocytes derived from sphere cells were 6–9%. Green in A-2, B-2, C-2 and D-1 are Oct-4, nest and GFAP. Blue, DAPI. Scale bars, 100 mm. doi:10.1371/journal.pone.0054332.g002 Figure 2. Characterization of sphere cells. (A) The sphere cells isolated by the end of the long term culture in stage 4 still contain Oct-4 positive cells. (B) The sphere cells are nestin positive. (C) After 5 stages of differentiation, the sphere cells alone gave rise to 35–40% Tuj-1positive neurons. (D) After 5 stages of differentiation, GFAP positive astrocytes derived from sphere cells were 6–9%. Green in A-2, B-2, C-2 and D-1 are Oct-4, nestin, Tuj-1, and GFAP. Blue, DAPI. Scale bars, 100 mm. doi:10.1371/journal.pone.0054332.g002 This suggests that proliferation of sphere cells in vivo are not autonomous but rather requires supporting cells, monolayer cells in our experiment, for growth and maximum proliferation (Fig. 5A). This suggests that proliferation of sphere cells in vivo are not autonomous but rather requires supporting cells, monolayer cells in our experiment, for growth and maximum proliferation (Fig. 5A). such as MEFs also promote sphere cell proliferation albeit not as much as monolayer cells. Thus, the ability to promote sphere cell growth is not unique to monolayer cells. We further studied whether the effect of monolayer cells or MEFs on sphere cell proliferation is a result of cell-cell contact or secreted cytokines. The BrdU incorporation rate in sphere cells cultured in medium conditioned with both monolayer cells and/or MEFs is significantly higher than in fresh stage 4 medium. This implies that secreted cytokines contribute to the proliferation of sphere cells. However, the BrdU incorporation rate in sphere cells cultured in the monolayer or MEF conditioned medium was not as In the next step, we tested if the proliferation effect of monolayer cells can be replaced by other cells. 4. The Effect of Monolayer Cells on the Proliferation of Sphere Cells For this, week 5 in stage 4 was selected as the time point and CF1 mouse embryonic fibroblasts (MEFs), which is a commonly used feeder cell in ES culture, was used as substitution for monolayer cells. We found that MEFs also significantly increased the rate of BrdU incorporation (56.867.2%) in sphere cells. This showed that cells January 2013 \| Volume 8 \| Issue 1 \| e54332 January 2013 \| Volume 8 \| Issue 1 \| e54332 PLOS ONE \| www.plosone.org 4 Effect of Surrounding Cells on Neural Progenitors Figure 3. Characterization of confluent monolayer cells. (A) At the end of 8 weeks long culture of stage 4, the separated monolayer cells are negative for Oct-4. (B) The flat monolayer cells are positive for nestin. (C) The flat monolayer cells are negative for neuron marker, Tuj-1, and negative for astrocyte marker GFAP (D). Green, nestin; Red, DAPI; Blue, DAPI. Scale bars, 100 mm (A, C and D) and 50 mm (B). doi:10.1371/journal.pone.0054332.g003 Figure 3. Characterization of confluent monolayer cells. (A) At the end of 8 weeks long culture of stage 4, the separated monolayer cells are negative for Oct-4. (B) The flat monolayer cells are positive for nestin. (C) The flat monolayer cells are negative for neuron marker, Tuj-1, and negative for astrocyte marker GFAP (D). Green, nestin; Red, DAPI; Blue, DAPI. Scale bars, 100 mm (A, C and D) and 50 mm (B). doi:10.1371/journal.pone.0054332.g003 high as that cultured with monolayer cells. This clearly indicates that cell-cell contact is also involved in sphere cell growth (Fig. 5B). cells, the percentage of neurons in sphere cells was significantly (p,0.05) higher in the last three time points (17.366.4, 48.163.0, and 68.6611.3%) (Fig. 5C). January 2013 \| Volume 8 \| Issue 1 \| e54332 5. The Effect of Monolayer Cells on the Differentiation of Sphere Cells The BrdU incorporation rate of sphere cells cultured in conditioned medium of both monolayer cells and MEFs were significantly higher than sphere cells cultured in fresh stage 4 medium, but was not as high as that in combined monolayer and sphere cells. This indicates that secreted cytokines contribute to the proliferation of sphere cells mostly. (C) During the first 14 days and the next 6 weeks of stage 4, the percentage of neurons derived from sphere cells alone was always significantly lower than that in combined sphere and monolayer cells. This suggests that differentiation of neural progenitors into neurons is prompted by monolayer cells. (D) MEFs did not prompt the differentiation of sphere cells. The sphere cells cultured with the conditioned medium from monolayer cells or MEFs gave rise to similar percentages of neurons as sphere cells alone in fresh medium, but still dramatically lower than the sphere cells on monolayer cells (* p,0.05, n = 4–5). (Abbreviations: MEF: mouse embryonic fibroblast, M: monolayer cells; S: sphere cells; M-media: monolayer cell conditioned medium; MEF-media: MEF conditioned medium). doi:10.1371/journal.pone.0054332.g005 Figure 5. The effect of monolayer cells on the proliferation and differentiation of sphere cells. (A) At the time point of day 7 and 14, week 5 and 8 in stage 4, the BrdU incorporation rate of spheres alone was significantly lower than in the combined sphere cells and monolayer cells. (B) MEFs had a similar effect on sphere cell proliferation as monolayer cells at week 5. The BrdU incorporation rate of sphere cells cultured in conditioned medium of both monolayer cells and MEFs were significantly higher than sphere cells cultured in fresh stage 4 medium, but was not as high as that in combined monolayer and sphere cells. This indicates that secreted cytokines contribute to the proliferation of sphere cells mostly. (C) During the first 14 days and the next 6 weeks of stage 4, the percentage of neurons derived from sphere cells alone was always significantly lower than that in combined sphere and monolayer cells. This suggests that differentiation of neural progenitors into neurons is prompted by monolayer cells. (D) MEFs did not prompt the differentiation of sphere cells. 5. The Effect of Monolayer Cells on the Differentiation of Sphere Cells The percentages of GFAP positive astrocytes in sphere cells that began to differentiate on days 3, 7, 10 and 14 in stage 4 without monolayer cells were 0, 7.363.7, 5.363.3, and 2.761.3% respectively. When monolayer cells were included in the culture, the sphere cells differentiated into 0, 12.865.2, 10.963.4 and 7.761.4% of astrocytes. In the last three time points, astrocyte percentage in the presence of monolayer cells are significantly higher than in their absence (p,0.05) (Fig. S1). In the following long term culture of stage 4, the Tuj-1 positive neurons after differentiation of spheres alone on weeks 5 and 8 were 37.366.7 and 34.866.3%, which was lower when compared The percentages of GFAP positive astrocytes in sphere cells that began to differentiate on days 3, 7, 10 and 14 in stage 4 without monolayer cells were 0, 7.363.7, 5.363.3, and 2.761.3% respectively. When monolayer cells were included in the culture, the sphere cells differentiated into 0, 12.865.2, 10.963.4 and 7.761.4% of astrocytes. In the last three time points, astrocyte percentage in the presence of monolayer cells are significantly higher than in their absence (p,0.05) (Fig. S1). To determine the effect of monolayer cells on sphere cell differentiation, sphere colonies were isolated as described above and differentiated in stage 5 medium. Sphere cells in combined cultures (spheres and monolayer cell) also underwent a similar process of differentiation. During the first 14 days in stage 4, we picked sphere cells on days 3, 7, 10 and 14 for differentiation to stage 5 with or without monolayer cells. Percentages of Tuj-1 positive neurons from sphere cells alone were 0, 12.365.2, 22.467.9 and 39.965.7% respectively. However, in the combined cultures with monolayer In the following long term culture of stage 4, the Tuj-1 positive neurons after differentiation of spheres alone on weeks 5 and 8 were 37.366.7 and 34.866.3%, which was lower when compared January 2013 \| Volume 8 \| Issue 1 \| e54332 PLOS ONE \| www.plosone.org 5 Effect of Surrounding Cells on Neural Progenitors PLOS ONE \| www.plosone.org 6 January 2013 \| Volume 8 \| Is January 2013 \| Volume 8 \| Issue 1 \| e54332 January 2013 \| Volume 8 \| Issue 1 \| e54332 PLOS ONE \| www.plosone.org PLOS ONE \| www.plosone.org Effect of Surrounding Cells on Neural Progenitors Effect of Surrounding Cells on Neural Progenitors Figure 4. 5. The Effect of Monolayer Cells on the Differentiation of Sphere Cells Two weeks in stage 4 is necessary for sphere cells to obtain complete ability to differentiate into neuron. The sphere cells (A, C, E, G) on days 3, 7, 10 and 14 in stage 4, were separated from monolayer cells and then differentiated alone or combined with monolayer cells. The Tuj-1 positive neurons in spheres isolated on day 3(A), 7(C), 10(E) and 14(G) reached the percentages of 0, 12.365.2, 22.467.9, and 39.965.7%. Each time point was significantly higher than the previous (p,0.05). In the presence of monolayer cells, the spheres isolated on day 3(B), 7(D), 10(F), and 14(H) in stage 4 gave rise to increasing numbers of neurons with percentages of 0, 17.366.4, 48.163.0, and 68.6611.3%. The later time points were significantly higher than the previous. Green,Tuj-1; Blue, DAPI. Scale bars, 50 mm. doi:10.1371/journal.pone.0054332.g004 to 63.665.1 and 61.265.4% when combined with monolayer cells (Fig. 5C). the unique potential to promote sphere cell differentiation into neurons. To understand the role of cell-cell contact and secreted factors of sphere and monolayer cells in neurogenesis, we compared the neuron percentage after differentiation using a conditioned medium method. We also tested whether the ability of monolayer cells to promote differentiation could be replaced by MEFs, during long term culture of stage 4 cells. When co-cultured with MEFs, sphere cells gave rise to only 40–45% Tuj-1 positive neurons and 4.5–5% GFAP positive astrocytes which are similar to the percentages of differentiation with sphere cells alone. This showed that MEFs do not have the ability to promote differentiation of neurons or astrocytes as monolayer cells. In other words, monolayer cells have The results showed that the percentage of neurons was 68.663.5% in stage 4 cells that includes normally formed sphere and monolayer cells. In the group with sphere cells cultured with conditioned medium from monolayer cells, the percentage of Figure 5. The effect of monolayer cells on the proliferation and differentiation of sphere cells. (A) At the time point of day 7 and 14, week 5 and 8 in stage 4, the BrdU incorporation rate of spheres alone was significantly lower than in the combined sphere cells and monolayer cells. (B) MEFs had a similar effect on sphere cell proliferation as monolayer cells at week 5. 7. Interaction of Monolayer and Sphere Cells Leads to Tumor Formation A critical issue to be considered in cell transplantation therapy is the risk of teratoma formation. The ability of pre-differentiated ES cells to form teratoma remains unclear and conflicting results have been reported so far. Because Oct-4 staining is observed in sphere cells, we previously hypothesized that sphere cells may be the cause of tumor formation after transplantation. Thirdly, cells in stage 4 require 14 days to gain complete ability to give rise to neurons as high as 60–70%. If stage 4 cells as a whole or sphere cells alone started to differentiate into stage 5 before day 14, the percentage of Tuj-1 positive neurons after differentiation reduces significantly. During the following 6 weeks in vitro culture, stage 4 cells require only 5–7 days to form new sphere cells when passaged. These new aggregates, once formed, can give rise to nearly 60–70% Tuj-1 positive neurons after stage 5 differentiation. Dihne´ [30] was the first to report the formation of substrate adherent embryonic stem cell induced cell cluster (SENA), which are considered to be neural progenitors after 18 days of culture in stage 4 medium. The difference in stage 4 medium used in his study and ours is the replacement of B27 by N2 in our protocol. Here, sphere cells cultured in stage 4 medium for 14 rather than 18 days were able to differentiate to , 70% neurons. While Dihne´’s study did not include the effect of long term culture on the ability of stage 4 cells to give rise to neurons, we found that during Thirdly, cells in stage 4 require 14 days to gain complete ability to give rise to neurons as high as 60–70%. If stage 4 cells as a whole or sphere cells alone started to differentiate into stage 5 before day 14, the percentage of Tuj-1 positive neurons after differentiation reduces significantly. During the following 6 weeks in vitro culture, stage 4 cells require only 5–7 days to form new sphere cells when passaged. These new aggregates, once formed, can give rise to nearly 60–70% Tuj-1 positive neurons after stage 5 differentiation. To test the role of sphere and monolayer cells in tumorigenesis, we increased the cell number for transplantation to 56106. After transplantation, monolayer cells did not form any tumors while sphere cells formed tumors in 2 out of 10 mice. Effect of Surrounding Cells on Neural Progenitors neurons was significantly lower (39.665.4%) and was the similar (41.866.7%) when sphere cells differentiated alone without any supporting cells. Culture of sphere cells together with MEF or in MEF conditioned medium didn’t increase the percentage of neurons either, which were 40.563.8% and 39.665.4% (Fig. 5D). To demonstrate that the interaction between sphere and monolayer cells is important in tumor formation, we separated the two types of cells and cultured them independently for 1 week. This was followed by transplantation together into nude mice at a ratio of 10:1. Tumors appeared in all transplanted animals. Tumor formation latency, which is the time between trans- plantation and appearance of palpable tumor, was 1563 days when sphere cells and monolayer cells were transplanted together after 7 days in independent culture. This is significantly longer (963 days) when the cells were cultured independently for 3 days before transplantation. This indicates that when sphere cells and monolayer cells are cultured separately for longer time before transplantation, more time is needed for them to interact and form tumor after transplantation. Within the neoplasma, tissues of three germ layers were observed including glandular epithelial (endo- derm) (Fig. 7A), connective tissue (Fig. 7B) (mesoderm), and immature epithelial (ectoderm) (Fig. 7C). 6. The Effect of Monolayer Cells on Apoptosis of Sphere Cells To detect apoptosis, active caspase 3 antibody staining was used. During the first 14 days in stage 4 culture, the apoptosis rates of sphere cells isolated on days 3, 7, 10 and 14 and expanded in stage 4 medium for another 2 days without monolayer cells were 82.867.8, 88.464.6, 30.068.7 and 34.766.6%. When sphere cells and monolayer cells were separated on days 3, 7, 10, 14 followed by culturing as a whole for 2 days, apoptosis rates of 37.269.4, 24.8612.3, 17.166.1, and 12.565.8% (Fig. 6A–6H) were achieved(Fig. 6I). During the next 6 weeks of continuous culture and passing, the apoptosis rate of combined sphere cells and monolayer cells was 8.061.9 and 6.562.3% on weeks 5 and 8. At the same time points in stage 4, the apoptosis rate of sphere cells which were cultured without monolayer cells for 2 days, increased to 11.562.6 and 10.962.7% (Fig. 6I). At all time points, apoptosis was significantly higher with sphere cells alone compared to that when sphere and monolayer cells were combined. Discussion ESCs have attracted the attention of researchers as potential sources of cell replacement in various diseases. In regenerative research, neural progenitors are needed more often than mature neurons. In previous study, we found that on day 14, stage 4 cells derived from mouse ES can be divided into sphere cells that are neural progenitors, and monolayer cells that are supporting cells playing an important role in accelerating proliferation, reducing apoptosis and prompting differentiation [29]. We also examined whether the anti-apoptotic effect is achieved by cell-cell contact or secreted factors. Apoptosis rate of sphere cells cultured in the conditioned medium from either monolayer cells (11.862.0%) or from MEF (11.361.6%) are not different from the apoptosis rates of sphere cells cultured in fresh medium(11.562.6%), but higher than that in sphere cells cultured together with monolayer cells (8.061.8%)or MEF(9.663.7%) (Fig. 6K). This highlights that it is more important of cell-cell contact than secreted factors in the anti-apoptotic effect. The fact that the effects of MEF in preventing sphere cells from undergoing apoptosis is similar to the monolayer cells suggest that the anti- apoptotic effect is not unique to monolayer cells. All these results suggest that without monolayer cells, the sphere cells were more vulnerable to apoptosis. The results were also confirmed by annexin V and propidium iodide (PI) flow cytometry detection (Fig. 6L). In this study, we aimed to address the following questions: can stage 4 cells be expanded for as long as 8 weeks and maintain the main characteristics, such as, ability of differentiation into neurons? Do sphere cells and monolayer cells exist during the first 14 days and the subsequent 6 weeks in stage 4 culture? Are sphere cells and monolayer cells in culture neural progenitors and supporting cells respectively? If so, when do the real progenitors in stage 4 become competent to differentiate into neurons and what is the function of cells other than neural progenitors towards neural development? In our study, using the 5-stage method, stage 4 cells can be expanded for 8 weeks and differentiated into neurons and glial cells. Therefore, stage 4 cells are in general termed as neural- progenitors. Next, we primarily observed that during the first 14 days and the subsequent 6 weeks long culture, stage 4 cells could be divided into two types based on morphology. Discussion These two cell types termed sphere cells and monolayer cells differ not only in morphology but also in their ability to differentiate into neurons. Sphere cells are the true neural-progenitors. Effect of Surrounding Cells on Neural Progenitors These results and the decreasing apoptosis tendency of sphere cells isolated on days 3, 7, 10 and 14 were confirmed by western blot using active caspase-3 antibody (Fig. 6J). 5. The Effect of Monolayer Cells on the Differentiation of Sphere Cells The sphere cells cultured with the conditioned medium from monolayer cells or MEFs gave rise to similar percentages of neurons as sphere cells alone in fresh medium, but still dramatically lower than the sphere cells on monolayer cells (* p,0.05, n = 4–5). (Abbreviations: MEF: mouse embryonic fibroblast, M: monolayer cells; S: sphere cells; M-media: monolayer cell conditioned medium; MEF-media: MEF conditioned medium). doi:10.1371/journal.pone.0054332.g005 January 2013 \| Volume 8 \| Issue 1 \| e54332 7 PLOS ONE \| www.plosone.org January 2013 \| Volume 8 \| Issue 1 \| e54332 7. Interaction of Monolayer and Sphere Cells Leads to Tumor Formation If sphere and monolayer cells were separated and the same amount (56106) was transplanted together 3 days later at a ratio of 1:10, 100% tumor formation was observed. These results suggest that tumor formation in vivo occurs easily when sphere and monolayer cells are together (Fig. 7D). The ES and the stage 4 cells transplanted in same amounts, developed into tumors in 90 and 100% of the animals respectively within 9 weeks. Dihne´ [30] was the first to report the formation of substrate adherent embryonic stem cell induced cell cluster (SENA), which are considered to be neural progenitors after 18 days of culture in stage 4 medium. The difference in stage 4 medium used in his study and ours is the replacement of B27 by N2 in our protocol. Here, sphere cells cultured in stage 4 medium for 14 rather than 18 days were able to differentiate to , 70% neurons. While Dihne´’s study did not include the effect of long term culture on the ability of stage 4 cells to give rise to neurons, we found that during PLOS ONE \| www.plosone.org January 2013 \| Volume 8 \| Issue 1 \| e54332 8 Effect of Surrounding Cells on Neural Progenitors Figure 6. The effect of monolayer cells on the apoptosis of sphere cells. The sphere cells were separated from monolayer cells on days 3, 7, 10 and 14 in stage 4 and then were cultured alone or combined with monolayer cells for another 2 days before active caspase3 staining. (A), (C), (E), (G) are sphere cells alone, (B), (D), (F), (H) are combined sphere and monolayer cells. (I) During the first 14 days and the next 6 weeks of continuous culture and passage the sphere cells were isolated on days 3, 7, 10 and 14 and, weeks 5 and 8. At all time points, the apoptosis of sphere cells alone was significantly higher than that together with monolayer cells. (J) Analysis for procaspase-3 cleavage and active caspase 3 by immuneblotting confirmed these tendencies. (K) On week 5 in stage 4, MEFs had a similar effect as monolayer cells in preventing sphere cells from apoptosis. Therefore, the anti-apoptotic effect is not unique to monolayer cells. Effect of Surrounding Cells on Neural Progenitors (11.160.2%). (Abbreviations: MEF: mouse embryonic fibroblast, M: monolayer cells; S: sphere cells; M-media: monolayer cell conditioned medium; MEF-media: MEF conditioned medium). Green, active caspase 3; Blue, DAPI. Scale bar, 20 mm. doi:10.1371/journal.pone.0054332.g006 (11.160.2%). (Abbreviations: MEF: mouse embryonic fibroblast, M: monolayer cells; S: sphere cells; M-media: monolayer cell conditioned medium; MEF-media: MEF conditioned medium). Green, active caspase 3; Blue, DAPI. Scale bar, 20 mm. doi:10 1371/journal pone 0054332 g006 cells underwent apoptosis in stage 5 differentiation before completing the first 14 days in stage 4. the subsequent 6-week culture, the ability of NP cells to give rise to 60, 70% neurons did not change. Furthermore, Dihne´ did not study the function of cells surrounding the neural progenitors. The characteristics of sphere cells in the first 14 days and the subsequent 6 weeks are tightly related to the presence of the surrounding monolayer cells. A novel finding of our study is the supporting role played by monolayer cells in accelerating growth, inhibiting apoptosis and promoting differentiation of neural- progenitors. the subsequent 6-week culture, the ability of NP cells to give rise to 60, 70% neurons did not change. Furthermore, Dihne´ did not study the function of cells surrounding the neural progenitors. The characteristics of sphere cells in the first 14 days and the subsequent 6 weeks are tightly related to the presence of the surrounding monolayer cells. A novel finding of our study is the supporting role played by monolayer cells in accelerating growth, inhibiting apoptosis and promoting differentiation of neural- progenitors. The anti-apoptotic effect of monolayer cell is important because in cell transplantation research, usually up to 90% of the transplanted cells die within one week [31,32], and a large proportion of cell death in transplants appears to be apoptotic [33]. Monolayer cells are essential for stage 4 cells to acquire and maintain the ability for neurogenesis. Both during the first 14 days in stage 4 cultures (when sphere cells become competent to give rise to neurons) and the subsequent 6-week long term culture, the presence of monolayer cell results in a higher percentage of differentiated neurons. Although the differentiation protocol we used focused primarily on neurons, the percentage of astrocyte was also higher in the presence of monolayer cells. Identified by the BrdU incorporation rate, we found that without monolayer cells, the growth of sphere cells was slower during the first 14 days and the subsequent 6 weeks in culture. 7. Interaction of Monolayer and Sphere Cells Leads to Tumor Formation (L) Besides cell counting, the percentages of apoptotic (Annexin V-positive, PI negative) cells in sphere cells cultured in the conditioned medium from monolayer cells (10.160.8%) and from fibroblasts (11.060.7%) are higher than that in sphere cells cultured together with monolayer cells (7.860.8%) or fibroblasts (8.160.7%) (p,0.05, n = 5), but not differ from the apoptosis rates of sphere cells cultured in fresh medium January 2013 \| Volume 8 \| Issue 1 \| e54332 PLOS ONE \| www.plosone.org 9 Effect of Surrounding Cells on Neural Progenitors 7. Interaction of Monolayer and Sphere Cells Leads to Tumor Formation The apoptosis rate of sphere cells cultured in the conditioned medium from monolayer cells or fibroblasts, are 11.7961.99 and 11.3461.57%, which are not different from the apoptosis rate of sphere cells cultured in fresh medium (* p,0.05, n = 5). (L) Besides cell counting, the percentages of apoptotic (Annexin V-positive, PI negative) cells in sphere cells cultured in the conditioned medium from monolayer cells (10.160.8%) and from fibroblasts (11.060.7%) are higher than that in sphere cells cultured together with monolayer cells (7.860.8%) or fibroblasts (8.160.7%) (p,0.05, n = 5), but not differ from the apoptosis rates of sphere cells cultured in fresh medium PLOS ONE \| www.plosone.org 9 January 2013 \| Volume 8 \| Issue 1 \| e54332 Figure 6. The effect of monolayer cells on the apoptosis of sphere cells. The sphere cells were separated from monolayer cells on days 3, 7, 10 and 14 in stage 4 and then were cultured alone or combined with monolayer cells for another 2 days before active caspase3 staining. (A), (C), (E), (G) are sphere cells alone, (B), (D), (F), (H) are combined sphere and monolayer cells. (I) During the first 14 days and the next 6 weeks of continuous culture and passage the sphere cells were isolated on days 3, 7, 10 and 14 and, weeks 5 and 8. At all time points, the apoptosis of sphere cells alone was significantly higher than that together with monolayer cells. (J) Analysis for procaspase-3 cleavage and active caspase 3 by immuneblotting confirmed these tendencies. (K) On week 5 in stage 4, MEFs had a similar effect as monolayer cells in preventing sphere cells from apoptosis. Therefore, the anti-apoptotic effect is not unique to monolayer cells. The apoptosis rate of sphere cells cultured in the conditioned medium from monolayer cells or fibroblasts, are 11.7961.99 and 11.3461.57%, which are not different from the apoptosis rate of sphere cells cultured in fresh medium (* p,0.05, n = 5). 1. ES Cell Culture and Differentiation by the 5-stage Method 1. ES Cell Culture and Differentiation by the 5-stage Method We used the mES line R1, passed 30–50 times, in our study. The usage of ES cells and laboratory animals was approved by the Ethical Committee of Xuanwu Hospital at Capital Medical University, Beijing. Mouse ESCs were cultured in knockout Dulbecco’s modified eagle’s medium (DMEM; Invitrogen, Carlsbad, CA, USA) supplemented with 15% serum (Invitrogen), 2 mM GlutaMax (Invitrogen), 1% non-essential amino acids (Cambrex Bio Science, New Jersey, NJ, USA), 50 U/mL penicillin/streptomy- cin (Invitrogen), 0.1 mm b-mercaptoethanol (Invitrogen, Min- neapolis, MN, USA) on STO feeder cell layer (ATCC, USA). The cultures were free of mycoplasma throughout the exper- iment. Karyotype of the ES cells was normal, 46XX, as analyzed in passages 30 and 50. ES cells were termed stage 1 in the 5-stage differentiation method. In our current study, we increased the number of cells used in transplantation to 56106 in order to further validate the function of sphere and monolayer cells in tumorigenesis. With higher number of cells transplanted we observed that monolayer cells did not form any tumors while sphere cells formed tumor in 2 out of 10 mice. In contrast, when transplanted together (in quantities of 56106) at a ratio of 10:1, sphere and monolayer cells resulted in 100% tumor formation. This suggests that tumor formation in vivo occurs only when monolayer cells and sphere cells are together. In addition, tumor formation latency experiment also confirmed the importance of the interaction between monolayer cells and sphere cells in tumor formation. For neural differentiation, the mESC colonies were dissected with the use of 0.25% trypsin-EDTA into single cells and plated on to gelatin coated dishes and cultured for three passages. The cells were then transferred to bacterial petri dishes to form embryonic bodies in the mES medium without leukemia inhibitory factor, LIF. The embryonic bodies were termed stage 2. Four days later, only the EBs in suspension were transferred into gelatin coated tissue culture dish and the selection of neural progenitors was initiated by replacing the medium with ITSFn (insulin, transferrin, selenium and fibronectin). The cells in ITSFn medium were termed stage 3 that lasted for 10 days. Our results showed that the separation of monolayer cells and sphere cells is a good model to study the interaction of NPCs and their niche. Effect of Surrounding Cells on Neural Progenitors cell line PA6 (PA6-CM) can induce dopaminergic differentiation in neural stem cells (NSCs) derived from hESCs but not directly from hESCs, indicating that soluble factors produced by PA6 cells act at the NSC stage to specify a dopaminergic fate. They also found Shh and FGF8 can substitute for PA6-CM at the NSC induction stage. Shh is indeed one of the active agents in PA6-CM and is likely an important soluble dopaminergic inducing factor secreted by stromal cells and acts after the neural fate determination [36]. Furthermore, to test whether the effect of monolayer cells on sphere cells was unique, we choose mouse fibroblast to replace the monolayer cells. We found that mouse embryonic fibroblasts also had the proliferating and anti-apoptotic abilities of monolayer cells, showing evidence that these two functions are not specific to monolayer cells. However, the effect of monolayer cells on the differentiation of sphere cells to neurons could not be substituted by MEFs. This suggests that the main function of monolayer cells is to promote differentiation of sphere cells into neural lineages. Cell-cell contact and secreted cytokines were the probable mechanisms by which monolayer cells promoted proliferation of sphere cells and their induction to neurons. The anti-apoptotic function of monolayer cells was mainly due to cell-cell contact. In summary, stage 4 culture creates two distinct cell populations that influence each other. Sphere cells are committed to form neural cells and monolayer cells are the supporting cells that prompt proliferation, decrease apoptosis and enhance differenti- ation of sphere cells. Significantly, we found the importance of the interaction between the two cell types in neurogenesis and tumorigenesis. Based on all of these results, we can make it for sure that the niche, mainly composed of monolayer cells in our experiments, is great helpful in the differentiation of sphere cells into neurons. Stem cell niches are defined as microenvironments that maintain survival, self-renewal, activation, proliferation and regenerative capacity of stem cells. Both in the developing embryo and in vitro, signaling transferred via soluble factors, and or direct cell-cell interactions, contribute to the appropriate regulation of stem cell function [34,35]. Further work is needed in particular to determine the function of monolayer cells on development of functional neurons, such as tyrosine hydroxylase (TH) positive neurons, GABA neurons, etc. Ethics Statement All animal protocols used in the present study were reviewed and approved by the Ethics Committee on Laboratory Animal Care at Capital Medical University. In our previous study [29], we tested tumorgenicity of sphere cells and monolayer cells by transplanting106 cells each into nude mice. About 30% of mice that received only 106 sphere cells developed tumors. In contrast, 106 monolayer cells did not result in tumor formation when injected alone into nude mice. However, when both sphere and monolayer cells were injected into nude mice, the rate of tumor formation did not increase. Therefore, we concluded that monolayer cells did not influence tumorgenicity of sphere cells. Effect of Surrounding Cells on Neural Progenitors In addition, sphere cells may also be a heterogeneous group of cells and it is yet to be determined which cell types and genotyping markers are more committed to become neural precursors or neurons. Although sphere cells can differentiate into more neurons with the help of monolayer cells in vitro, it is not known if the same is true in vivo. Monolayer cells are also important for tumor formation that occurs after transplantation of sphere cells. The existence of Oct-4 immune-staining during stage 4 is often considered as evidence that the remaining ES cells lead to teratoma formation after transplantation. Since sphere cells contain Oct-4 positive cells, we presumed that tumor formation would occur from sphere cells but not monolayer cells. On the contrary, sphere cells formed tumors only in 2 out of 10 mice while ES and stage 4 cells caused 90 and 100% tumor formation respectively. Effect of Surrounding Cells on Neural Progenitors Similarly, monolayer cells exert an anti-apoptotic effect on the neural progenitors both during the first 14 days and the next 6 weeks long culture period of stage 4. This was exemplified when Figure 7. Teratoma formation after transplantation of sphere cells and monolayer cells. Injection of sphere cells and monolayer cells together into immune-deficient mice leads to 100% tumor formation. Sections from excised tumors were stained by H&E for histology. (A) Glandular epithelial (endoderm). (B) Connective tissue (mesoderm). (C) Immature epithelium (ectoderm). (D) When 56106 cells were used in transplantation, monolayer cells did not form any tumors, while sphere cells formed tumor in 20% of the mice. Transplantation of the sphere cells and monolayer cells together at same quantity of 56106 led to 100% tumor formation. The same amount of ES cells transplanted turned into tumors in 90% of the animals within 9 weeks. Scale bar, 100 mm. doi:10.1371/journal.pone.0054332.g007 Figure 7. Teratoma formation after transplantation of sphere cells and monolayer cells. Injection of sphere cells and monolayer cells together into immune-deficient mice leads to 100% tumor formation. Sections from excised tumors were stained by H&E for histology. (A) Glandular epithelial (endoderm). (B) Connective tissue (mesoderm). (C) Immature epithelium (ectoderm). (D) When 56106 cells were used in transplantation, monolayer cells did not form any tumors, while sphere cells formed tumor in 20% of the mice. Transplantation of the sphere cells and monolayer cells together at same quantity of 56106 led to 100% tumor formation. The same amount of ES cells transplanted turned into tumors in 90% of the animals within 9 weeks. Scale bar, 100 mm. doi:10.1371/journal.pone.0054332.g007 January 2013 \| Volume 8 \| Issue 1 \| e54332 10 PLOS ONE \| www.plosone.org January 2013 \| Volume 8 \| Issue 1 \| e54332 8. Effect of Monolayer Cells on Sphere Cells In this section, we compared the proliferation, apoptosis, and neurogenesis of stage 4 cells as a whole and sphere cells alone. To avoid the impact of collagenase disassociation on the character- istics of sphere cells and monolayer cells, the stage 4 cells are actually ‘‘co-culture’’ cells, which was a combination of sphere and monolayer cells after collagenase disassociation. To prepare the ‘‘co-culture’’ cells, the two cell types in stage 4 were separated at first, then after counting, we concluded that the quantitative ratio of sphere cells to monolayer cells is approximately 10:1. We first plated 26104/well monolayer cells in each well of the 6-well plate, followed by 26105 sphere cells onto the monolayer cells. For the combined cells to setup crosstalk again, the ‘‘co-culture’’ cells were kept in stage 4 medium for 2 days before further experiments were performed. 6. Cell Counting Percentage of neurons or astrocytes was determined by counting the numbers of Tuj-1 positive or GFAP positive cells and the numbers of DAPI stained nuclei using a fluorescent microscope at 206 magnification (TE2000U, Nikon, Japan). Ten visual fields were randomly selected and counted. Effect of Surrounding Cells on Neural Progenitors supplemented with N2 (Invitrogen), 2 mM GlutaMax (Invitro- gen) and 10 ng/mL basic fibroblast growth factor, bFGF (R&D systems) in 6-well plates pre-coated with 10 mg/mL polyor- nithine (Sigma, St. Louis, http://www.sigmaaldrich.com). The cells in the medium with bFGFwere termed stage 4. In this study, the mouse stage 4 cells, containing spheres and mono- layer cells, were passed once every 5–7 days and the medium was changed every day. Each time after passing, the stage 4 cells were induced into stage 5 differentiation to test their ability to giving rise to neurons. numbers in the spheres. The spheres were gently dissociated by accutase to avoid cell death. All cells were centrifuged, re- suspended in PBS, plated in 2% gelatin and 0.05% chromic potassium sulfate coated slides and fixed with 4% PFA before immuno-histochemistry staining was performed. 7. Teratoma Formation In our study, we divided stage 4 into the first 14 days, and the subsequent 6 weeks of continuous culture. The monolayer cells which adhered to the dish and cell aggregates that were scattered over the monolayer cells formed on day 2–3 in stage 4. To obtain these pure sphere cells, stage 4 cells were digested with 0.3 mg/ml collagenase IV (Sigma-Aldrich, St. Louis, http://www. sigmaaldrich.com) for 5–8 minutes at 37uC. During the collage- nase treatment, dishes were gently shaken to detach the spheres from the monolayer cells. We separated the sphere cells and monolayer cells and injected them into nude mice to find their respective roles in tumor formation. To study their interaction, the cells were initially cultured for 1 week independently followed by combining them at a 10:1 ratio (sphere cells: monolayer cells) before infusing into the right flank of nude mice. Sixty male BALB/c-nude mice (Laboratory of Animal Science Institute of Chinese Academy of Medical Sciences, Beijing, http:// www.cnilas.org/html/en/) were used in the teratoma formation test. Mice were kept in a specific pathogen-free (SPF) unit at the Capital Medical University. All experiments were performed after approval of the Committee for Animal Care and Use of the Faculty of Sciences at the Capital Medical University. 3. Identification of Neural Progenitors To identify whether neural progenitors reside in sphere cells or the monolayer cells in the first 14 days and the subsequent 6 weeks culture, the sphere cells were separated at different time points. The spheres and monolayer cells were then differentiated in stage 5 medium for 7 days before staining with neural and glial lineage antibodies. The animals were divided into 6 groups and injected with (1) sphere cells, (2) monolayer cells, (3) sphere cells+monolayer cells immediately after combining, (4) sphere cells+monolayer cells one week after combining, (5) ES cells and (6) stage 4 cells. For each animal, 56106 cells were suspended in 100 mL PBS and injected into the right flank [20]. Tumor latency was monitored over a period of 90 days. At the end of the observation period, mice were sacrificed. 4. The Maturation of NPCs in Stage 4 Previous studies have described the duration of stage 4 as 7 days [8–10]. In our experiment, we wanted to determine when the sphere cells get the full potential to form neurons. We started differentiation of stage 5 on days 3, 7, 10, and 14 in stage 4. Longer time points were not chosen because the cells cultured in stage 4 are nearly 90–100% confluent by day 14, and if not passed, many cells would die. After the final differentiation to stage 5, the percentage of neurons and astrocytes were quantified by cell counting. 1. ES Cell Culture and Differentiation by the 5-stage Method To study the interaction of NPC and its niche is good to find more soluble factors and transcript factors useful in increasing the percentage of specific neurons in differentiation. Some research made breakthrough recently. For example, Vazin et al. found PA6 cell surface activity is required for neural differentiation of hESC, but secreted factors are required for the specific DA neuron-inducing effect. Using more comprehensive array studies, they have identified a large number of candidate molecules potentially responsible for the SDIA effect [25]. Swistowska showed that medium conditioned by the stromal The cell clusters formed in stage 3 were then disassociated by using 0.25% trypsin-EDTA and plated at the density of 16105/ cm2 and cultured for 14 days in DMEM/F-12 (Invitrogen) January 2013 \| Volume 8 \| Issue 1 \| e54332 11 PLOS ONE \| www.plosone.org January 2013 \| Volume 8 \| Issue 1 \| e54332 Effect of Surrounding Cells on Neural Progenitors References 1. Bjorklund A, Lindvall O (2000) Cell replacement therapies for central nervous system disorders. Nat Neurosci 3: 537–544. 11. Buhnemann C, Scholz A, Bernreuther C, Malik CY, Braun H, et al. (2006) Neuronal differentiation of transplanted embryonic stem cell-derived precursors in stroke lesions of adult rats. Brain 129: 3238–3248. 1. Bjorklund A, Lindvall O (2000) Cell replacement therapies for central nervous system disorders. Nat Neurosci 3: 537–544. 2. Brustle O, McKay RD (1996) Neuronal progenitors as tools for cell replacement in the nervous system. Curr Opin Neurobiol 6: 688–695. 12. Chung S, Shin BS, Hwang M, Lardaro T, Kang UJ, et al. (2006) Neural precursors derived from embryonic stem cells, but not those from fetal ventral mesencephalon, maintain the potential to differentiate into dopaminergic neurons after expansion in vitro. Stem Cells 24: 1583–1593. 3. Brignier AC, Gewirtz AM (2010) Embryonic and adult stem cell therapy. J Allergy Clin Immunol 125: S336–344. J gy 4. Ringden O, Le Blanc K, Hovatta O (2003) Transplantation of embryonic stem cells: possibilities and challenges. Transplantation 76: 1011–1012. neurons after expansion in vitro. Stem Cells 24: 1583–1593. 13. Kim JH, Auerbach JM, Rodriguez-Gomez JA, Velasco I, Gavin D, et al. (2002) Dopamine neurons derived from embryonic stem cells function in an animal model of Parkinson’s disease. Nature 418: 50–56. 5. Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, Swiergiel JJ, et al. (1998) Embryonic stem cell lines derived from human blastocysts. Science 282: 1145–1147. 14. Kawasaki H, Mizuseki K, Nishikawa S, Kaneko S, Kuwana Y, et al. (2000) Induction of midbrain dopaminergic neurons from ES cells by stromal cell- derived inducing activity. Neuron 28: 31–40. 6. Bjorklund LM, Sanchez-Pernaute R, Chung S, Andersson T, Chen IY, et al. (2002) Embryonic stem cells develop into functional dopaminergic neurons after transplantation in a Parkinson rat model. Proc Natl Acad Sci U S A 99: 2344– 2349. g y 15. Barberi T, Klivenyi P, Calingasan NY, Lee H, Kawamata H, et al. (2003) Neural subtype specification of fertilization and nuclear transfer embryonic stem cells and application in parkinsonian mice. Nat Biotechnol 21: 1200–1207. 7. Goldman S (2005) Stem and progenitor cell-based therapy of the human central nervous system. Nat Biotechnol 23: 862–871. 16. Kawasaki H, Suemori H, Mizuseki K, Watanabe K, Urano F, et al. (2002) Generation of dopaminergic neurons and pigmented epithelia from primate ES cells by stromal cell-derived inducing activity. Acknowledgments We thank Dr. Mohammed Hussain (Sinai Grace Hospital, Detroit, USA) for helpful discussions. We thank Dr. Mohammed Hussain (Sinai Grace Hospital, Detroit, USA) for helpful discussions. Supporting Information Figure S1 Sphere cells began to differentiate on days 3, 7, 10 and 14 in stage 4 with or without monolayer cells. In the later 3 time points, astrocyte percentage in the presence of monolayer cells were more than in their absence (p,0.05). Green,GFAP; Blue, DAPI. Scale bar, 20 mm. (TIF) p y 8.3 Effect of monolayer cells on apoptosis of the sphere cells. We separated sphere cells on days 3, 7, 10, 14 and weeks 5 and 8. Cells used for the combined culture were brought together after separation. Two days later all cells were stained by active caspase 3 antibody and positive cells were counted. The results were confirmed by western blot. For analysis of protein levels, cells were rinsed twice with ice-cold PBS, lysed with ice-cold lysis buffer and were centrifuged at 14,000 g for 10 minutes at 4uC; the supernatant was then mixed with SDS sample buffer, boiled and separated through 12% SDS-PAGE gels. After electrophoresis and being transferred to nylon membranes by electrophoretic transfer, the proteins were blocked and incubated overnight at 4uC with primary antibody and 2 hours with horseradish peroxidase- conjugated secondary antibodies. Bands were visualized by enhanced chemoluminescence. 9. Statistical Analysis Microsoft Excel (Microsoft, http://www.microsoft.com) soft- ware was used for calculating mean and SD, and constructing histogram plots. Samples that passed the normal distribution test were subjected to t-test. The t-tests were used to establish null hypothesis between any two groups based on unpaired, unequal variance and two-sided model. Statistical significance was considered if p value was ,0.05. 8.2 The effect of monolayer cells on the neurogenesis of sphere cells. Day 3, 7, 10, 14, and weeks 5 and 8 in stage 4 were selected to study the effect of monolayer cells on the neurogenesis of sphere cells. The cells were transferred into stage 5 medium for 7 days differentiation. We further selected week 5 to study the effect of cell-cell contact and secreted cytokines on the neurogenesis of sphere cells, and whether the effect of monolayer cells can be replaced by MEFs. Author Contributions Conceived and designed the experiments: YQG LC YAZ. Performed the experiments: YQG QAD WWZ CLZ DW. Analyzed the data: YQG QAD LC YAZ. Contributed reagents/materials/analysis tools: LC YAZ. Wrote the paper: YQG. Conceived and designed the experiments: YQG LC YAZ. Performed the experiments: YQG QAD WWZ CLZ DW. Analyzed the data: YQG QAD LC YAZ. Contributed reagents/materials/analysis tools: LC YAZ. Wrote the paper: YQG. Besides active caspase 3 staining, the ‘‘FITC Annexin V Apoptosis Detection Kit I ’’ (BD Pharmingen, http://www. Besides active caspase 3 staining, the ‘‘FITC Annexin V Apoptosis Detection Kit I ’’ (BD Pharmingen, http://www. Effect of Surrounding Cells on Neural Progenitors To study the effect of cell-cell contact and secreted factors on the proliferation, apoptosis and differentiation of sphere cells, sphere cells at week 5 were divided into 5 groups; one group was cultured with monolayer cells in stage 4 medium, the second with a conditioned medium from monolayer cells, the third only with stage 4 medium, the fourth with a conditioned medium from MEFs, and the fifth with MEFs in stage 4 medium. bdbiosciences.com/) was also be used to study the effect of cell-cell contact and secreted cytokines on the apoptosis of sphere cells, and whether the effect of monolayer cells can be replaced by MEFs. The apoptosis were analyzed according to the manufacturer’s instructions. In brief, cells were rinsed with ice-cold PBS and resuspended in binding buffer at a concentration of 16106 cells/ ml. Approximately 16105 cells was transferred to a 5 ml culture tube, then 5 uL FITC-Annexin V and 5 uL propidine iodide (PI) was added to the cells and incubated for 15 minutes at room temperature in the dark. The cells were immediately analyzed on a FACSC-LSR equipped with CellQuest software(BD, http:// www.bd.com). g 8.1 The effect of monolayer cells on the proliferation of sphere cell. The time points of day 7, 10, and week 5, 8 in stage 4 were selected to study the effect of monolayer cells on the proliferation of sphere cell. Cells used for the combined culture were brought together after separation. The separated and re- combined cells were then cultured for 2 days. During the last 15 hours of the co-culture, 10 mM 5-bromo-2-deoxyuridine (BrdU) was used to identify the number of cells still in proliferation. Cells were fixed in 4% PFA and immune-histological staining was performed to test the percentage of BrdU positive cells. 5. Immunocytochemical Characterization of Differentiated Cells Cells after 5 stage differentiation were fixed in 4% para- formaldehyde (PFA) at room temperature for immunocytochem- ical analysis. After fixation, cells were treated with 0.1% Triton X- 100 and 1% bovine serum albumin (BSA) for 30 min and then blocked with 10% normal goat serum in 0.1% phosphate-buffered saline (PBS). Cells were incubated overnight in PBS with primary antibodies, 0.1% Triton X-100 and 1% BSA at 4uC. The next day, cells were washed with 0.1% PBS and incubated in the same solution with secondary antibodies (1:400). Finally, cells were washed with PBS and counterstained with 496-diamidino-2- phenylindole (DAPI; Vector Laboratories, Peterborough, UK). To analyze whether the effect of proliferation, apoptosis and differentiation on sphere cells was exclusive to the monolayer cells, we selected week 5 as the time point for examination and another monolayer feeder cells, the CF1 mouse embryonic fibroblasts (MEFs) (Shanghai Institute of Biochemistry and Cell biology, http://www.sibcb.ac.cn/eindex.asp), as the substitution of mono- layer cells. To exclude the influence of MEF or monolayer cell growth on the characteristics of sphere cells, the MEFs or monolayer cells were treated with 10 mg/ml mitomycin-C for 150 min (Sigma, St. Louis, MO) and then washed three times with fresh medium. The MEFs or monolayer cells were then permitted to recover overnight before sphere cells were seeded. The antibodies included polyclonal goat anti- mouse glial fibrillary acidic protein (GFAP; 1:600, AF2594; R&D Systems) and monoclonal mouse anti-mouse Tuj-1 (1:400, MAB1637, Millipore, USA). The secondary antibodies were Cy2 or Texas red conjugated goat anti-mouse or goat anti-rabbit IgG (Immuno- Jackson, Baltimore, PA, USA). To exclude the false absorption of secondary fluorescent antibody by cell aggregates, we accurately measured the cell PLOS ONE \| www.plosone.org January 2013 \| Volume 8 \| Issue 1 \| e54332 January 2013 \| Volume 8 \| Issue 1 \| e54332 12 Effect of Surrounding Cells on Neural Progenitors Effect of Surrounding Cells on Neural Progenitors 28. Shihabuddin LS, Horner PJ, Ray J, Gage FH (2000) Adult spinal cord stem cells generate neurons after transplantation in the adult dentate gyrus. J Neurosci 20: 8727–8735. 19. Hong S, Kang UJ, Isacson O, Kim KS (2008) Neural precursors derived from human embryonic stem cells maintain long-term proliferation without losing the potential to differentiate into all three neural lineages, including dopaminergic neurons. J Neurochem 104: 316–324. 29. Du Q, Guan Y, Ji H, Chen Z, Zhang YA (2011) Effect of monolayer cells on sphere cells–two types of cells that emerge during the neural differentiation of mouse embryonic stem cells. Neurosci Lett 504: 285–289. 20. Chung S, Moon JI, Leung A, Aldrich D, Lukianov S, et al. (2011) ES cell- derived renewable and functional midbrain dopaminergic progenitors. Proc Natl Acad Sci U S A 108: 9703–9708. 30. Dihne M, Bernreuther C, Hagel C, Wesche KO, Schachner M (2006) Embryonic stem cell-derived neuronally committed precursor cells with reduced teratoma formation after transplantation into the lesioned adult mouse brain. Stem Cells 24: 1458–1466. 21. Hayashi MA, Guerreiro JR, Cassola AC, Lizier NF, Kerkis A, et al. (2010) Long- term culture of mouse embryonic stem cell-derived adherent neurospheres and functional neurons. Tissue Eng Part C Methods 16: 1493–1502. 22. Doi D, Morizane A, Kikuchi T, Onoe H, Hayashi T, et al. (2012) Prolonged maturation culture favors a reduction in the tumorigenicity and the dopaminergic function of human ESC-derived neural cells in a primate model of Parkinson’s disease. Stem Cells 30: 935–945. 31. Boonman Z, Isacson O (1999) Apoptosis in neuronal development and transplantation: role of caspases and trophic factors. Exp Neurol 156: 1–15. 32. Schierle GS, Hansson O, Leist M, Nicotera P, Widner H, et al. (1999) Caspase inhibition reduces apoptosis and increases survival of nigral transplants. Nat Med 5: 97–100. 23. Brederlau A, Correia AS, Anisimov SV, Elmi M, Paul G, et al. (2006) Transplantation of human embryonic stem cell-derived cells to a rat model of Parkinson’s disease: effect of in vitro differentiation on graft survival and teratoma formation. Stem Cells 24: 1433–1440. 33. Caldero J, Prevette D, Mei X, Oakley RA, Li L, et al. (1998) Peripheral target regulation of the development and survival of spinal sensory and motor neurons in the chick embryo. J Neurosci 18: 356–370. 24. Kriks S, Shim JW, Piao J, Ganat YM, Wakeman DR, et al. References Proc Natl Acad Sci U S A 99: 1580–1585. 8. Lumelsky N, Blondel O, Laeng P, Velasco I, Ravin R, et al. (2001) Differentiation of embryonic stem cells to insulin-secreting structures similar to pancreatic islets. Science 292: 1389–1394. 9. Lee SH, Lumelsky N, Studer L, Auerbach JM, McKay RD (2000) Efficient generation of midbrain and hindbrain neurons from mouse embryonic stem cells. Nat Biotechnol 18: 675–679. 17. Perrier AL, Tabar V, Barberi T, Rubio ME, Bruses J, et al. (2004) Derivation of midbrain dopamine neurons from human embryonic stem cells. Proc Natl Acad Sci U S A 101: 12543–12548. 10. Friling S, Andersson E, Thompson LH, Jonsson ME, Hebsgaard JB, et al. (2009) Efficient production of mesencephalic dopamine neurons by Lmx1a expression in embryonic stem cells. Proc Natl Acad Sci U S A 106: 7613–7618. 18. Zeng X, Cai J, Chen J, Luo Y, You ZB, et al. (2004) Dopaminergic differentiation of human embryonic stem cells. Stem Cells 22: 925–940. January 2013 \| Volume 8 \| Issue 1 \| e54332 13 PLOS ONE \| www.plosone.org Effect of Surrounding Cells on Neural Progenitors (2011) Dopamine neurons derived from human ES cells efficiently engraft in animal models of Parkinson’s disease. Nature 480: 547–551. 34. Solozobova V, Wyvekens N, Pruszak J (2012) Lessons from the embryonic neural stem cell niche for neural lineage differentiation of pluripotent stem cells. Stem Cell Rev 8: 813–829. 25. Vazin T, Becker KG, Chen J, Spivak CE, Lupica CR, et al. (2009) A novel combination of factors, termed SPIE, which promotes dopaminergic neuron differentiation from human embryonic stem cells. PLoS One 4: e6606. 35. Keung AJ, Kumar S, Schaffer DV (2010) Presentation counts: microenviron- mental regulation of stem cells by biophysical and material cues. Annu Rev Cell Dev Biol 26: 533–556. y 26. Suhonen JO, Peterson DA, Ray J, Gage FH (1996) Differentiation of adult hippocampus-derived progenitors into olfactory neurons in vivo. Nature 383: 624–627. 36. Swistowska AM, da Cruz AB, Han Y, Swistowski A, Liu Y, et al. (2010) Stage- specific role for shh in dopaminergic differentiation of human embryonic stem cells induced by stromal cells. Stem Cells Dev 19: 71–82. 27. Yandava BD, Billinghurst LL, Snyder EY (1999) ‘‘Global’’ cell replacement is feasible via neural stem cell transplantation: evidence from the dysmyelinated shiverer mouse brain. Proc Natl Acad Sci U S A 96: 7029–7034. PLOS ONE \| www.plosone.org January 2013 \| Volume 8 \| Issue 1 \| e54332 PLOS ONE \| www.plosone.org 14
https://openalex.org/W3037511471	https://ruj.uj.edu.pl/xmlui/bitstream/handle/item/257785/zalewski_undas_et-al_delayed_thrombin_generation_is_associated_2020.pdf?sequence=1&isAllowed=y	English	null	Delayed Thrombin Generation Is Associated with Minor Bleedings in Venous Thromboembolism Patients on Rivaroxaban: Usefulness of Calibrated Automated Thrombography	Journal of clinical medicine	2,020	cc-by	11,423	Received: 11 May 2020; Accepted: 22 June 2020; Published: 27 June 2020 Abstract: Bleeding is the most feared and diﬃcult to predict adverse event of anticoagulation. We sought to investigate whether calibrated automated thrombography (CAT) parameters are associated with minor bleeding (MB) in anticoagulated patients following venous thromboembolism (VTE). Enrolled were 132 patients on rivaroxaban, 145 on vitamin K antagonists (VKA) and 31 controls who stopped anticoagulation. Prior to the next dose of the anticoagulant, we measured CAT parameters, along with rivaroxaban concentration and INR. During a median follow-up of 10 months, we recorded minor and major bleedings. On rivaroxaban, 27 (20.5%) patients with MB had longer time to start thrombin generation, lower peak thrombin generation and lower endogenous thrombin potential compared with subjects without MB (all p < 0.001). All CAT parameters, except for peak thrombin generation (p = 0.049), were similar in VKA patients with (n = 25, 17.2%) vs. without MBs. By logistic regression, time to start thrombin generation (p = 0.007) and unprovoked VTE (p = 0.041) independently predicted MBs on rivaroxaban. Major bleedings were more frequent in patients with MBs (17.3% vs. 1.8%, p < 0.001). Abnormal CAT parameters characterize VTE patients prone to MBs on rivaroxaban, but not on VKA. Time to start thrombin generation measured about 24 h since the last rivaroxaban dose might help predict MBs. Keywords: venous thromboembolism; minor bleeding; calibrated automated thrombography; rivaroxaban; vitamin K antagonist Jaroslaw Zalewski 1,, Konrad Stepien 1 , Karol Nowak 1, Sandi Caus 2, Saulius Butenas 2 and Anetta Undas 3 1 Department of Coronary Artery Disease and Heart Failure, Jagiellonian University Medical College, John Paul II Hospital, 80 Pradnicka Street, 31-202 Krakow, Poland; konste@interia.eu (K.S.); karol.nowak@student.uj.edu.pl (K.N.) 1 Department of Coronary Artery Disease and Heart Failure, Jagiellonian University Medical College, John Paul II Hospital, 80 Pradnicka Street, 31-202 Krakow, Poland; konste@interia.eu (K.S.); karol.nowak@student.uj.edu.pl (K.N.) j p ( ) 2 Department of Biochemistry, University of Vermont, 89 Beaumont Avenue, Burlington, VT 05405-0068, USA sandi.caus@uvm.edu (S.C.); sbutenas@uvm.edu (S.B.) 2 Department of Biochemistry, University of Vermont, 89 Beaumont Avenue, Burlington, VT 05405-0068, USA; sandi.caus@uvm.edu (S.C.); sbutenas@uvm.edu (S.B.) 3 Department of Experimental Cardiac Surgery, Anesthesiology and Cardiology, Institute of Cardiology, Jagiellonian University Medical College, 80 Pradnicka Street, 31-202 Krakow, Poland; mmundas@cyf-kr.edu.pl C d j l @ it lj 2 k k l T l 48 12 6143004 3 Department of Experimental Cardiac Surgery, Anesthesiology and Cardiology, Institute of Cardiology, Jagiellonian University Medical College, 80 Pradnicka Street, 31-202 Krakow, Poland; mmundas@cyf-kr.edu.pl * Correspondence: jzalews@szpitaljp2.krakow.pl; Tel.: +48-12-6143004 Journal of Clinical Medicine Journal of Clinical Medicine Journal of Clinical Medicine Journal of Clinical Medicine Delayed Thrombin Generation Is Associated with Minor Bleedings in Venous Thromboembolism Patients on Rivaroxaban: Usefulness of Calibrated Automated Thrombography Jaroslaw Zalewski 1,, Konrad Stepien 1 , Karol Nowak 1, Sandi Caus 2, Saulius Butenas 2 and Anetta Undas 3 Jaroslaw Zalewski 1,, Konrad Stepien 1 , Karol Nowak 1, Sandi Caus 2, Saulius Butenas 2 and Anetta Undas 3 1. Introduction Minor bleeding (MB) is usually deﬁned, based on the International Society on Thrombosis and Hemostasis (ISTH) guidance, as a clinically overt event not meeting the criteria of major or clinically relevant non-major (CRNM) bleeding [1,2]. MB represents a common, unappreciated and poorly understood adverse event in patients receiving anticoagulation. The RIETE [3], GARFIELD-VTE [4] and XALIA [5] registries performed in VTE patients treated with rivaroxaban, as well both EINSTEIN trials with rivaroxaban in patients with symptomatic deep-vein thrombosis (DVT) [6] and those with J. Clin. Med. 2020, 9, 2018; doi:10.3390/jcm9072018 www.mdpi.com/journal/jcm www.mdpi.com/journal/jcm 2 of 17 J. Clin. Med. 2020, 9, 2018 pulmonary embolism (PE) [7], have not speciﬁcally reported MBs. A high prevalence of MBs have been reported in observational studies. Själander et al. found increased incidence of easy bruising from 17.8% to 75.6%, menorrhagia from 44.2% to 70.8%, gingival bleedings from 22.2% to 48.3%, bleedings after tooth extraction from 3.0% to 45.2% and epistaxis from 11.1% to 23.6% after intiation of anticoagulation with vitamin K antagonist (VKA) among women [8]. In turn, in VTE patients treated mostly with non-vitamin K antagonist oral anticoagulants (NOAC) MBs were frequently found including heavy menstrual bleeding in 23.2%, easy bruising in 15.4% and gingival bleeding in 8.8% of subjects [9,10]. The Dresden NOAC registry [11] showed that ISTH-deﬁned MBs occur with the frequency of 37.8 per 100 person-years in real-life VTE subjects treated with rivaroxaban and they accounted for 58.9% of all observed bleedings. In turn, MBs in VTE patients on apixaban or rivaroxaban deﬁned as outpatient claims not requiring hospital admission were reported in 20 or 34 per 100 person-years, respectively [12]. It has been suggested that MB might predict subsequent major adverse events in patients on chronic anticoagulation. Veeger et al. [13] found that in patients on VKA, MBs were associated with a 3-fold increased risk of subsequent major bleeding as compared with the remainder, and this relationship was independent of quality of anticoagulation. Moreover, in a large cohort of patients on VKA, Van Rein et al. [14] demonstrated that increased risk of major bleeds in patients following MBs is due to ﬁxed and currently unknown underlying risk factors. 1. Introduction In patients with AF, occurrence of only MBs on apixaban or VKA was associated with higher risk of subsequent major bleeding (hazard ratio (HR) 1.53, 95% conﬁdence interval (CI) 1.37–2.26), ischemic stroke (HR 1.69, 95% CI 1.09–2.61) and all-cause mortality (HR 1.25, 95% CI 1.02–1.55) [15]. Evidence for similar minor–major bleeding relationships in VTE patients treated with NOAC is not available. Calibrated automated thrombography (CAT) parameters reﬂect the overall function of the blood clotting system. Increased endogenous thrombin potential (ETP) and peak thrombin concentration, together with shorter time to thrombin peak, reﬂect an enhanced and faster activation of blood coagulation and is typical of a prothrombotic state in vivo [16]. In turn, a decreased thrombin generation predicts a bleeding tendency in patients with von Willebrand disease [17] and with hemophilia A [18]. This assay was also used to evaluate the eﬀect of prothrombin complex concentrates in reversing the anticoagulant eﬀect of rivaroxaban [19,20]. CAT proﬁles used for many years to assess the overall function of the blood clotting system [21,22], are signiﬁcantly modiﬁed by anticoagulants, including NOAC [23,24]. Both rivaroxaban and warfarin eﬀectively inhibit thrombin generation in patients with VTE as evidenced by reduced ETP, peak thrombin as well as prolonged lag time and time to thrombin peak, as compared to normal controls [25]. However, their impact on individual CAT parameters diﬀers. Rivaroxaban, a direct inhibitor of factor Xa, mainly inﬂuences the initiation and propagation phases of thrombogram, which leads to the protraction of thrombin generation curves [23]. As a result, there is a lower than expected decrease in ETP. In turn, VKA reducing the activity of vitamin-K-dependent coagulation factors have an uniform and generalized eﬀect on all CAT parameters [23,26]. In this study, we tested the hypothesis that CAT parameters may help identify anticoagulated VTE patients prone to developing bleeding tendencies during long-term treatment with rivaroxaban, the most commonly used NOAC in VTE patients, as well as with VKA. 2. Materials and Methods We screened 353 consecutive adult patients with documented symptomatic VTE aged from 22 to 71 years, referred to outpatient clinics between January 2012 and January 2016. The diagnosis of DVT was established on the basis of positive ﬁndings on color duplex ultrasound, whereas the diagnosis of PE was performed by positive result of high resolution computed tomography. Patients with a history of portal vein thrombosis or cerebral sinus vein thrombosis conﬁrmed by imaging were also eligible. Following documented VTE episodes, all patients were treated with unfractionated or low molecular weight heparins followed by VKA or rivaroxaban, for at least six months in patients with 3 of 17 J. Clin. Med. 2020, 9, 2018 VTE provoked by transient risk factor, or longer in patients with unprovoked VTE. The decision as to whether VKA or rivaroxaban would be continued was left at the discretion of the treating physician based on the patient preferences. A VTE episode was deﬁned as provoked if the patient had a history of surgery requiring general anesthesia, major trauma, pregnancy or delivery, hospitalization within the prior three months, hormone replacement therapy or oral contraception. A proximal DVT was deﬁned as the presence of thrombus in the popliteal, femoral or iliac veins. We excluded patients diagnosed with cancer, chronic kidney disease (CKD) stage 4 and 5, acute coronary syndrome or ischemic stroke within the previous 6 months, any acute infection, all the states known to aﬀect thrombin generation [21] or with the time since last dose of rivaroxaban of less than 20 h. All patients on anticoagulation were asked to show up prior to administration of the next dose of the anticoagulant. Of 353 VTE patients, 45 patients were excluded from the analysis because of unavailable CAT data or follow-up data (n = 21) or a time interval of <20 h since the last rivaroxaban dose (n = 24). Obesity was deﬁned as a body mass index (BMI) over 30 kg/m2. Diabetes mellitus was stated as a history of diabetes, use of antidiabetic agents, or fasting plasma glucose ≥126 mg/dL (7 mmol/L) on two separate occasions. Hypertension was deﬁned as oﬃce systolic blood pressure values ≥140 mmHg and/or diastolic blood pressure values ≥90 mmHg or current antihypertensive treatment. Chronic kidney disease stage 3 was diagnosed when creatinine clearance, calculated using the Cockcroft–Gault formula, was lower than 60 mL/min. 2.1. Plasma Preparation Fasting blood samples were drawn between 8 and 10 a.m. from an antecubital vein directly into tubes containing citrated anticoagulant after at least 3 months of anticoagulation. The blood samples were centrifuged at 2.500 g at a temperature of 18 ◦C to 22 ◦C for 20 min and processed immediately or stored in aliquots at −80 ◦C until analysis. 2. Materials and Methods Heart failure referred to a symptomatic condition with relevant structural heart disease with diastolic dysfunction or reduced left ventricular ejection fraction below 40%. The study protocol complied with the Declaration of Helsinki and was approved by the Bioethical Commission of Local Medical Chamber (approval number 135/KBIL/OIL/2013). All included patients gave informed consent before they participated in the study. 2.2. Laboratory Measurements The laboratory investigations, including international normalized ratio (INR) and activated partial thromboplastin time (APTT), were assayed by routine hospital techniques. The quality of anticoagulation with VKA was assessed by time in the therapeutic range according to the method of Rosendaal (TTR). Fibrinogen was determined using the Clauss assay. High-sensitivity C-reactive protein (CRP) was measured by immunoturbidimetry (Roche Diagnostics GmbH, Mannheim, Germany). Plasma D-dimer was measured with the Innovance D-dimer assay (Siemens, Marburg, Germany). 2.3. Thrombophilia Testing Thrombophilia screening was performed in all study participants. Factor V Leiden and prothrombin G20210A polymorphisms were determined using TaqMan Genotyping Assays (Assay ID: C_11975250_10 and C_8726802_20, respectively; ThermoFisher Scientiﬁc, Waltham, MA, USA) on QuantStudio Dx Real-Time PCR Instrument (ThermoFisher Scientiﬁc) as described previously [27]. Plasma Protein C activity was quantiﬁed using a chromogenic assay (HemosIL Protein C Instrumentation Laboratory, Milan, Italy or Berichrom Protein C, Siemens Healthcare Diagnostics). Free protein S levels were measured with an immunoturbidimetric assay (INNOVANCE® Free PS Ag, Siemens Healthcare Diagnostic) [28]. Antithrombin activity was measured using an assay based on FXa inhibition (INNOVANCE™ATIII, Siemens Healthcare Diagnostics, Marburg, Germany) [29]. Antiphospholipid syndrome was diagnosed according to the current recommendations [30]. All coagulation tests were performed twice and yielded positive results. J. Clin. Med. 2020, 9, 2018 4 of 17 2.4. Rivaroxaban Concentration Rivaroxaban concentration was measured by the anti-Xa chromogenic assay, Biophen DiXaI (Hyphen Biomed, Neuilly-sur-Oise, France) according to the manufacturer’s instructions. Brieﬂy, 200 µL of diluted plasma (1:8 with Tris-NaCl EDTA buﬀer at pH 7.85) were incubated with 50 µL of human FXa (Hyphen BioMed, Neuilly-sur-Oise, France) for 120 s at 37 ◦C, then 50 µL of a speciﬁc FXa substrate were added to start the reaction and the color development was measured at 405 nm [31,32]. The calibration was performed with standards for low plasma concentrations of rivaroxaban (Biophen Rivaroxaban Calibrator, Neuilly-sur-Oise, France) [33]. 2.5. Thrombin Generation Assay The thrombin generation assay was performed as previously described [21,22]. The assay was performed in untreated, polystyrene 96-well plates (Costar, Lowell, MA, USA). Citrated plasma samples were thawed at 37 ◦C for 3 min and 5 mg/mL corn trypsin inhibitor was immediately added to achieve a 0.1 mg/mL ﬁnal concentration. Eighty µL of each plasma sample, in duplicate, was added to a 96-well plate followed by addition of relipidated tissue factor at a ﬁnal concentration of 5 pM. The ﬂuorogenic substrate used was benzyloxycarbonyl-Gly-Gly-Arg-7-amido-4methyl-coumarin·HCl (Z-GGR-AMC) (Bachem, Torrance, CA, USA). Twenty µL of a 2.5 mM Z-GGR-AMC/90 mM CaCl2 solution in HEPES-buﬀered saline was added to plasma samples to achieve ﬁnal concentrations of 417 µM and 15 mM, respectively. A 3 min incubation period at 37 ◦C followed to allow for recalciﬁcation of the plasma. Twenty µL of a 120 µM phospholipid vesicle solution (25% dioleoyl-sn-glycero-3-phospho-l serine and 75% 1,2-dioleoyl-sn-glycero-3-phosphocholine) (Avanti Polar Lipids, Inc, Alabaster, Al, USA) in HEPES-buﬀered saline was then added to plasma samples to achieve a ﬁnal concentration of 20 µM, thus initiating thrombin generation. Fluorescence readings began immediately and hydrolysis of the AMC (7-amino-4-methylcoumarin) substrate (at 370 nm excitation and 460 nm emission wavelengths) was followed over a 60 min period. Changes in ﬂuorescence were converted to thrombin concentration using a calibration curve built by sequential dilutions of human thrombin (Haematologic Technologies, Inc., Essex Junction, VT, USA). The plate reader used was the BioTek Synergy 4 and analysis was performed using the Gen5 plate reader software (BioTek, Winooski, VT, USA). The CAT parameters including Lag time (s, time to start thrombin generation), TTP (s, time to the highest rate of thrombin generation), max IIa (nM, peak thrombin generation), max rate (nM/s, the highest rate of thrombin generation) and ETP (nM × s, endogenous thrombin potential, calculated as the area under the thrombin generation curve) were calculated. 2.6. Follow-up All patients were assessed in the outpatient clinic every 6 months or by phone. They were instructed to report symptoms that suggested recurrent VTE or bleeding requiring appropriate conﬁrmatory diagnostic and/or laboratory tests. We recorded MB deﬁned as bleeding, not requiring outpatient visits, medical interventions or hospitalizations including easy bruising, skin ecchymosis, conjunctiva bleeds, gingival bleeds, epistaxis, haematuria, or vaginal bleeds. Easy bruising was deﬁned as the self-reported, frequent (on 2 or more separate occasions) occurrence of bruises of at least 2.5 cm in diameter as well as the presence of one or more bruises on the day of any clinic visit during follow-up conﬁrmed by an investigator. Major bleedings were deﬁned according to the ISTH deﬁnition [1,2]. We also recorded recurrent symptomatic VTE. The diagnosis of recurrent symptomatic DVT was established based on positive ﬁndings of color duplex ultrasonography. In cases of suspected DVT recurrence in the same leg, non-compressibility of a previously compressible venous segment or an increase of at least 4 mm in the residual diameters was applied to conﬁrm the diagnosis. PE was each time conﬁrmed by computed tomography angiography. J. Clin. Med. 2020, 9, 2018 5 of 17 2.7. Statistical Analysis The study was powered to have a 90% chance of detecting a 15% diﬀerence in lag time between patients with and without MB using a p value of 0.05. Based on the CAT values in the previous study [31], in order to demonstrate such a diﬀerence or greater, 23 patients were required in each group. For a p value of 0.001, 45 patients per group were required. The power analysis was done based on a two-tailed t-test. Statistical analyses were performed with IBM SPSS Statistics version 25 software. Continuous variables are expressed as mean ± standard deviation or median and interquartile range (IQR), whereas categorical variables are expressed as numbers and percentages. Continuous variables were ﬁrst checked for normal distribution by the Shapiro–Wilk test and compared by Student t test when normally distributed or by the Mann–Whitney U test for non-normally distributed variables. Categorical variables were compared by the Fisher’s exact test. The Pearson or Spearman rank correlation coeﬃcients were calculated to test the association between two variables with a normal or non-normal distribution, respectively. Diﬀerences in CAT parameters between groups were adjusted for rivaroxaban concentration or INR in patients on VKA with regression analysis. Receiver operating characteristic (ROC) curves were used to determine the optimal cut-oﬀvalue of CAT parameters and their sensitivity and speciﬁcity in prediction of minor bleedings. All independent variables with their potential for confounding both the exposure and the outcome and a lack of signiﬁcant correlation with other variables were included in the multivariate logistic regression model to determine independent predictors of minor bleeding. A two-tailed p-value of less than 0.05 was considered statistically signiﬁcant. 3.1. General Characteristics In this study, 308 (87.3%) VTE patients were enrolled. Of them, 132 (42.8%) were on rivaroxaban, and 145 (47.1%) on VKA (77 on warfarin and 68 on acenocoumarol), while 31 (10.1%) patients following provoked VTE who stopped anticoagulation at least three months before inclusion served as controls. Baseline characteristics of the studied groups are shown in Table 1. One hundred and nineteen (38.6%) subjects had a history of symptomatic DVT alone, 74 patients had PE (24.0%) alone, 95 (30.9%) patients DVT combined with PE, and 20 (6.5%) patients had thrombosis of atypical location including portal vein thrombosis and cerebral venous sinus thrombosis. In 172 (55.8%) patients, VTE was unprovoked, whereas among provoked VTE patients, 37 (12.0%) subjects had VTE related to surgery or trauma, 23 (7.5%) to pregnancy, and 61 (19.8%) to hormone use. Thrombophilia screening yielded positive results in similar proportions of patients on rivaroxaban and VKA, including factor V Leiden (22.7 vs. 15.9%, p = 0.17), prothrombin G20210A (4.5 vs. 7.6%, p = 0.33) and deﬁciencies in natural anticoagulants (9.1 vs. 15.9%, p = 0.11), respectively. There were no diﬀerences in the distribution of thrombophilias between patients with and without MB on rivaroxaban, while in VKA patients prothrombin G20210A mutation was more frequent in subjects with MB as compared to those free of MB (Table 1). The median time of anticoagulation with rivaroxaban was 8 (interquartile range, 5–15) months, while that for VKA was 9 (6–16) months (p = 0.47). 6 of 17 J. Clin. Med. 2020, 9, 2018 Table 1. Baseline characteristics of the studied patients. Table 1. Baseline characteristics of the studied patients. 3.1. General Characteristics Controls n = 31 Rivaroxaban without MB n = 105 Rivaroxaban with MB n = 27 VKA without MB n = 120 VKA with MB n = 25 Age, year 42 (36–48) 46 (36–57) 47 (42–52) 47 (37–57) 50 (38–61) Male, n (%) 12 (38.7) 38 (36.2) 15 (55.6) 55 (45.8) 10 (40.0) Body mass index, kg/m2 25.1 (22.0–28.3) 27.3 (22.3–30.7) 26.6 (25.6–29.8) 28 (24.1–30.5) * 26.8 (23.6–30.4) Currently smoking, n (%) 7 (22.6) 15 (14.3) 7 (25.9) 31 (25.8) 7 (28.0) VTE characteristics, n (%) Baseline VTE diagnosis DVT alone 15 (48.4) 40 (38.1) 9 (33.3) 43 (35.8) 12 (48.0) Pulmonary embolism alone 6 (19.3) 25 (23.8) 9 (33.3) 28 (23.4) 6 (24.0) Pulmonary embolism and DVT 7 (22.6) 29 (27.6) 9 (33.3) 43 (35.8) 7 (28.0) Other 3 (9.7) 11 (10.5) 0 6 (5.0) 0 Unprovoked VTE 15 (48.4) 55 (52.3) 24 (88.9) # 67 (55.8) 11 (44.0) Proximal DVT 16 (51.6) 62 (59.0) 16 (59.3) 71 (59.2) 19 (76) Bilateral DVT 3 (9.7) 5 (4.8) 0 3 (2.5) 1 (4) Time since last VTE, month 10 (6–22) 10 (6–17) 8 (6–17) 12 (8–23) 8 (7–16) Duration of anticoagulation, month 6 (4–12) 7 (5–14) 8 (6–17) 10 (6–18) 8 (6–13) Family history of VTE 14 (45.2) 34 (32.4) 8 (29.6) 35 (29.2) 4 (16.0) Varices 11 (35.5) 37 (35.2) 11 (40.7) 37 (30.8) 9 (36.0) Comorbidities, n (%) Coronary artery disease 0 3 (2.9) 0 3 (2.5) 1 (4.0) Prior ischemic stroke 0 6 (5.7) 1 (3.7) 3 (2.5) 1 (4.0) Hypertension 4 (12.9) 31 (29.5) 7 (25.9) 39 (32.5) * 10 (40.0) Heart failure 1 (3.2) 1 (1.0) 0 2 (1.7) 1 (4.0) Diabetes mellitus 2 (6.5) 4 (3.8) 0 5 (4.2) 4 (16.0) Chronic kidney disease 1 (3.2) 3 (2.9) 0 0 0 Use of aspirin 4 (12.9) 8 (7.6) 0 7 (5.8) 2 (8.0) Use of statin 2 (6.5) 15 (14.3) 1 (3.7) 17 (14.2) 3 (12.0) Proton pump inhibitor 3 (9.7) 24 (22.9) 1 (3.7) ## 25 (20.8) 4 (16.0) Thrombophilia, n (%) Factor V Leiden 11 (35.5) 26 (24.8) 6 (22.2) 23 (19.2) 2 (8.0) Prothrombin G20210A mutation 1 (3.2) 6 (5.7) 0 6 (5.0) 5 (20.0) Deﬁciencies in natural anticoagulants 1 (3.2) 9 (9.6) 3 (11.1) 16 (13.3) 7 (28.0) Antiphospholipid syndrome 2 (6.5) 7 (6.7) 5 (18.5) 10 (8.3) 2 (8.0) Rivaroxaban without MB n = 105 Rivaroxaban with MB n = 27 VKA without MB n = 120 VKA with MB n = 25 7 of 17 J. 3.1. General Characteristics Clin. Med. 2020, 9, 2018 Table 1. Cont. Controls n = 31 Rivaroxaban without MB n = 105 Rivaroxaban with MB n = 27 VKA without MB n = 120 VKA with MB n = 25 Laboratory Investigations White blood cells, 103/µL 5.90 (4.92–7.54) 5.89 (4.85–7.1) 6.07 (5.69–6.94) 6.36 (5.27–7.66) 6.08 (5.46–7.19) Hemoglobin, g/dL 13.6 (12.9–14.5) 13.9 (13.1–14.9) 13.9 (13.1–15.4) 14.4 (13.4–15.5) * 13.4 (13.1–14.5) Platelets, 103/µL 232 (195–265) 245 (206–284) 216 (186–265) 244 (201–285) 243 (212–293) Glucose, mmol/L 5.1 (4.8–5.3) 5.1 (4.9–5.8) 5.2 (5.0–5.4) 5.2 (4.9–5.6) 5.4 (5.0–5.7) Creatinine, µmol/L 70 (64–85) 70 (61–79) 74 (67–89) 72 (65–83) 70 (64–82) eGFR, ml/min/1.73 m2 92 (83–114) 101 (89–111) 94 (86–106) 98 (88–107) 99 (90–1017) hsCRP, ng/mL 1.12 (0.71–1.87) 1.1 (0.6–3.43) 1.98 (0.9–5.6) 1.76 (0.85–3.11) 1.52 (0.74–3.58) D-Dimer, ng/mL 275 (183–447) 265 (171–375) 209 (171–348) 203 (171–341) 177 (171–226) Rivaroxaban concentration, µg/L - 25 (13–44) 35 (6–66) - - APTT, s 25.8 (24.7–29.1) 25.8 (24.1–28.4) 27.1 (24.9–34.1) 28.7 (25.1–34.1) * 31.0 (25.4–35.7) INR 0.99 (0.97–1.04) 1.03 (0.99–1.14) 1.04 (0.98–1.10) 1.47 (1.02–2.20) 2.19 (1.46–2.45) ** Fibrinogen, g/L 3.03 (2.61–4.0) 3.06 (2.64–3.64) 3.3 (2.8–4.05) 3.13 (2.71–363) 3.24 (2.91–3.48) Abbreviations: data are shown as numbers (percentages) or median (interquartile range), DVT: deep vein thrombosis, eGFR: glomerular ﬁltration rate, MB: minor bleeding, VKA: vitamin K antagonist, VTE: venous thromboembolism, APTT: activated partial thromboplastin time, INR: international normalized ratio. There were no signiﬁcant diﬀerences between the following groups: (1) controls and patients on rivaroxaban without MB, (2) controls and patients on VKA without MB, (3) patients on rivaroxaban with MB and without MB, and (4) patients on VKA with MVB and without MB. The only signiﬁcant intergroup diﬀerences were marked as follows: * p < 0.05 vs. controls, ** p < 0.05 vs. VKA without MB, # p < 0.001 vs rivaroxaban without MB, ## p < 0.05 vs rivaroxaban without MB. Abbreviations: data are shown as numbers (percentages) or median (interquartile range), DVT: deep vein thrombosis, eGFR: glomerular ﬁltration rate, MB: minor bleeding, VKA: vitamin K antagonist, VTE: venous thromboembolism, APTT: activated partial thromboplastin time, INR: international normalized ratio. There were no signiﬁcant diﬀerences between the following groups: (1) controls and patients on rivaroxaban without MB, (2) controls and patients on VKA without MB, (3) patients on rivaroxaban with MB and without MB, and (4) patients on VKA with MVB and without MB. 3.3. Minor Bleeding on Rivaroxaban there were no differences in CAT Subjects treated with rivaroxaban, who experienced MBs or not, were similar in terms of demographics, clinical and laboratory variables, including the duration of anticoagulation, aspirin use, and ﬁbrinogen levels (Table 1). Patients with MBs more often experienced unprovoked VTE as compared with subjects without MB (88.9 vs. 53.3%, p < 0.001). A plasma rivaroxaban concentration was measured in 116 (87.9%) patients and CAT parameters in this group were not diﬀerent from those in whom this concentration was unavailable, except for ETP, which was larger (150472 (129249–169270) vs. 132293 (77811-156610) nM × s, respectively, p = 0.003). A median time since last dose of rivaroxaban in patients with vs. without MB was similar (23 (20–24) vs. MB 24 (24–26) hours, respectively; p = 0.27). There was also no diﬀerence in rivaroxaban concentrations between patients with MB (35 (6–66) µg/L) and without this complication (25 (13–44) µg/L, p = 0.73). 3.3. Minor Bleeding on Rivaroxaban Subjects treated with rivaroxaban, who experienced MBs or not, were similar in terms of demographics, clinical and laboratory variables, including the duration of anticoagulation, aspirin use, and fibrinogen levels (Table 1). Patients with MBs more often experienced unprovoked VTE as compared with subjects without MB (88.9 vs. 53.3%, p < 0.001). A plasma rivaroxaban concentration was measured in 116 (87.9%) patients and CAT parameters in this group were not different from those in whom this concentration was unavailable, except for ETP, which was larger (150472 (129249– 169270) vs. 132293 (77811-156610) nM × s, respectively, p = 0.003). A median time since last dose of rivaroxaban in patients with vs. without MB was similar (23 (20–24) vs. MB 24 (24–26) hours, respectively; p = 0.27). There was also no difference in rivaroxaban concentrations between patients In the studied patients, there were no correlations between rivaroxaban concentrations and CAT parameters with the exception of TTP (r = 0.27, p = 0.02). After adjustment for rivaroxaban concentration, patients on rivaroxaban with MB had longer lag time (p < 0.001), longer TTP (p < 0.001), lower max IIa (p < 0.001), lower max rate (p < 0.001) and lower ETP (p < 0.001) as compared with subjects without MB (Figure 1, Table S2). Representative thrombin generation curves are shown in Figure 2. with MB (35 (6–66) µg/L) and without this complication (25 (13–44) µg/L, p = 0.73). 3.1. General Characteristics The only signiﬁcant intergroup diﬀerences were marked as follows: * p < 0.05 vs. controls, ** p < 0.05 vs. VKA without MB, # p < 0.001 vs rivaroxaban without MB, ## p < 0.05 vs rivaroxaban without MB. J. Clin. Med. 2020, 9, 2018 8 of 17 3.2. Bleeding and Thromboembolic Events During Follow-up J. Clin. Med. 2020, 9, x FOR PEER REVIEW 3.2. Bleeding and Thromboembolic Events During Follow-up J. Clin. Med. 2020, 9, x FOR PEER REVIEW During a median follow-up of 10 (interquartile range, 7–19) months, MBs were observed in 27 (20.5%) patients on rivaroxaban, in 25 (17.2%) on VKA and in 2 (6.5%) controls (Table S1). In patients on rivaroxaban, major bleedings occurred in seven (5.3%) subjects, including ﬁve gastrointestinal bleeds and one menorrhagia. In ﬁve patients (71.4%) major bleeding was observed in individuals who also reported MB. In VKA patients, major bleedings occurred in six (4.1%) subjects, including two gastrointestinal bleeds and two vaginal bleeds (Table S1). In four VKA patients (66.7%), major bleed was preceded by MB. In the whole anticoagulated group, major bleeding was observed more often in patients reporting MBs (9 (17.3%) vs. 4 (1.8%), p < 0.0001) with risk ratio of 9.7 (95% CI 3.1–30.4). 3.2. Bleeding and Thromboembolic Events During Follow-up During a median follow-up of 10 (interquartile range, 7–19) months, MBs were observed in 27 (20.5%) patients on rivaroxaban, in 25 (17.2%) on VKA and in 2 (6.5%) controls (Table S1). In patients on rivaroxaban, major bleedings occurred in seven (5.3%) subjects, including five gastrointestinal bleeds and one menorrhagia. In five patients (71.4%) major bleeding was observed in individuals who also reported MB. In VKA patients, major bleedings occurred in six (4.1%) subjects, including two gastrointestinal bleeds and two vaginal bleeds (Table S1). In four VKA patients (66.7%), major bleed was preceded by MB. In the whole anticoagulated group, major bleeding was observed more f ( ( %) ( %) ) h k f ( % CI p p g p Recurrent thromboembolic events were reported in ﬁve (3.8%) patients on rivaroxaban and in three (2.1%) on VKA, without any relationship with MBs. Irrespective of the anticoagulant treatment, there were no diﬀerences in CAT parameters between patients with and without recurrent VTE. often in patients reporting MBs (9 (17.3%) vs. 4 (1.8%), p < 0.0001) with risk ratio of 9.7 (95% CI 3.1– 30.4). Recurrent thromboembolic events were reported in five (3.8%) patients on rivaroxaban and in three (2 1%) on VKA without any relationship with MBs Irrespective of the anticoagulant treatment 3.3. Minor Bleeding on Rivaroxaban there were no differences in CAT In the studied patients, there were no correlations between rivaroxaban concentrations and CAT parameters with the exception of TTP (r = 0.27, p = 0.02). After adjustment for rivaroxaban concentration, patients on rivaroxaban with MB had longer lag time (p < 0.001), longer TTP (p < 0.001), lower max IIa (p < 0.001), lower max rate (p < 0.001) and lower ETP (p < 0.001) as compared with subjects without MB (Figure 1, Table S2). Representative thrombin generation curves are shown in Figure 2 g No MB MB No MB MB 0 500 1000 1500 2000 2500 p < 0.0011 p < 0.0012 p = 0.1021 p = 0.1953 Controls Rivaroxaban VKA Lag, s (A) Figure 1. Cont. g No MB MB No MB MB 0 500 1000 1500 2000 2500 p < 0.0011 p < 0.0012 p = 0.1021 p = 0.1953 Controls Rivaroxaban VKA Lag, s (A) Figure 1. Cont. 9 of 17 f 1 J. Clin. Med. 2020, 9, 2018 No MB MB No MB MB 0 1000 2000 3000 4000 p < 0.0011 p < 0.0012 p = 0.1911 p = 0.3423 Controls Rivaroxaban VKA TTP, s (B) No MB MB No MB MB 0 100 200 300 400 500 600 p < 0.0011 p < 0.0012 p = 0.0231 p = 0.0493 Controls Rivaroxaban VKA Max IIa, nM (C) No MB MB No MB MB 0 2 4 6 8 p < 0.0011 p < 0.0012 p = 0.0351 p = 0.0723 Controls Rivaroxaban VKA Max rate, nM/s (D) No MB MB No MB MB 0 5.0×104 1.0×105 1.5×105 2.0×105 2.5×105 3.0×105 p < 0.0011 p < 0.0012 p = 0.0881 p = 0.1663 Controls Rivaroxaban VKA ETP, nM x s (E) The calibrated automated thrombography parameters in the studied groups. 3.3. Minor Bleeding on Rivaroxaban there were no differences in CAT P an with MB had longer lag time (A), longer TTP (B), lower max IIa (C), lower ma (B) No MB MB No MB MB 0 100 200 300 400 500 600 p < 0.0011 p < 0.0012 p = 0.0231 p = 0.0493 Controls Rivaroxaban VKA Max IIa, nM (B) (C) No MB MB No MB MB 0 2 4 6 8 p < 0.0011 p < 0.0012 p = 0.0351 p = 0.0723 Controls Rivaroxaban VKA Max rate, nM/s (C) (D) No MB MB No MB MB 0 5.0×104 1.0×105 1.5×105 2.0×105 2.5×105 3.0×105 p < 0.0011 p < 0.0012 p = 0.0881 p = 0.1663 Controls Rivaroxaban VKA ETP, nM x s (E) (D) (E) Figure 1. The calibrated automated thrombography parameters in the studied groups. Patients on rivaroxaban with MB had longer lag time (A), longer TTP (B), lower max IIa (C), lower max rate (D) 10 of 17 18 10 of 17 18 J. Clin. Med. 2020, 9, 2018 J. Clin. Med. 2020, 9, x F and lower endogenous thrombin potential (E) as compared with subjects without MB. Patients on VKA with MBs had similar lag time (A), TTP (B), max rate (D), ETP (E) and lower max IIa (C) as compared with subjects without MB. Abbreviations: box plot shows median and interquartile range (IQR). Whiskers are drawn at Q3 + 1.5 × IQR, Q1−1.5 × IQR, MB: minor bleeding, VKA: vitamin K antagonist, lag time: time to start thrombin generation, TTP: time to peak thrombin generation, max IIa: peak thrombin generation, max rate: the highest rate of thrombin generation, ETP: endogenous thrombin potential, 1 Before adjustment diﬀerences compared by Student t test or by the Mann–Whitney U test and after adjustment with regression analysis for 2 rivaroxaban concentration or for 3 INR. and lower endogenous thrombin potential (E) as compared with subjects without MB. Patients on VKA with MBs had similar lag time (A), TTP (B), max rate (D), ETP (E) and lower max IIa (C) as compared with subjects without MB. Abbreviations: box plot shows median and interquartile range (IQR). 3.3. Minor Bleeding on Rivaroxaban there were no differences in CAT Whiskers are drawn at Q3 + 1.5 × IQR, Q1−1.5 × IQR, MB: minor bleeding, VKA: vitamin K antagonist, lag time: time to start thrombin generation, TTP: time to peak thrombin generation, max IIa: peak thrombin generation, max rate: the highest rate of thrombin generation, ETP: endogenous thrombin potential, 1 Before adjustment differences compared by Student t test or by the Mann– Whitney U test and after adjustment with regression analysis for 2 rivaroxaban concentration or for 3 INR. Figure 2. Representative thrombin generation curves in the calibrated automated thrombogram. (A). Rivaroxaban vs. control, (B). VKA vs. control. Abbreviations: MB: minor bleeding, VKA: vitamin K antagonist. Figure 2. Representative thrombin generation curves in the calibrated automated thrombogram. (A). Rivaroxaban vs. control, (B). VKA vs. control. Abbreviations: MB: minor bleeding, VKA: vitamin K antagonist. Figure 2. Representative thrombin generation curves in the calibrated automated thrombogram. (A). Rivaroxaban vs. control, (B). VKA vs. control. Abbreviations: MB: minor bleeding, VKA: vitamin K antagonist. Figure 2. Representative thrombin generation curves in the calibrated automated thrombogram. (A). Rivaroxaban vs. control, (B). VKA vs. control. Abbreviations: MB: minor bleeding, VKA: vitamin K antagonist. In 40 patients on rivaroxaban, CAT parameters were measured twice after 3–6 months. As expected, there was a correlation between repeated CAT parameters, including Lag time (r = 0.48), max IIa (r = 0.51), ETP (r = 0.37), TTP (r = 0.33) and max rate (r = 0.32) (for all p < 0.05). Baseline characteristics of patients on rivaroxaban without MBs and controls who stopped In 40 patients on rivaroxaban, CAT parameters were measured twice after 3–6 months. As expected there was a correlation between repeated CAT parameters, including Lag time (r = 0.48), max IIa (r = 0.51), ETP (r = 0.37), TTP (r = 0.33) and max rate (r = 0.32) (for all p < 0.05). Baseline characteristics of patients on rivaroxaban without MBs and controls who stopped anticoagulation were comparable. There were no differences in CAT parameters between those two groups except for TTP (p = 0.025) (Figure 1, Table S2). Patients on rivaroxaban with unprovoked VTE were significantly older, more frequently male, Baseline characteristics of patients on rivaroxaban without MBs and controls who stopped anticoagulation were comparable. There were no diﬀerences in CAT parameters between those two groups except for TTP (p = 0.025) (Figure 1, Table S2). 3.3. Minor Bleeding on Rivaroxaban there were no differences in CAT p g y , q y , had higher BMI, hemoglobin and creatinine as well as a greater prevalence of deficiencies in natural Patients on rivaroxaban with unprovoked VTE were signiﬁcantly older, more frequently male, had higher BMI, hemoglobin and creatinine as well as a greater prevalence of deﬁciencies in natural anticoagulants (20.8 vs. 1.3%, p < 0.001) as compared with the provoked VTE group (Table S3). Patients with unprovoked VTE treated with rivaroxaban had lower max IIa and ETP as compared with those with provoked episodes (Table S3). J. Clin. Med. 2020, 9, 2018 ding aspirin use, exc 4 2% 0 048 T 11 of 17 h type 2 ti t 3.4. Minor Bleeding on VKA mpared with subjects wi th TTR l MBs were reported by 25 patients (17.2%) on VKA, including 9 (11.7%) on warfarin and 16 (23.5%) on acenocoumarol (p = 0.08). MBs on VKA was not associated with patient characteristics, including aspirin use, except for the overrepresentation of individuals diagnosed with type 2 diabetes (16.0 vs. 4.2%, p = 0.048; Table 1). Laboratory investigations showed higher INR in patients with MB as compared with subjects without MB (2.19 (1.46–2.45) vs. 1.47 (1.02–2.20), respectively, p = 0.01), however the TTR values were almost identical in both groups (60% (43–80%) vs. 60% (46–80%), respectively, p = 0.81). ver the TTR values were almost identical in both groups (60% (43–80%) vs. 60% (4 ctively, p = 0.81). After adjustment for INR, patients treated with VKA with and without MBs had compa TTP time, max rate and ETP. Only lower max IIa (p = 0.049) was found in patients with ared with the remaining subjects on VKA (Figure 1, Table S2). In groups with INR < 2 T parameters were not different between MB and non-MB subjects. There were si e ces bet ee all CAT pa a ete s of patie ts o VKA ithout MBs as co pa ed ith After adjustment for INR, patients treated with VKA with and without MBs had comparable lag time, TTP time, max rate and ETP. Only lower max IIa (p = 0.049) was found in patients with MBs as compared with the remaining subjects on VKA (Figure 1, Table S2). In groups with INR < 2 or INR ≥2, CAT parameters were not diﬀerent between MB and non-MB subjects. There were signiﬁcant diﬀerences between all CAT parameters of patients on VKA without MBs as compared with controls who stopped anticoagulation (Figure 1, Table S2). ences between all CAT parameters of patients on VKA without MBs as compared with topped anticoagulation (Figure 1, Table S2). AT Parameters in Patients with Major Bleeding n 13 (4.7%) patients with major bleeding on rivaroxaban or VKA, lag time was 703 (4 II / 3.5. CAT Parameters in Patients with Major Bleeding P was 950 (646 1501) s, max IIa was 130 (60 25101 (44445–143410) nM × s, without any d In 13 (4.7%) patients with major bleeding on rivaroxaban or VKA, lag time was 703 (413–1302) s, TTP was 950 (646–1501) s, max IIa was 130 (60–225) nM, max rate was 1.1 (0.9–2.2) nM/s and ETP was 125101 (44445–143410) nM × s, without any diﬀerences as compared to the subjects with MB on rivaroxaban or VKA. There was a lower max IIa (p = 0.039) in patients who suﬀered from major bleeding during follow-up as compared with those free of any bleeding event. No other diﬀerences in CAT parameters were noted. ( ) , y p j xaban or VKA. There was a lower max IIa (p = 0.039) in patients who suffered fro ng during follow-up as compared with those free of any bleeding event. No other di T parameters were noted. OC Curves 3.6. ROC Curves n patients on riv In patients on rivaroxaban, both lag time and TTP predicted MBs with the area under the ROC curve of 0.85 (p < 0.0001 for both; Figure 3). The cut-oﬀvalue in prediction of MBs for lag time was 693.5 s with sensitivity of 81.5% and speciﬁcity of 76.2%, while the cut-oﬀvalue for TTP was 1017 s with sensitivity of 77.8% and speciﬁcity of 77.1% and for max IIa was 173.5 nM (sensitivity, 74.1% and speciﬁcity, 71.4%) and for max rate was 1.55 nM/s (sensitivity, 70.4% and speciﬁcity, 70.5%). Finally, the corresponding values for ETP were 132772 nM × s, 66.7% and 69.5%. of 0.85 (p < 0.0001 for both; Figure 3). The cut-off value in prediction of MBs for lag t s with sensitivity of 81.5% and specificity of 76.2%, while the cut-off value for TTP w sensitivity of 77.8% and specificity of 77.1% and for max IIa was 173.5 nM (sensitivit pecificity, 71.4%) and for max rate was 1.55 nM/s (sensitivity, 70.4% and specificity y, the corresponding values for ETP were 132772 nM × s, 66.7% and 69.5%. atie t o VKA the hi he t but ode ate edi ti e alue ea hed a a IIa a d p g In patients on VKA, the highest, but moderate predictive values reached a max IIa and max rate with the area under the ROC curve of 0.63 (p < 0.05 for both) and both sensitivity and speciﬁcity did not exceed 60% (Figure 3). n patients on VKA, the highest, but moderate predictive values reached a max IIa and he area under the ROC curve of 0.63 (p < 0.05 for both) and both sensitivity and specif ceed 60% (Figure 3). Lag time on rivaroxaban 0 50 100 0 50 100 AUC 0.85 (0.79-0.92), p < 0.0001 100% - Specificity% Sensitivity, % (A) Lag time on VKA 0 50 100 0 50 100 AUC 0.59 (0.45-0.72), p = 0.17 100% - Specificity% Sensitivity, % (B) Figure 3. Cont. (A) (B) Figure 3. Cont. J. Clin. Med. 2020, 9, 2018 12 o Med. 3.6. ROC Curves n patients on riv 2020, 9, x FOR PEER REVIEW TTP on rivaroxaban 0 50 100 0 50 100 AUC 0.85 (0.79-0.92), p < 0.0001 100% - Specificity% Sensitivity, % (C) TTP on VKA 0 50 100 0 50 100 AUC 0.60 (0.47-0.73), p = 0.12 100% - Specificity% Sensitivity, % (D) Max IIa on rivaroxaban 0 50 100 0 50 100 AUC 0.77 (0.67-0.87), p < 0.0001 100% - Specificity% Sensitivity, % (E) Max IIa on VKA 0 50 100 0 50 100 AUC 0.63 (0.52-0.74), p = 0.038 100% - Specificity% Sensitivity, % (F) Max rate on rivaroxaban 0 50 100 0 50 100 AUC 0.76 (0.66-0.87), p < 0.0001 100% - Specificity% Sensitivity, % (G) Max rate on VKA 0 50 100 0 50 100 AUC 0.63 (0.52-0.75), p = 0.034 100% - Specificity% Sensitivity, % (H) Figure 3. Cont. 12 of 17 J. Clin. Med. 2020, 9, 2018 Med. eneration, max IIa: peak throm TP: endogenous thrombin pote 3.7. Predictors of Minor Bleedings TP: endogenous thrombin potential. redictors of Minor Bleedings n patients on rivaroxaban, age, gender, BMI, creatinine level, INR, rivaroxaban conce voked VTE and CAT parameters were identified as potentially associated with In patients on rivaroxaban, age, gender, BMI, creatinine level, INR, rivaroxaban concentration, unprovoked VTE and CAT parameters were identiﬁed as potentially associated with MBs. By multivariate analysis, lag time (odds ratio (OR) 1.006, 95% CI 1.002–1.010 per 1 s, p = 0.007) and unprovoked VTE (OR 19.61, 95% CI 1.13–59.31, p = 0.041) were independently associated with this adverse event in VTE patients on rivaroxaban (Table 2). variate analysis, lag time (odds ratio (OR) 1.006, 95% CI 1.002–1.010 per 1 s, p = 0. voked VTE (OR 19.61, 95% CI 1.13–59.31, p = 0.041) were independently associated se event in VTE patients on rivaroxaban (Table 2). mong VTE patients on VKA, age, gender, BMI, creatinine, INR, diabetes mellitus a eters were identified as potentially associated with MBs. By multivariate analy bin generation (OR 0.995, 95% CI 0.989–0.999 per 1 nM, p = 0.045) was the only inde associated with this kind of bleeding on VKA (Table 2). Table 2. The independent predictors of minor bleedings on rivaroxaban or on VKA. Univariate Model Multivariate Model dependent Variable p-Value OR 95% CI for OR p-Value OR 95% MB on Rivaroxaban Age, per 1 year 0.586 1.010 0.975–1.045 0.996 1.000 0.9 ale gender, yes/no 0.071 0.454 0.193–1.069 0.828 0.818 0.1 mass index, per 1 kg/m2 0.935 0.997 0.927–1.073 0.787 0.977 0.8 atinine, per 1 µmol/L 0.069 1.029 0.998–1.061 0.503 0.980 0.9 INR, per 0.01 0.427 0.982 0.940–1.026 0.542 0.969 0.8 oxaban concentration, per 1 µg/L 0.349 1.011 0.989–1.033 0.947 1.001 0.9 k d VTE / 0 002 7 246 2 062 25 641 0 041 19 607 1 13 Table 2. The independent predictors of minor bleedings on rivaroxaban or on VKA. 3.6. ROC Curves n patients on riv 2020, 9, x FOR PEE 12 of 17 TTP on rivaroxaban 0 50 100 0 50 100 AUC 0.85 (0.79-0.92), p < 0.0001 100% - Specificity% Sensitivity, % (C) TTP on VKA 0 50 100 0 50 100 AUC 0.60 (0.47-0.73), p = 0.12 100% - Specificity% Sensitivity, % (D) Max IIa on rivaroxaban 0 50 100 0 50 100 AUC 0.77 (0.67-0.87), p < 0.0001 100% - Specificity% Sensitivity, % (E) Max IIa on VKA 0 50 100 0 50 100 AUC 0.63 (0.52-0.74), p = 0.038 100% - Specificity% Sensitivity, % (F) Max rate on rivaroxaban 50 100 Sensitivity, % Max rate on VKA 50 100 Sensitivity, % TTP on rivaroxaban 0 50 100 0 50 100 AUC 0.85 (0.79-0.92), p < 0.0001 100% - Specificity% Sensitivity, % (C) TTP on VKA 0 50 100 0 50 100 AUC 0.60 (0.47-0.73), p = 0.12 100% - Specificity% Sensitivity, % (D) (C) Max IIa on VKA 0 50 100 0 50 100 AUC 0.63 (0.52-0.74), p = 0.038 100% - Specificity% Sensitivity, % (F) Max IIa on rivaroxaban 0 50 100 0 50 100 AUC 0.77 (0.67-0.87), p < 0.0001 100% - Specificity% Sensitivity, % (E) Sensitivity, % (E) (F) Max rate on rivaroxaban 0 50 100 0 50 100 AUC 0.76 (0.66-0.87), p < 0.0001 100% - Specificity% Sensitivity, % (G) Max rate on VKA 0 50 100 0 50 100 AUC 0.63 (0.52-0.75), p = 0.034 100% - Specificity% Sensitivity, % (H) (F) (E) Max rate on VKA 0 50 100 0 50 100 AUC 0.63 (0.52-0.75), p = 0.034 100% - Specificity% Sensitivity, % Max rate on VKA Sensitivity, % Max rate on rivaroxaban Sensitivity, % Sensitivity, % (H) (G) Figure 3. Cont. 13 of 17 J. Clin. Med. 2020, 9, 2018 Med. 2020, 9, x FOR PEE ETP on rivaroxaban 0 50 100 0 50 100 AUC 0.73 (0.61-0.84), p = 0.0003 100% - Specificity% Sensitivity, % (I) ETP on VKA 0 50 100 0 50 100 AUC 0.61 (0.50-0.73), p = 0.08 100% - Specificity% Sensitivity, % (J) gure 3. The receiver operating characteristics curves for calibrated automated thrombo rameters in prediction of minor bleedings. 3.6. ROC Curves n patients on riv In patients on VKA, moderate predictive values reached max IIa (F) and max rate (H) whereas lag time (B), TTP (D) and ETP (J) had no predictive value. Abbreviations: AUC: area under the curve, VKA: vitamin K antagonist, lag time: time to start thrombin generation, TTP: time to peak thrombin generation, max IIa: peak thrombin generation, max rate: the highest rate of thrombin generation, ETP: endogenous thrombin potential. 3.6. ROC Curves n patients on riv In patients on rivaroxaban, the highest predictiv minor bleedings had lag time (A), TTP (C), max IIa (E) and max rate (G) and moderate ETP tients on VKA, moderate predictive values reached max IIa (F) and max rate (H) whereas la , TTP (D) and ETP (J) had no predictive value. Abbreviations: AUC: area under the curve amin K antagonist, lag time: time to start thrombin generation, TTP: time to peak th ti II k th bi ti t th hi h t t f th bi Figure 3. The receiver operating characteristics curves for calibrated automated thrombography parameters in prediction of minor bleedings. In patients on rivaroxaban, the highest predictive value for minor bleedings had lag time (A), TTP (C), max IIa (E) and max rate (G) and moderate ETP (I). In patients on VKA, moderate predictive values reached max IIa (F) and max rate (H) whereas lag time (B), TTP (D) and ETP (J) had no predictive value. Abbreviations: AUC: area under the curve, VKA: vitamin K antagonist, lag time: time to start thrombin generation, TTP: time to peak thrombin generation, max IIa: peak thrombin generation, max rate: the highest rate of thrombin generation, ETP: endogenous thrombin potential. ETP on rivaroxaban 0 50 100 0 50 100 AUC 0.73 (0.61-0.84), p = 0.0003 100% - Specificity% Sensitivity, % (I) ETP on VKA 0 50 100 0 50 100 AUC 0.61 (0.50-0.73), p = 0.08 100% - Specificity% Sensitivity, % (J) (J) (I) re 3. The receiver operating characteristics curves for calibrated automated thromb meters in prediction of minor bleedings. In patients on rivaroxaban, the highest predicti minor bleedings had lag time (A), TTP (C), max IIa (E) and max rate (G) and moderate ET nts on VKA, moderate predictive values reached max IIa (F) and max rate (H) whereas TTP (D) and ETP (J) had no predictive value. Abbreviations: AUC: area under the curv min K antagonist, lag time: time to start thrombin generation, TTP: time to peak th Figure 3. The receiver operating characteristics curves for calibrated automated thrombography parameters in prediction of minor bleedings. In patients on rivaroxaban, the highest predictive value for minor bleedings had lag time (A), TTP (C), max IIa (E) and max rate (G) and moderate ETP (I). , µg/L 0.349 1.011 0.989–1.033 0.947 1.001 o s o oe o s , go s , g e e o s o ge e o , pe ombin generation. For model with rivaroxaban Negelkerke R2 was 0.49 and for VKA 0.16 (p < 0.001 for both). 4. Discussion We demonstrated that CAT parameters, measured prior to the next dose of rivaroxaban, could be useful in the prediction of minor bleedings in patients with VTE. We found that both lag time reﬂecting the time necessary to start thrombin generation as well as time to peak thrombin generation are valuable predictors of minor bleedings in patients on rivaroxaban, but not in those on VKA. Moreover, both time to start thrombin generation as well as the absence of any identiﬁable cause of VTE should be considered as potential predictors of elevated risk of MBs. Our ﬁndings regarding minor bleeds indicate that in the real-life setting, CAT assessment could be useful in the optimization of anticoagulation strategy in VTE patients treated on a long-term basis. Nevertheless, despite the fact that major bleedings, irrespective of the type of anticoagulation, were more frequent in patients with MB, CAT parameters were not identiﬁed as predictors of major bleedings. MBs are commonly observed on anticoagulation and are clinically relevant, though hardly predictable, in the context of recent studies [13–15]. To our knowledge, studies in which the risk of bleeding was assessed using CAT have been performed solely in patients on VKA [34]. Bloemen et al. [35], in a prospective cohort study involving 129 patients on VKA, found signiﬁcantly lower values of ETP and thrombin peak concentrations in patients with diverse bleeding complications. In turn, Dargaud et al. [36] showed that patients on warfarin with INR within the recommended range admitted with hemorrhage were characterized by markedly lower ETP. Our VTE patients with MBs on VKA had only lower peak thrombin generation without any signiﬁcant diﬀerences after adjustment for INR in other CAT parameters. These discrepancies might be associated with both diﬀerent thrombin generation assays applied as well as with diﬀerent clinical and laboratory characteristics of the studied patients. In Bloemen’s study [35], signiﬁcant diﬀerences were found only in the whole blood thrombin generation assay but not in plasma-based assays. Moreover, in both cited studies [35,36], INR values were higher as compared with those determined in our patients, particularly in the group without minor bleeds. Finally, the study by Dargaud et al. [36] included patients who were admitted to the emergency department with acute illnesses including major bleedings or thrombosis. In turn, of particular importance in real-life are the present ﬁndings regarding VTE patients on rivaroxaban showing association of CAT parameters with MBs during follow-up. eneration, max IIa: peak throm TP: endogenous thrombin pote 3.7. Predictors of Minor Bleedings Univariate Model Multivariate Model Independent Variable p-Value OR 95% CI for OR p-Value OR 95% CI for OR MB on Rivaroxaban Age, per 1 year 0.586 1.010 0.975–1.045 0.996 1.000 0.929–1.076 Male gender, yes/no 0.071 0.454 0.193–1.069 0.828 0.818 0.134–5.005 Body mass index, per 1 kg/m2 0.935 0.997 0.927–1.073 0.787 0.977 0.827–1.155 Creatinine, per 1 µmol/L 0.069 1.029 0.998–1.061 0.503 0.980 0.926–1.039 INR, per 0.01 0.427 0.982 0.940–1.026 0.542 0.969 0.877–1.071 Rivaroxaban concentration, per 1 µg/L 0.349 1.011 0.989–1.033 0.947 1.001 0.970–1.034 Unprovoked VTE, yes/no 0.002 7.246 2.062–25.641 0.041 19.607 1.131–59.311 Lag time, per 1 s <0.001 1.004 1.002–1.006 0.007 1.006 1.002–1.010 MB on VKA Age, per 1 year 0.811 1.004 0.969–1.042 0.851 0.996 0.956–1.038 Male gender, yes/no 0.594 1.269 0.528–3.051 0.928 0.948 0.297–3.024 Body mass index, per 1 kg/m2 0.971 0.998 0.917–1.088 0.972 0.998 0.905–1.101 Creatinine, per 1 µmol/L 0.563 0.991 0.960–1.022 0.365 0.980 0.938–1.024 INR, per 0.01 0.027 1.007 1.001–1.013 0.054 1.006 1.000–1.013 Diabetes mellitus, yes/no 0.038 4.386 1.086–17.544 0.059 4.098 0.948–17.857 Max IIa, per 1 nM 0.023 0.994 0.989–0.999 0.045 0.995 0.989–0.999 Abbreviations: OR: odds ratio, CI: confidence interval, MB: minor bleeding, INR: international normalized ratio, VTE: venous thromboembolism, VKA: vitamin K antagonist, lag time: time to start thrombin generation, max IIa: peak thrombin generation. For model with rivaroxaban Negelkerke R2 was 0.49 and for VKA 0.16 (p < 0.001 for both). alysis, lag time (odds ratio (OR) 1.006, 95% CI 1.002–1.010 per Table 2. The independent predictors of minor bleedings on rivaroxaban or on VKA. J. Clin. Med. 2020, 9, 2018 14 of 17 Among VTE patients on VKA, age, gender, BMI, creatinine, INR, diabetes mellitus and CAT parameters were identiﬁed as potentially associated with MBs. By multivariate analysis, peak thrombin generation (OR 0.995, 95% CI 0.989–0.999 per 1 nM, p = 0.045) was the only independent factor associated with this kind of bleeding on VKA (Table 2). 4. Discussion Precisely lag time and TTP had the highest predictive value with diﬀerences also after adjustment for the residual rivaroxaban concentration. This observation is novel and might have practical implications, if corroborated in larger cohort studies. Our unexpected ﬁnding linking unprovoked VTE to MBs in VTE patients on rivaroxaban deserves a comment. Given the recurrence rate after unprovoked VTE episodes reaching 11% at 1 year and 30% at 5 years after cessation of anticoagulation [37], in patients with low or moderate bleeding risk, indeﬁnite anticoagulation longer than 3 months is recommended with regular bleeding risk assessment [38,39]. Prolonged treatment increases the risk of various types of bleedings and might contribute to the higher MBs incidence observed in our cohort. Moreover, patients with unprovoked VTE were older, more often male and with higher body mass index. These factors represent recognized risk factors in scales used to predict bleeding in anticoagulated VTE patients such as a VTE-BLEED score [40]. This study suggests that unprovoked VTE patients, while anticoagulated, should be screened for MBs given the risk of non-compliance in the case of persistent, or even mild, bleeding complications. Our study has several limitations. Firstly, the sample size is relatively small, however adequately powered. The low number of major bleedings hampers analysis of their association with CAT, and we cannot exclude that CAT curves display a diﬀerent pattern in larger groups of VTE patients with major bleeds compared with those without this complication, as shown previously for patients treated 15 of 17 J. Clin. Med. 2020, 9, 2018 with warfarin [35,36]. Secondly, since our patients had rivaroxaban concentrations undetectable or lower than 100 ng/mL, it remains to be established whether CAT parameters measured at peak drug concentration might have a similar predictive value. Thirdly, we did not determine other potential modulators of blood coagulation and ﬁbrinolysis, which might aﬀect the CAT parameters, e.g., prothrombin, antithrombin, α2-macroglobulin or α2-antiplasmin [41]. Finally, clinical relevance of altered CAT parameters in anticoagulated VTE patients using VKA or rivaroxaban, as well as apixaban, dabigatran or edoxaban, in terms of both life-threatening bleeding as well as thromboembolic events, remains to be conﬁrmed in larger studies. 5. Conclusions In VTE patients requiring chronic anticoagulation, both time to start thrombin generation and time to peak thrombin generation derived from CAT may predict the risk of minor bleedings on rivaroxaban. Together with unprovoked etiology of VTE, time to start thrombin generation were found to be independently associated with a predisposition to minor bleedings. However, CAT parameters are not accurate enough in the prediction of minor bleedings in patients on VKA. Our ﬁndings require further validation in the larger studies. Supplementary Materials: The following are available online at http://www.mdpi.com/2077-0383/9/7/2018/s1, Table S1. Minor and major bleedings in the studied groups during follow-up, Table S2: The calibrated automated thrombography parameters, Table S3: Baseline characteristics of patients on rivaroxaban with provoked and unprovoked VTE. Author Contributions: Conceptualization, J.Z., S.B. and A.U.; methodology, A.U.; software, J.Z., K.S. and K.N.; validation, J.Z., K.S., K.N., S.C., S.B. and A.U.; formal analysis, J.Z., K.S. and K.N.; investigation, S.C., S.B. and A.U.; resources, S.C., S.B. and A.U.; data curation, S.C., S.B. and A.U.; writing—original draft preparation, J.Z., K.S. and K.N.; writing—review and editing, J.Z., K.S., K.N. and A.U.; visualization, J.Z., K.S. and K.N.; supervision, S.B. and A.U.; project administration, S.B. and A.U.; funding acquisition, S.B. and A.U. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by Jagiellonian University Medical College, grant number N41/DBS/000184 to A.U. and by the National Institutes of Health, grant number UM1 HL120877 TACTIC Grant to S.B. The publication of this article was funded by the Priority Research Area qLife under the program “Excellence Initiative – Research University” at the Jagiellonian University in Krakow to J.Z. (application number 06/IDUB/2019/94). Conﬂicts of Interest: A.U. received lecture honoraria from Bayer, Boehringer Ingelheim and Pﬁzer. The remaining authors have no conﬂict of interest. The indicated funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. References 1. Schulman, S.; Angerås, U.; Bergqvist, D.; Eriksson, B.; Lassen, M.R.; Fisher, W. Subcommittee on Control of Anticoagulation of the Scientiﬁc and Standardization Committee of the International Society on Thrombosis and Haemostasis. Deﬁnition of major bleeding in clinical investigations of antihemostatic medicinal products in surgical patients. J. Thromb. Haemost. 2010, 8, 202–204. [CrossRef] 2. Kaatz, S.; Ahmad, D.; Spyropoulos, A.C.; Schulman, S. Subcommittee on Control of Anticoagulation. Deﬁnition of clinically relevant non-major bleeding in studies of anticoagulants in atrial ﬁbrillation and venous thromboembolic disease in non-surgical patients: Communication from the SSC of the ISTH. J. Thromb. Haemost. 2015, 13, 2119–2126. [CrossRef] [PubMed] 3. Ruíz-Giménez, N.; Suárez, C.; González, R.; Nieto, J.A.; Todolí, J.A.; Samperiz, A.L.; Monreal, M. RIETE Investigators. Predictive variables for major bleeding events in patients presenting with documented acute venous thromboembolism. Findings from the RIETE Registry. Thromb Haemost. 2008, 100, 26–31. [CrossRef] 3. Ruíz-Giménez, N.; Suárez, C.; González, R.; Nieto, J.A.; Todolí, J.A.; Samperiz, A.L.; Monreal, M. RIETE Investigators. Predictive variables for major bleeding events in patients presenting with documented acute venous thromboembolism. Findings from the RIETE Registry. Thromb Haemost. 2008, 100, 26–31. [CrossRef] 4. Weitz, J.I.; Haas, S.; Ageno, W.; Angchaisuksiri, P.; Bounameaux, H.; Nielsen, J.D.; Goldhaber, S.Z.; Goto, S.; g g y 4. Weitz, J.I.; Haas, S.; Ageno, W.; Angchaisuksiri, P.; Bounameaux, H.; Nielsen, J.D.; Goldhaber, S.Z.; Goto, S.; Kayani, G.; Mantovani, L.; et al. Global Anticoagulant Registry in the Field—Venous Thromboembolism (GARFIELD-VTE). Rationale and design. Thromb. Haemost. 2016, 116, 1172–1179. [CrossRef] 4. Weitz, J.I.; Haas, S.; Ageno, W.; Angchaisuksiri, P.; Bounameaux, H.; Nielsen, J.D.; Goldhaber, S.Z.; Goto, S.; Kayani, G.; Mantovani, L.; et al. Global Anticoagulant Registry in the Field—Venous Thromboembolism (GARFIELD-VTE). Rationale and design. Thromb. Haemost. 2016, 116, 1172–1179. [CrossRef] 5. Ageno, W.; Mantovani, L.G.; Haas, S.; Kreutz, R.; Monje, D.; Schneider, J.; van Eickels, M.; Gebel, M.; Zell, E.; Turpie, A.G. Safety and eﬀectiveness of oral rivaroxaban versus standard anticoagulation for the treatment of symptomatic deep-vein thrombosis (XALIA): An international, prospective, non-interventional study. Lancet Haematol. 2016, 3, e12–e21. [CrossRef] 5. Ageno, W.; Mantovani, L.G.; Haas, S.; Kreutz, R.; Monje, D.; Schneider, J.; van Eickels, M.; Gebel, M.; Zell, E.; Turpie, A.G. Safety and eﬀectiveness of oral rivaroxaban versus standard anticoagulation for the treatment of symptomatic deep-vein thrombosis (XALIA): An international, prospective, non-interventional study. Lancet Haematol. 2016, 3, e12–e21. [CrossRef] J. Clin. Med. 2020, 9, 2018 16 of 17 16 of 17 6. References Einstein Investigators; Bauersachs, R.; Berkowitz, S.D.; Brenner, B.; Buller, H.R.; Decousus, H.; Gallus, A.S.; Lensing, A.W.; Misselwitz, F.; Prins, M.H.; et al. Oral rivaroxaban for symptomatic venous thromboembolism. N. Engl. J. Med. 2010, 363, 2499–2510. [CrossRef] 7. Einstein–PE Investigators; Büller, H.R.; Prins, M.H.; Lensin, A.W.; Decousus, H.; Jacobson, B.F.; Minar, E.; Chlumsky, J.; Verhamme, P.; Wells, P.; et al. Oral rivaroxaban for the treatment of symptomatic pulmonary embolism. N. Engl. J. Med. 2012, 366, 1287–1297. [CrossRef] 8. Själander, A.; Friberg, B.; Svensson, P.; Stigendal, L.; Lethagen, S. Menorrhagia and minor bleeding symptoms in women on oral anticoagulation. J. Thromb. Thrombolysis. 2007, 24, 39–41. [CrossRef] [PubMed] 9. Konieczy´nska, M.; Bijak, P.; Desteghe, L.; Heidbuchel, H.; Undas, A. Knowledge gaps in patients with venous thromboembolism: Usefulness of a new questionnaire. Pol. Arch. Intern. Med. 2019, 129, 28–35. [CrossRef] [PubMed] 10. Konieczy´nska, M.; Sobieraj, E.; Bryk, A.H.; D˛ebski, M.; Polak, M.; Podolec, P.; Małecka, B.; Paj ˛ak, A.; Desteghe, L.; Heidbuchel, H.; et al. Diﬀerences in knowledge among patients with atrial ﬁbrillation receiving non-vitamin K antagonist oral anticoagulants and vitamin K antagonists. Kardiol. Pol. 2018, 76, 1089–1096. [CrossRef] 11. Beyer-Westendorf, J.; Förster, K.; Pannach, S.; Ebertz, F.; Gelbricht, V.; Thieme, C.; Michalski, F.; Köhler, C.; Werth, S.; Sahin, K.; et al. Rates, management, and outcome of rivaroxaban bleeding in daily care: Results from the Dresden NOAC registry. Blood 2014, 124, 955–962. [CrossRef] [PubMed] 12. Dawwas, G.K.; Brown, J.; Dietrich, E.; Park, H. Eﬀectiveness and safety of apixaban versus rivaroxaban for prevention of recurrent venous thromboembolism and adverse bleeding events in patients with venous thromboembolism: A retrospective population-based cohort analysis. Lancet Haematol. 2019, 6, e20–e28. [CrossRef] 13. Veeger, N.J.; Piersma-Wichers, M.; Meijer, K.; Hillege, H.L. Minor bleeds alert for subsequent major bleeding in patients using vitamin K antagonists. Br. J. Haematol. 2011, 153, 508–514. [CrossRef] [PubMed] 14. van Rein, N.; le Cessie, S.; van Vliet, I.P.; Reitsma, P.H.; van der Meer, F.J.; Lijfering, W.M.; Cannegieter, S.C. Increased risk of major bleeding after a minor bleed during treatment with vitamin K antagonists is determined by ﬁxed common risk factors. J. Thromb. Haemost. 2016, 14, 948–952. [CrossRef] 15. Bahit, M.C.; Lopes, R.D.; Wojdyla, D.M.; Held, C.; Hanna, M.; Vinereanu, D.; Hylek, E.M.; Verheugt, F.; Goto, S.; Alexander, J.H.; et al. Non-major bleeding with apixaban versus warfarin in patients with atrial ﬁbrillation. Heart 2017, 103, 623–628. [CrossRef] 16. Kremers, R.M.; Peters, T.C.; Wagenvoord, R.J.; Hemker, H.C. © 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/). References Altered plasma clot properties increase the risk of recurrent deep vein thrombosis: A cohort study Blood 2018 131 797–807 [CrossRef] [PubMed] 27. Cieslik, J.; Mrozinska, S.; Broniatowska, E.; Undas, A. Altered plasma clot properties increase the risk of recurrent deep vein thrombosis: A cohort study. Blood 2018, 131, 797–807. [CrossRef] [PubMed] 28. Wypasek, E.; Corral, J.; Alhenc-Gelas, M.; Sydor, W.; Iwaniec, T.; Celi´nska-Lowenhoﬀ, M.; Potaczek, D.P.; Blecharczyk, A.; Zawilska, K.; Musiał, J.; et al. Genetic characterization of antithrombin, protein C, and protein S deﬁciencies in Polish patients. Pol. Arch. Intern. Med. 2017, 127, 512–523. [CrossRef] [PubMed] 29. Stepien, K.; Nowak, K.; Wypasek, E.; Zalewski, J.; Undas, A. High prevalence of inherited thrombophilia and antiphospholipid syndrome in myocardial infarction with non-obstructive coronary arteries: Comparison with cryptogenic stroke. Int. J. Cardiol. 2019, 290, 1–6. [CrossRef] 30. Miyakis, S.; Lockshin, M.D.; Atsumi, T.; Branch, D.W.; Brey, R.L.; Cervera, R.; Derksen, R.H.; DE Groot, P.G.; Koike, T.; Meroni, P.L.; et al. International consensus statement on an update of the classiﬁcation criteria for deﬁnite antiphospholipid syndrome (APS). J. Thromb. Haemost. 2006, 4, 295–306. [CrossRef] 31. Zalewski, J.; Rychlak, R.; Góralczyk, T.; Undas, A. Rivaroxaban concentration in patients with deep vein thrombosis who reported thrombus progression or minor hemorrhagic complications: First Polish experience. Pol. Arch. Med. Wewn. 2014, 124, 553–555. [CrossRef] [PubMed] 32. Tripodi, A.; Braham, S.; Scimeca, B.; Moia, M.; Peyvandi, F. How and when to measure anticoagulant eﬀects of direct oral anticoagulants? Practical issues. Pol. Arch. Intern. Med. 2018, 128, 379–385. [CrossRef] [PubMed] 32. Tripodi, A.; Braham, S.; Scimeca, B.; Moia, M.; Peyvandi, F. How and when to measure anticoagulant eﬀects of direct oral anticoagulants? Practical issues. Pol. Arch. Intern. Med. 2018, 128, 379–385. [CrossRef] [PubMed] 33. Fr ˛aczek, P.; Krzysztoﬁk, M.; Stanisz, A.; Undas, A. Clinical outcomes and plasma clot permeability and l bilit i ti t ith th b b li i b A h t t d Pol Arch Intern Med 33. Fr ˛aczek, P.; Krzysztoﬁk, M.; Stanisz, A.; Undas, A. Clinical outcomes and plasma clot permeability and lysability in patients with venous thromboembolism on rivaroxaban: A cohort study. Pol. Arch. Intern. Med. 2019, 129, 377–385. [CrossRef] [PubMed] 34. Zwaveling, S.; Bloemen, S.; de Laat, B.; Ten Cate, H.; Ten Cate-Hoek, A. Calibrated automated thrombinography (CAT), a tool to identify patients at risk of bleeding during anticoagulant therapy: A systematic review. TH Open 2018, 2, e291–e302. [CrossRef] [PubMed] 35. References The balance of pro- and anticoagulant processes underlying thrombin generation. J. Thromb. Haemost. 2015, 13, 437–447. [CrossRef] 17. Al Dieri, R.; Peyvandi, F.; Santagostino, E.; Giansily, M.; Mannucci, P.M.; Schved, J.F.; Béguin, S.; Hemker, H.C. The thrombogram in rare inherited coagulation disorders: Its relation to clinical bleeding. Thromb. Haemost. 2002, 88, 576–582. [CrossRef] 18. Matsumoto, T.; Shima, M.; Takeyama, M.; Yoshida, K.; Tanaka, I.; Sakurai, Y.; Giles, A.R.; Yoshioka, A. The measurement of low levels of factor VIII or factor IX in hemophilia A and hemophilia B plasma by clot waveform analysis and thrombin generation assay. J. Thromb. Haemost. 2006, 4, 377–384. [CrossRef] 19. Barco, S.; Whitney Cheung, Y.; Coppens, M.; Hutten, B.A.; Meijers, J.C.; Middeldorp, S. In vivo reversal of the anticoagulant eﬀect of rivaroxaban with four-factor prothrombin complex concentrate. Br. J. Haematol. 2016, 172, 255–261. [CrossRef] 20. Crowther, M.; Cuker, A. How can we reverse bleeding in patients on direct oral anticoagulants? Kardiol. Pol. 2019, 77, 3–11. [CrossRef] 21. Hemker, H.C.; Giesen, P.; Al Dieri, R.; Regnault, V.; de Smedt, E.; Wagenvoord, R.; Lecompte, T.; Béguin, S. Calibrated automated thrombin generation measurement in clotting plasma. Pathophysiol. Haemost. Thromb. 2003, 33, 4–15. [CrossRef] [PubMed] 22. Mann, K.G.; Whelihan, M.F.; Butenas, S.; Orfeo, T. Citrate anticoagulation and the dynamics of thrombin generation. J. Thromb. Haemost. 2007, 5, 2055–2061. [CrossRef] [PubMed] 23. Orfeo, T.; Gissel, M.; Butenas, S.; Undas, A.; Brummel-Ziedins, K.E.; Mann, K.G. Anticoagulants and the propagation phase of thrombin generation. PLoS ONE 2011, 6, e27852. [CrossRef] [PubMed] 24. Rigano, J.; Ng, C.; Nandurkar, H.; Ho, P. Thrombin generation estimates the anticoagulation eﬀect of direct oral anticoagulants with signiﬁcant interindividual variability observed. Blood Coagul. Fibrinolysis. 2018, 29, 148–154. [CrossRef] [PubMed] J. Clin. Med. 2020, 9, 2018 17 of 17 25. Arachchillage, D.R.; Efthymiou, M.; Mackie, I.J.; Lawrie, A.S.; Machin, S.J.; Cohen, H. Rivaroxaban and warfarin achieve eﬀective anticoagulation, as assessed by inhibition of TG and in-vivo markers of coagulation activation, in patients with venous thromboembolism. Thromb. Res. 2015, 135, 388–393. [CrossRef] [PubMed] 6. Barcellona, D.; Vannini, M.L.; Fenu, L.; Balestrieri, C.; Marongiu, F. Warfarin or acenocoumarol: Whic better in the management of oral anticoagulants? Thromb. Haemost. 1998, 80, 899–902. [CrossRef] [PubM better in the management of oral anticoagulants? Thromb. Haemost. 1998, 80, 899–902. [CrossRef] [PubMed] 27. Cieslik, J.; Mrozinska, S.; Broniatowska, E.; Undas, A. References Bloemen, S.; Zwaveling, S.; Ten Cate, H.; Ten Cate-Hoek, A.; de Laat, B. Prediction of bleeding risk in taking vitamin K antagonists using thrombin generation testing. PLoS ONE 2017, 12, e0176967. [Cr 36. Dargaud, Y.; Hoﬀman, M.; Lefrapper, L.; Lin, F.C.; Genty, A.; Chatard, B.; Marin, S.; Négrier, C.; Monroe, D.M. Bleeding risk in warfarinized patients with a therapeutic international normalized ratio: The eﬀect of low factor IX levels. J. Thromb. Haemost. 2013, 11, 1043–1052. [CrossRef] [PubMed] 37. Prandoni, P.; Noventa, F.; Ghirarduzzi, A.; Pengo, V.; Bernardi, E.; Pesavento, R.; Iotti, M.; Tormene, D.; Simioni, P.; Pagnan, A. The risk of recurrent venous thromboembolism after discontinuing anticoagulation in patients with acute proximal deep vein thrombosis or pulmonary embolism. A prospective cohort study in 1,626 patients. Haematologica 2007, 92, 199–205. [CrossRef] 38. Tritschler, T.; Wells, P.S. Extended therapy for unprovoked venous thromboembolism: When is it indicated? Blood Adv. 2019, 3, 499. [CrossRef] 39. Kearon, C.; Akl, E.A.; Ornelas, J.; Blaivas, A.; Jimenez, D.; Bounameaux, H.; Huisman, M.; King, C.S.; Morris, T.A.; Sood, N.; et al. Antithrombotic Therapy for VTE Disease: CHEST Guideline and Expert Panel Report. Chest 2016, 149, 315–352. [CrossRef] 40. Klok, F.A.; Barco, S.; Turpie, A.G.G.; Haas, S.; Kreutz, R.; Mantovani, L.G.; Gebel, M.; Herpers, M.; Bugge, J.P.; Kostantinides, S.V.; et al. Predictive value of venous thromboembolism (VTE)-BLEED to predict major bleeding and other adverse events in a practice-based cohort of patients with VTE: Results of the XALIA study. Br. J. Haematol. 2018, 183, 457–465. [CrossRef] 41. Hemker, H.C.; Al Dieri, R.; De Smedt, E.; Beguin, S. Thrombin generation, a function test of the haemostatic-thrombotic system. Thromb. Haemost. 2006, 96, 553–561. [PubMed] © 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/W2528473158	https://journal.binus.ac.id/index.php/BBR/article/download/989/859	Indonesian	null	Mengembangkan Human Resource Management yang Strategis untuk Menunjang Daya Saing Organisasi: Perspektif Manajemen Kinerja (Performance Management) di Bank Syariah	Binus business review	2,015	cc-by-sa	5,767	MENGEMBANGKAN HUMAN RESOURCE MANAGEMENT YANG STRATEGIS UNTUK MENUNJANG DAYA SAING ORGANISASI: PERSPEKTIF MANAJEMEN KINERJA (PERFORMANCE MANAGEMENT) DI BANK SYARIAH Ahmad Azmy Management Programme, Tanri Abeng University Jl Swadarma Raya No.58 Ulujami-Pesanggrahan Jakarta Selatan 12250 Indonesia azmy33@gmail.com ABSTRACT This article discusses performance management in sharia banks related to human resources. The method used is descriptive in providing a comprehensive explanation based on factual data. Performance management in sharia banks should be applied based on the characteristics of human resources by sharia. The characteristics of the human resources required by sharia banks are different from conventional banks. Human resources in sharia banks should have different performance indicators to conventional banks. Performance indicators serve as a guide in the process of implementing sustainability performance as an effort of sharia banking in the banking industry in Indonesia. Keywords: performance management, performance indicators, sharia banking, sharia human resources h t i ti Keywords: performance management, performance indicators, sharia banking, sharia human resources characteristics y p f characteristics kunci: manajemen kinerja, indikator kinerja, bank syariah, karakteristik sumber daya manusia syariah ABSTRAK Artikel ini membahas tentang manajemen kinerja di bank syariah berkaitan dengan sumber daya manusia. Metode penelitian yang digunakan adalah deskriptif dalam memberikan penjelasan komprehensif berdasarkan data faktual. Manajemen kinerja di bank syariah harus diterapkan berdasarkan karakteristik sumber daya manusia secara syariah. Karakteristik sumber daya manusia yang dibutuhkan oleh bank syariah berbeda dengan bank konvensional. Sumber daya manusia di bank syariah harus memiliki indikator kinerja yang berbeda dengan bank konvensionalIndikator kinerja dijadikan sebagai panduan dalam proses penerapan kinerja sebagai upaya sustainabilitas perbankan syariah di industry perbankan di Indonesia. Kata kunci: manajemen kinerja, indikator kinerja, bank syariah, karakteristik sumber daya manusia syariah 78 78 BINUS BUSINESS REVIEW Vol. 6 No. 1 Mei 2015: 78-90 PENDAHULUAN Sebagai negara dengan penduduk muslim terbesar, sudah selayaknya Indonesia menjadi pelopor dalam pengembangan keuangan syariah di dunia. Hal ini bukan merupakan impian yang mustahil karena potensi Indonesia untuk menjadi global player dalam keuangan syariah sangat besar, antara lain adalah: (1) Jumlah penduduk muslim yang besar menjadi potensi nasabah industri keuangan syariah. (2) Prospek ekonomi yang cerah, tercermin dari pertumbuhan ekonomi yang relatif tinggi (kisaran 6,0%-6,%) yang ditopang oleh fundamental ekonomi yang solid. (3) Peningkatan sovereign credit rating Indonesia menjadi investment grade yang akan meningkatkan minat investor untuk berinvestasi di sektor keuangan domestik, termasuk industri keuangan syariah. (4) Memiliki sumber daya alam yang melimpah yang dapat dijadikan sebagai underlying transaksi industri keuangan syariah (Alamsyah, 2012). Di bawah ini indeks keuangan syariah dunia yang dirilis oleh Islamic Finance Country Index Tahun 2011: Gambar 1 Islamic Finance Country Index 2011 Gambar 1 Islamic Finance Country Index 2011 Dalam penilaian Global Islamic Financial Report (GIFR) tahun 2011, Indonesia menduduki urutan keempat negara yang memiliki potensi dan kondusif dalam pengembangan industri keuangan syariah setelah Iran, Malaysia dan Saudi Arabia (Gambar 1). Dengan melihat beberapa aspek dalam penghitungan indeks, seperti jumlah bank syariah, jumlah lembaga keuangan nonbank syariah, maupun ukuran aset keuangan syariah yang memiliki bobot terbesar, maka Indonesia diproyeksikan akan menduduki peringkat pertama dalam beberapa tahun ke depan. Optimisme ini sejalan dengan laju ekspansi kelembagaan dan akselerasi pertumbuhan aset perbankan syariah yang sangat tinggi, ditambah dengan volume penerbitan sukuk yang terus meningkat (Alamsyah: 2012). Pertumbuhan bank syariah di Indonesia saat ini cukup pesat karena jumlah mayoritas penduduk beragama islam di Indonesia sehingga menyebabkan perkembangan bisnis perbankan syariah menjadi daya tarik tersendiri bagi konsumen untuk menabung dan menikmati jasa layanan syariah. Bank syariah sudah terbukti menjadi lembaga yang tahan krisis ekonomi di Indonesia. Pada saat itu Bank Muamalat Indonesia menjadi bukti bahwa sistem perbankan syariah tahan dari krisis ekonomi yang menjadi virus bagi industri perbankan nasional. Bank syariah memiliki ciri tersendiri dibandingkan bank konvensional bahwa tidak ada unsur bunga dalam setiap penghitungan bagi hasil dan membuat nasabah yang khususnya beragama islam menjadi aman untuk menyimpan uangnya di bank syariah. PENDAHULUAN Di bawah ini jumlah pertumbuhan bank syariah di Indonesia sebagai berikut: Mengembangkan Human Resource Management …… (Ahmad Azmy) 79 Tabel 1 Jumlah Kantor Perbankan Syariah Tabel 1 Jumlah Kantor Perbankan Syariah Keterangan 2007 2008 2009 2011 2012 2013 Bank Umum Syariah Jumlah Kantor 401 541 711 1215 1780 1998 Jumlah Bank 3 5 6 11 11 11 Unit Usaha Syariah Afiliasi dengan Bank Konvensional 24 17 25 24 24 23 Jumlah Kantor 196 241 287 292 517 590 Bank Pembiayaan Syariah Jumlah Bank 114 138 150 155 158 168 Jumlah Kantor 185 202 225 364 401 402 Sumber: Statistik Perbankan Syariah 2013 Sumber: Statistik Perbankan Syariah 2013 Perkembangan perbankan syariah cukup pesat dengan melihat banyaknya pertumbuhan bank- bank baru di Indonesia. Pada tahun 2013 jumlah bank umum syariah tumbuh sebanyak 11 bank dengan jumlah kantor sebanyak 1998. Unit usaha syariah yang berafiliasi dengan bank konvensional berjumlah 23 unit dan jumlah kantor berjumlah 590. Bank Pembiayaan Syariah berjumlah 168 dan jumlah kantor 402. Melihat pertumbuhan bank syariah dipastikan membutuhkan banyak tenaga kerja. Sebuah organisasi dipastikan dapat tumbuh dan berkembang sesuai dengan ekspektasi konsumen ditentukan oleh jumlah anggota organisasi. Jumlah pertumbuhan bank syariah diiringi dengan pertumbuhan tenaga kerja. Di bawah ini jumlah tenaga kerja di bank syariah yaitu: Tabel 2 Jumlah Pekerja di Perbankan Syariah Keterangan 2007 2008 2009 2011 2012 2013 Bank Umum Syariah 4.311 6.609 10.348 15.224 24.111 26.717 Unit Usaha Syariah 2.266 2.562 2.296 1.868 3.108 11.223 Bank Pembiayaan Rakyat Syariah 2.108 2.581 2.799 3.172 4.359 4.824 Sumber: Statistik Perbankan Syariah 2013 Tabel 2 Jumlah Pekerja di Perbankan Syariah Sumber: Statistik Perbankan Syariah 2013 Sumber: Statistik Perbankan Syariah 2013 Tabel di atas menunjukkan bahwa pertumbuhan bank syariah diikuti oleh peningkatan jumlah tenaga kerja di bank syariah. Pada tahun 2013 mengalami peningkatan dimulai dari bank umum syariah berjumlah 26.717 orang, kemudian unit usaha syariah berjumlah 11.223 orang, dan bank pembiayaan rakyat syariah berjumlah 4.824 orang. Perkembangan jumlah tenaga kerja di bank syariah membuktikan bahwa bisnis perbankan syariah mengalami peningkatan yang signifikan. Pelayanan bank syariah harus menjamin bahwa semua produk yang ditawarkan kepada konsumen harus berbasis syariah. Daya saing organisasi bank syariah mutlak diperlukan untuk meningkatkan persaingan bisnis keuangan syariah. Daya saing bank syariah harus berlomba untuk menciptakan berbagai macam inovasi yang sesuai dengan tuntutan bisnis. Menurut Radenakers (2005) membagi inovasi ke dalam beberapa tipe yang mempunyai karakteristik masing-masing seperti disajikan pada tabel berikut: BINUS BUSINESS REVIEW Vol. 6 No. PENDAHULUAN 1 Mei 2015: 78-90 80 Tabel 3 Tipe dan Karakteristik Inovasi Tipe Inovasi Karakteristik 1 Inovasi Produk Produk, jasa, atau kombinasi keduanya yang baru 2 Inovasi Proses Metode baru dalam menjalankan kegiatan bernilai tambah (misalnya distribusi atau produksi) yang lebih baik atau lebih murah 3 Inovasi Organisasional Metode baru dalam mengelola, mengkoordinasi, dan mengawasi pegawai, kegiatan, dan tanggung jawab 4 Inovasi bisnis Kombinasi produk, proses, dan sistem organisasional yang baru (dikenal juga sebagai model bisnis) Beberapa tipe inovasi dibutuhkan untuk melihat apakah kebutuhan bisnis bank syariah sudah mampu menjawab kebutuhan konsumen. Inovasi produk merupakan tantangan bagaimana perbankan syariah mampu menciptakan produk, jasa, atau kombinasi sesuai kebutuhan nasabah dalam menjalankan mobilitas aktivitas sehingga bank syariah selalu digunakan dalam setiap kegiatan konsumen. Inovasi proses merupakan kesempatan bank syariah untuk menemukan proses bisnis yang baru yang memiliki nilai tambah bagi proses bisnis bank syariah. Inovasi organisasional harus melakukan adaptasi dengan produk dan proses bisnis melalui metode dalam mengelola sumber daya manusia yang memahami inti proses bisnis sehingga pegawai mampu memberikan kontribusi positif dan kinerja maksimal bagi sustainabilitas organisasi. Terakhir, inovasi bisnis bagaimana industri perbankan syariah bisa menggabungkan ketiga inovasi ini menjadi keunggulan kompetitif dan model bisnis sehingga mampu bersaing pada industri secara global dan internasional. Jadi, daya saing organisasi merupakan strategi bisnis harus sejalan dengan strategi pengembangan sumber daya manusia sehingga mampu memberikan kontribusi maksimal bagi pertumbuhan bisnis perbankan syariah dan menciptakan keunggulan kompetitif. Keseimbangan pertumbuhan bank syariah diikuti dengan kebutuhan peningkatan jumlah tenaga kerja. Ini bertujuan untuk memberikan pelayanan prima kepada konsumen dan edukasi produk perbankan syariah. Walau demikian, ada sejumlah permasalahan dalam sumber daya manusia di bank syariah. Menurut Permana (2012) salah satu masalah terbesar sumber daya manusia syariah adalah pihak perbankan kesulitan untuk mencari SDM perbankan syariah yang kompeten dan mumpuni. Perbankan syariah cenderung mengambil sumber daya manusia dari luar perguruan tinggi syariah karena SDM di bank syariah biasanya justru lebih mudah diberikan pengetahuan tentang perbankan syariah. Bank Indonesia memproyeksi industri perbankan syariah bisa memiliki pangsa pasar sebesar 15 persen pada 10 tahun mendatang (atau sekitar tahun 2022) apabila bisa mengalami pertumbuhan yang stabil seperti beberapa tahun terakhir. Deputi Gubernur Bank Indonesia (BI) Halim Alamsyah yang saat ini menjadi anggota Otoritas Jasa Keuangan (OJK) mengatakan industri perbankan syariah mengalami pertumbuhan dengan rerata 40,5 persen per tahun, dalam setengah dasawarsa terakhir. Pertumbuhan tersebut dua kali lebih cepat dibandingkan dengan perbankan konvensional sehingga pangsa pasarnya terus meningkat dalam beberapa tahun terakhir. METODE Metode penulisan pada artikel ini adalah dengan menggunakan deskriptif. Menurut Sugiyono (2011) penelitian desktiptif adalah sebuah penelitian yang bertujuan untuk memberikan atau menjabarkan suatu keadaan atau fenomena yang terjadi saat ini dengan menggunakan prosedur ilmiah untuk menjawab masalah secara aktual. Artikel ini akan mengacu pada sejumlah teori dan data pendukung untuk menjelaskan sasaran dari aspek-aspek kinerja yang dibutuhkan oleh institusi perbankan syariah. Bank syariah memiliki sejumlah tantangan untuk pengembangan sumber daya manusia insani (human capital) yang sesuai dengan tuntutan ekspektasi konsumen. Bisnis syariah harus membuktikan bahwa implementasi bisnis harus sesuai dengan nilai-nilai organisasi. Kinerja organisasi sangat ditentukan dengan sumber daya manusia. Manusia atau anggota organisasi yang bekerja di bank syariah harus mampu memahami target dari bank syariah. Artikel ini akan menggunakan sejumlah data yang mendukung untuk menjelaskan tujuan pembahasan berkaitan dengan manajemen kinerja sumber daya manusia. Data yang digunakan adalah buku, jurnal, laporan perkembangan bank syariah, dan dokumen lain yang mendukung dalam penjelasan secara detail dan komprehensif. Analisa data digunakan secara deskriptif, dikaji secara teoritis, dijelaskan secara terinci, dan dibuat kesimpulan berdasarkan tiga topik masalah yang dikaji dalam artikel ini. PENDAHULUAN Namun saat ini pangsa pasarnya (berdasarkan aset) masih sekitar 4 persen. Institusi perbankan syariah merupakan bank yang beroperasi dengan tidak mengandalkan pada bunga. Bank Islam atau biasa disebut bank tanpa bunga adalah lembaga keuangan perbankan yang operasional produknya dikembangkan berlandaskan pada Al-Quran dan Al-Hadist (Muhammad: 2005). Sebuah organisasi atau institusi membutuhkan sumber daya manusia yang mampu menerjemahkan visi, misi, dan target organisasi. Kinerja organisasi merupakan indikator tingkatan prestasi yang dapat dicapai dan mencerminkan keberhasilan suatu organisasi, serta merupakan hasil yang dicapai dari perilaku anggota organisasi. Mengembangkan Human Resource Management …… (Ahmad Azmy) 81 Kinerja bisa juga dikatakan sebagai sebuah hasil (output) dari suatu proses tertentu yang dilakukan oleh seluruh komponen organisasi terhadap sumber-sumber tertentu yang digunakan (input). Selanjutnya, kinerja juga merupakan hasil dari serangkaian proses kegiatan yang dilakukan untuk mencapai tujuan tertentu organisasi. Bagi suatu organisasi, kinerja merupakan hasil dari kegiatan kerjasama antara anggota atau komponen organisasi dalam rangka mewujudkan tujuan organisasi. Kesuksesan perbankan syariah sangat ditentukan oleh kinerja sumber daya manusia yang ada pada organisasi. Melihat sejumlah masalah yang ada pada perbankan syariah ditinjau dari aspek sumber daya manusia. Artikel ini akan membahas dan mengupas lebih dalam dalam hal: (1) Karakteristik sumber daya manusia syariah yang dibutuhkan oleh institusi perbankan syariah demi menjamin pelayanan dan penerapan bisnis sesuai nilai-nilai syariah. (2) Indikator kinerja (Key Performance Indicator) sumber daya manusia syariah yang dijadikan sebagai patokan dalam manajemen kinerja di perbankan syariah. (3) Strategi pengembangan sumber daya manusia di bank syariah agar mampu memenuhi ekspektasi konsumen dalam mendapatkan edukasi produk syariah dan bisnis syariah secara komprehensif. Manajemen Kinerja Secara etimologi, kinerja berasal dari kata prestasi kerja (performance). Sebagaimana dikemukakan oleh Mangkunegara (2006) bahwa isitilah kinerja dari kata job performance atau actual performance (prestasi kerja atau prestasi sesungguhnya yang dicapai oleh seseorang) yaitu hasil kerja secara kualitas dan kuantitas yang dicapai oleh seorang pegawai dalam melaksanakan tugasnya sesuai dengan tanggung jawab yang diberikan padanya. Lebih lanjut Mangkunegara (2006) menyatakan bahwa pada umumnya kinerja dibedakan menjadi dua, yaitu kinerja individu dan kinerja organisasi. Kinerja individu adalah hasil kerja pegawai baik dari segi kualitas maupun kuantitas berdasarkan standar kerja yang telah ditentukan, sedangkan kinerja organisasi adalah gabungan dari kinerja individu dan kinerja kelompok. BINUS BUSINESS REVIEW Vol. 6 No. 1 Mei 2015: 78-90 82 Pendapat lain yang mengemukakan manajemen kinerja adalah Nawawi (2003) menyatakan bahwa, kinerja adalah hasil pelaksanaan suatu pekerjaan, baik bersifat fisik/ material maupun non fisik/ non material. Menurut Simanjutak (2005), kinerja adalah tingkatan pencapaian hasil atas pelaksanaan tugas tertentu. Simanjuntak juga mengartikan kinerja individu sebagai tingkat pencapaian atau hasil kerja seseorang dari sasaran yang harus dicapai atau tugas yang harus dilaksanakan dalam kurun waktu tertentu. Foster dan Seeker (2001) menyatakan bahwa, “Kinerja adalah hasil yang dicapai seseorang menurut ukuran yang berlaku untuk pekerjaan yang bersangkutan”. Kinerja organisasi mencerminkan bagaimana suatu institusi membentuk kesepahaman kepada semua anggota organisasi dalam mencapai target-target tertentu demi keberlanjutan dalam jangka pendek dan jangka panjang. Bank syariah merupakan sebuah organisasi yang harus menginformasikan sejumlah target kepada pegawai atau karyawan dalam upaya membentuk komitmen dan konsistensi terhadap usaha pencapaian target bisnis. Target-target bisnis tersebut dapat dicapai melalui usaha individu dan kelompok tergantung pekerjaan yang akan dikerjakan dalam waktu tertentu. Kinerja individu akan memberikan gambaran bagaimana pergerakan anggota organisasi menerjemahkan dan memberikan kontribusi positif bagi organisasi. Usaha pencapaian tersebut dapat dimulai dari proses perencanaan, pengorganisasian, implementasi, dan pengawasan secara menyeluruh untuk menjamin proses pencapaian target sesuai yang diharapkan oleh organisasi. Penelitian yang dilakukan oleh Sari, Muis, dan Hamid (2012) menjelaskan bahwa kinerja pegawai di bank syariah dipengaruhi oleh tiga faktor yaitu kepemimpinan, motivasi, dan tekanan pekerjaan. Penelitian ini dilakukan di Bank Negara Indonesia Syariah (BNI Syariah) cabang Makassar. Data dikumpukan melalui kuesioner dengan jumlah responden sebanyak 77 (tujuh puluh tujuh) orang. Hasil penelitian ini menunjukkan bahwa kepemimpinan, motivasi dan stress kerja secara simultan berpengaruh signifikan terhadap kinerja karyawan dengan determinasi sebesar 0,345 atau 34,5%. Kepemimpinan, motivasi dan stress kerja secara parsial berpengaruh signifikan terhadap kinerja karyawan. Variabel yang dominan berpengaruh terhadap kinerja karyawan adalah variabel kepemimpinan. Manajemen Kinerja Penelitian yang dilakukan oleh Markos dan Sandhya (2010) tentang kunci kesuksesan kinerja adalah keterlibatan karyawan (employee engagement). Hasil penelitian menjelaskan bahwa keterlibatan karyawan merupakan faktor kuat dari kinerja yang positif dari sebuah organisasi. Hubungan dua arah antara atasan dan bawahan dibandingkan tiga variabel yang digunakan yaitu kepuasan kerja, komitmen karyawan, dan perilaku anggota organisasi. Keterlibatan karyawan secara emosional melekat pada organisasi dan sangat terlibat pada pekerjaan dengan antusiasme yang besar untuk keberhasilan target dari organisasi. Karyawan akan bekerja ekstra di luar perjanjian kontrak kerja. Keterlibatan karyawan pada bank syariah diperlukan untuk melihat apa yang menjadi target organisasi baik jangka pendek dan jangka panjang. Komunikasi diperlukan untuk menghindari dan mengurangi kesalahpahaman informasi agar terjadi kesepahaman antara level atas dan level bawah. Semua lini organisasi dapat terlibat untuk usaha pencapaian target organisasi. Faktor penunjang lain dalam meningkatkan kinerja pegawai di bank syariah adalah komitmen organisasi. Penelitian yang dilakukan oleh Ansari, Sanaullah, Abbas dan Arsyad (2012) menjelaskan tentang strategi menciptakan nilai organisasi yang superior di bank syariah untuk peningkatan sumber daya manusia insani (human capital). Studi kasus di bank syariah Pakistan. Hasil penelitian menunjukkan bahwa faktor psikologis memainkan peran penting terhadap komitmen organisasi. Faktor lain yang menunjang dalam peningkatan kinerja adalah proses pembelajaran dan motivasi. Hal ini menunjukkan bahwa perbankan syariah yang unggul dapat dibuat di Pakistan dengan mengembangkan sumber daya manusia dalam peran positif dan efektif. Proses pemberian keterampilan yang dibutuhkan oleh industry perbankan syariah. Kemudian penelitian lain yang dilakukan oleh Dwipasari (2007) menjelaskan bahwa kompensasi memberikan efek tak langsung pada kepuasan kerja. Akan tetapi, kompensasi tidak memiliki efek secara langsung pada kinerja pekerjaan. Mengembangkan Human Resource Management …… (Ahmad Azmy) 83 Kedisiplinan memiliki efek langsung terhadap kinerja karyawan sehingga mampu memberikan manfaaat positif kepada organisasi. Jadi dapat disimpulkan bahwa manajemen kinerja memainkan peran yang efektif dalam proses pengembangan sumber daya manusia. Beberapa penelitian menunjukkan bahwa kinerja karyawan dipengaruhi oleh psikologis, kedisiplinan, kepemimpinan, proses pembelajaran, keterlibatan karyawan, dan motivasi. Faktor psikologis membuat seorang karyawan harus nyaman dalam bekerja dan memiliki rasa kepemilikan yang tinggi. Seorang karyawan harus disiplin dalam bekerja sesuai dengan prosedur yang ditetapkan oleh pekerjaan. Kepemimpinan membuat seorang karyawan menjadi bertanggung jawab dan harus mampu memberikan contoh yang baik sesuai dengan ajaran syariah. Organisasi harus membuat proses pembelajaran dengan melibatkan karyawan sehingga terjadi kesepahaman antara organisasi dan karyawan. Organisasi harus selalu memotivasi karyawan untuk selalu memberikan yang terbaik bagi perusahaan dan mengeluarkan kemampuan terbaik dalam mencapai target perusahaan. Manajemen Kinerja Kinerja individu dan organisasi memberikan pengaruh yang cukup luas untuk meningkatkan daya saing organisasi. Faktor-faktor yang mempengaruhi dalam manajemen kinerja yang disebutkan pada beberapa penelitian memberikan gambaran yang jelas bahwa kinerja pegawai bank syariah memberikan peran penting dalam kesuksesan industri bank syariah. Karakteristik Sumber Daya Manusia Syariah Menurut Siamat (2005) kegiatan usaha bank secara umum menuntut adanya profesionalisme yang tinggi guna mendukung proses pengambilan keputusan dan pengendalian resiko usaha sekecil mungkin. Sesuai dengan karakteristik kegiatan usahanya, sumber daya manusia perbankan syariah selain harus mempunyai kemampuan teknis di bidang perbankan juga dituntut untuk memiliki pengetahuan mengenai ketentuan dan prinsip syariah secara baik, serta memiliki akhlak dan moral yang Islami. Akhlak dan moral yang Islami dalam bekerja mempunyai empat ciri pokok yaitu: shiddiq (benar dan jujur), tabligh (mengembangkan lingkungan/bawahan menuju kebaikan), amanah (dapat dipercaya), dan fathonah (komperten dan profesional) keempat ciri pokok tersebut hendaknya dapat menjadi ketentuan umum yang bersifat normatif dalam penetapan kualitas sumber daya manusia baik pimpinan maupun pelaksana pada bank syariah. Secara khusus Bank Indonesia mengatur bahwa pimpinan bank syariah dan pimpinan kantor cabang bank syariah diharuskan memenuhi persyaratan sebagai berikut: (1) Memiliki komitmen dalam menjalankan operasional bank berdasarkan prinsip syariah secara konsisten. (2) Memiliki integritas dan moral yang baik. (3) Mempunyai pengalaman operasional perbankan syariah atau telah mendapatkan pendidikan atau pelatihan perbankan syariah baik di dalam maupun di luar negeri. Oleh karena itu, bank syariah memerlukan kepercayaan masyarakat bahwa dalam pelaksanaan kegiatan usahanya tidak menyimpang dari ketentuan dan prinsip syariah serta mempertimbangkan aspek sosio-kultural masyarakat muslim Indonesia, maka sebaiknya dalam tahap awal pengangkatan pimpinan unit usaha syariah dan pimpinan kantor cabang syariah beragama Islam. Key Performance Indicator (KPI) SDM Syariah Indikator kinerja merupakan landasan sebuah organisasi untuk melakukan evaluasi secara menyeluruh berkaitan dengan peningkatan atau penurunan bisnis. Kinerja individu yang dilakukan oleh seorang karyawan harus sesuai dengan strategi pencapaian target organisasi. Sebuah penelitian yang dilakukan oleh Sumantri (2014) tentang Pengembangan Kapasitas Institusi Perbankan Syariah Dalam Penyediaan Infrastruktur Jaringan, SDM, dan Produk. Salah satu pembahasan adalah membahas mengenai indikator kinerja sumber daya manusia di bank syariah. Indikator kinerja (Key Performance Indicator) yang harus dilakukan adalah melakukan integrasi dengan pelayanan berbasis syariah dan diharapkan dapat meningkatkan pertumbuhan perbankan syariah. Untuk selanjutnya SDM BINUS BUSINESS REVIEW Vol. 6 No. 1 Mei 2015: 78-90 84 Perbankan Syariah diharapkan memiliki kemampuan yang baik yang berdasarkan Kompetensi Multi Disiplin Ilmu dan Multi Dimensi sebagai berikut: (1) Kontrak-kontrak Muamalah dalam bisnis sesuai Undang-Undang Syariah/Nilai-nilai islam. (2) Pengetahuan di Produk-Produk Perbankan dan Produk- Produk Perbankan Syariah. (3) Keahlian Investasi Managemen Kekayaan. (4) Keahlian dalam Struktur Keuangan Perbankan dan Produk-Produk Finansial yang lain. (5) Memiliki jejaringan sosial ekonomi yang kuat dan kemampuan membangun jaringan baru. (6) Keahlian dalam berkomunikasi. Enam keahlian yang harus dimiliki sumber daya manusia di bank syariah harus dimiliki oleh setiap karyawan. Seorang pegawai harus mampu memahami kontrak muamalat dalam bisnis syariah. Indikator kinerja di atas bisa dijadikan rujukan oleh institusi perbankan syariah untuk melakukan peningkatan keahlian sumber daya manusia sehingga dapat dipersiapkan dalam proses pelayanan produk syariah. Semua anggota organisasi yang bekerja di bank syariah harus memiliki semua keahlian tersebut demi menjamin pengembangan bisnis syariah sesuai dengan ekspektasi konsumen. Penelitian yang dilakukan oleh Muda et. al. (2014) mengenai faktor-faktor kinerja karyawan di bank syariah menggunakan tiga variabel yaitu tekanan pekerjaan (job stress), motivasi (motivation), dan komunikasi (communication). Hasil penelitian menunjukkan bahwa 59.3 % ketiga faktor tersebut dapat mempengaruhi kinerja karyawan di bank syariah dan sisanya 40.7% dapat dijelaskan oleh faktor lain. Uji F simultan menjelaskan bahwa secara komprehensif tekanan pekerjaan (job stress), motivasi (motivation), dan komunikasi (communication) dapat mempengaruhi kinerja karyawan. Uji T parsial menjelaskan bahwa tekanan pekerjaan (job stress) dan motivasi (motivation) tidak mempengaruhi kinerja karyawa, sedangkan komunikasi (communication) mempengaruhi kinerja karyawan. Kemampuan komunikasi berkaitan bagaimana karyawan dapat menjelaskan dengan baik produk-produk syariah dan membina hubungan baik dengan nasabah dalam jangka panjang. Jadi dapat disimpulkan kemampuan komunikasi merupakan salah satu kemampuan yang harus dimiliki oleh karyawan sebagai standar kinerja di bank syariah. Sebuah organisasi harus memiliki indikator kinerja sebagai panduan keputusan untuk pengembangan karir karyawan. Key Performance Indicator (KPI) SDM Syariah Indikator kinerja atau Key Performance Indicator (KPI) merupakan standar kinerja yang dijadikan oleh organisasi dan karyawan untuk terus melakukan evaluasi dalam pengembangan sumber daya manusia. Kebijakan promosi, demosi, transfer jabatan, dan rotasi harus berdasarkan kinerja yang telah dilakukan oleh karyawan. Standar kinerja harus dirumuskan secara baik dan bertujuan untuk meningkatkan produktifitas secara berkelanjutan dan pencapaian target organisasi. Strategi Pengembangan SDM Syariah Peningkatan bisnis perbankan syariah membutuhkan sumber daya manusia yang handal dan memiliki beberapa indikator kinerja dalam mendukung pengembangan organisasi. Karakteristik sumber daya manusia syariah dan enam indikator kinerja memberikan penjelasan yang jelas bagaimana institusi syariah membutuhkan sumber daya manusia yang memiliki beberapa keahlian dalam menunjang bisnis jangka panjang. Menurut Heri (2013) perguruan tinggi memiliki peran strategis dalam penyediaan sumber daya manusia di bank syariah. Data menunjukkan bahwa ada sebanyak 850 institusi pendidikan menyelenggarakan program studi perbankan syariah untuk mempersiapkan kebutuhan sdm dalam menjamin pelayanan dan peningkatan bisnis perbankan syariah. Jumlah lulusan yang dihasilkan sebanyak 404.198 alumni yang siap berkarir di bank syariah. Bank Indonesia memiliki panduan rencana untuk mengembangkan sumber daya manusia di bank syariah. Ada beberapa sasaran target yang dijadikan untuk peningkatan sumber daya manusia antara lain adalah pertama, sumber daya manusia berkualitas tinggi di bisnis syariah harus dilakukan sebagai upaya peningkatan bisnis bank syariah. Ini dimaksudkan bahwa bank syariah harus memiliki sumber daya manusia yang mumpuni dengan bekerja sama pada lembaga pendidikan atau pelatihan sehingga karyawan dapat memberikan kontribusi positif kepada organisasi. Beberapa institusi Mengembangkan Human Resource Management …… (Ahmad Azmy) 85 perbankan syariah sudah melakukan kerja sama dengan lembaga pendidikan seperti STIE TAZKIA, UIN Syarif Hidayatullah Jakarta, Universitas Indonesia, dan berbagai macam perguruan tinggi yang sudah memiliki program studi perbankan syariah untuk mengisi posisi-posisi jabatan yang ada di bank syariah. Lembaga pendidikan dituntut untuk membuat standar pendidikan yang sesuai dengan kebutuhan bank syariah meliputi pengetahuan, skill, dan perilaku (Knowledge, Skill, Attitude) sesuai dengan kebutuhan bank syariah. Ini bertujuan untuk membangun konsep Link and Match dalam pengembangan sumber daya manusia yang memiliki daya saing tinggi sebagai upaya peningkatan bisnis bank syariah. Kedua, regulasi dan supervisi yang efektif bagi bank syariah. Bank Indonesia sebagai regulator harus mampu mengakomodasi kebutuhan aturan pengembangan bisnis syariah. Supervisi harus dilakukan untuk menjamin nilai-nilai bisnis syariah dapat diterapkan pada semua bank syariah melalu dewan pengawas yang bisa ditunjuk langsung oleh pemerintah. Ini sudah dilakukan oleh Bank Indonesia dengan menerbitkan aturan Nomor: 12/ 21 /PBI/2010 bahwa setiap bank harus memiliki rencana bisnis dan memetakan tahapan implementasi baik jangka pendek, menengah, dan jangka panjang. Kemudian langkah ini dilanjutkan dengan menerbitkan Kodifikasi Peraturan Bank Indonesia Kelembagaan Rencana Bisnis Tahun 2012 yang mengatur beberapa lampiran laporan rencana bisnis untuk mempermudah pengawasan dan penerapan regulasi secara konsisten. Ginting dkk (2012) menyusun beberapa kodifikasi laporan kelembagaan mengenai rencana bisnis bank dengan beberapa item yang harus dipenuhi oleh perbankan. Strategi Pengembangan SDM Syariah Keenam, kepatuhan pada prinsip syariah yang tinggi. Bank Indonesia sebagai regulator harus membuat aturan yang tegas dalam penerapan prinsip-prinsip bisnis syariah. Asas kepatuhan harus dijalankan oleh bank syariah demi menjamin kepercayaan masyarakat sebagai nasabah dalam menyimpan uang dan investasi atas dasar syariah. Laporan Tahunan 2013 yang dimiliki oleh Bank Syariah Mandiri sudah mampu menerapkan standar kesehatan perbankan nasional. Non Performing Financing (NPF) tidak lebih dari 5 % hanya berkisar 4.32 %. Capital Adequacy Ratio (CAR) merupakan rasio kecukupan modal berjalan dengan baik sebesar 14.10% dan beberapa rasio keuangan yang lain bisa menunjukkan kinerja bank cukup stabil. Ini menandakan bahwa prinsip kepatuhan dalam melaksanakan aturan bank syariah yang ditetapkan Bank Indonesia sudah cukup baik dan dapat ditingkatkan seiring kinerja bisnis dapat mencapai target pencapaian yang ditetapkan oleh organisasi. Aliansi strategis yang sinergis. Pemerintah dan bank syariah harus menjadi mitra kerja yang proporsional demi kontribusi terhadap ekonomi nasional. Walaupun kontribusi bank syariah masih dibawah 5%. Ini bukan merupakan hambatan dalam melakukan sebuah terobosan baru dalam bisnis syariah dala membentuk sebuah aliansi bisnis dan sinergitas yang positif. Bank Indonesia (BI) sebagai regulator mampu mengakomodasi kebutuhan aturan-aturan untuk pengembangan produk syariah. Bank syariah harus mampu menerapkan aturan yang sudah ditetapkan oleh Bank Indonesia dengan menjaga kinerja bisnis, meningkatkan kesehatan perbankan, dan edukasi produk syariah kepada nasabah. Ini dilakukan untuk memunculkan kepercayaan masyarakat dalam menggunakan bank syariah sebagai alternatif keuangan sehingga pangsa pasar menjadi tinggi dan sustainabilitas industri perbankan syariah dalam jangka panjang. Beberapa strategi pengembangan sdm syariah yang bisa dilakukan adalah pertama, perguruan tinggi harus menyediakan program pendidikan yang dimulai dari Strata Satu sampai Strata 3. Ini mutlak dilakukan untuk menyediakan sumber daya manusia bukan hanya untuk entry level, tetapi sampai level manajemen puncak sehingga bank syariah tidak kesulitan untuk mencari sumber daya manusia insani sesuai dengan karakteristik bisnis bank syariah. Peningkatan kualitas pendidikan perbankan syariah sudah dilakukan oleh beberapa perguruan tinggi baik swasta maupun negeri. Penelitian yang dilakukan oleh Munthe (2012) melakukan pemetaan antara supply dan demand kebutuhan sumber daya manusia bank syariah. Hasil penelitian ini menunjukkan bahwa kebutuhan sumber daya manusia bank syariah untuk melakukan akselerasi berjumlah 179.646 orang pegawai dari 37.356 orang pada akhir tahun 2012. Jumlah tersebut akan terdiri dari 165.274 orang pegawai kategori low Syariah quality, dan 14.372 orang kategori middle to high syariah quality. Supply gap SDM syariah kategori low syariah quality akan terjadi sampai dengan tahun 2016 dan kategori middle to high syariah quality, hingga tahun 2020. Strategi Pengembangan SDM Syariah Salah satu output kodifikasi adalah laporan kondisi dan rencana kebutuhan sumber daya manusia serta rencana pendidikan dan pelatihan sumber daya manusia. Ini dilakukan untuk meningkatkan daya saing sumber daya manusia dalam peningkatan kemampuan dan menghasilkan inovasi bisnis sehingga terjadi kecocokan antara rencana bisnis dengan strategi pemenuhan sumber daya manusia di bank syariah. Ketiga, struktur perbankan yang efektif dalam melakukan bisnis syariah. Struktur organisasi mencerminkan bagaimana tata laksana bisnis dan budaya organisasi. Bank syariah harus mampu membuat struktur organisasi yang efektif sehingga prosedur bisnis dan peningkatan pengetahuan dapat menjamin bahwa karyawan memahami target organisasi. Beberapa bank syariah sudah melakukan pembuatan struktur organisasi sesuai dengan aturan yang ditetapkan Bank Indonesia. Bank Syariah Mandiri sudah menerapkan struktur organisasi yang efektif untuk beradaptasi pada trend bisnis keuangan syariah. Nilai-nilai organisasi yang diterapkan adalah Excellent, Teamwork, Humanity, Integrity, dan Customer Focus. Bank Syariah Mandiri berusaha untuk memberikan pelayanan terbaik bagi nasabah dan kepuasan menjadi jaminan sehingga daya saing perusahaan memiliki keunggulan kompetitif. Keempat, infrastruktur yang mendukung bisnis syariah. Bank Indonesia memiliki rencana untuk membangun system yang terintegrasi dengan semua bank syariah. Ini dilakukan untuk mendukung upaya kontribusi bank syariah terhadap perekonomian nasional. Bank syariah harus mampu mengakomodasi infrastruktur yang direncanakan oleh pemerintah sehingga dapat memberikan efek positif pada bank syariah. Bank Indonesia memiliki beberapa grand strategi untuk membangun infrastruktur perbankan syariah. Bank Indonesia menerapkan dual-banking system dalam bank syariah untuk menghadirkan alternatif jasa perbankan yang semakin lengkap kepada masyarakat Indonesia. Secara bersama-sama, sistem perbankan syariah dan perbankan konvensional secara sinergis mendukung mobilisasi dana masyarakat secara lebih luas untuk meningkatkan kemampuan pembiayaan bagi sektor-sektor perekonomian nasional. Contoh Bank Central Asia (BCA) membuka BCA Syariah sebagai implementasi dual-banking system dan Bank Rakyat Indonesia (BRI) membuka BRI Syariah sebagai perwujudan dukungan peningkatan infrastruktur bisnis perbankan syariah. Kelima, pemberdayaan nasabah yang efektif. Nasabah merupakan objek konsumen yang harus dilayani dengan baik. Bank syariah harus mampu membuktikan bahwa bisnis yang dijalankan sesuai dengan prinsip syariah. Edukasi produk syariah harus dilakukan kepada nasabah sehingga mengetahui secara lengkap dan komprehensif tentang produk syariah. Ini sudah dilakukan oleh Bank Syariah BINUS BUSINESS REVIEW Vol. 6 No. 1 Mei 2015: 78-90 86 Mandiri dalam melakukan edukasi bisnis bank syariah dalam websitenya. Bank Syariah Mandiri melakukan pembuatan artikel ilmiah dan memperkenalkan konsep Islamic Wealth Management kepada masyarakat dengan bertujuan untuk menjelaskan konsep produk dan manfaat yang didapat nasabah. Artikel-artikel tersebut menggunakan bahasa yang mudah dipahami sehingga masyarakat memahami bisnis bank syariah dan edukasi berjalan efektif. Strategi Pengembangan SDM Syariah Ini berarti bahwa peran perguruan tinggi sangat vital dalam penyediaan sumber daya manusia. Sinergitas antara bank syariah sebagai pemberi kerja dan perguruan tinggi sebagai penyedia tenaga kerja harus memiliki komitmen bersama untuk mewujudkan akselerasi perbankan syariah pada tahun 2020 yang sudah ditetapkan oleh Bank Indonesia dimana tingkat pertumbuhan bank syariah harus diatas 5%. Kedua, bank syariah harus melakukan kerjasama dengan perguruan tinggi untuk rekrutmen dan seleksi karyawan. Efek positif dari kerjasama ini bahwa bank syariah bisa melakukan efisiensi pelatihan dan pengembangan sumber daya manusia. Ini disebabkan selama proses pendidikan calon pegawai sudah mengetahui target organisasi dan kebutuhan skill sesuai dengan industri syariah. Bank Syariah Mandiri sudah memiliki sistem Early Development Program (ERP) dimana perusahaan menjalin kerjasama dengan 38 perguruan tinggi baik negeri dan swasta dalam proses rekrutmen pegawai. Bank Syariah Mandiri tidak mau salah dalam merekrut kandiat pekerja jika tidak sesuai Mengembangkan Human Resource Management …… (Ahmad Azmy) 87 nilai-nilai perbankan syariah. Implementasi sudah dilakukan dalam jangka panjang dan sustainabilitas organisasi bisa terjaga dengan baik. Ketiga, pengembangan sumber daya manusia dilakukan secara berkelanjutan.Bank Syariah Mandiri menggunakan sistem pengembangan bakat (Talent Management). Sistem ini digunakan untuk memantau dan menyeleksi bakat-bakat terbaik dari pegawai bank syriah untuk diberikan program pengembangan kenaikan jabatan yang dibutuhkan oleh organisasi. Bank Syariah Mandiri memiliki kepercayaan jika sumber daya manusia dikembangkan secara baik dan benar serta penempatan bakat yang tepat, maka dapat menghasilkan calon-calon pemimpin yang berkualitas. Beberapa program yang sudah dijalankan seperti Officer Development Program (ODP), Middle Management Development Program (MMDP), dan Senior Management Development Program (SMDP). Ketiga program inilah yang dijadikan andalan bagi Bank Syariah Mandiri (BSM) dalam mendapatkan sumber daya manusia yang terbaik dan menerapkan sistem kompensasi yang kompetitif sebagai upaya mewujudkan retensi pegawai yang baik. Keempat, pendidikan dan pelatihan yang diberikan hanya sesuai dengan kebutuhan. Ini dilakukan bahwa selama proses perkuliahan sudah diberikan materi dasar bisnis syariah. Bank Syariah Mandiri memiliki beberapa program pendidikan dan pelatihan bagi sumber daya manusia yaitu Banking Staff Program (BSP), Banking Academy, dan Enhancement Program. Ketiga program ini dijadikan sebagai bekal para pegawai dengan berbagai macam kemampuan dan kompetensi yang sesuai dengan bisnis perbankan syariah. Ini dilakukan proses pendidikan dan pelatihan terfokus pada skill dan kemampuan yang dibutuhkan sehingga dapat menghasilkan output sesuai harapan organisasi. Kelima, bank syariah dengan lembaga pendidikan harus melakukan diskusi tentang sinergitas kurikulum yang akan diberikan kepada mahasiswa. Strategi Pengembangan SDM Syariah Ini harus dilakukan untuk mempersiapkan calon mahasiswa untuk berkarir di bank syariah sesuai dengan kebutuhan bank syariah dalam melayani ekspektasi konsumen. Bank syariah dengan perguruan tinggi mengadakan seminar atau workshop dengan mengundang para praktisi dan peneliti bank syariah. Pokok pembahasan mengenai isu-isu terkini yang menjadi topik hangat pada industri bank syariah. Perguruan tinggi baik swasta dan negeri sering melakukan diskusi ilmiah dalam pembahasan penelitian dan artikel sehingga dapat meningkatkan wawasan serta pengetahuan tentang industri perbankan syariah. Mahasiswa menjadi peserta untuk melihat dan menganalisis secara kritis apa yang harus dilakukan untuk membangun industri perbankan syariah di Indonesia menjadi lebih baik, kompetitif, dan memiliki daya saing yang tinggi. SIMPULAN Bisnis perbankan syariah memiliki keunikan tersendiri. Peningkatan bisnis syariah di Indonesia membuktikan bahwa nasabah memerlukan pelayanan berbasis syariah. Ini harus diimbangi dengan ketersediaan sumber daya manusia. Karakteristik sumber daya manusia yang dibutuhkan oleh bank syariah berbeda dengan bank konvensional. Sumber daya manusia di bank syariah harus memiliki indikator kinerja yang berbeda dengan bank konvensional. Jadi, bank syariah memiliki ekspektasi yang tinggi dalam mengembangkan sumber daya manusia sehingga membuat keunggulan kompetitif dibandingkan mengambil dari bank konvensional. Bank syariah harus memiliki strategi pengembangan jangka panjang dalam pengembangan sumber daya manusia. Karakteristik sumber daya manusia yaitu shiddiq (benar dan jujur), tabligh (mengembangkan lingkungan/bawahan menuju kebaikan), amanah (dapat dipercaya), dan fathonah (kompeten dan profesional). Keempat karakteristik ini harus menjadi pilar dalam membentuk sumber daya manusia berbasis syariah. Strategi pengembangan ini harus dihubunkan dengan indikator kinerja yang dijadikan sebagai sarana pengukuran pengembangan karir sehingga semua karyawan dapat meningkatkan kinerja dan kontribusi positif kepada organisasi. BINUS BUSINESS REVIEW Vol. 6 No. 1 Mei 2015: 78-90 88 Prospek bisnis syariah sangat menjanjikan di Indonesia. Kesuksesan organisasi tergantung pada sumber daya manusia dalam mengadopsi dan menerjemahkan bisnis utama dalam perbankan syariah. Indikator kinerja memberikan gambaran yang jelas target organisasi baik jangka pendek dan jangka panjang. Manajemen kinerja harus diberikan standar kinerja sehingga karyawan dapat memberikan ide-ide baru dalam bisnis dan pelayanan terbaik kepada nasabah. Tantangan ke depan bank syariah harus menjawan bahwa penerapan bisnis sesuai dengan prinsip-prinsip syariah sehingga mampu memberikan kenyaman kepada nasabah dalam menyimpan uangnya di bank syariah. DAFTAR PUSTAKA Alamsyah, H. (2012). Perkembangan dan Prospek Perbankan Syariah di Indonesia: Tantangan Dalam Menyongsong MEA 2015. Ceramah Ilmiah Ikatan Ahli Ekonomi Islam (IAEI). Ansari, S., Abbas, M. A. (2012). Creating the Superior Islamic Banking Through Improving Quality of Human Resources in Pakistan. Proceedings of the 8th European Conference on Management, Leadership and Governance. Academic Conferences Limited. Bank Indonesia. (2010). Peraturan Nomor: 12/ 21 /PBI/2010 Tentang Rencana Bisnis Bank. Ja Bank Syariah Mandiri. (2013). Stronger Fundamental for Greater Indonesia. Laporan Tahunan Dahlan, S. (2005). Manajemen Lembaga Keuangan. Kebijakan Moneter dan Perbankan. (edisi kesatu) Jakarta: Fakultas Ekonomi Universitas Indonesia. Dwipasari, L. (2007). Kompensasi dan Kedisiplinan Sebagai Faktor Yang Berpengaruh Terhadap Kinerja dan Kepuasan Kerja Karyawan Bank. Jurnal Keuangan dan Perbankan, 12(3): 494. Foster, B., Seeker, K. R. (2001). Pembinaan Untuk Meningkatkan Kinerja Karyawan. Penerjemah. Jakarta: PPM. Ginting, R., Iskandar, D., Wuryandani, G., Hutabarat, C., Rosdiana, R. (2012). Kodifikasi Peraturan Bank Indonesia Kelembagaan Rencana Bisnis Bank. Pusat Riset dan Edukasi Bank Sentral. Bank Indonesia. Heri, P. (2013). Peran Strategis Perguruan Tinggi dalam Meningkatkan Kualitas Sumber Daya Manusia Ekonomi Syariah. KARYA DOSEN. Fakultas Ekonomi Universitas Negeri Malang. Mangkunegara, A. P. (2006). Evaluasi Kinerja Sumber Daya Manusia. Bandung: PT. Repika Utama. Markos, S., Sandhya, S. M. (2010). Employee Engagement: The Key to Improving Performance. International Journal of Business and Management, 5(12). Muhammad. (2005). Manajemen Bank Syariah. UPP AMP YKPN, Yogyakarta. Muda, I., Rafiki, A., Harahap, M. R. (2014). Factors Influencing Employees’ Performance: A Study on the Islamic Banks in Indonesia. International Journal of Business and Social Sciences, 5 (2):73-80. Munthe, G. J. (2012). Proyeksi Kebutuhan Sumber Daya Manusia (SDM) Perbankan Syariah Dan Skenario Pemenuhannya.Tesis. Fakultas Ekonomi Program Magister Manajemen Universitas Indonesia. Mengembangkan Human Resource Management …… (Ahmad Azmy) 89 Mengembangkan Human Resource Management …… (Ahmad Azmy) 89 Nawawi, H. (2003). Kepemimpinan Mengefektifkan Organisasi. Yogyakarta: Gadjah Mada University Press. Permana, K. A. (2012). Tiga Masalah Terbesar di Bank Syariah. Diakses 10 Nivember 2014 dari http://forpiko.com/berita-192-tiga-masalah-terbesar-di-bank-syariah.html Rademakers, M. (2005), Corporate universities: driving force of knowledge innovation. Journal of Workplace Learning, 17: 130. Sari, R., Muis, M., Hamid, N. (2012). Pengaruh Kepemimpinan, Motivasi, dan Stress Kerja Terhadap Kinerja Karyawan Pada Bank Syariah Mandiri Kantor Cabang Makassar. Jurnal Analisis, 1(1): 87-93. Sugiyono. (2011). Metode Penelitian Bisnis dan R & D. Bandung: Alfabeta Simanjuntak, P. J. (2005). Manajemen dan Evaluasi Kinerja. Jakarta: FE UI. Sumantri, B. A. (2014). Pengembangan Kapasitas Institusi Perbankan Syariah Dalam Penyediaan Jaringan, SDM, dan Produk. Jurnal Eksyar. DAFTAR PUSTAKA 01(01): 1-17. 90 BINUS BUSINESS REVIEW Vol. 6 No. 1 Mei 2015: 78-90 90
https://openalex.org/W2087831282	https://openresearch-repository.anu.edu.au/bitstream/1885/16902/1/01_Michel_Correlation_of_Clinical_2011.pdf	English	null	Correlation of Clinical Trachoma and Infection in Aboriginal Communities	PLoS neglected tropical diseases	2,011	cc-by	9,375	Correlation of Clinical Trachoma and Infection in Aboriginal Communities 1 Diagnostics Development Unit, Department of Haematology, University of Cambridge, National Health Service Blood and Transplant Site, Cambridge, United Kingdom, 2 Master of Applied Epidemiology Program, National Centre for Epidemiology and Population Health, College of Medicine and Health Sciences, Australian National University, Canberra, Australia, 3 Indigenous Eye Health Unit, Melbourne School of Population Health, University of Melbourne, Melbourne, Australia, 4 Vision CRC, Kensington, Australia Abstract This is an open-access article distributed under the terms of the Creative Commons Attributi unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Funding: The study was funded by the Wellcome Trust and the NIHR Cambridge Biomedical Research Centre to the University of Cambridge and by Christian Blind Mission International, Bennelong Foundation, Fred Hollows Foundation and Myer Foundation to the Centre for Eye Research Australia. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. MAD and CEM are equity holders of a spin-off company, Diagnostics for the Real World Ltd, based on rapid test technologies f Cambridge. Both the University of Cambridge and the Wellcome Trust are also equity holders of the company. mpeting Interests: HHL, MAD and CEM are equity holders of a spin-off company, Diagnostics for the Real World Ltd, based on r veloped at the University of Cambridge. Both the University of Cambridge and the Wellcome Trust are also equity holders of the com * E-mail: h.taylor@unimelb.edu.au response that resolves much more slowly than the infection [4,5,9,10,11,12,13]. It is further compounded by the occurrence of repeated episodes of infection. Also important is the relative lack of precision in assessing clinical status with the WHO simplified trachoma grading system [14,15], which was designed to be learnt and used by local health workers and generally has a high level of reproducibility [16]. Abstract Background: Trachoma is the leading infectious cause of blindness due to conjunctival infection with Chlamydia trachomatis. The presence of active trachoma and evidence of infection are poorly correlated and a strong immunologically- mediated inflammatory response means that clinical signs last much longer than infection. This population-based study in five Aboriginal communities endemic for trachoma in northern Australia compared a fine grading of clinical trachoma with diagnostic positivity and organism load. Methods: A consensus fine grading of trachoma, based on clinical assessment and photograding, was compared to PCR, a lipopolysacharide (LPS)-based point-of-care (POC) and a 16S RNA-based nucleic acid amplification test (NAAT). Organism load was measured in PCR positive samples. Results: A total of 1282 residents, or 85.2% of the study population, was examined. Taking the findings of both eyes, the prevalence of trachomatous inflammation-follicular (TF) in children aged 1–9 years was 25.1% (96/383) of whom 13 (13.7%) were PCR positive on the left eye. When clinical data were limited to the left eye as this was tested for PCR, the prevalence of TF decreased to 21.4% (82/383). The 301 TF negative children, 13 (4.3%) were PCR positive. The fine grading of active trachoma strongly correlated with organism load and disease severity (rs = 0.498, P = 0.0004). Overall, 53% of clinical activity (TF1 or TF2) and 59% of PCR positivity was found in those with disease scores less than the WHO simplified grade of TF. Conclusion: Detailed studies of the pathogenesis, distribution and natural history of trachoma should use finer grading schemes for the more precise identification of clinical status. In low prevalence areas, the LPS-based POC test lacks the sensitivity to detect active ocular infection and nucleic acid amplification tests such as PCR or the 16S-RNA based NAAT performed better. Trachoma in the Aboriginal communities requires specific control measures. Citation: Michel C-EC, Roper KG, Divena MA, Lee HH, Taylor HR (2011) Correlation of Clinical Trachoma and Infection in Aboriginal Communities. PLoS Negl Trop Dis 5(3): e986. doi:10.1371/journal.pntd.0000986 Editor: Jeremiah M. Ngondi, University of Cambridge, United Kingdom Received April 19, 2010; Accepted February 14, 2011; Published March 15, 2011 eceived April 19, 2010; Accepted February 14, 2011; Published hel 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: 2011 Michel et al. www.plosntds.org www.plosntds.org Author Summary Repeated episodes of C. trachomatis infection lead to active trachoma clinically characterised by an often intense inflammatory response to chlamydial antigens with later scarring and distortion of the eyelid leading to blindness. However, the clinical signs of trachoma do not correlate well with laboratory tests to detect the presence of Chlamydia. The WHO simplified clinical grading scheme currently used for assessment of trachoma has a poor correlation with C. trachomatis genomic test findings, even though the detection of bacterial genome is strongly correlated with the prevalence and severity of active trachoma. A detailed assessment of the clinical signs using a finer grading system was studied in a population-based survey in five Australian Aboriginal communities. Much clinical activity and infection was found in those with clinical signs below the threshold used in the current WHO grading scheme. Future studies of the distribution of infection and pathogenesis should use finer grading methods than the current WHO scheme. The prevalence of trachoma in these communities confirms that trachoma remains of public health impor- tance and sustained interventions to control trachoma are warranted. During the clinical assessment, both examiners wore two pairs of gloves. To limit the risk of cross-contamination between specimens and to prevent the exposure to transmittable diseases to consecutive subjects, utensils and surfaces [5], the outer pair was removed between successive participants and the inner pair disinfected and regularly disposed of. The examiner graded the clinical signs of trachoma using a fine grading scheme (Table 1) [15,17]. Digital photographs were taken of the left inverted upper lid that were subsequently graded independently using the fine grading scheme and any discrepancies were adjudicated to give the final consensus grading [15,17]. The clinical grading was also expressed in terms of the WHO simplified grading system as ‘‘Trachomatous inflammation-follicular’’ (referred to as TFWHO), ‘‘Trachomatous inflammation-intense’’ (TIWHO) and ‘‘Trachoma- tous scarring’’ (TSWHO) [14]. as the reference test. Organism load was quantified with real-time quantitative PCR (qPCR) in CT positive individuals [20,21]. Methods After the clinical assessment and photography, two ocular swabs from the left eye were collected consecutively under stringent conditions to limit cross-contamination [5], and rigorous photo- graphic cataloguing and sample labelling systems. Over 95% of Correlation of Clinical Trachoma and Infection Workers from five Aboriginal communities with endemic trachoma in the Katherine region of the Northern Territory Australia during a five week period over July and August 2007 as described previously [15,17]. These communities had not received any recent azithromycin or other mass antibiotic treatment, although local health services do prescribe and dispense a range of broad-spectrum antibiotics on an individual patient basis that might indirectly have had an impact on trachoma and our findings. However, without access to confidential patient history, the recipients of these antibiotic prescriptions were not identifiable. In principle, school children in remote communities receive one annual health check that includes trachoma screening, but this process is patchy at best. Despite recommendations, very little screening for trichiasis in elders has been conducted in area where trachoma is endemic [15]. Author Summary Introduction Trachoma is the leading infectious cause of blindness [1,2,3], and results from repeated episodes of conjunctival infection by Chlamydia trachomatis (CT) serovars A, B, Ba and C. It is a major public health problem associated with poverty in environments with inadequate sanitation, poor personal hygiene and poor water supply and is now largely confined to developing countries, particularly in Sub- Saharan Africa [2,4,5,6]. Nucleic acid amplification tests (NAATs) require appropriate facilities and skilled staff, but a assay designed for use in resource-limited settings may offer some advantages for the diagnosis of infection over clinical assessment [7,8]. In general, irrespective of the diagnostic methodology, there is a relatively poor correlation between clinically active trachoma and biological evidence of infection, in part because signs of the disease are induced by a strong immunologically-mediated inflammatory We sought to compare a fine consensus grading of trachoma combining clinical and photographic grading [15,17] with a commercially available polymerase chain reaction (PCR), the CT/ NG Amplicor test (Roche Diagnostic Corporation, IN, USA). We sought to compare field performance of a previously described POC assay [7] and a sensitive in-house 16S-RNA NAAT using an improved visual detection of nucleic acid by dipstick [18,19] using the CT/NG Amplicor assay targeting one sequence coding for ORF1 (Open Reading Frame 1) of the Chlamydia cryptic plasmid March 2011 \| Volume 5 \| Issue 3 \| e986 1 March 2011 \| Volume 5 \| Issue 3 \| e986 Correlation of Clinical Trachoma and Infection Correlation of Clinical Trachoma and Infection The anti-biotin monoclonal antibodies (clone BII-10A12A9A1, Diagnostic Development Unit, University of Cambridge, Cam- bridge, UK) conjugated to colloidal gold (British Biocell International, Cardiff, UK) by passive adsorption specifically bind to the lyophilised signal amplification reagents, consisting of a biotinylated monoclonal antibody to chlamydial LPS detection antibody (clone CTIII-10B9A10A4D28, Diagnostic Development Unit) biotinylated with the BAC-Sulfo-NHS-LC-biotin reagent (Sigma, St Louis, MO, USA) at a ratio of nine biotins per antibody molecule. samples were collected by the same swabber throughout the study to minimise any sampling variability. Additionally, gloves, surfaces, loupes, the camera and other utensils were swabbed twice a day to detect possible cross-contamination. We obtained approval for the study from the Human Research Ethics Committees of the Royal Victorian Ear and Eye Hospital, the Australian National University, the Northern Australian National University and the Northern Territory Government Department of Health & Communities Services and Menzies School of Health Research. Signed written consent was obtained from each person, with consent for children under 18 years of age being provided by a parent or guardian [15,17]. For LPS-POC testing, ocular swabs were placed in the sample extraction tube with a tapered bottom to facilitate extraction of the swab and a cap that allows it to also function as a dropper. The lysis reagent and analyte stabiliser were added sequentially as previously described [7]. Briefly, the lysis reagent (400 mL, Diagnostic Development Unit) and analyte stabiliser (300 mL, Diagnostic Development Unit) were added sequentially and mixed by gently dipping the swab to the bottom of the extraction tube three times after addition of each reagent. Two hundred microlitres of the above extract were immediately transferred to 800 mL of pre-dispensed Amplicor sample dilution buffer (Roche) for PCR testing. Thereafter, the signal enhancer reagent (33 mL, Diagnostic Development Unit) was added to each extract. This allows the release of chlamydial-LPS for detection. Five drops of the resulting extract (100 mL) were transferred to the detection tube into which the dipstick is placed. Two hundred microlitres of the above extract were immediately transferred to 800 mL of pre- dispensed Amplicor sample dilution buffer (Roche) for PCR testing. Thereafter, the signal enhancer reagent was added to each extract. This allows the release of chlamydial-LPS for detection. Five drops of the resulting extract were transferred to the detection tube into which the dipstick is placed. Laboratory assays The POC test was performed on site using the first left-eye ocular swab collected by only one experienced technician throughout the study. The assay detects chlamydial lipopolysac- charide (LPS) as previously described [7] with the following modifications for field use: 1) an alternative nitrocellulose membrane was used as the manufacturer discontinued the membrane previously used, 2) the ratio of lyophilised signal amplification system was modified for the test to function at high ambient temperature and 3) increased length of the conjugate tube which houses the dipstick to minimise the evaporation of the reagents during wicking and to protect the membrane against dust. In addition to the lyophilised signal amplification reagents consisting of a biotin-labelled monoclonal antibody to chlamydial LPS and an anti-biotin monoclonal antibody conjugated to colloidal gold particles as colour indicator [7], the nitrocellulose- based membranes are the heart of lateral or vertical flow assays. The wicking rate, pore size, residual surfactants and detergents present on the matrix affect the characteristics of nitrocellulose- based membranes and reaction kinetics. Therefore, changing this porous substrate matrix and the addition of some features (i.e. shape of the conjugate tube) require a systematic adjustment of ratio of the lyophilized signal amplification reagents. The POC test was performed on site using the first left-eye ocular swab collected by only one experienced technician throughout the study. The assay detects chlamydial lipopolysac- charide (LPS) as previously described [7] with the following modifications for field use: 1) an alternative nitrocellulose membrane was used as the manufacturer discontinued the membrane previously used, 2) the ratio of lyophilised signal amplification system was modified for the test to function at high ambient temperature and 3) increased length of the conjugate tube which houses the dipstick to minimise the evaporation of the reagents during wicking and to protect the membrane against dust. For PCR testing, 200 mL of the POC extract, obtained before adding 6% H2O2, were mixed with 800 mL of Amplicor sample dilution buffer (Roche) and placed at 4 uC within 1 hr, and frozen at –20 uC within 2 days until transport to Cambridge, UK in dry- ice. These samples were stored at –80 uC until blind-tested by Amplicor. The second matched swab was stored dry on cold packs, frozen at –20 uC within 2 days of collection and transported to Cambridge, UK in dry ice, and stored at –80 uC until tested to minimise any target degradation. Clinical examination Clinical assessment can be difficult and inconsistent when conducted by poorly trained or inexperienced staff [5,22]. Taking digital photographs, was previously described as an alternative method and compared to clinical assessment [15,22]. Briefly, the majority of the examinations and taking digital photographs was done by examiner A (96%) to minimise inter-observer variability while the remaining examination were performed by examiner B who also examined and graded independently digital photographs without prior knowledge of the clinical assessment. In a masked fashion, both examiners re-examined photographs and gave an adjudicated score when either the clinical grade or the photographic grade was 3 or greater. A total of 88, 29 and 93 photographs were re-examined for the presence of follicles, inflammation and scarring, respectively. Weighted kappa analysis was previously reported to determine the concordance between methods. The data indicated that there was 79.7% agreement (kappa = 0.40) between clinical assessment, clinical grading and photographic assessment of trachomatous follicles (TF1-TF4) and 96.1% agreement (k = 0.71) when the fine score was translated to TFWHO. The agreement for TIWHO, and TSWHO was 89.3% (k = 0.67) and 92.7% (k = 0.67), respectively [15]. Previous studies have shown the advantages of using finer scales to enhance the sensitivity of clinical measurement, although finer grading schemes may reduce the concordance, or frequency of perfect agreement, between the grades assigned by pairs of independent observations [23]. Correlation of Clinical Trachoma and Infection The detection tube contains lyophilised reagents of the signal amplification system consisting of biotinylated monoclonal antibody to chlamydial LPS and anti- biotin monoclonal antibodies conjugated to colloidal gold particles as the colour indicator. The dipstick contains a nitrocellulose membrane, lined with another monoclonal antibody to chlamyd- ial-LPS (clone CVII-105A5A8, Diagnostic Development Unit) at the capture zone, which captures the immune complex formed between the chlamydial-LPS and signal amplification system reagents, if present. The accumulation of coloured conjugate at the capture line of the dipstick generates a visible colour change as previously described [7]. To generate a visual signal on the parallel to and above the capture zone, the dipstick was lined with the anti- biotin antibody described above, which served as the procedural control zone. All antibodies were produced in-house and purified by affinity chromatography to more than 95% purity before use. Patients Patients were recruited using the medical clinical list, the local council housing list and local knowledge of the Aboriginal Health Table 1. Detailed grading scheme for clinical assessment of trachoma. Grading Definition of the finer grading TF - Trachomatous follicular: TF0 No visible follicles in the upper tarsal conjunctiva TF1 One or two small follicles in the upper tarsal conjunctiva TF2 More than two but less than 5 follicles of 0.5 mm in diameter in the upper tarsal conjunctiva TF3 Five or more follicles of 0.5 mm in diameter in the upper tarsal conjunctiva and equivalent to WHO simplified grading of TF TF4 Extensive large follicles of 0.5 mm in diameter in the upper tarsal conjunctiva TI - Trachomatous inflammation – intense TI0 No visible inflammation of the tarsal conjunctiva TI1 Mild inflammation of the tarsal conjunctiva without obstruction of the vessels TI2 Moderate inflammation of the tarsal conjunctiva with less than half of the deep tarsal vessels being obscured TI3 Pronounced inflammatory thickening of the tarsal conjunctiva that obscures more than half of the normal deep tarsal vessels and equivalent to WHO simplified grading of TI TI4 Very pronounced inflammation of the tarsal conjunctiva TS - Trachomatous scarring TS0 No visible scarring of the tarsal conjunctiva TS1 Small amount of early scarring apparent, but not clearly visible TS2 Moderate amount of early scarring apparent, but not clearly visible TS3 Presence of clearly visible scarring apparent in the upper tarsal conjunctiva and equivalent to WHO simplified grading of TS TS4 Extensive clearly visible scarring involving most of the tarsal conjunctiva doi:10.1371/journal.pntd.0000986.t001 Table 1. Detailed grading scheme for clinical assessment of trachoma. Definition of the finer grading Definition of the finer grading March 2011 \| Volume 5 \| Issue 3 \| e986 March 2011 \| Volume 5 \| Issue 3 \| e986 2 www.plosntds.org Correlation of Clinical Trachoma and Infection www.plosntds.org www.plosntds.org Laboratory assays Amplicor CT/NG Specimen Collection tube (Roche) containing the ocular swab were aliquoted into a DNase/RNase free siliconized tube (BioQuote, North Yorkshire, UK). Specimens were incubated at room temperature for 10 min prior to centrifugation at 17,860 g (max speed: 15,000 rpm) for 15 min at 25 uC (1.0R Megafuge). Supernatants obtained from diluted M4RT-media were decanted with sterile filter tips and the resulting pellets were re-suspended in 1 ml of cell culture grade Dulbecco’s phosphate-buffered saline (DPBS) lacking Ca2+ and Mg2+ (BioWhittaker, Walkersville, MD) by vortexing. The re- suspended pellets were re-centrifuged as indicated above and re- suspended in 100 mL of 2M solution of ammonium hydroxide (obtained from a diluted 5N ammonium hydroxide volumetric standard, Sigma-Aldrich, St. Louis, MO, USA). Specimens were vortexed vigorously, incubated at room temperature for 10 minutes and vortexed again. If the pellet had not dissolved, it was solubilized by repeat pipetting and continuous cycle of vortexing until dissolved. Each specimen was placed into a heating block and heat-treated at 95–100 uC for 1 hour, or until the ammonia had evaporated (dry tubes). Dried specimens were re- suspended in 500 mL of molecular reagent-grade water and, vigorously vortexed and incubated at room temperature for $30 minutes to ensure that any precipitate had re-dissolved. The extracts were stored at 4 uC and tested within 24 h. The above extracted samples and standard curves were prepared and amplified in duplicate on two different days (4 data points) by Real-time qPCR. In brief, after amplification, the amplicon was incubated in a 2 mL microcentrifuge tube at 41 uC on a heating block. 20 mL of amplification product were added to a proprietary detection mixture and the dipstick was inserted in the reaction mixture. The test results were examined after 25 min of incubation and signal on the dipstick scored by an experienced operator according to the in-house scoring chart [25]. Real-time qPCR was performed using a previously described method [20,21] targeting one sequence coding for ORF1 of the Chlamydia cryptic plasmid [20,21]. This method was previously demonstrated highly reproducible (R2 = 0.998) and with analytical sensitivity of ,10 copies per amplification [21]. The previously described reproducibility was established against eleven standard curves constructed for the EB standard on different days. Each curve was generated from seven serial 10-fold dilutions of the pCTL12A plasmid amplified in duplicate. Laboratory assays For chlamydial and internal control testing, they were placed overnight in the Amplicor M4RT-transport medium (3 mL, Roche) and tested by one experienced technician according to the manufacturer’s instruc- tion (Roche). In addition to the lyophilised signal amplification reagents consisting of a biotin-labelled monoclonal antibody to chlamydial LPS and an anti-biotin monoclonal antibody conjugated to colloidal gold particles as colour indicator [7], the nitrocellulose- based membranes are the heart of lateral or vertical flow assays. The wicking rate, pore size, residual surfactants and detergents present on the matrix affect the characteristics of nitrocellulose- based membranes and reaction kinetics. Therefore, changing this porous substrate matrix and the addition of some features (i.e. shape of the conjugate tube) require a systematic adjustment of ratio of the lyophilized signal amplification reagents. All samples yielding a positive PCR result were quantified by previously described ethanol precipitation and qPCR methods [20,21]. Briefly, homogenized M4RT-media (500 mL) from an March 2011 \| Volume 5 \| Issue 3 \| e986 March 2011 \| Volume 5 \| Issue 3 \| e986 3 Correlation of Clinical Trachoma and Infection on a dipstick as described previously [18,19,24,25]. The test designated as SAMBA (Simple AMplification-Based Assay) is based on a proprietary technology [18,19]. Primer and probe target conserved sequences for all the 16sRNA Chlamydia trachomatis serovars obtained from the American Type Culture Collection (ATCC; MD, USA). Regions are conserved for all Chlamydia trachomatis serovars and were selected as previously described for the diagnosis of 2009 pandemic influenza (H1N1) [25] with sequences obtained from the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov) and analysed with Jalview 2.3 (University of Dundee, UK). Detector and capture probes [24] were also designed to target similarly these specific regions. The primers and probes were compared with the Nucleotide Collection database at NCBI with the use of the Basic Local Alignment Search Tool (BLAST). Specificity was established against a panel of microorganisms commonly associated with human eye and skin (e.g. Staphylococcus, Pseudomonas, Strepto- coccus, Escherichia, Proteus and Candida, obtained from ATCC). The SAMBA Chlamydia in a closed device to prevent amplicon contamination has the same unique characteristics of the previously described SAMBA HIV-1 test chemistry render it suitable for near-patient testing in both developed and developing countries because the test uses thermostable reagents and a simplified protocol with minimum sample processing [24]. Laboratory assays In addition, previously published data showed that 7.7260.68 (mean 6 SD) plasmid copies corresponded to one elementary body of C. trachomatis (serovar L1), consistent with previously obtained values [20]. Analysis of genital clinical specimens revealed a strong correlation (R2 = 0.929) between elementary body counts determined by a quantitative ligase chain reaction (LCR)–based Chlamydia tracho- matis LCx Assay (Abbott Laboratories) which targets a conserved region of the cryptic plasmid and those determined by the current qPCR method [20]. Although most of the infected patients were likely to harbour C. trachomatis serovars A, B, Ba and C, the primer sets for both Amplicor and qPCR assays correspond to conserved regions of the C. trachomatis cryptic plasmid and are therefore able to detect all C. trachomatis serovars. Through the present analysis, the organism load was expressed in number of plasmid per swab and not in EB per swab even though, to the knowledge of the authors, it has not been reported that the number of cryptic plasmid significantly varies between serovars. In addition, the second swab from patients identified as Amplicor-positive and 50 randomly selected Amplicor-negative samples with or without clinical signs were tested in duplicate with the 16S-RNA assay. The second swabs were tested in a masked fashion (randomised order) by Amplicor and the 16S-RNA assay. Sample that yielded a positive result on the first swab, but was negative on the second swab, was re-tested in a chessboard manner in presence of known positive and negative samples and, positive and negative controls. To assess the quality of the sampling procedure, human genomic DNA was quantified with Double-Dye Taqman kit according to the manufacturer’s instruction (Primer Design, Southampton, UK) in all PCR positive samples and 41 randomly selected PCR-negative samples. The primers of human genomic DNA kit detect a single copy region of non-transcribed DNA. Statistical analysis Statistical analysis was performed with SAS v9.1 software. Confidence intervals (CI) were calculated as exact binomials. The geometric mean of organism load as well as its respective standard deviation (SD) and 95% confidence interval (CI) for the ocular swabs were calculated from the natural log transformation of the organism load obtained for each swab. The organism load of the ocular samples was compared between the first and second grades of the clinical signs by the Student’s t-test, unequal variance t-test Satterthwaite and equal variance pooled t-test. The correlation between organism load and the fine grading scheme, the load first swabs and the organism load of seconds swabs, and organism load between the different population was obtained using Spearman Rho (R) coefficients and paired Wilcoxon rank tests. Reliability of PCR positivity of both swabs or with 16S-RNA positivity was assessed with the kappa coefficient and its 95% confidence intervals. A p-value of ,0.05 was considered statistically significant. www.plosntds.org March 2011 \| Volume 5 \| Issue 3 \| e986 Fine grading scheme of the clinical signs We examined 1316 of 1545 potential participants, giving an overall examination rate of 85.2% (Figure 1). A total of 1282 participants were eligible for this analysis with a median age of 17.1 years (range: 0.1–95). Each participant was assessed for Amplification of RNA extracted samples (total RNA RNeasy Mini Kit, QIAGEN Inc. Valencia, CA, USA) were performed by isothermal amplification and amplified products detected visually March 2011 \| Volume 5 \| Issue 3 \| e986 March 2011 \| Volume 5 \| Issue 3 \| e986 4 Correlation of Clinical Trachoma and Infection Figure 1. Recruitment algorithm for participants in five Aboriginal communities. doi:10.1371/journal.pntd.0000986.g001 Figure 1. Recruitment algorithm for participants in five Aboriginal communities. doi:10.1371/journal.pntd.0000986.g001 Figure 1. Recruitment algorithm for participants in five Aboriginal communities. doi:10.1371/journal.pntd.0000986.g001 clinical signs of trachoma using a fine grading scheme (Table 1, [15]), by PCR (Roche) and by the LPS-based POC assay. and second swabs tested with PCR (kappa coefficient 0.97; 95% CI: 0.93–1.00) and between the first swab PCR and the 16S-RNA result (kappa coefficient 0.98; 95% CI: 0.95–1.0). and second swabs tested with PCR (kappa coefficient 0.97; 95% CI: 0.93–1.00) and between the first swab PCR and the 16S-RNA result (kappa coefficient 0.98; 95% CI: 0.95–1.0). and second swabs tested with PCR (kappa coefficient 0.97; 95% CI: 0.93–1.00) and between the first swab PCR and the 16S-RNA result (kappa coefficient 0.98; 95% CI: 0.95–1.0). On the basis of clinical examination of both eyes of each subject, 135 participants had active trachoma (10.5%; 95% CI: 8.9–12.2), 130 with TFWHO and five had TIWHO without TFWHO. Taking the findings of both eyes the highest age-specific prevalence of TF was in children 2–4 year-old (33/121; 27.3%) followed by 5–9 year-old (51/229; 22.3%) and those with less than 2 years of age (13/61; 21.3%). The overall prevalence of TFWHO in children aged 1–9 was 25.1% (96/383, 95% CI: 20.7–29.4) and the prevalence of the five communities ranged from 9.8% to 38.5% (4/41 [9.8%], 5/49 [10.2%], 23/115 [20.0%], 27/82 [32.9%] and 37/96 [38.5%], respectively). The CT organism load was analysed by qPCR in all PCR- positive swabs (Figure 3). The geometric mean organism load was 55,585 (95%CI: 801–3,811,754) pCTL12A plasmid per swab for the first swab and 4,355 (95%CI: 98–193,602) for the second. The mean organism load for the second swab was 12.8 times lower (95%CI: 0.79–566.3) than for the first (paired Wilcoxon rank sum test, P,0.0001). Fine grading scheme of the clinical signs The organism load in the first swab was strongly correlated with the load in the second swab (Spearman Rho = 0.74, P,0.0001). To confirm the adequacy of specimen collection, human genomic DNA was quantified for the 46 PCR-positive (1.2560.69 mg of genomic DNA/swab) and 41 random PCR-negative (including nine participants presenting signs of TFWHO – 1.3360.70 mg of genomic DNA/swab). The amount of genomic DNA was not significantly different between positive and negative ocular samples (two-tailed P = 0.6), nor was there a correlation between the organism load and the quantity of genomic DNA/swab. All of the 32 control swabs of potential formites were negative by PCR. In contrast, when clinical data were limited to the left eye in order to directly compare with the PCR and POC testing results, the frequency of clinical signs of active trachoma decreased to 8.6% (110/1282), 108 with TFWHO and two had TIWHO without TFWHO (Table 2). The resulting prevalence in the five commu- nities in children aged 1–9 was 0/41 (0%), 4/49 (8.2%), 21/115 (18.3%), 21/82 (25.6%) and 36/96 (37.5%). www.plosntds.org www.plosntds.org NAAT-positivity versus clinical signs The PCR (Amplicor) positivity rate in the population was 3.6% (46/1282, 95% CI: 2.6–4.6) and, in children aged 1–9, 6.8% (26/ 383, 95% CI: 4.3–9.3). Of the PCR positive participants, the highest rate was in children 5–9 year-old (17/46; 37%) followed by 2–4 year-old (9/46; 19.6%) and 10–14 year-old (8/46; 17.4%, Figure 2). Of the 46 people for whom the first swab from the left eye was PCR-positive, on testing of the second swab, 43 (93.5%, 95%CI: 86.3–100) were PCR-positive and 44 (95.7%) were 16S- RNA-positive. Two of the three PCR-negative second swabs were 16S-RNA-positive and one of those had the lowest organism load on the first swab. There was a good agreement between the first A significant correlation was observed between PCR-positivity (Amplicor) and TFWHO (Wilcoxon rank sum tests P,0.0001) and, between PCR-positivity and the fine grading scheme (Spearman Rho = 0.98 and P = 0.0004) (Table 2). A higher proportion of people were PCR positive as clinical disease, as assessed by the fine grading, became more severe. However, it should be noted that 59% (27/46) of PCR positive results occurred in people with TFWHO, although only 13% (6/46) occurred in people with TF0. As result, the agreement between PCR and TFWHO was poor for children of #9 year of age (k = 0.15; 95% CI: 0.01–0.25) and March 2011 \| Volume 5 \| Issue 3 \| e986 5 Correlation of Clinical Trachoma and Infection Table 2. Clinical signs of active trachoma compared with the presence of C. trachomatis. Clinical signs of active trachoma (TF and TI) in the left eye compared with the presence of C. NAAT-positivity versus clinical signs trachomatis DNA by PCR and the POC test positivity Clinical sign PCR Positive (n = 46) POC Positive (n = 14) Total and photograding n % 95% CI1 n % 95% CI1 (n = 1282) TF0 6 0.8 0.3–1.7 0 0.0 0.0–0.4 793 TF1 16 6.2 3.8–9.9 2 0.8 0.0–3.0 257 TF2 5 4.0 1.5–9.3 1 0.8 0.0–4.9 124 TF3 16 16.3 10.2–25.0 9 9.2 4.7–16.7 98 TF4 3 30.0 10.3–60.8 2 20.0 4.6–52.1 10 TFWHO absent 27 2.3 1.6–3.3 3 0.3 0.1–0.8 1,174 TFWHO present 19 17.6 11.5–25.9 11 10.2 5.6–17.5 1082 TFWHO & TIWHO present 6 66.7 35.1–88.3 3 33.3 11.7–64.9 9 TIWHO present without TFWHO 1 50.0 9.5–90.6 1 50.0 9.5–90.6 2 Active trachoma 20 18.2 12.0–26.5 12 10.9 6.2–18.3 1103 195% Confidence intervals were calculated with the adjusted Wald interval method. 2108 participants had TFWHO, 99 with TFWHO and nine had TFWHO with TIWHO. 3110 participants had active trachoma, 108 with TFWHO and two had TIWHO without TFWHO. doi:10.1371/journal.pntd.0000986.t002 195% Confidence intervals were calculated with the adjusted Wald interval method. 2108 participants had TFWHO, 99 with TFWHO and nine had TFWHO with TIWHO. 3110 participants had active trachoma, 108 with TFWHO and two had TIWHO without TFWHO. doi:10.1371/journal.pntd.0000986.t002 195% Confidence intervals were calculated with the adjusted Wald interval method. 2108 participants had TFWHO, 99 with TFWHO and nine had TFWHO with TIWHO. 3110 participants had active trachoma, 108 with TFWHO and two had TIWHO without TFWHO. doi:10.1371/journal.pntd.0000986.t002 swab whose house number was missing so her household contacts could not be identified. Therefore, with the exception of the last girl, a plausible case can be made for exposure to infection and four of five had some signs of inflammation (TI of some degree). still poor (k = 0.23; 95% CI: 0.08–0.37) for older participants. Figure 4 describes the age-specific prevalence of the left eye fine grading of TF1 (Figure 4A), TF2 (Figure 4B), TF3 (Figure 4C) and TF4 (Figure 4D) versus PCR positivity. Of particular interest were six PCR positive participants who had not have active follicular disease and were graded as TF0. All were female whose ages were 9, 16, 21, 57, 66 and 73 years. Two lived in houses with children who were PCR positive. Another two lived in houses in which three or more children had TF3. The fifth woman was aged 73 and had TI1 and TS3 with 30,918 plasmid/ swab. NAAT-positivity versus clinical signs She shared a house with two men, one aged 28 who had TF2 and TI1 and the other aged 58 with TF1 and TI1. The sixth was a 9 year-old girl with a normal exam and 37,074 plasmid/ POC assay The LPS-based POC assay yielded 18 positive individuals, 14 were PCR positive and 4 were young adults who were PCR negative and without clinical disease (TF0 and TI0). Using PCR as the comparator test, the sensitivity and specificity of the POC was 30.4% (14/46, 95% CI: 17.1–43.7) and 99.7% (1232/1236, 95% CI: 99.4–100), respectively. To reduce the likelihood of over-grading of clinical disease, the assessment was made both in the field using frequent reference to the WHO grading card and by independent photo-grading. Although over-grading can still occur, this combined approach reduces the risk. A comparison of the performance data of the LPS-based POC with an analytical sensitivity of 2,500 chlamydial elementary bodies [7] from Tanzania and the current study in Australia is interesting. Although the prevalence rates of TFWHO in 1 to 9 year olds are roughly comparable; 28% and 21% respectively, the intensity of disease (the proportion of those with TF and/or TI who have TI) was nearly three times higher in Tanzania (25% compared to 9.5%) as was the mean organism load (147,267 compared to 55,585 plasmids/swab). The unequal variance t-test Satterwaite using load (P = 0.0057) and equal variance pooled t-test using natural log transformation of the organism load indicated (P = 0.0333) that organism load difference between those samples collected in Aboriginal communities and those samples collected in the Masai Figure 3. Distribution of the organism load1 in first and second swabs showing the geometric mean2. doi:10.1371/journal.pntd.0000986.g003 Figure 3. Distribution of the organism load1 in first and second swabs showing the geometric mean2. doi:10.1371/journal.pntd.0000986.g003 mean organism loads for the WHO grades were: TFWHO present 133,252 plasmid/swab (95% CI: 2,173–8.26106), TFWHO absent 26,903 plasmid/swab (95% CI: 534–1.46106), TIWHO present 400,312 plasmid/swab (95% CI: 38,101–4.26106) and TIWHO absent 40,135 plasmid/swab (95% CI: 655–2.56106). disease may take up to 9 months or so [5,27]. As previously observed [7], the prevalence of active trachoma varies when one or both eyes are considered, in this study from 8.6% to 10.5%, respectively [7,28]. For practical and economic reasons, swabs for PCR, LPS-based and 16S-RNA testing, and photographs were only collected from the left eye. Therefore, clinical/laboratory diagnostic comparisons were made only for the left eye using the consensus grading based on clinical and photographic data. www.plosntds.org Organism load versus clinical sign The fine grading of TF0–4 (Figure 5A), TI0–4 in presence of TFWHO (Figure 5B), TI0–4 in absence of TFWHO (Figure 5C) and TS0–4 in presence of TFWHO was positively correlated with the organism load whereas there was no correlation for TS0–4 in absence of TFWHO (Spearman Rho (R) = 0.498 and P = 0.0004, R = 0.473 and P = 0.0009, R = 0.438 and P = 0.0023, R = 0.449 and P = 0.0017 and R = 20.039 and P = 0.7946, respectively). The Figure 2. Age distribution of PCR-positive subjects (n = 46/1282). doi:10.1371/journal.pntd.0000986.g002 Figure 2. Age distribution of PCR-positive subjects (n = 46/1282). doi:10.1371/journal.pntd.0000986.g002 March 2011 \| Volume 5 \| Issue 3 \| e986 6 www.plosntds.org Correlation of Clinical Trachoma and Infection Figure 3. Distribution of the organism load1 in first and second swabs showing the geometric mean2. doi:10.1371/journal.pntd.0000986.g003 www.plosntds.org www.plosntds.org www.plosntds.org Discussion The poor correlation between the prevalence of clinically active trachoma and evidence of infection is not new [12,13], especially in low prevalence communities [5]. As with any infectious disease, there is an initial incubation period (4–8 days) between inoculation and the development of clinical disease [5]. This is followed by frank disease when both bacterium and clinical signs co-exist, and a later stage when the infection is no longer present or cannot be detected by diagnostic tests, yet the clinical signs persist as disease slowly resolve [5,26]. In humans, the lag period between the last detectable bacterial shedding and the resolution of the active March 2011 \| Volume 5 \| Issue 3 \| e986 March 2011 \| Volume 5 \| Issue 3 \| e986 7 Correlation of Clinical Trachoma and Infection Figure 4. Age-specific prevalence of the left eye fine grading of TF and PCR positivity. Fig. 4A: Sign of TF1; Fig. 4B: Sign of TF2; Fig. 4C: Sign of TF3; Fig. 4D: Sign of TF4. doi:10.1371/journal.pntd.0000986.g004 Figure 4. Age-specific prevalence of the left eye fine grading of TF and PCR positivity. Fig. 4A: Sign of TF1; Fig. 4B: Sign of TF2; Fig. 4C: Sign of TF3; Fig. 4D: Sign of TF4. doi:10.1371/journal.pntd.0000986.g004 Figure 4. Age-specific prevalence of the left eye fine grading of TF and PCR positivity. Sign of TF3; Fig. 4D: Sign of TF4. doi:10.1371/journal.pntd.0000986.g004 Figure 4. Age-specific prevalence of the left eye fine grading of TF and PCR positivity. Fig. 4A: Sign of TF1; Fig. 4B: Sign of TF2; Fig. 4C: Sign of TF3; Fig. 4D: Sign of TF4. doi:10 1371/journal pntd 0000986 g004 infection was found in people who did not have the WHO grade of TF but who still had some milder clinical changes (TF1 or TF2). As mentioned, organism load also correlated with the fine grading of trachoma. Similar findings come from an earlier study that used a roughly similar finer grading scheme and that used both tissue culture and direct fluorescent antibody cytology to detect infection [13]. That study also found the load of infection was higher in those with more severe disease (WHO grade TF) than in those with less severe clinical disease. communities were significantly different. The lower intensity of disease in Australia is reflected in the two to three times lower rates of PCR positivity in both those with TFWHO (16% in Australia and 44% in Tanzania) and those without TFWHO (4.3% and 9.7%). Similarly the reduced performance of the LPS-based POC assay in Australia may in part reflect the lower organism load and in part the modification of the test. The reduced disease severity and infectious load observed in Australian Aboriginal communities may reflect the dramatic differences in medical, environment and living conditions between the Masai and Aboriginal people. In that study with a less sensitive assessment of infection 12% of infection was in those who did not have the simplified WHO grade of TF or TI. A rapid, simple and affordable POC test capable of accurate identification of active infection would nevertheless be a useful tool in trachoma control. Correlation of Clinical Trachoma and Infection As demonstrated in this study, a rational explanation for a possible source of infection was identified in the household in all but one case. In this case an elderly woman who was PCR positive shared a house with two other adults. A treatment program that only targeted households with children would miss treating households such as this. However, a significant problem in these communities is the frequent sharing of houses and it is not unusual for children to sleep in two or three different houses each week [5]. We were not able to track this sort of movement to identify extended families. More precise data of people’s movements and behaviour would be needed to explore this in further detail. However, the key messages of this study are that the use of a binary grading system like the simplified WHO grading system will miss many of the more subtle issues around the identification of chlamydial infection and clinical status [13]. Specifically, 53% of the clinical disease (TF1 or TF2) and 59% of PCR positivity occurred in people with disease less severe than the WHO grade of TF. These data suggest that detailed studies of the pathogenesis, distribution and natural history of trachoma should make use of both a finer grading scheme and also quantify the infectious load. The fine grading also revealed a high proportion of people with clinical disease who were below the WHO threshold for TF. Most of the active trachoma was seen in the younger children who also had the more severe disease. Both the severity and prevalence of TF decreased with age. However, the finding of a significant proportion of people with few follicles, TF1 is noteworthy. Although some of this will reflect the waxing and waning of active trachoma, it suggests that in some people at least a few follicles, once formed, may persist for a long period of time or indefinitely. It is unclear whether these ‘‘persistent’’ follicles reflect occasional exposure to chlamydial antigens or infection or if they are a permanent tissue change. The strengths of this study include firstly the careful documentation and quantification of the two outcomes, the presence of infection and clinical trachoma. Careful specimen collection and handling limited sampling cross-contamination critical when highly sensitive NAATs capable of detecting low level of targets are used. Correlation of Clinical Trachoma and Infection Poor sampling methods and risk of cross- contamination in the field have caused concern about the validity of some early NAAT findings [5,31]. We observed no significant difference between the quantity of human genomic DNA contained in negative and positive PCR swabs that makes inadequate sample collection unlikely. The repeat negative control swabs of fomites suggest that the field precautions taken to prevent cross-contamination were successful. Further, the positioning of the samples in the test plate was verified to ensure that neither positive specimens nor positive controls were adjacent on repeat testing. Quantitative PCR was used to assess the infectious load in all positive specimens. Organism load varies in areas with different levels of endemicity and intensity of disease. In areas with a lower prevalence or intensity, laboratory tests may be of limited used for community- based assessment. However, it is in these situations that these tests would be of most use for the confirmation of sporadic cases with clinical disease. In addition, when the upper tarsal conjunctiva was swabbed transversally to collect an appropriate specimen for testing, the organism load of the consecutive swab was dramatically decreased even though the amount of collected cells was similar. This great disparity in organism load raises concerns about the use of consecutive specimens in preference to split specimens. The second generation NAATs capable of detecting high multiple-copy of targets such as ribosomal rRNA should enhance analytical sensitivity [32]. This may enable to detect some low level infection previously missed by PCR [32,33] or other less sensitive methods such as Real-time qPCR targeting a single copy genomic sequence such as the major outer membrane protein gene (omp1) or the outer membrane complex B protein gene (omcB) instead of the multiple-copy sequences (Chlamydia cryptic plasmid), LPS-based rapid test [7], culture and direct immuno-fluorescence (DFA), and so extend the period of detectable infection. A point- of-care nucleic acid amplification test based on targets with multiple copies such as 16S-RNA would be a more appropriate Secondly, this study used a fine or semi-quantitative scale to grade the clinical severity of trachoma and developed a consensus- based grade that used both clinical and photographic findings. This allowed for a much more detailed assessment of the presence of clinical signs and the recognition of less advanced disease than the more frequently used WHO simplified trachoma grading system [14]. Correlation of Clinical Trachoma and Infection dipstick, offers several advantages and with its inherent sensitivity and thermostability and therefore has the potential to meet this need [29,30]. Because of the sealed containment of amplified sample a dedicated laboratory is not require. Assay reagents have been converted into thermostable formulas (stability tested at 55 uC for one month and 37 uC for 12 months, data not shown) and the test uses a simplified protocol with minimum operation and sample processing made possible by a disposable modular cartridge previously described by Lee et al [24]. This technology, Simple Amplification-Based Assay (SAMBA) has a substantial advantage over currently available NATs, in that it is able to provide results quickly and on-site, thereby facilitating appropriate clinical support. The simplicity of SAMBA tests will allow their use in resource-poor settings in developing countries and for near- patient testing in the developed world [24]. These features may offer advantages over the LPS-based POC, although further evaluation of the 16S-RNA Chlamydia SAMBA test in areas of varying trachoma prevalence is required to better assess its performance and cost effectiveness. ‘‘asymptomatic’’. A finer grading scheme allows the identification of subjects with clinical signs below the threshold set by the simplified WHO grading scheme because many people with less advanced disease do not fit in the clear-cut definitions used in the WHO simplified trachoma grading system. Finally, this study was a population-based study with high community coverage that included people of all ages. Potential weaknesses include the inability to precisely link the exposure to infection of individual participants as in these communities children in particular may move frequently from one house to house another. In addition, there were inevitably some missing data for both people and house numbers. The study did not assess the sampling, testing, clinical-grading and photo- grading agreement between swabbers, laboratories and examiners because the same swabber, laboratory and clinical-grading or photo-grading examiners were used throughout the study. The inter-operator agreement and intra-operator reproducibility of the LPS-based test [7] and clinical- versus photo-grading agreement [15] have been described elsewhere. Of considerable interest is the small number of individuals who had a positive PCR in the absence of a detectable follicular response (TF0). Some were older people with scarred conjunctiva who may be incapable of mounting such an immune response. Others may have had a transitory infection or may have been in an incubatory phase of infection. Figure 4. Age-specific prevalence of the left eye fine grading of TF and PCR positivity. Fig. 4A: Sign of TF1; Fig. 4B: Sign of TF2; Fig. 4C: Sign of TF3; Fig. 4D: Sign of TF4. doi:10.1371/journal.pntd.0000986.g004 The 16S-RNA test, a closed- system device based on visual detection of nucleic acid on a Even though the detection of infection by PCR is a poor predictor of the presence of clinical disease and equally clinical disease was poorly correlated with infection, organism load was strongly correlated with the prevalence and severity of active trachoma as graded by the finer grading scheme and that 46% of March 2011 \| Volume 5 \| Issue 3 \| e986 March 2011 \| Volume 5 \| Issue 3 \| e986 8 Correlation of Clinical Trachoma and Infection Figure 5. Organism load versus fine left eye clinical grading. Fig. 5A: TF0–4 vs. load1; Fig. 5B: TI0–4 in presence of TFWHO TI0–4 in absence of TFWHO vs. load3. doi:10.1371/journal.pntd.0000986.g005 www plosntds org 9 March 2011 \| Volume 5 Figure 5 Organism load versus fine left eye clinical grading Fig 5A: TF0 4 vs load1; Fig 5B: TI0 4 in presence of TFWH Figure 5. Organism load versus fine left eye clinical grading. Fig. 5A: TF0–4 vs. load1; Fig. 5B: TI0–4 in presence of TFWHO vs. load TI0 4 in absence of TFWHO vs load3 Figure 5. Organism load versus fine left eye clinical grading. Fig. 5A: TF0–4 vs. load1; Fig. 5B: TI0–4 in presence of TFWHO vs. load2; Fig. 5C: TI0–4 in absence of TFWHO vs. load3. doi:10.1371/journal.pntd.0000986.g005 March 2011 \| Volume 5 \| Issue 3 \| e986 March 2011 \| Volume 5 \| Issue 3 \| e986 9 www.plosntds.org www.plosntds.org www.plosntds.org References y y ( ) the effects of scaling. Invest Ophthalmol Vis Sci 32: 422–432. 24. Lee HH, Dineva MA, Chua YL, Ritchie AV, Ushiro-Lumb I, et al. (2010) Simple amplification-based assay: a nucleic acid-based point-of-care platform for HIV-1 testing. J Infect Dis 201 Suppl 1: S65–72. 8. Solomon AW, Harding-Esch E, Alexander ND, Aguirre A, Holland MJ, et al. (2008) Two doses of azithromycin to eliminate trachoma in a Tanzanian community. N Engl J Med 358: 1870–1871. 25. Wu LT, Curran MD, Ellis J, Parmar S, Ritchie AV, et al. (2010) Nucleic Acid Dipstick Test: Molecular Diagnosis of Pandemic H1N1. J Clin Microbiol. y g J 9. Miller K, Schmidt G, Melese M, Alemayehu W, Yi E, et al. (2004) How reliable is the clinical exam in detecting ocular chlamydial infection? Ophthalmic Epidemiol 11: 255–262. 26. Taylor HR, Johnson SL, Prendergast RA, Schachter J, Dawson CR, et al. (1982) An animal model of trachoma II. The importance of repeated reinfection. Invest Ophthalmol Vis Sci 23: 507–515. p 10. Solomon AW, Foster A, Mabey DC (2006) Clinical examination versus Chlamydia trachomatis assays to guide antibiotic use in trachoma control programmes. Lancet Infect Dis 6: 5–6; author reply 7–8. p 27. Taylor HR, Siler JA, Mkocha HA, Munoz B, West S (1992) The natural history of endemic trachoma: a longitudinal study. Am J Trop Med Hyg 46: 552–559. 28 Mi h l CE (2007) I f id i f f h l f 11. Liu DT, Chan AY, Tai MH (2006) Clinical examination and laboratory tests for estimation of trachoma prevalence in remote settings. Lancet Infect Dis 6: 7; author reply 7–8. 28. Michel CE (2007) Impact of rapid point-of-care assays for the control Chlamydia trachomatis infections. Cambridge: Open University. Chlamydia trachomatis infections. Cambridge: Open Universit 29. Dineva MA, MahiLum-Tapay L, Lee H (2007) Sample preparation: a challenge in the development of point-of-care nucleic acid-based assays for resource- limited settings. Analyst 132: 1193–1199. 12. Wright HR, Taylor HR (2005) Clinical examination and laboratory tests for estimation of trachoma prevalence in a remote setting: what are they really telling us? Lancet Infect Dis 5: 313–320. 30. Mabey D, Peeling RW, Ustianowski A, Perkins MD (2004) Diagnostics for the developing world. Nat Rev Microbiol 2: 231–240. 13. Taylor HR, Rapoza PA, West S, Johnson S, Munoz B, et al. (1989) The Epidemiology of Infection in Trachoma. Investigative Ophthalmology & Visual Science 30: 1823–1833. 31. STARD checklist STARD checklist for reporting of studies of diagnostic accuracy. STARD checklist STARD checklist for reporting of studies of diagnostic accuracy. Found at: doi:10.1371/journal.pntd.0000986.s002 (0.05 MB DOC) Found at: doi:10.1371/journal.pntd.0000986.s002 (0.05 MB DOC) Finally, the current prevalence of both active trachoma and trichiasis are roughly similar to that reported 30 year ago in this region by National Trachoma Eye Health Program [15] and both of which exceed the thresholds set by WHO to define blinding trachoma as a public health problem indicate the need for appropriate interventions to control trachoma and prevent blindness in these five Aboriginal communities. Acknowledgments We thank the community members and clinic staff in the Katherine Region and field staff, M. Cheang of University of Manitoba, J-P. Allain and members of the Diagnostic Development Unit, University of Cambridge. STARD flowchart STARD flowchart for reporting of studies of diagnostic accuracy. STARD flowchart STARD flowchart for reporting of studies of diagnostic accuracy. STARD flowchart STARD flowchart for reporting of studies of diagnostic accuracy. Conceived and designed the experiments: HHL HRT. Performed the experiments: C-ECM KGR HRT. Analyzed the data: C-ECM HHL HRT. Contributed reagents/materials/analysis tools: C-ECM KGR MAD HHL HRT. Wrote the paper: C-ECM HHL HRT. Found at: doi:10.1371/journal.pntd.0000986.s001 (0.05 MB DOC) Found at: doi:10.1371/journal.pntd.0000986.s001 (0.05 MB DOC) Correlation of Clinical Trachoma and Infection The use of a finer grading scheme is especially important in detailed research studies looking for correlation between clinical disease, environmental or genetic risk factors and the distribution of infection in a community when attempts are made to separate those with ‘‘disease’’ from those who are March 2011 \| Volume 5 \| Issue 3 \| e986 March 2011 \| Volume 5 \| Issue 3 \| e986 10 Correlation of Clinical Trachoma and Infection tool to detect low level of infection previously missed by LPS-based rapid test [7]. STARD checklist STARD checklist for reporting of studies of diagnostic accuracy. References 1. Resnikoff S, Pascolini D, Etya’ale D, Kocur I, Pararajasegaram R, et al. (2004) Global data on visual impairment in the year 2002. Bull World Health Organisation 82: 844–851. 1. Resnikoff S, Pascolini D, Etya’ale D, Kocur I, Pararajasegaram R, et al. (2004) Global data on visual impairment in the year 2002. Bull World Health Organisation 82: 844–851. 18. Lee HH, Dineva MA, inventors. Diagnostics for the Real World, assignee (2002) Improved detection signal and capture In dipstick assays. International patent WO 02/04667. g 2. Wright HR, Turner A, Taylor HR (2008) Trachoma. Lancet 371: 1945–1954. g 2. Wright HR, Turner A, Taylor HR (2008) Trachoma. Lancet 371: 1945–1954. 19. Lee HH, Dineva MA, Fletcher-Brown F, inventors. Cambridge Enterprise, assignee (2008) Nucleic acid amplification and testing. International patent WO 2008090340. g , , y ( ) 3. Mariotti SP, Pascolini D, Rose-Nussbaumer J (2009) Trachoma: global magnitude of a preventable cause of blindness. Br J Ophthalmol 93: 563–568. 4. Solomon AW, Peeling RW, Foster A, Mabey DC (2004) Diagnosis and assessment of trachoma. Clin Microbiol Rev 17: 982–1011. 20. Michel CE, Sonnex C, Carne CA, White JA, Magbanua JP, et al. (2007) Chlamydia trachomatis load at matched anatomic sites: implications for screening strategies. J Clin Microbiol 45: 1395–1402. 5. Taylor HR (2008) Trachoma: a blinding scourge from the Bronze Age to the twenty-first centory East Melbourne, Viv. Centre for Eye Research Australia, Haddington Press Pty Ltd. 21. Wisniewski CA, White JA, Michel CE, Mahilum-Tapay L, Magbanua JP, et al. (2008) Optimal method of collection of first-void urine for diagnosis of Chlamydia trachomatis infection in men. J Clin Microbiol 46: 1466–1469. g y 6. Wright HR, Turner A, Taylor HR (2007) Trachoma and poverty: unnecessary blindness further disadvantages the poorest people in the poorest countries. Clin Exp Optom 90: 422–428. 22. West SK, Taylor HR (1990) Reliability of Photographs for Grading Trachoma in Field Studies. British Journal of Ophthalmology 74: 12–13. in Field Studies. British Journal of Ophthalmology 74: 12–13. 7. Michel CE, Solomon AW, Magbanua JP, Massae PA, Huang L, et al. (2006) Field evaluation of a rapid point-of-care assay for targeting antibiotic treatment for trachoma control: a comparative study. Lancet 367: 1585–1590. 23. Bailey IL, Bullimore MA, Raasch TW, Taylor HR (1991) Clini 23. Bailey IL, Bullimore MA, Raasch TW, Taylor HR (1991) Clinical grading and the effects of scaling. Invest Ophthalmol Vis Sci 32: 422–432. References de Barbeyrac B, Goldschmidt P, Malembic S, Raherison S, Clerc M, et al. (2007) Quality assessment of conjunctival specimens for detection of Chlamydia trachomatis by PCR in children with active trachoma. Clin Microbiol Infect 13: 689–694. 14. Thylefors B, Dawson CR, Jones BR, West SK, Taylor HR (1987) A simple system for the assessment of trachoma and its complications. Bull World Health Organ 65: 477–483. 32. Schachter J, Hook EW, Martin DH, Willis D, Fine P, et al. (2005) Confirming positive results of nucleic acid amplification tests (NAATs) for Chlamydia trachomatis: all NAATs are not created equal. J Clin Microbiol 43: 1372–1373. g 15. Roper KG, Taylor HR (2009) Comparison of clinical and photographic assessment of trachoma. Br J Ophthalmol 93: 811–814. 16. Taylor HR (1987) Trachoma grading: a new grading sche Pathol Ocul Trop Subtrop Sante Publique. pp 175–181. 16. Taylor HR (1987) Trachoma grading: a new grading sch 33. Chernesky M, Jang D, Luinstra K, Chong S, Smieja M, et al. (2006) High analytical sensitivity and low rates of inhibition may contribute to detection of Chlamydia trachomatis in significantly more women by the APTIMA Combo 2 assay. J Clin Microbiol 44: 400–405. Pathol Ocul Trop Subtrop Sante Publique. pp 175–18 17. Roper K, Michel CE, Kelly PM, Taylor HR (2008) Prevalence of trachoma in Aboriginal communities in the Katherine Region of the Northern Territory in 2007. Med J Aust 189: 409. March 2011 \| Volume 5 \| Issue 3 \| e986 11 www.plosntds.org
https://openalex.org/W3196532164	https://iris.unimore.it/bitstream/11380/1275297/2/fgene-12-705019.pdf	English	null	Hologene 5: A Phase II/III Clinical Trial of Combined Cell and Gene Therapy of Junctional Epidermolysis Bullosa	Frontiers in genetics	2,021	cc-by	10,247	STUDY PROTOCOL published: 01 September 2021 doi: 10.3389/fgene.2021.705019 Hologene 5: A Phase II/III Clinical Trial of Combined Cell and Gene Therapy of Junctional Epidermolysis Bullosa Laura De Rosa 1†, Elena Enzo 2†, Giulia Zardi 3, Christine Bodemer 4, Cristina Magnoni 5, Holm Schneider 6 and Michele De Luca 1,2* Laura De Rosa 1†, Elena Enzo 2†, Giulia Zardi 3, Christine Bodemer 4, Cristina Magnoni 5, Holm Schneider 6 and Michele De Luca 1,2* 1 Holostem Terapie Avanzate, s.r.l, Modena, Italy, 2 Centre for Regenerative Medicine “Stefano Ferrari”, University of Modena and Reggio Emilia, Modena, Italy, 3 Department of Statistical Sciences, University of Bologna, Bologna, Italy, 4 Department of Dermatology, Necker Enfants Malades Hospital, APHP, University Paris Centre, ERN-Skin Network (European Network for Rare Skin Disorders), Paris, France, 5 Unit of Dermatology, University of Modena and Reggio Emilia, Modena, Italy, 6 Department of Pediatrics, University Hospital Erlangen, Erlangen, Germany Q , Boston Children’s Hospital and Harvard Medical School, United States Gerarda Cappuccio, University of Naples Federico II, Italy Cristina Has, University of Freiburg Germany †These authors share first authorship Specialty section: This article was submitted to Genetics of Common and Rare Diseases, a section of the journal Frontiers in Genetics Specialty section: This article was submitted to Genetics of Common and Rare Diseases, Received: 04 May 2021 Accepted: 28 July 2021 Published: 01 September 2021 Keywords: epidermolysis bullosa, cell and gene therapy, epidermal stem cell, clinical trial, skin, genodermatoses STUDY PROTOCOL published: 01 September 2021 doi: 10.3389/fgene.2021.705019 Edited by: Desheng Liang, Central South University, China Edited by: Desheng Liang, Central South University, China Reviewed by: Qifei Li, Boston Children’s Hospital and Harvard Medical School, United States Gerarda Cappuccio, University of Naples Federico II, Italy Cristina Has, University of Freiburg, Germany Correspondence: Michele De Luca michele.deluca@unimore.it †These authors share first authorship Edited by: Desheng Liang, Central South University, China Reviewed by: Qifei Li, Boston Children’s Hospital and Harvard Medical School, United States Gerarda Cappuccio, University of Naples Federico II, Italy Cristina Has, University of Freiburg, Germany Correspondence: Michele De Luca michele.deluca@unimore.it †These authors share first authorship Epidermolysis bullosa (EB) is a group of devastating genetic diseases characterized by skin and mucosal fragility and formation of blisters, which develop either spontaneously or in response to minor mechanical trauma. There is no definitive therapy for any form of EB. Intermediate junctional EB (JEB) caused by mutations in the gene LAMB3 has been the first genetic skin disease successfully tackled by ex vivo gene therapy. Here, we present a multicenter, open-label, uncontrolled phase II/III study that aims at confirming the efficacy of Hologene 5, a graft consisting of cultured transgenic keratinocytes and epidermal stem cells and meant to combine cell and gene therapy for the treatment of LAMB3-related JEB. Autologous clonogenic keratinocytes will be isolated from patients’ skin biopsies, genetically corrected with a gamma-retroviral vector (γRV) carrying the full-length human LAMB3 cDNA and plated onto a fibrin support (144 cm2). The transgenic epidermis will be transplanted onto surgically prepared selected skin areas of at least six JEB patients (four pediatric and two adults). Evaluation of clinical efficacy will include, as primary endpoint, a combination of clinical parameters, such as percentage of re-epithelialization, cellular, molecular, and functional parameters, mechanical stress tests, and patient- reported outcome (PRO), up to 12 months after transplantation. Safety and further efficacy endpoints will also be assessed during the clinical trial and for additional 15 years in an interventional non-pharmacological follow-up study. If successful, this clinical trial would provide a therapeutic option for skin lesions of JEB patients with LAMB3 mutations and pave the way to a combined cell and gene therapy platform tackling other forms of EB and different genodermatoses. INTRODUCTION There is no cure for EB (Mellerio et al., 2020). Available therapies are largely palliative and aim to prevent blisters and/ or promote their healing, but only partially alleviate the devastating clinical manifestations. Long-lasting, curative therapies are urgently needed and, nowadays, gene therapy in combination with stem cell-based strategies is still at the experimental stage. Epidermolysis bullosa (EB) is a heterogeneous group of rare, genetic disorders caused by molecular defects in genes encoding several structural proteins that form the epidermal-dermal junction. EB is characterized by mechanical fragility and recurrent blistering of the skin and other stratified epithelia (Bardhan et al., 2020).i i Based on the level of skin cleavage, EB can be classified into four major types – EB simplex (EBS), junctional EB (JEB), dystrophic EB (DEB), and Kindler syndrome – that differ in severity and prevalence. Clinical manifestations range from mild to severe with local or generalized skin and mucosal involvement, depending on phenotypic and molecular factors (Fine, 2010; Fine et al., 2014; Has et al., 2020). The main clinical features of EB, such as blistering, erosions, and recurrent infections, can be associated with scarring, nail dystrophy, teeth abnormalities, milia, oral lesions, esophageal strictures, and thriving impairment. The severe scarring typical for recessively inherited DEB (RDEB) induces pseudosyndactyly, which contributes to the poor quality of life of RDEB patients. Patients suffering from intermediate JEB and RDEB are highly prone to develop aggressive squamous cell carcinomas (SCC) leading to premature death (Mallipeddi et al., 2004; Fine et al., 2009; Fine, 2010; Montaudie et al., 2016). Thus, depending upon the subtype, the prognosis of EB may range from a virtually normal life with only minor skin lesions to neonatal death due to severe skin and mucosal defects. Citation: De Rosa L, Enzo E, Zardi G, Bodemer C, Magnoni C, Schneider H and De Luca M (2021) Hologene 5: A Phase II/III Clinical Trial of Combined Cell and Gene Therapy of Junctional Epidermolysis Bullosa. Front. Genet. 12:705019. doi: 10.3389/fgene.2021.705019 Clinical Trial Registration: EudraCT Number: 2018-000261-36. Keywords: epidermolysis bullosa, cell and gene therapy, epidermal stem cell, clinical trial, skin, genodermatoses September 2021 \| Volume 12 \| Article 705019 1 Frontiers in Genetics \| www.frontiersin.org Hologene 5 De Rosa et al. RATIONALE AND PREVIOUS CLINICAL STUDIES the proliferative pressure on epidermal cells induce the development of cancer in the wounded skin. Indeed, JEB patients have an increased risk (1:4) of developing SCC (Yuen and Jonkman, 2011; Montaudie et al., 2016). These tumors have a high recurrence rate and an aggressive course that results in premature death in approximately 20% of patients (Yuen and Jonkman, 2011). Transgenic epidermal cultures have proven to be lifesaving in Patient 3, a 7-year-old boy carrying a homozygous acceptor splice site mutation (c.1977-1G > A) within intron 14 of LAMB3 (Hirsch et al., 2017). Due to the severity of the disease and to infection with Staphylococcus aureus and Pseudomonas aeruginosa shortly after admission to hospital, he suffered complete epidermal loss on approximately 80% of his total body surface and the prognosis was very poor. Virtually his entire skin was replaced by approximately 1 m2 of transgenic cultured epidermis, the majority of which consisted of Hologene 5 (Hirsch et al., 2017). At the last follow-up (5 years after transplantation), the patient’s newly formed epidermis expressed normal levels of laminin 332, had an intact basement membrane, normal thickness, and morphology of hemidesmosomes, did not develop blisters or erosions, remained robust and resistant to mechanical stress and unveiled normal wound healing upon injury. The regenerated epidermis was entirely transgenic, as LAMB3 mRNA and laminin 332 were uniformly and seamlessly detected in all the analyzed skin sections. No adverse events, immune response or inflammation were observed (Kueckelhaus et al. manuscript submitted). Intermediate JEB caused by mutations in the gene LAMB3 (LAMB3-JEB) was the first genetic skin disease successfully tackled by combined ex vivo cell and gene therapy. Autologous transgenic epidermal cultures were grafted on selected skin areas in two adult and one pediatric patients (Mavilio et al., 2006; Bauer et al., 2017; Hirsch et al., 2017; Table 1). The first two patients were treated using transgenic epidermal cells grown on plastic. Cohesive epidermal sheets were detached from the vessel and mounted on petrolatum gauze (Mavilio et al., 2006; Bauer et al., 2017). Patient 3 was also treated with grafts derived from the patient’s own epidermal stem cells corrected by retroviral gene transfer and with transgenic keratinocytes grown on a fibrin substrate (Hologene 5). RATIONALE AND PREVIOUS CLINICAL STUDIES Previous in vitro experiments and clinical application of plastic- and fibrin-cultured grafts on full-thickness skin burns have unambiguously shown that these culture systems are fully equivalent, both in terms of biological parameters (including preservation of epidermal stem cells) and clinical performance. Indeed, both plastic- and fibrin-cultured transgenic epidermal sheets were able to engraft and permanently and fully restore a functional epidermis (Pellegrini et al., 1999). Cultivation on fibrin has multiple advantages, including a better coordination between culture timing and surgery, an easier handling and long-distance transportation of the grafts, no time-consuming procedure for the detachment of the cultures, and, more importantly, it avoids the dispase- dependent shrinking of the epidermal sheet. The last feature allows preparation of 144 cm2 grafts using the same number of clonogenic keratinocytes and the same amount of culture media needed to produce 50–60 cm2 plastic-cultured grafts (Pellegrini et al., 1999). Thus, for this trial, we will use only Hologene 5. Patients 1 and 3 are currently included in an observational study (HTA-HG5-01-OBS) started in July 2018 and still ongoing to collect further safety and efficacy data (Table 1). fi These findings do not only attest a most likely permanent (16 years so far) functional restoration of the epidermal-dermal junction but also reveal a promising safety profile. Indeed, despite the genotoxic risks linked to gamma-retroviral vector (γRV) insertional mutagenesis reported for hematopoietic stem cells (Wu and Dunbar, 2011), no clonal selection or immortalization events have been observed in epidermal stem cells. Altogether, the three JEB patients received ~4 × 108 transgenic clonogenic keratinocytes, none of which provided evidence of clonal expansion, within a time frame that entailed over 150 epidermal renewal cycles (Mavilio et al., 2006; De Rosa et al., 2014; Bauer et al., 2017; Hirsch et al., 2017). Indeed, neither tumor development nor other adverse events have been observed in any of the treated areas. These data suggest that insertional mutagenesis is an extremely rare event (if present at all) in human epidermal cells. Most likely, to trigger tumor formation, insertional mutagenesis will require other oncogenic factors, which could be related to cell type, patient’ s individual genetic background, age, disease, transgene, or other mutations (Howe et al., 2008; Cavazza et al., 2013). In fact, we have never observed any immortalization or transformation event in cultured primary clonogenic keratinocytes despite hundreds of transduction procedures performed in 3 decades of basic research. RATIONALE AND PREVIOUS CLINICAL STUDIES Recessively inherited JEB is caused by mutations in three genes, LAMA3, LAMB3, or LAMC2, that jointly encode laminin 332 (a heterotrimeric protein consisting of α3, β3, and γ2 chains, also known as laminin 5) and in genes encoding collagen XVII, α6β4, and α3 integrins. Laminin 332 mutations are usually associated with the most severe forms of JEB and mainly affect LAMB3 (Fine et al., 2014; Has et al., 2020). Absence of laminin 332 due to deleterious (nonsense) mutations causes early lethality and defines severe JEB (also known as Herlitz form). In non-lethal intermediate JEB, laminin 332 is strongly reduced and hemidesmosomes are rudimentary or absent, which lead to recurrent blisters and chronic erosions greatly impairing patients’ quality of life (Kiritsi et al., 2013). More than 40% of patients suffering from intermediate JEB die before adolescence (Fine et al., 2008). The continuous inflammation, the disruption of the adhesive machinery and TABLE 1 \| Summary of clinical data on previously treated patients. 1° Mavilio et al., 2006 2° Bauer et al., 2017 3° Hirsch et al., 2017 Type of application Phase I/II Single-case study Compassionate use Year of transplantation 2005 2014 2015 Type of mutation Compound heterozygous: c.628G>A; c.29insC Compound heterozygous: c.1903C>T; c.3009C>T Homozygous: c.1977-1G>A Total graft size Ca. 500 cm2 (0.05 m2) Ca. 80 cm2 (0.008 m2) 0.85 m2 Transplanted area Right and left legs Right leg Arms, legs, flanks, back skin, and thigh % of re-epithelization 100% 100% 100% Admission After treatment Presence of the transgene (mRNA and protein) YES YES YES Years of follow-up post treatment 16 2 (Patient not included in the observational study) 6 Figures from Mavilio et al., 2006 and Bauer et al., 2017 published with permission of Elsevier and Springer Nature. 0.85 m2 Arms, legs, flanks, back skin, and thigh 100% d YES YES YES 16 2 (Patient not included in the observational study) 6 YES 6 6 (Patient not included in the observational study) Figures from Mavilio et al., 2006 and Bauer et al., 2017 published with permission of Elsevier and Springer Nature. September 2021 \| Volume 12 \| Article 705019 2 De Rosa et al. Hologene 5 this patient and no adverse events were reported after 2 years follow-up (Bauer et al., 2017; Table 1). this patient and no adverse events were reported after 2 years follow-up (Bauer et al., 2017; Table 1). Frontiers in Genetics \| www.frontiersin.org Trial Design In March 2015, Hologene 5 received an orphan designation (EU/3/15/1465) by the European Commission. This trial is a multicenter, open-label, uncontrolled phase II/III study designed to confirm efficacy and safety of Hologene 5 in patients with LAMB3-related JEB. RATIONALE AND PREVIOUS CLINICAL STUDIES the rationale for performing this trial also in children – besides the obvious advantage of promptly restoring a functional epidermis, hence, preventing recurrent blisters, erosion, and chronic inflammation – arises from the likely assumption that gene therapy could actually decrease the risk of skin cancer development in the transplanted areas. Despite these encouraging preliminary results, such experimental approaches are still far from a routine therapy. Therefore, the Hologene 5 trial aims to confirm the efficacy and to strengthen the safety of combined ex vivo cell and gene therapy of LAMB3-related JEB. If successful, this clinical trial would pave the way for the approval of Hologene 5 as a therapeutic option for at least one subgroup of patients with intermediate JEB. Hologene 5 will be delivered to the clinical center using a temperature controlled proprietary packaging system that preserves the graft up to 36 h after packaging. Hologene 5 is intended for transplantation onto surgically prepared blistering skin areas of JEB patients to obtain a permanent regeneration of a healthy, functional, and renewing epidermis. Each patient will be treated with one or more grafts according to wound size and number, and each transplanted area will be considered as a single treatment regardless the number of grafts used (see below). RATIONALE AND PREVIOUS CLINICAL STUDIES Similarly, a group at Stanford University treated seven patients with epidermal cultures genetically corrected with a γRV carrying the COL7A1 gene, and no adverse events were observed after 5 years of follow-up, irrespective of patients’ age (Siprashvili et al., 2010, 2016; Eichstadt et al., 2019). Patient 1, a 36-year-old man, carries a null allele of LAMB3 and a second allele with the LAMB3 missense mutation E210K, which impaired normal assembly of laminin-332 (Mellerio et al., 1998; Posteraro et al., 2004). He showed blisters, arising spontaneously or upon minimal mechanical stress or trauma, on the majority of his body surface. In October 2005, the anterior upper parts of his legs were successfully treated with transgenic epidermal cultures (Mavilio et al., 2006; Table 1). A fully functional epidermis has been restored and no adverse events (AEs) were recorded during 6.5 years follow-up (De Rosa et al., 2014; Table 1). An ongoing observational study confirms both the stability of the transgenic epidermis and the absence of AEs 16 years after transplantation. Patient 2, a 49-year-old woman, is a compound-heterozygous carrier of splice-site mutation (c.3009C > T) and a nonsense mutation (R635X) in LAMB3 (Buchroithner et al., 2004). She had a large non-healing chronic wound on her lower right leg that was treated in June 2014. A functional epidermis could be restored also in Given the high likelihood of JEB and RDEB patients to develop aggressive SCC, often leading to their premature death, September 2021 \| Volume 12 \| Article 705019 Frontiers in Genetics \| www.frontiersin.org 3 De Rosa et al. Hologene 5 144 cm2 epidermal graft. It consists of fibrin-cultured epidermal keratinocytes transduced with a γRV carrying a LAMB3 cDNA (Figure 1). Autologous primary clonogenic keratinocytes will first be expanded ex vivo and genetically corrected with the LAMB3-γRV (Drug Substance). Hologene 5 will then be prepared by plating 50.000–150.000 transduced cells/cm2 on a 144 cm2 fibrin support to generate transplantable grafts, each containing 20–30.000.000 viable transgenic keratinocytes (Drug Product; Hirsch et al., 2017; Figure 1). Such grafts must contain an adequate number of transduced epidermal stem cells, which are instrumental to achieve a long-lasting skin regeneration and proper wound healing. Recruitment, Trial Population, and Eligibility Criteriah Product Review and Drug Formulation Hologene 5 is an Investigational Medicinal Product (IMP; classified as Advanced Therapy Medicinal Product, ATMP), manufactured under Good Manufacturing Practice (GMP) conditions by Holostem Terapie Avanzate s.r.l. as a sterile, Product Review and Drug Formulation In case only one criterion is met, a patient-reported outcome (PRO) will be considered (2 step). If PRO is positive, the treatment can still be defined as Success. FIGURE 2 \| Flowchart describing assessment of Hologene 5 efficacy. The evaluation of Hologene 5 efficacy is the primary endpoint of this clinical trial and will follow a two-step rule. At the first step, clinical (percentage of re- epithelialization) and functional (presence of a transgenic functional epidermis) parameters will be evaluated. If both criteria are met, the treatment will be defined as Success; if none of them is fulfilled, the treatment will be scored as Failure. In case only one criterion is met, a patient-reported outcome (PRO) will be considered (2 step). If PRO is positive, the treatment can still be defined as Success. • Presence of any skin cancers in the area(s) qualifying for treatment • Female subject who is pregnant • Presence of (i) systemic diseases, (ii) clinically significant or unstable concurrent disease, (iii) other concomitant medical conditions, and (iv) other clinical contraindications to stem cell transplantation, based on investigator’s judgment, and consultation with the sponsor's medical expert (exploratory endpoint), will be confirmed at Visit 5 (see below). If multiple control areas are used and the clinical performance of the treated areas is satisfactory, some of the control areas can be considered for Hologene 5 transplantation, even before the end of the trial.i • Previous treatments or clinical trials envisaging the use of cells (including bone marrow transplantation) and/or in vivo or ex vivo gene therapy products • Previous treatments or clinical trials envisaging the use of cells (including bone marrow transplantation) and/or in vivo or ex vivo gene therapy products Hologene 5 will be administered by qualified surgeons under standard sterile operating conditions. Transplantation can be done under local or general anesthesia. To remove remnants of affected epidermis, Erbium YAG laser or copper filament sponge or surgical knife can be used. Hologene 5 will be applied on the prepared wound bed; the grafts will be covered with petrolatum gauze overlaid by a firm but non-compressive standard bandage to ensure proper engraftment and fibrin absorption. After treatment, patients should preferably remain hospitalized and the treated areas are immobilized up to 2 weeks (if deemed necessary). The immobilization of the treated body part is required to ensure proper engraftment. Control skin areas will be treated with standard medications and bandages. Study Treatment Skin areas suitable for transplantation will be identified and evaluated during the screening visit and monitored at each study visit. The selection of such areas shall be based on specific parameters like severity (areas with open wounds and blistering should be chosen; see inclusion criteria), impact on quality of life (for instance pain and recurrent infections), accessibility for local anesthesia, ease of immobilization, and management after grafting. The patient’s opinion will also be taken into account.h Product Review and Drug Formulation The selection will be carried out according to stringent inclusion/ exclusion criteria, as summarized in Table 2. If the criteria are met, patients of both genders and of any ethnic origin will be considered for the study. Selected patients will receive a personal invitation from the local investigators and will only be included after written informed consent. They will be followed at the study site where they were enrolled. Patients can withdraw from the study at any time. Frontiers in Genetics \| www.frontiersin.org Product Review and Drug Formulation TABLE 2 \| Eligibility criteria for Hologene 5. TABLE 2 \| Eligibility criteria for Hologene 5. Inclusion criteria • Male and female patients between 6 months and 65 years old • Diagnosis of intermediate LAMB3- related JEB, confirmed by DNA sequencing and/or immunofluorescence analysis of skin biopsy sample(s) • Presence of blisters and/or erosions ≥ 6 cm2, persistent or recurrent for more than 3 months • Cooperative attitude and patient's compliance with scheduled visits • Detectable residual expression of laminin-332 and its beta-3 chain by immunofluorescence and/or Western blot analysis Exclusion criteria • Known or suspected intolerance to anaesthesia or to study medication excipients or other material required by study protocol • Bad general condition (ECOG index >1) and/or acute systemic infection at the time of screening (patient can be re- screened after appropriate treatment) • Presence of any skin cancers in the area(s) qualifying for treatment • Female subject who is pregnant • Presence of (i) systemic diseases, (ii) clinically significant or unstable concurrent disease, (iii) other concomitant medical conditions, and (iv) other clinical contraindications to stem cell transplantation, based on investigator’s judgment, and consultation with the sponsor's medical expert • Previous treatments or clinical trials envisaging the use of cells (including bone marrow transplantation) and/or in vivo or ex vivo gene therapy products • Diagnosis of intermediate LAMB3- related JEB, confirmed by DNA sequencing and/or immunofluorescence analysis of skin biopsy sample(s) • Presence of blisters and/or erosions ≥ 6 cm2, persistent or recurrent for more than 3 months • Cooperative attitude and patient's compliance with scheduled visits • Detectable residual expression of laminin-332 and its beta-3 chain by immunofluorescence and/or Western blot analysis Exclusion criteria • Known or suspected intolerance to anaesthesia or to study medication excipients or other material required by study protocol • Bad general condition (ECOG index >1) and/or acute systemic infection at the time of screening (patient can be re- screened after appropriate treatment) FIGURE 2 \| Flowchart describing assessment of Hologene 5 efficacy. The evaluation of Hologene 5 efficacy is the primary endpoint of this clinical trial and will follow a two-step rule. At the first step, clinical (percentage of re- epithelialization) and functional (presence of a transgenic functional epidermis) parameters will be evaluated. If both criteria are met, the treatment will be defined as Success; if none of them is fulfilled, the treatment will be scored as Failure. Product Review and Drug Formulation This trial is designed to enroll at least six LAMB3-JEB patients (four pediatric and two adult). Their enrollment will occur at study sites, where the respective patients have already been registered and cared for. Each center will identify and pre-screen suitable patients taking into account their clinical history. FIGURE 1 \| Scheme describing the Hologene 5 clinical trial. Junctional EB (JEB) patients will be enrolled at the clinical centers and skin biopsies will be sent to Holostem Terapie Avanzate s.r.l for Good Manufacturing Practice (GMP) manufacturing of Hologene 5. Clonogenic epidermal cells will be cultivated and genetically corrected with a gamma-retroviral vector (γRV) expressing LAMB3. Hologene 5 will be delivered to the clinical center for transplantation in V5. Safety and efficacy of Hologene 5 will be assessed for 12 months based on a combined evaluation of clinical performance, molecular and functional parameters, and patient perception. Treated patients will be monitored for additional 15 years in an interventional study. FIGURE 1 \| Scheme describing the Hologene 5 clinical trial. Junctional EB (JEB) patients will be enrolled at the clinical centers and skin biopsies will be sent to Holostem Terapie Avanzate s.r.l for Good Manufacturing Practice (GMP) manufacturing of Hologene 5. Clonogenic epidermal cells will be cultivated and genetically corrected with a gamma-retroviral vector (γRV) expressing LAMB3. Hologene 5 will be delivered to the clinical center for transplantation in V5. Safety and efficacy of Hologene 5 will be assessed for 12 months based on a combined evaluation of clinical performance, molecular and functional parameters, and patient perception. Treated patients will be monitored for additional 15 years in an interventional study. September 2021 \| Volume 12 \| Article 705019 4 Frontiers in Genetics \| www.frontiersin.org Frontiers in Genetics \| www.frontiersin.org Hologene 5 De Rosa et al. FIGURE 2 \| Flowchart describing assessment of Hologene 5 efficacy. The evaluation of Hologene 5 efficacy is the primary endpoint of this clinical trial and will follow a two-step rule. At the first step, clinical (percentage of re- epithelialization) and functional (presence of a transgenic functional epidermis) parameters will be evaluated. If both criteria are met, the treatment will be defined as Success; if none of them is fulfilled, the treatment will be scored as Failure. In case only one criterion is met, a patient-reported outcome (PRO) will be considered (2 step). If PRO is positive, the treatment can still be defined as Success. Primary Endpointh The primary endpoint is the proportion of transplantation success after 12 months of treatment, defined through a two-steps rule (Figure 2). An appropriate assessment of the efficacy of Hologene 5 cannot simply rely on re-epithelialization (as stand-alone primary endpoint), since JEB skin lesions can heal spontaneously, though temporarily. Thus, in order to rigorously evaluate Hologene 5 efficacy, the primary endpoint (Figure 2) will take into account not only clinical parameters (as the percentage of re-epithelialization measured by the Investigator) but also cellular, molecular, and functional parameters (by means of in situ hybridization, PCR, immunofluorescence and transmission electron microscopy) to verify the presence of LAMB3 mRNA, integrated transgene, laminin 332 and β3 proteins and mature hemidesmosomes, mechanical stress assays (as stripping tests), and PRO [Likert scale to provide their perception of treatment success on the transplanted area(s)]. At Visit 1, the complete medical and surgical history of the patient is recorded. Skin physical examination will be performed and digital skin photographs will be collected using digital imaging software (iMitoWound; Figure 3). All areas suitable for transplantation will be scored using EBDASI (Visit 1 to Visit 4). After ICF/IAF signature, AE will be registered and assessed through the study (up to Visit 11) as well as any changes in concomitant medications. A self-monitoring disease assessment has been introduced in this study in order to provide a fast track record and report every adverse event. The patient is provided with a tablet containing a user-friendly App (HoloApp, developed by the Sponsor) as patients’ diary for self-monitoring of their skin. HoloApp is designed for remote collection of digital images of the area(s) before and after transplantation. HoloApp will help patients to record any changes in the concomitant medications and wounds management and any adverse event to foster a quick Investigator’s action, if needed. Pictures of the transplanted and control area(s) are directly collected in a structured and pseudo-anonymized archive. HoloApp is not aimed to provide any diagnosis or to evaluate the Quality of Life. Study Protocol Outline At least six patients, four pediatric (6 month- to 17-year-old), and two adults will be enrolled in the study. During the screening phase (Visit 1–2), the subjects will be examined at the centers involved to confirm the clinical diagnosis of intermediate JEB. They will be properly and fully informed and invited to sign the informed consent/assent (ICF/IAF). For children only, parents shall act as deputy. In the absence of a molecular diagnosis, a blood sample will be taken to determine the mutation through NGS analysis and confirm the involvement of the LAMB3 gene. Two skin punch biopsies will be taken in order to verify the residual expression of laminin 332 and β3 chain by both immunofluorescence on skin sections and Western blot on primary keratinocytes cultures (Table 2, inclusion criteria). All clinical and molecular data previously collected from the patients will be analyzed (Visit 1–2). Whenever possible, previous testing that is documented and satisfies screening requirements will be considered, in order to limit the number of skin biopsies and the quantity of blood required from the patient. The exploratory objectives are the assessment of safety and efficacy of Hologene 5 based on intra-patient’s comparison and the percentage of patients with successfully treated areas, absence of immune response against the therapy and amelioration of patients’ quality of life. Secondary Endpointsh The percentage of re-epithelialization will be assessed through clinical inspection and computer-assisted image analysis (iMitoWound digital imaging software), at 1, 3, 6, 9, and 12 months by the Investigator and a trained Independent Assessor. The absence of sub-blisters in the treated areas will be evaluated by Optical Coherence Tomography (OCT) at the end of the study. Changes in quality of treated area(s), from baseline to 12 months follow-up, will be analyzed according to the Epidermolysis Bullosa Disease Activity and Scarring Index (EBDASI) validated questionnaire (Loh et al., 2014). The incidence of adverse events (TEAEs), serious adverse events (SAEs), event of special interest (AESI), and adverse drug reactions (ADRs), vital signs and laboratory parameters will also be analyzed as secondary safety endpoints. At Visit 3 (Day −60 ± 30 before transplantation, Figure 3), one or two skin biopsy sample(s) of 2–9 cm2 will be taken from non-blistering areas and sent to Holostem Terapie Avanzate s.r.l for Hologene 5 manufacturing. If needed, one or two (depending on the quality of the donor’s epidermal tissue) additional punch biopsies will be performed on the affected area for transmission electron microscopy, in situ hybridization and/or PCR analysis. Primary, Secondary, and Exploratory Objectivesh The primary objective of the clinical trial is to evaluate the efficacy of Hologene 5, defined as proportion of success based on clinical performance, cellular, molecular, and functional parameters, and patient-reported outcome (PRO).fi tth The area(s) to be treated by transplantation, as well as the control area to be used for intra-patient treatment comparison Secondary objectives are the efficacy and the safety of Hologene 5 according to combined visual and microscopic September 2021 \| Volume 12 \| Article 705019 Frontiers in Genetics \| www.frontiersin.org 5 De Rosa et al. Hologene 5 analysis of the re-epithelialized area, measured both by the Investigator and an Independent Assessor.h Frontiers in Genetics \| www.frontiersin.org Exploratory Endpoints As exploratory endpoints, patients will be monitored for the presence of any immune response (B- or T-cell-mediated) against the transgene. Change in patients quality of life will also be evaluated through a specific questionnaire, Quality of Life Evaluation in Epidermolysis Bullosa (QOLEB; Frew et al., 2009), in case of multiple treatments involving a large body area. Finally, an intra-patient comparison will be performed between treated and non-treated areas and the percentage of patients with one or more successful transplantation(s) among all treated patients will be determined. At Visit 4 (3–7 days before transplantation), the patient will be hospitalized and prepared for the surgical procedure. Areas to be treated by transplantation will be analyzed through iMitoWound and smear/swabs for bacterial cultures will be taken from the selected areas as well as from axilla, groin, and nose. In case of any positivity, based on Investigator’s judgment, a specific antibiotic treatment will be initiated. QOLEB will September 2021 \| Volume 12 \| Article 705019 Frontiers in Genetics \| www.frontiersin.org 6 De Rosa et al. Hologene 5 FIGURE 3 \| Study visit schedule. This scheme describes the main activities performed at each visit. Pre-treatment phase is from approximately 6 months (V1) to approximately 7 days before the transplantation (V4). Hologene 5 transplantation will be performed during V5 (Day 1). Follow-up visits are planned from approximately 10 days after the grafting (V6) to the endpoint, 12 months after the transplantation (V11). FIGURE 3 \| Study visit schedule. This scheme describes the main activities performed at each visit. Pre-treatment phase is from approximately 6 months (V1) to approximately 7 days before the transplantation (V4). Hologene 5 transplantation will be performed during V5 (Day 1). Follow-up visits are planned from approximately 10 days after the grafting (V6) to the endpoint, 12 months after the transplantation (V11). patients treated with transgenic epidermal cultures (see previous clinical data) did not require such medications.ttt be administered to patients undergoing transplantation on large body areas (or candidates for multiple transplantations) and its outcome will be assessed 12 months after the last transplantation. At Visit 6 (8–10 days after grafting), Hologene 5 engraftment will be evaluated both through physical examination and iMitoWound analysis. A blood sample will be collected to monitor potential immune reactions (presence of IgM and/or IgG to Laminin 332). The EBDASI questionnaire will be administered at each visit, from Visit 6 to Visit 11. Exploratory Endpoints p At Visit 5, Hologene 5 transplantation will be performed (Day 1). Physical examination [e.g., Eastern Cooperative Oncology Group (ECOG) index] and vital signs collection (SBP, DBP, PR, and temperature) will be performed. A collection of digital images of control and areas to be treated will be taken through iMitoWound. These images will be considered as the reference baseline for the calculation of the percentage of re-epithelialization from Visit 6 onward. Clinical staff in the operating room will take the pictures before and after preparation of the wound bed. The patient’s immune profile (presence of IgM and/or IgG against laminin 332 and T cell profile) will be assessed. A peripheral blood sample and one punch biopsy from peri-lesional skin wound areas selected for transplantation will be obtained and specific tests will be performed to monitor the immune response (see below).i At Visit 7 (approximately 30 days after transplantation), the Investigator will evaluate the percentage of re-epithelialization and the presence of blisters and erosions through physical examination and iMitoWound analysis, both in treated and control area(s). In case of any adverse reaction (e.g., inflammation, turgor, or any other suspicious signs), a punch biopsy and a blood sample may be taken for immunological assessment (Figure 3).t At Visit 8 (approximately 90 days after transplantation), the Investigator will evaluate the re-epithelialization of both treated and control area(s) through physical examination and iMitoWound analysis. To confirm and validate Hologene 5 efficacy, punch biopsies of the treated area will be used to investigate the expression of the transgene and the proper assembly of laminin 332. In situ hybridization and/or PCR analysis will be performed using a transgene-specific probe and immunofluorescence investigations will be conducted on 7-μm skin sections using both β3- and laminin 332-specific antibodies. A blood sample will be collected for immunological tests (presence of IgM and/or IgG against laminin 332). Hologene 5 will be transplanted by qualified surgeons under standard sterile operating condition on the surgically prepared wound bed. Transplantation can be done under local or general anesthesia. Smear/swabs for bacterial cultures from all areas to be treated by transplantation will be taken to ensure, in the case of positivity, proper post-treatment antibiotic medication. At each administration, one or more Hologene 5 graft will be transplanted. Each graft will be adapted to the size of the skin lesion by removing the remnants using surgery scissors. Frontiers in Genetics \| www.frontiersin.org Safety and Laboratory Assessments Safety and Laboratory Assessments As indicated at each Visit, a set of morphological, functional, and molecular tests will be performed to evaluate the safety and efficacy of Hologene 5. Standard Operating Procedures are set in the specialized laboratories involved in the study, in agreement with current Good Laboratory Practices rules. Immunofluorescence, in situ hybridization and PCR analyses will take place at the Centre for Regenerative Medicine (CMR) at the University of Modena and Reggio Emilia. The absence of replication-competent retrovirus (RCR) will be confirmed by qPCR on blood samples at CMR. Transmission Electron Microscopy will be performed at the laboratory of Biology and Pathology of soft Connective Tissues (BioPaCT) at the same University. Exploratory Endpoints At this time point, the Investigator will evaluate the final clinical efficacy through physical examination and iMitoWound analysis, both in treated and control area(s). The skin architecture will be investigated through OCT, which allows the detection of epidermal sub-blisters or microscopic blisters that are not identifiable by visual inspection. The regeneration of a transgenic epidermis will be confirmed at a molecular level by in situ hybridization and/or PCR (using a transgene-specific probe) and immunofluorescence (using β3- and laminin 332-specific antibodies) performed on punch biopsies. In addition, the proper assembly of a correct number of mature hemidesmosomes will be evaluated by transmission electron microscopy. Finally, mechanical properties of the regenerated transgenic epidermis will be assessed by a stripping test assay. A PRO will be collected using a 1–5 Likert scale based on patient’s self-perceived level of improvement of the treated area(s) and the QOLEB will be administered to patients who underwent transplantation on large body areas or multiple grafting procedures. Exploratory Endpoints The treated areas will be immobilized and covered with Adaptic petrolatum gauze overlaid by non-compressive standard bandage. At Visit 9 (day 180) and Visit 10 (day 270), the patient will be clinically monitored through physical examination and iMitoWound analysis. To verify the mechanical strength of the regenerated skin, a stripping test will be performed, but only in case of evident clinical success of the treatment. Based on Investigator’s judgment, in consultation with the Sponsor Based on Investigator’s judgment, in consultation with the Sponsor Medical Expert, patients may receive (if needed) post- transplantation medications, as corticosteroids, immune- suppressants, and/or immune-modulators. In fact, the first three September 2021 \| Volume 12 \| Article 705019 7 De Rosa et al. Hologene 5 Patient-specific baseline reactivity will be defined from plasma samples taken before grafting (Visit 4) and compared with samples collected after transplantation, from Visit 6 to Visit 11. To investigate a potential local immune response, direct immunofluorescence (DIF) for IgM or IgG detection will be performed on skin sections derived from the selected areas, before and after transplantation. To evaluate patients’ β3-specific T-cell reactivity, a set of cytokines, including IFN-γ, IL-4/IL-13, and IL-17, and CD107a will be investigated before and after transplantation through ELISpot and flow cytometry assays. Percentage of CD4+ T regulatory and CD8+ T follicular helper cells will be also assessed. A minimum of 10–15 ml of blood (three aliquots) is required from pediatric patients (based on age) and 30 ml (six aliquots) from adult patients. Minors below the age of 2 years are excluded from this test. Medical Expert, additional punch biopsy samples can be taken to confirm the expression of the transgene and the proper assembly of laminin 332 by in situ hybridization and/or PCR and immunofluorescence. Such biopsies will be avoided in case of evident failure or if they are considered unnecessary. The study will end at Visit 11, 12 months after the transplantation (Figure 3). At this time point, the Investigator will evaluate the final clinical efficacy through physical examination and iMitoWound analysis, both in treated and control area(s). The skin architecture will be investigated through OCT, which allows the detection of epidermal sub-blisters or microscopic blisters that are not identifiable by visual inspection. hi The study will end at Visit 11, 12 months after the transplantation (Figure 3). Frontiers in Genetics \| www.frontiersin.org Data Storage All parties will ensure the protection of patients’ personal data and identifiable personal data will not be included in any report, publication, or other disclosure, except when required by law. Data will be securely stored, both on paper and on password-protected computers, with locked doors and shelves, accessible only to dedicated personnel. After study completion, all documents and study data will be kept by the Investigators in a secure and ordered study file for a minimum of 30 years. It is the responsibility of the Sponsor to inform the Investigators on timing and modalities of data storage. The Investigators will contact the Sponsor before destroying any trial- related documentation. t Punch biopsies on transplanted areas will be avoided in case of evident failure of the treatment or if considered unnecessary. If needed, additional punch biopsies, as well as blood samples, can be taken and used to assess a potential immune reaction (presence of IgM and/or IgG against laminin 332). CONCLUSION Hologene 5 is an ATMP consisting of fibrin-cultured epidermal sheets generated by transgenic clonogenic keratinocytes, including epidermal stem cells. Given the complexity of the technology, the clinical features of JEB and the quite high cost of ATMPs, the cumbersome development of a product like Hologene 5 becomes acceptable if a full and permanent restoration of a functional epidermis is achieved (Mavilio et al., 2006; De Rosa et al., 2014; Bauer et al., 2017; Hirsch et al., 2017). Although, previous studies have shown that this can indeed be the case, we will be very stringent in determining the criteria for claiming efficacy of Hologene 5 and use a two-steps rule based on combined evaluation of primary endpoints, such as clinical performance, molecular and functional parameters, and patient perception. In fact, because of the intrinsic nature of the pathology, re-epithelialization of the treated areas, though being a necessary and crucial parameter, per se is not sufficient to score a positive clinical outcome, hence to claim efficacy of treatment. Nowadays, JEB (and RDEB) patients rely on many sophisticated supportive medications fostering a satisfactory, yet only temporary, wound closure (especially when patients are hospitalized) leading to continuous rounds of blistering and healing (Mellerio et al., 1998; Goldschneider et al., 2014; Bruckner-Tuderman, 2019). Hence, the simple measurement of the percentage of re-epithelialization in treated areas at a given moment could lead to a misleading positive score of Hologene 5 efficacy.fi Treatment emergent events will be tabulated by system organ class (SOC) and preferred term (PT) using the MedDRA dictionary. Vital signs (actual values and change from baseline, if applicable) will be presented by time point. Safety laboratory data (actual values and change from baseline, if applicable) will be presented by time point.i A detailed statistical analysis plan will be finalized before database locking. Sample Size Calculationh This study is designed to enroll at least six patients, with no formal estimation of sample size performed. The total number of six patients was defined based on the incidence of LAMB3- dependent intermediate JEB in Europe as indicated by Orphanet (Fine, 2010; Bardhan et al., 2020). Given the exploratory nature and the characteristics of this study, rather than testing a formal hypothesis, a justification based on epidemiologic data was considered more appropriate than a power calculation in the traditional fashion or using an estimation approach through the precision of confidence intervals. The target population is a subgroup of the JEB, referred to as intermediate JEB. Intermediate JEB affects about 0.1 in 1 million people. Approximately 80% of such patients have a mutation in LAMB3, the gene encoding the β3 chain of laminin 332 (Kiritsi et al., 2013). This makes the estimate to be around 0.06–0.08 patients per million, i.e., no more than 40 patients in the total EU population (data at 1-01-2019; EUROSTAT, 2019). Intermediate JEB patients show an increased risk of developing SCC, especially in adulthood (Fine et al., 2008; Yuen and Jonkman, 2011; Fine, 2012). Since the presence of SCC in the areas chosen for transplantation is an exclusion criterion and only patients expressing detectable amounts of laminin 5 are enrolled, the total number of six patients (four Immunological tests to monitor both immune response and specific T-cell activation against β3 chain and/or laminin 332 will take place at the MiTiCi laboratory, San Raffaele Hospital in Milan (Italy). Blood samples will be collected both pre- and post-transplantation for standard blood tests and detection of infectious agents (including SARS-CoV-2). The volume of blood samples will be compliant with European Ethical rules, in particular for pediatric patients.l To evaluate the B cell response, indirect immunofluorescence (IIF), and ELISA will be used to test for presence of autoantibodies against laminin 332 (Abs-LM332) in the patient’s plasma. September 2021 \| Volume 12 \| Article 705019 Frontiers in Genetics \| www.frontiersin.org 8 Hologene 5 De Rosa et al. laboratory-specific standards. All legal and regulatory requirements will be accomplished. laboratory-specific standards. All legal and regulatory requirements will be accomplished. pediatric and two adults) can be considered as a realistic estimate of the eligible patients, although this number is exceptionally small. Governance and Monitoring Holostem Terapie Avanzate s.r.l. is responsible for the governance of the trial. A monitoring plan has been established. All monitoring will be performed by an independent contract research organization (CRO). The CRO will contact the investigator/center before and during the study and survey the study records (including eCFR, study file, and source data) to monitor the progress and the management of the study and to address and discuss any emergent problem. Planned Statistical Analyses In patients with JEB, skin wounds are more likely to become infected than in otherwise healthy individuals, and entire epithelial grafts may get lost due to secondary infection despite successful transplantation. This might have impact on the primary endpoint of the trial. Therefore, rigorous systemic antibiotic treatment of any documented or suspected wound infection will be required. General descriptive statistics for continuous variables will include number of observations (n), mean, SD, median, interquartile range (IQR), minimum and maximum values. Categorical data will be summarized by frequency and percentages.h The primary endpoint as the proportion of transplantation success (according to the two-steps rule) at 12 months will be presented together with the two-sided 95% CI. The analysis of the proportion of transplantation successes will be performed in the Full Analysis Set (FAS; as primary analysis) at transplantation level, that is, all transplantations of all patients who undergo at least one application of the investigational product and in the Per Protocol Set (as sensitivity analysis) at transplantation level. Registration and Trial Statushi g The clinical study protocol is identified by the EUDRACT number 2018-000261-36 and has been included in the EU Clinical Trial Registry. Additional registration in the ClinicalTrials.gov database has also been initiated. g The clinical study protocol is identified by the EUDRACT number 2018-000261-36 and has been included in the EU Clinical Trial Registry. Additional registration in the ClinicalTrials.gov database has also been initiated. fi An overview of AEs with the number of patients affected, as well as the number of events, with any treatment emergent events (TEAEs), serious TEAEs, drug-related TEAEs, serious drug-related TEAEs, TEAEs leading to withdrawal, and TEAEs leading to death will be presented. Ethics and Disseminationh The study will be conducted in agreement with the Declaration of Helsinki for patient care. Before beginning the trial, all ethical and regulatory approvals will be obtained. Results of the study, both negative and positive, will be published and/ or presented at scientific meetings after the end of the study. The proportion of patients with one or more transplantation success (according to the two-steps rule) will be presented together with the two-sided 95% CI. This analysis will be performed on the FAS set at patient level. In addition to descriptive statistics, when applicable, appropriate non-parametric statistical tests will be provided for secondary efficacy variables.f Good Practices Previous clinical data generated using Hologene 5 in LAMB3-JEB patients have indeed shown that the long-term stability of a fully functional transgenic epidermis is solely sustained by a defined population of stem cells, which have been identified as holoclone-forming cells, represent only a small percentage (approximately 5%) of all clonogenic keratinocytes and continuously generate pools of short-lived progenitors eventually producing terminally differentiated suprabasal cells (Hirsch et al., 2017). Thus, transgenic holoclone-forming cells must be contained in Hologene 5 as well in any other transgenic epidermal culture aimed at ex vivo gene therapy of any form of EB. Hologene 5 represents probably the most advanced gene therapy approach for JEB. Hologene 5 is a complex product that combines cell therapy, gene transfer, and tissue engineering. This pivotal phase II/III clinical trial aims to strengthen its safety and confirm its efficacy, to obtain a formal approval of its use in patients affected by intermediate LAMB3-JEB. The knowledge and experience acquired during the development of Hologene 5 is currently driving several clinical trials tackling other forms of epidermolysis bullosa. ACKNOWLEDGMENTS The authors would like to thank Michele Palamenghi for his contribution to the Figure 1 layout. The authors would like to thank Michele Palamenghi for his contribution to the Figure 1 layout. gene in a junctional epidermolysis bullosa patient reveals exonic splicing and allele-specific nonsense-mediated mRNA decay. Lab. Investig. 84, 1279–1288. doi: 10.1038/labinvest.3700164 gene in a junctional epidermolysis bullosa patient reveals exonic splicing and allele-specific nonsense-mediated mRNA decay. Lab. Investig. 84, 1279–1288. doi: 10.1038/labinvest.3700164 FUNDING This work was supported by POR-FESR 2007-2013 and 2014-2020 Regione Emilia-Romagna (E8IJ10000120007, E92I16000220005, and E51F18000380009). AUTHOR CONTRIBUTIONS LDR and MDL designed the study protocol. CB, CM, and HS revised the study protocol. LDR and EE wrote the first draft of the manuscript. GZ planned the statistical analysis. All authors contributed to the article and approved the submitted version. g py y Keratinocyte cultures have long been safely used worldwide for life-saving permanent regeneration of a functional epidermis in massive full-thickness skin burns (Gallico et al., 1984; Pellegrini et al., 1999; De Luca et al., 2006) and for corneal restoration in ocular chemical burns associated to limbal stem cell deficiency (Pellegrini et al., 1997, 2013). Serious adverse events have never been reported. The transplanted cultured epidermis can be easily monitored throughout the lifetime of the patients and removed anytime, should adverse events occur. Accordingly, previous clinical studies using transgenic epidermal cultures have shown a good safety profile for both JEB and RDEB ex vivo gene therapy (Mavilio et al., 2006; De Rosa et al., 2014; Siprashvili et al., 2016; Bauer et al., 2017; Hirsch et al., 2017; Eichstadt et al., 2019). No adverse events have been reported, particularly no insertional mutagenesis (see Rationale for the clinical trial) and immune reaction against Good Practices Hologene 5 will be manufactured by Holostem Terapie Avanzate following full current GMP rules. All study procedures will adhere to Good Clinical Practices (GCP) and Good Clinical Laboratory Practice (GCLP) according to manufacturing or fi In order for Hologene 5 to be judged as efficacious, the newly formed epidermis should (i) be entirely transgenic, September 2021 \| Volume 12 \| Article 705019 Frontiers in Genetics \| www.frontiersin.org 9 De Rosa et al. Hologene 5 (ii) produce physiologic amounts of laminin 332 properly and seamlessly located within the basement membrane at the epidermal-dermal junction, (iii) restore a normal number of mature and functional hemidesmosomes, (iv) be robust, non-blistering, fully resistant to mechanical stress and able to heal wounds, and (v) be long-lasting (ideally for the lifetime of the patient), hence, contain engrafted transgenic epidermal stem cells allowing continuous self-renewal. In the three JEB patients treated so far, we have used these criteria for the evaluation of efficacy (Mavilio et al., 2006; De Rosa et al., 2014; Bauer et al., 2017; Hirsch et al., 2017; unpublished own data). the cultured cells. Nevertheless, we will include in this trial only patients carrying LAMB3 mutations that allow a residual, though minimal, expression of laminin 332. A potential immune reaction to Hologene 5 will be anyhow analyzed. In fact, the outcome of these studies might pose the basis for a potential use of Hologene 5 also in some selected severe forms of JEB, which are characterized by undetectable expression of laminin 332 (Kopp et al., 2005; Hammersen et al., 2016). In this respect, it has been shown that gentamicin can induce read-through of nonsense mutations and partially restore the expression of laminin 332 in patients with double null mutations in LAMB3 alleles, without triggering any immune reaction (Hammersen et al., 2019; Kwong et al., 2020). The remarkable clinical outcomes of regenerative medicine in renewing tissues, such as blood and squamous epithelia, accrue from a thorough characterization of their specific stem cells (De Luca et al., 2019; De Rosa et al., 2020). Thus, targeting stem cells is the cornerstone for successful ex vivo gene therapy of genetic diseases affecting the epidermis. gene in a junctional epidermolysis bullosa patient reveals exonic splicing and allele-specific nonsense-mediated mRNA decay. Lab. Investig. 84, 1279–1288. doi: 10.1038/labinvest.3700164 REFERENCES J. Dermatol. 139, 325–331. doi: 10.1046/j.1365-2133.1998.02377.x Eichstadt, S., Barriga, M., Ponakala, A., Teng, C., Nguyen, N. T., Siprashvili, Z., et al. (2019). Phase 1/2a clinical trial of gene-corrected autologous cell therapy for recessive dystrophic epidermolysis bullosa. JCI Insight 4:e130554. doi: 10.1172/jci.insight.130554 Br. J. Dermatol. 139, 325–331. doi: 10.1046/j.1365-2133.1998.02377.x EUROSTAT (2019). EU population up to over 513 millionon 1 January 2019. Fine, J. D. (2010). Inherited epidermolysis bullosa. Orphanet J. Rare Dis. 5:12. doi: 10.1186/1750-1172-5-12 Mellerio, J. E., El Hachem, M., Bellon, N., Zambruno, G., Buckova, H., Autrata, R., et al. (2020). Emergency management in epidermolysis bullosa: consensus clinical recommendations from the European reference network for rare skin diseases. Orphanet J. Rare Dis. 15:142. doi: 10.1186/s13023-020-01403-x Fine, J. D. (2012). Squamous cell carcinoma and junctional epidermolysis bullosa. J. Am. Acad. Dermatol. 66, 856–857. doi: 10.1016/j.jaad.2012.01.045 Montaudie, H., Chiaverini, C., Sbidian, E., Charlesworth, A., and Lacour, J. P. (2016). Inherited epidermolysis bullosa and squamous cell carcinoma: a systematic review of 117 cases. Orphanet J. Rare Dis. 11:117. doi: 10.1186/s13023-016-0489-9 Fine, J. D., Bruckner-Tuderman, L., Eady, R. A., Bauer, E. A., Bauer, J. W., Has, C., et al. (2014). Inherited epidermolysis bullosa: updated recommendations on diagnosis and classification. J. Am. Acad. Dermatol. 70, 1103–1126. doi: 10.1016/j.jaad.2014.01.903 Pellegrini, G., Rama, P., Matuska, S., Lambiase, A., Bonini, S., Pocobelli, A., et al. (2013). Biological parameters determining the clinical outcome of autologous cultures of limbal stem cells. Regen. Med. 8, 553–567. doi: 10.2217/rme.13.43 Fine, J. D., Johnson, L. B., Weiner, M., Li, K. P., and Suchindran, C. (2009). Epidermolysis bullosa and the risk of life-threatening cancers: the national EB registry experience, 1986-2006. J. Am. Acad. Dermatol. 60, 203–211. doi: 10.1016/j.jaad.2008.09.035 Pellegrini, G., Ranno, R., Stracuzzi, G., Bondanza, S., Guerra, L., Zambruno, G., et al. (1999). The control of epidermal stem cells (holoclones) in the treatment of massive full-thickness burns with autologous keratinocytes cultured on fibrin. Transplantation 68, 868–879. doi: 10.1097/00007890-199909270-00021 Fine, J. D., Johnson, L. B., Weiner, M., and Suchindran, C. (2008). Cause- specific risks of childhood death in inherited epidermolysis bullosa. J. Pediatr. 152, 276–280. doi: 10.1016/j.jpeds.2007.06.039 i Pellegrini, G., Traverso, C. E., Franzi, A. T., Zingirian, M., Cancedda, R., and De Luca, M. (1997). Long-term restoration of damaged corneal surfaces with autologous cultivated corneal epithelium. Lancet 349, 990–993. doi: 10.1016/S0140-6736(96)11188-0 Frew, J. W., Martin, L. K., Nijsten, T., and Murrell, D. F. (2009). REFERENCES Quality of life evaluation in epidermolysis bullosa (EB) through the development of the QOLEB questionnaire: an EB-specific quality of life instrument. Br. J. Dermatol. 161, 1323–1330. doi: 10.1111/j.1365-2133.2009.09347.x Posteraro, P., De Luca, N., Meneguzzi, G., El Hachem, M., Angelo, C., Gobello, T., et al. (2004). Laminin-5 mutational analysis in an Italian cohort of patients with junctional epidermolysis bullosa. J. Invest. Dermatol. 123, 639–648. doi: 10.1111/j.0022-202X.2004.23302.x Gallico, G. G. 3rd., O’Connor, N. E., Compton, C. C., Kehinde, O., and Green, H. (1984). Permanent coverage of large burn wounds with autologous cultured human epithelium. N. Engl. J. Med. 311, 448–451. doi: 10.1056/NEJM198408163110706 Siprashvili, Z., Nguyen, N. T., Bezchinsky, M. Y., Marinkovich, M. P., Lane, A. T., and Khavari, P. A. (2010). Long-term type VII collagen restoration to human epidermolysis bullosa skin tissue. Hum. Gene Ther. 21, 1299–1310. doi: 10.1089/hum.2010.023 Goldschneider, K. R., Good, J., Harrop, E., Liossi, C., Lynch-Jordan, A., Martinez, A. E., et al. (2014). Pain care for patients with epidermolysis bullosa: best care practice guidelines. BMC Med. 12:178. doi: 10.1186/s12916-014-0178-2 Hammersen, J., Has, C., Naumann-Bartsch, N., Stachel, D., Kiritsi, D., Söder, S., et al. (2016). Genotype, clinical course, and therapeutic decision-making in 76 infants with severe generalized junctional epidermolysis bullosa. J. Invest. Dermatol. 136, 2150–2157. doi: 10.1016/j.jid.2016.06.609 Siprashvili, Z., Nguyen, N. T., Gorell, E. S., Loutit, K., Khuu, P., Furukawa, L. K., et al. (2016). Safety and wound outcomes following genetically corrected autologous epidermal grafts in patients With recessive dystrophic epidermolysis bullosa. JAMA 316, 1808–1817. doi: 10.1001/jama.2016.15588 Hammersen, J., Neuner, A., Wild, F., and Schneider, H. (2019). Attenuation of severe generalized junctional epidermolysis bullosa by systemic treatment with gentamicin. Dermatology 235, 315–322. doi: 10.1159/000499906 Wu, C., and Dunbar, C. E. (2011). Stem cell gene therapy: the risks of insertional mutagenesis and approaches to minimize genotoxicity. Front. Med. 5, 356–371. doi: 10.1007/s11684-011-0159-1 Yuen, W. Y., and Jonkman, M. F. (2011). Risk of squamous cell carcinoma in junctional epidermolysis bullosa, non-Herlitz type: report of 7 cases and a review of the literature. J. Am. Acad. Dermatol. 65, 780–789. doi: 10.1016/j.jaad.2010.07.006 Has, C., Bauer, J. W., Bodemer, C., Bolling, M. C., Bruckner-Tuderman, L., Diem, A., et al. (2020). Consensus reclassification of inherited epidermolysis bullosa and other disorders with skin fragility. Br. J. Dermatol. 183, 614–627. doi: 10.1111/bjd.18921 Hirsch, T., Rothoeft, T., Teig, N., Bauer, J. W., Pellegrini, G., De Rosa, L., et al. (2017). REFERENCES Regeneration of the entire human epidermis using transgenic stem cells. Nature 551, 327–332. doi: 10.1038/nature24487 Conflict of Interest: MDL is co-founder and member of the Board of Directors of Holostem Terapie Avanzate (HTA), s.r.l, Modena, Italy, as well as consultants for J-TEC-Japan Tissue Engineering, Ltd. LDR is HTA employee since 2018. Howe, S. J., Mansour, M. R., Schwarzwaelder, K., Bartholomae, C., Hubank, M., Kempski, H., et al. (2008). Insertional mutagenesis combined with acquired somatic mutations causes leukemogenesis following gene therapy of SCID-X1 patients. J. Clin. Invest. 118, 3143–3150. doi: 10.1172/JCI35798 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. Kiritsi, D., Has, C., and Bruckner-Tuderman, L. (2013). Laminin 332 in junctional epidermolysis bullosa. Cell Adhes. Migr. 7, 135–141. doi: 10.4161/cam.22418 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. Kopp, J., Stachel, K. D., Holter, W., Kandler, M., Hertzberg, H., Campean, V., et al. (2005). Haematopoietic stem cell transplantation and subsequent 80% skin exchange by grafts from the same donor in a patient with Herlitz disease. Transplantation 79, 255–256. doi: 10.1097/01.TP.0000144325.01925.BE Kwong, A., Cogan, J., Hou, Y., Antaya, R., Hao, M., Kim, G., et al. (2020). Gentamicin induces Laminin 332 and improves wound healing in junctional epidermolysis bullosa patients with nonsense mutations. Mol. Ther. 28, 1327–1338. doi: 10.1016/j.ymthe.2020.03.006 Copyright © 2021 De Rosa, Enzo, Zardi, Bodemer, Magnoni, Schneider and De Luca. 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. Loh, C. C., Kim, J., Su, J. C., Daniel, B. S., Venugopal, S. S., Rhodes, L. M., et al. (2014). Development, reliability, and validity of a novel epidermolysis bullosa disease activity and scarring index (EBDASI). J. Am. Acad. Dermatol. 70, 89.e81-13–97.e81-13. REFERENCES Bardhan, A., Bruckner-Tuderman, L., Chapple, I. L. C., Fine, J. D., Harper, N., Has, C., et al. (2020). Epidermolysis bullosa. Nat. Rev. Dis. Primers 6:78. doi: 10.1038/s41572-020-0210-0 Cavazza, A., Cocchiarella, F., Bartholomae, C., Schmidt, M., Pincelli, C., Larcher, F., et al. (2013). Self-inactivating MLV vectors have a reduced genotoxic profile in human epidermal keratinocytes. Gene Ther. 20, 949–957. doi: 10.1038/ gt.2013.18 Bauer, J. W., Koller, J., Murauer, E. M., De Rosa, L., Enzo, E., Carulli, S., et al. (2017). Closure of a large chronic wound through transplantation of gene- corrected epidermal stem cells. J. Investig. Dermatol. 137, 778–781. doi: 10.1016/j.jid.2016.10.038 De Luca, M., Aiuti, A., Cossu, G., Parmar, M., Pellegrini, G., and Robey, P. G. (2019). Advances in stem cell research and therapeutic development. Nat. Cell Biol. 21, 801–811. doi: 10.1038/s41556-019-0344-z Bruckner-Tuderman, L. (2019). Newer treatment modalities in Epidermolysis Bullosa. Indian Dermatol. Online J. 10, 244–250. doi: 10.4103/idoj.IDOJ_287_18 De Luca, M., Pellegrini, G., and Green, H. (2006). Regeneration of squamous epithelia from stem cells of cultured grafts. Regen. Med. 1, 45–57. doi: 10.2217/17460751.1.1.45 Buchroithner, B., Klausegger, A., Ebschner, U., Anton-Lamprecht, I., Pohla-Gubo, G., Lanschuetzer, C. M., et al. (2004). Analysis of the LAMB3 September 2021 \| Volume 12 \| Article 705019 Frontiers in Genetics \| www.frontiersin.org 10 De Rosa et al. Hologene 5 De Rosa, L., Carulli, S., Cocchiarella, F., Quaglino, D., Enzo, E., Franchini, E., et al. (2014). Long-term stability and safety of transgenic cultured epidermal stem cells in gene therapy of junctional epidermolysis bullosa. Stem Cell Rep. 2, 1–8. doi: 10.1016/j.stemcr.2013.11.001 Mallipeddi, R., Keane, F. M., McGrath, J. A., Mayou, B. J., and Eady, R. A. (2004). Increased risk of squamous cell carcinoma in junctional epidermolysis bullosa. J. Eur. Acad. Dermatol. Venereol. 18, 521–526. doi: 10.1111/j.1468-3083.2004.00968.x Mavilio, F., Pellegrini, G., Ferrari, S., Di Nunzio, F., Di Iorio, E., Recchia, A., et al. (2006). Correction of junctional epidermolysis bullosa by transplantation of genetically modified epidermal stem cells. Nat. Med. 12, 1397–1402. doi: 10.1038/nm1504 De Rosa, L., Latella, M. C., Secone Seconetti, A., Cattelani, C., Bauer, J. W., Bondanza, S., et al. (2020). Toward combined cell and gene therapy for genodermatoses. Cold Spring Harb. Perspect. Biol. 12:a035667. doi: 10.1101/cshperspect.a035667 Mellerio, J. E., Eady, R. A., Atherton, D. J., Lake, B. D., and McGrath, J. A. (1998). E210K mutation in the gene encoding the beta3 chain of laminin-5 (LAMB3) is predictive of a phenotype of generalized atrophic benign epidermolysis bullosa. Br. REFERENCES doi: 10.1016/j.jaad.2013.09.041 September 2021 \| Volume 12 \| Article 705019 11 Frontiers in Genetics \| www.frontiersin.org
https://openalex.org/W3088843732	https://www.nature.com/articles/s41467-020-18507-4.pdf	English	null	A theoretical model of Polycomb/Trithorax action unites stable epigenetic memory and dynamic regulation	Nature communications	2,020	cc-by	20,680	ARTICLE 1 Humboldt Universität zu Berlin, IRI- Lifesciences, Philippstr. 13, 10115 Berlin, Germany. 2 IMBA, Institute of Molecular Biotechnology, Dr. Bohr- Gasse 3, 1030 Vienna, Austria. 3 John Innes Centre, Norwich Research Park, Norwich NR4 7UH, UK. 4These authors contributed equally: Jeannette Reinig, Frank Ruge. ✉email: leonie.ringrose@hu-berlin.de A theoretical model of Polycomb/Trithorax action unites stable epigenetic memory and dynamic regulation Jeannette Reinig1,4, Frank Ruge 2,4, Martin Howard 3 & Leonie Ringrose 1,2✉ Polycomb and Trithorax group proteins maintain stable epigenetic memory of gene expression states for some genes, but many targets show highly dynamic regulation. Here we combine experiment and theory to examine the mechanistic basis of these different modes of regulation. We present a mathematical model comprising a Polycomb/Trithorax response element (PRE/TRE) coupled to a promoter and including Drosophila developmental timing. The model accurately recapitulates published studies of PRE/TRE mediated epigenetic memory of both silencing and activation. With minimal parameter changes, the same model can also recapitulate experimental data for a different PRE/TRE that allows dynamic reg- ulation of its target gene. The model predicts that both cell cycle length and PRE/TRE identity are critical for determining whether the system gives stable memory or dynamic regulation. Our work provides a simple unifying framework for a rich repertoire of PRE/TRE functions, and thus provides insights into genome-wide Polycomb/Trithorax regulation. 1 Humboldt Universität zu Berlin, IRI- Lifesciences, Philippstr. 13, 10115 Berlin, Germany. 2 IMBA, Institute of Molecular Biotechnology, Dr. Bohr- Gasse 3, 1030 Vienna, Austria. 3 John Innes Centre, Norwich Research Park, Norwich NR4 7UH, UK. 4These authors contributed equally: Jeannette Reinig, Frank Ruge. ✉email: leonie.ringrose@hu-berlin.de 1 NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 E pigenetic memory is essential to many biological systems, allowing maintenance of gene expression states over mul- tiple cell generations in the absence of the initiating signals1. Polycomb/Trithorax response elements (PRE/TREs) are cis- regulatory elements that can maintain epigenetic memory of repressed gene expression states over many cell generations. In transgenic reporter assays using PRE/TREs from Homeobox (Hox) genes, maintenance of repression depends on the Poly- comb group (PcG) proteins2–5. Several Drosophila Hox PRE/ TREs have also been shown to maintain the memory of tran- siently activated gene expression states, in a manner dependent on the Trithorax group (TrxG) proteins6. Several different Dro- sophila PRE/TREs are interchangeable in these assays, suggesting that epigenetic memory is a general property of PRE/TREs2,7–10. The above examples from Drosophila have several features in common: First, the responsiveness of the system to reporter gene expression state decreases as development proceeds3,6. Second, the initial repressors or activators disappear during development, and expression status is maintained in their absence. Results A model for PcG/TrxG regulation links promoter and PRE/ TRE. To examine the regulatory interactions between PRE/TREs and their target promoters during Drosophila development, we used minimal stochastic models for a promoter and a PRE/TRE, and introduced coupling between them. The model, its imple- mentation and the assumptions used are described in detail in Supplementary Methods. Here we summarise the most important features. The promoter was modelled as an array of DNA binding sites for a transcription factor (Fig. 1a), each of which can be either free (F) or bound (B). The probabilities of binding and unbinding are represented by the parameters p1 and p2, respectively, which can be modiﬁed to reﬂect different promoter strengths (Fig. 1b). The promoter state (active or repressed) is given by the proportion of sites that are in the B or F conﬁgurations, respectively. However, there are also results that argue against classifying all PRE/TREs as epigenetic memory elements. Genome-wide studies of PcG and TrxG target genes in ﬂies and vertebrates have identiﬁed several hundred targets beyond the Hox genes, many of which do not conform to the above criteria12–16. These genes include many that switch late in development, or that switch dynamically several times. This raises the question of how these genes overcome the restrictions imposed by a memory system that gains stability as development proceeds. In addition the expression patterns of many of these PcG/TrxG target genes are far more complex than a simple on or off state14,17–20. Finally for a large number of these genes, the transcription factors that regulate them do not disappear but are present throughout the time window of their expression or repression. Thus these genes do not appear to be subject to epigenetic memory in the classic sense, raising the intriguing question of how the PcG and TrxG proteins are involved in their regulation (reviewed in ref. 13). g , p y The PRE/TRE was modelled as an array of nucleosomes, as described in refs. 21,22 (Fig. 1a). Despite its complexity, the PcG/ TrxG system is robustly bistable23, thus a simple bistable model is reasonable. Each nucleosome can be in a silent (M), neutral (U) or active (A) conﬁguration. The PRE/TRE state (active or silent) is given by the proportion of nucleosomes that are in the A or M conﬁgurations, respectively. Results Each of the A and M conﬁgurations represents all histone modiﬁcations and other bound molecules, such as TrxG and PcG proteins or non-coding RNAs that contribute to activation or silencing. Thus the model makes no assumptions about the molecular nature of memory. We use whole nucleosomes as the minimum unit, as in refs. 21,22 (see Supplementary Methods for more detail). In the model, a nucleosome in the M or A conﬁguration will attempt to convert other nucleosomes in the array towards its own conﬁguration, with probabilities p3 and p4 for M and A respectively (Fig. 1b). This feedback renders intermediate U nucleosome unstable, and A and M nucleosome conﬁgurations much more stable21,22. Thus the PRE/TRE tends to adopt a dominant A or M state, and is bistable. A further parameter (p5) gives the probability of interconversions between A, U and M conﬁgurations that are independent of feedback (Fig. 1b). This includes histone exchange and random noisy conversions as described previously21,22 but also includes speciﬁc conversions that do not require a previously existing modiﬁcation for a modifying enzyme to be recruited, such as direct recruitment by DNA binding proteins24,25. p g In summary, although some PcG/TrxG target genes are subject to epigenetic memory in the strict sense, many are regulated in a far more dynamic manner, suggesting a rich repertoire of PcG/ TrxG regulatory modes. How a given gene responds to PcG/TrxG regulation may depend on developmental timing, transcriptional status of the associated gene, and inherent PRE/TRE properties, determined by differences in their nucleic acid sequences4. Understanding the mechanistic basis for these different modes of regulation will be essential for understanding genome-wide PcG/ TrxG function in health and disease. A large gap in our under- standing of PcG/TrxG regulation has been the lack of a coherent theoretical framework that links developmental timing and transcriptional regulation to PRE/TRE activity. We are working to bridge this gap by using simple mathematical models. y g The separation of promoter and PRE/TRE in the model reﬂects the regulatory units that are experimentally tractable, and allows the effect of regulated coupling between them to be examined in simulations. The PRE/TRE and promoter were coupled to each other as described in Supplementary Methods, Fig. 1c and Supplementary Fig. 1. A theoretical model of Polycomb/Trithorax action unites stable epigenetic memory and dynamic regulation Finally, the state maintained by the PRE/TRE is either on or off, and is stable over the whole of development. This has given rise to a paradigm in which PRE/TREs are thought to be switchable elements that maintain stable epigenetic memory of both silent and active states, and do so more stably as development proceeds4,11. E recapitulate our own experimental data for a different PRE/TRE that allows late switching and highly dynamic regulation of its target gene. In summary, we show that a single simple model can account for profoundly different regulatory modes, and we identify parameters that govern those differences. Thus, this work has broad implications for understanding the molecular nature of locus-speciﬁc and developmental differences in stability and ﬂexibility of genome-wide PcG/TrxG regulation. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 F B p1 p2 M p4 p3 C b a C Promoter Promoter PRE/TRE PRE/TRE F: free B: bound M: silent U: neutral A: active Promoter state influences PRE/TRE PRE/TRE state influences promoter p5 p5 Promoter s = 10 sites PRE/TRE S = 40 nucleosomes 1. Evaluate promoter state 2. Adjust p3 and p4 3. Evaluate PRE/TRE state 4. Adjust p1 and p2 C A U Fig. 1 A simple model for Polycomb/Trithorax regulation. a The promoter and PRE/TRE are shown schematically. Left: each promoter site can be eit free (F) or bound (B). Right: each nucleosome in the PRE/TRE can be either silent (red, M), neutral (grey, U) or active (green, A). See main text for deta b The model is implemented stochastically, with probabilities for each of the transitions between F and B, and between A, U and M. For the promoter, probability of transcription factor binding at a single promoter binding site. p2: probability of transcription factor unbinding at a single site. For the PRE/T p3 and p4 (red and green arrows) denote feedback reactions in which nucleosomes in each of the M or A conﬁgurations convert other nucleosome towards that conﬁguration. The parameter p5 (black arrows) gives the probability of conversions between A, U and M that are independent of feedback The promoter and PRE/TRE are coupled. See also Supplementary Fig. 1. c At each iteration of the simulation, the promoter state is evaluated and used adjust the PRE/TRE parameters p3 and p4. Likewise the PRE/TRE state is evaluated and used to adjust the promoter parameters p1 and p2. These adjus p1, p2, p3 and p4 values are used in the next iteration. For coupling relationships see Supplementary Fig. 1. a Promoter PRE/TRE F: free B: bound M: silent U: neutral A: active Promoter state influences PRE/TRE PRE/TRE state influences promoter a Promoter state influences PRE/TRE U: neutral A: active PRE/TRE state influences promoter PRE/TRE state influences promoter F B p1 p2 M p4 p3 C b C Promoter PRE/TRE p5 p5 A U Promoter s = 10 sites PRE/TRE S = 40 nucleosomes 1. Evaluate promoter state 2. Adjust p3 and p4 3. Evaluate PRE/TRE state 4. Adjust p1 and p2 C b C Promoter s = 10 sites Fig. 1 A simple model for Polycomb/Trithorax regulation. a The promoter and PRE/TRE are shown schematically. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 14 different coupling models were explored (Supplementary Fig. 1), which differ in the number of model parameters (p1, p2 and (p3, p4)) that are adjusted at each iteration (Supplementary Fig. 1G, H), and the mathematical relationship between PRE/TRE state and promoter state (Supplementary Fig. 1A, B). This analysis showed that only two mathematical descriptions of coupling gave robust results in all tests, designated as models 1 and 2 in the rest of this paper. Models 1 and 2 have in common that they adjust all of the four model parameters at each iteration (p1, p2 and (p3, p4)). Model 1 is used for all results shown in the main ﬁgures, models 1 and 2 are compared in Supplementary Figs. 3 and 8. Coupling strength in the model is adjusted by the parameter C (Fig. 1b), which determines the magnitude of the response of the promoter to a given PRE/TRE state, and that of the response of the PRE/TRE to a given promoter state. model, but the mathematical description of the interaction between PRE/TRE state and promoter, and the strength of this interaction, were varied to identify the model that best ﬁts the data. 14 different coupling models were explored (Supplementary Fig. 1), which differ in the number of model parameters (p1, p2 and (p3, p4)) that are adjusted at each iteration (Supplementary Fig. 1G, H), and the mathematical relationship between PRE/TRE state and promoter state (Supplementary Fig. 1A, B). This analysis showed that only two mathematical descriptions of coupling gave robust results in all tests, designated as models 1 and 2 in the rest of this paper. Models 1 and 2 have in common that they adjust all of the four model parameters at each iteration (p1, p2 and (p3, p4)). Model 1 is used for all results shown in the main ﬁgures, models 1 and 2 are compared in Supplementary Figs. 3 and 8. Coupling strength in the model is adjusted by the parameter C (Fig. 1b), which determines the magnitude of the response of the promoter to a given PRE/TRE state, and that of the response of the PRE/TRE to a given promoter state. To take account of the changes in cell cycle length during Drosophila development, we curated published data on cell cycle timing for all developmental stages as described in detail in Supplementary Methods. These time constraints for each cell The model recapitulates memory of silencing. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Left: each promoter site can be either free (F) or bound (B). Right: each nucleosome in the PRE/TRE can be either silent (red, M), neutral (grey, U) or active (green, A). See main text for details. b The model is implemented stochastically, with probabilities for each of the transitions between F and B, and between A, U and M. For the promoter, p1: probability of transcription factor binding at a single promoter binding site. p2: probability of transcription factor unbinding at a single site. For the PRE/TRE, p3 and p4 (red and green arrows) denote feedback reactions in which nucleosomes in each of the M or A conﬁgurations convert other nucleosomes towards that conﬁguration. The parameter p5 (black arrows) gives the probability of conversions between A, U and M that are independent of feedback. C: The promoter and PRE/TRE are coupled. See also Supplementary Fig. 1. c At each iteration of the simulation, the promoter state is evaluated and used to adjust the PRE/TRE parameters p3 and p4. Likewise the PRE/TRE state is evaluated and used to adjust the promoter parameters p1 and p2. These adjusted p1, p2, p3 and p4 values are used in the next iteration. For coupling relationships see Supplementary Fig. 1. cycle were included in the model (Supplementary Table 1). To model cell divisions for the promoter, all sites were set to F (unbound) at the end of each cell cycle. This is based on the observation that many transcription factors dissociate from mitotic chromatin26 and that transcription is actively and globally repressed during mitosis27,28. For the PRE/TRE, cell division was modelled as described previously in refs. 21,22, by setting each nucleosome to U with a probability of 0.5, at the end of each cell cycle. Thus on average, half of the nucleosomes are set to U at the end of each cell cycle. This is based on the observation that parental histones and their modiﬁcations are partitioned randomly to the two daughter chromosomes after replication4,29. In summary, this simple model comprises several essential features of PRE/TRE mediated gene regulation during Drosophila development, namely a regulatable promoter coupled to a bistable PRE/TRE, and the known timing of cell cycles throughout development. model, but the mathematical description of the interaction between PRE/TRE state and promoter, and the strength of this interaction, were varied to identify the model that best ﬁts the data. Results The effect of coupling is to render the promoter and the PRE/TRE dependent on each other, so that that the more active or silent the PRE/TRE, the more active or silent the promoter, and vice versa. This reﬂects the regulatory interactions that have been observed in vivo2,3,6. The exact mechanism of this coupling is not known but may include looping, spreading of chromatin marks or interaction of homologs. These mechanisms are not explicitly included in the Here we present a mathematical model consisting of a PRE/ TRE coupled to a promoter. The system is subjected to replication cycles whose length and number reﬂect those that occur during Drosophila development. We combine theory and experiment to quantitatively dissect the contributions of developmental timing, transcriptional input and PRE/TRE identity to the output of the system. We show that this simple model can accurately recapi- tulate published studies of PRE/TRE mediated epigenetic memory of both silencing and activation in Drosophila, and that cell cycle length is an essential component of memory. Furthermore with minimal parameter changes, the model can also precisely NICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications 2 ARTICLE NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Right: PcG mutant was simulated by reducing p3 to 0.001 at the onset of the maintenance phase. (c) PRE/TRE state. Parameter values as in (b); (see also Supplementary Table 2 and Supplementary Fig. 2). For each condition, data from 50 independent simulations are shown. Promoter and PRE/TRE data from the same simulation run are shown. Each of the 50 simulations is continuous throughout the vertical scale. d, e Memory score in anterior compartment (1 indicates perfect memory, see Supplementary Fig. 2 for calculation). Boxplots of data for 400 independent simulations. Central mark on box plots: median; bottom and top edges of box: 25th and 75th percentiles, respectively. Whiskers extend to cover 99.3% of the data. Outliers are plotted as dots. f PRE/TRE states for the different early cycle conditions as indicated, averaged over cycles 12 and 13 for 1000 independent simulations. PRE/TRE state (on a scale of −1 to +1, where −1 is silent, and +1 is active). NA = number of A nucleosomes, NM = number of M nucleosomes, S = total number of nucleosomes. Boxplot parameters as in (d, e). a a Reporter with PRE/TRE A P A P LacZ enhancer PRE/TRE Reporter with PRE/TRE in PcG mutant A P A P c Initiation (Stage 5–9) A P A A A Maintenance (Stage 10–17) A P A P c P P A A Model PRE/TRE without promoter with promoter with promoter in PcG mutant c A P A P A P Cycle no. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 d e Cycles 1–13 (Stage 1–3) Reporter without PRE/TRE Reporter with PRE/TRE Reporter with PRE/TRE in PcG mutant Initiation (Stage 5–9) Maintenance (Stage 10–17) a A P A P A P A P A P A P F B A M Promoter state PRE/TRE state f PRE/TRE state cycles 12–13 (NA - NM ) /S –0.6 –0.4 –0.2 0 0.2 0.4 M A Ce Cyc 1–13 (min) 0 Silent LacZ enhancer PRE/TRE LacZ enhancer Model promoter Model PRE/TRE Time (h:m) without PRE/TRE without promoter with promoter 2:10 with PRE/TRE in PcG mutant with PRE/TRE with promoter in PcG mutant 4:50 7:30 b c A P A P A P A P A P A P p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 Ce = 0 Cycle no. Initiation Maintenance 1 4 7 10 13 14 15 Cycles 1–13 0 – – + + + + 0.2 0.4 0.6 0.8 1.0 Memory score anterior PRE/TRE PcG Memory score anterior U Initial PRE/TRE state 0 0.2 0.4 0.6 0.8 1.0 Active Ce = 0 Ce = 0 Ce = 0 Ce = 0 Ce = 0 Ci,m = 0 Ci,m = 4 Ci,m = 4 Ci,m = 0 Ci,m = 4 Ci,m = 4 M A 8 20 20 10 10 8 0 Fig. 2 The model recapitulates memory of silencing. a Experimental test of epigenetic memory of silencing2,3. See main text for details. Top: transgenic LacZ reporter constructs. Bottom: embryonic LacZ patterns. Active LacZ: blue. Anterior: A, posterior: P. b, c Simulated time courses of Drosophila development showing promoter state (b) and PRE/TRE state (c) over time. Active: blue; silent: white. Division cycles are indicated (right). Plots show results from model 1 (see also Supplementary Figs. 1 and 2). Initial conditions, promoter: all sites are F; PRE/TRE: all nucleosomes are U. b Promoter state. The input values of p1, Ce and Ci,m are shown on the plots, p2 (TF dissociation) was 0.1 in all simulations (see also Supplementary Table 2). Input values of p3 and p4 = 0.25, p5 = 0.04. Simulations were performed without coupling between promoter and PRE/TRE (left), and with coupling during the initiation and maintenance phases (middle). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Initiation Maintenance 1 4 7 10 13 14 15 Cycles 1–13 Ce = 0 Ce = 0 Ce = 0 Ci,m = 0 Ci,m = 4 Ci,m = 4 b Model promoter Time (h:m) without PRE/TRE 2:10 with PRE/TRE in PcG mutant with PRE/TRE 4:50 7:30 b A P A P A P p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 p1 = 0.6 Ce = 0 Ce = 0 Ce = 0 Ci,m = 0 Ci,m = 4 Ci,m = 4 c Model promoter f PRE/TRE state cycles 12–13 (NA - NM ) /S –0.6 –0.4 –0.2 0 0.2 0.4 M A Ce Cyc 1–13 (min) 0 8 20 20 10 10 8 0 d 0 – – + + + + 0.2 0.4 0.6 0.8 1.0 Memory score anterior PRE/TRE PcG e Memory score anterior U Initial PRE/TRE state 0 0.2 0.4 0.6 0.8 1.0 M A f d e Fig. 2 The model recapitulates memory of silencing. a Experimental test of epigenetic memory of silencing2,3. See main text for details. Top: transgenic LacZ reporter constructs. Bottom: embryonic LacZ patterns. Active LacZ: blue. Anterior: A, posterior: P. b, c Simulated time courses of Drosophila development showing promoter state (b) and PRE/TRE state (c) over time. Active: blue; silent: white. Division cycles are indicated (right). Plots show results from model 1 (see also Supplementary Figs. 1 and 2). Initial conditions, promoter: all sites are F; PRE/TRE: all nucleosomes are U. b Promoter state. The input values of p1, Ce and Ci,m are shown on the plots, p2 (TF dissociation) was 0.1 in all simulations (see also Supplementary Table 2). Input values of p3 and p4 = 0.25, p5 = 0.04. Simulations were performed without coupling between promoter and PRE/TRE (left), and with coupling during the initiation and maintenance phases (middle). Right: PcG mutant was simulated by reducing p3 to 0.001 at the onset of the maintenance phase. (c) PRE/TRE state. Parameter values as in (b); (see also Supplementary Table 2 and Supplementary Fig. 2). For each condition, data from 50 independent simulations are shown. Promoter and PRE/TRE data from the same simulation run are shown. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 We ﬁrst asked whether the model can recapitulate the epigenetic memory of silencing shown by a PRE/TRE in a transgenic Drosophila TURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications 3 3 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Thus the model was ﬁtted with only three free parameters: (p3, p4) (feedback); p5 (feedback-independent transition) and Ci,m (coupling during initiation and maintenance phases). An additional constraint in the parameter search was the require- ment for a uniform and spatially precise memory of the initiated pattern until the end of the maintenance phase under the inﬂuence of the PRE/TRE (Fig. 2b and Supplementary Fig. 2). This analysis revealed that longer initial cycles allowed the PRE/TRE to stabilise more often into random A or M states before being coupled to the promoter, causing variegation (because the PRE/TRE had more time to stabilise before being wiped by replication; Fig. 2f). Introducing coupling during these early cycles, even without a change in cycle length, resulted in a bias towards M states, resulting in a failure to adopt the A state in response to the active promoter in the posterior (because the PRE/TRE was coupled early to a strongly silenced promoter). This effect became more pronounced if cycle length was increased in addition to coupling (Fig. 2f). In summary this analysis demonstrates that in the model, the early rapid division cycles and an absence of coupling during this phase are essential for keeping the PRE/TRE in a naive state prior to receiving information from the promoter, and are thus instrumental in determining system ﬂexibility and ﬁdelity of epigenetic memory. Under these conditions, a range of values for the parameters (p3, p4), p5, and Ci,m were found, under which the system gave a precise memory of the promoter expression pattern established during the initiation phase, namely repression in the anterior and uniform high expression in the posterior (Supplementary Fig. 2 and Fig. 2b, middle panels). This maintenance was stable until 7h30 of development in the model (Fig. 2b, middle panels) and throughout adult development despite several further replication cycles (for examples of persistence of early established states until later development see Fig. 3d). Reduction of the parameter p3 to simulate PcG null mutants that display loss of repression at the onset of the maintenance phase led to a derepression of the model promoter in the anterior compartment after ~6 h of the simulation, consistent with published data (Fig. 2b, d and Supplementary Table 2)2,3. The same model recapitulates memory of activation. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 During the initiation phase, the Ubx enhancer is activated in the posterior part of the embryo and suppressed by repressors in the anterior (Fig. 2a, left and Supplementary Table 1). At the onset of stage 10, the repressors disappear and the reporter becomes active in the anterior unless coupled to a PRE/TRE (Fig. 2a, middle and Supplementary Table 1). The maintenance of silencing is dependent on the PRE/TRE, and is lost in a PcG mutant (Fig. 2a, right)2,3. To adapt the model to the experimental observations, we used the model promoter to represent the LacZ reporter gene described above. We ﬁrst established the pattern of the model promoter throughout development without the PRE/TRE by adjusting the input value of the parameter p1, representing promoter repression (small p1) or activation (large p1) (Fig. 2b, left panels and Supplementary Table 2). To ask whether the model PRE/TRE is able to maintain memory of anterior repression of the promoter that was established during the initiation phase, we coupled the PRE/TRE to the promoter and ran the simulation with the same promoter inputs, and various coupling regimes and PRE/TRE parameters (Fig. 2b, middle and Supplementary Fig. 2). Early rapid division cycles are essential ﬂexibility. In the above simulations, the early rapid cycles 1–13 and absence of coupling during these cycles had the effect of keeping the PRE/TRE pre- dominantly in the U conﬁguration prior to the initiation phase (on average approximately 90% of nucleosomes were in the U (unmodiﬁed) state at any given time, Fig. 2c). To gain further insight into these early stages, we varied the strength of coupling and the length of the early cycles in simulations. pp y g To reduce the number of free parameters for this analysis, we introduced several constraints (see Supplementary Fig. 2 and Supplementary Methods). The input values of all PRE/TRE parameters were kept constant throughout development, and the feedback parameters p3 and p4 (for silencing and activation respectively) were kept equal to each other. The coupling strength C, of PRE/TRE to promoter and promoter to PRE/TRE was kept equal in both directions, and this strength was kept constant during the initiation and maintenance phases (Supplementary Fig. 2). Coupling during the ﬁrst 13 cycles was set to 0 (this requirement was determined by ﬁtting, see Fig. 2). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Each of the 50 simulations is continuous throughout the vertical scale. d, e Memory score in anterior compartment (1 indicates perfect memory, see Supplementary Fig. 2 for calculation). Boxplots of data for 400 independent simulations. Central mark on box plots: median; bottom and top edges of box: 25th and 75th percentiles, respectively. Whiskers extend to cover 99.3% of the data. Outliers are plotted as dots. f PRE/TRE states for the different early cycle conditions as indicated, averaged over cycles 12 and 13 for 1000 independent simulations. PRE/TRE state (on a scale of −1 to +1, where −1 is silent, and +1 is active). NA = number of A nucleosomes, NM = number of M nucleosomes, S = total number of nucleosomes. Boxplot parameters as in (d, e). NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications 4 4 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 ARTICLE a stable A or M state within the time frame of the initiation phase (cycle 14, Fig. 2c, middle). However, if this conversion rate is too fast, then random A and M states that exist at the end of cycle 13 become ﬁxed before the PRE/TRE can receive information from the promoter, causing variegation. The parameter p5 (feedback- independent transitions) affects the stability of the system. If p5 is low, then feedback dominates, and the system has a high capacity for memory but is unable to respond to changes in the promoter state during early initiation (cycle 14). If p5 is high, the opposite is true: the PRE/TRE responds well to the promoter but is unable to maintain memory (Supplementary Fig. 2). A minimum value of Ci,m (coupling strength during initiation and maintenance phases) was required for optimum memory (see Supplementary Fig. 2D). At lower Ci,m-values the system responded too slowly to changes in promoter or PRE/TRE state, and memory was not correctly established during the initiation phase. Taken together these results demonstrate that the simple model is sufﬁcient to recapitulate several essential features of epigenetic memory of silencing that have been observed experimentally, and deﬁne a set of conditions under which memory is most effective. reporter assay2,3 (Fig. 2). In the experiment, an embryonic enhancer from a Hox gene (such as Ultrabithorax (Ubx)) is linked to a β-galactosidase (LacZ) reporter gene, with or without a PRE/ TRE (Fig. 2a). Transcription is generally silent during the ﬁrst 13 cycles before zygotic genome activation at cycle 1430. NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 mw UAS PRE/TRE LacZ GAL4 hsp a Time (h:m) 2:10 4:50 7:30 none Heat shock: Model promoter without PRE/TRE Model promoter with PRE/TRE Model PRE/TRE with promoter c e d none late early Heat shock: Embryo (LacZ) Larva (LacZ) Adult eye (miniwhite) mw UAS PRE/TRE LacZ GAL4 hsp a b F B A M Promoter state PRE/TRE state Embryo (LacZ) F B Promoter state A M PRE/TRE state Adult (mw) Eye disc cycles 22:00 100:00 Silent Silent Cycle no. Initiation Maintenance 1 4 7 10 13 14 15 Cycles 1–13 2 4 6 8 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.6 p1 = 0.6 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 Ce = 0 Ce = 0 Ce = 0 0 + + + + + – – – – – – – 0.2 0.4 0.6 0.8 1.2 0 0.2 0.4 0.6 0.8 1.2 f g PRE/TRE Heatshock late Heatshock early 0 0.2 0.4 0.6 0.8 1.2 0 0.2 0.4 0.6 0.8 1.0 1.2 1.0 1.0 1.0 Ce 0 Initial PRE/TRE state Length of cycle 1–13 (mins) PRE/ TRE h i Promoter (NB/s) Eye colour (NB/s) Eye colour (NB/s) Eye colour (NB/s) Eye colour (NB/s) early late none early late none early late Ci,m = 0 Ci,m = 4 Ci,m = 4 Active Active M A U 4 + + + + – 10 10 20 80 40 2 none late early Heat shock: Embryo (LacZ) Larva (LacZ) Adult eye (miniwhite) b b late early a late Time (h:m) 2:10 4:50 7:30 none Heat shock: Model promoter without PRE/TRE Model promoter with PRE/TRE Model PRE/TRE with promoter c e d y (LacZ) Larva (LacZ) Adult eye (miniwhite) mw UAS PRE/TRE LacZ GAL4 hsp F B A M Promoter state PRE/TRE state Embryo (LacZ) F B Promoter state A M PRE/TRE state Adult (mw) Eye disc cycles 22:00 100:00 Silent Silent Cycle no. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Initiation Maintenance 1 4 7 10 13 14 15 Cycles 1–13 2 4 6 8 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.6 p1 = 0.6 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 Ce = 0 Ce = 0 Ce = 0 0 + + + + + – – – – – – – 0.2 0.4 0.6 0.8 1.2 0 0.2 0.4 0.6 0.8 1.2 f g PRE/TRE Heatshock late Heatshock early 0 0.2 0.4 0.6 0.8 1.2 0 0.2 0.4 0.6 0.8 1.0 1.2 1.0 1.0 1.0 Ce 0 Initial PRE/TRE state Length of cycle 1–13 (mins) Early heat shock PRE/ TRE h i Promoter (NB/s) Eye colour (NB/s) Eye colour (NB/s) Eye colour (NB/s) Eye colour (NB/s) early late none early late none early late Ci,m = 0 Ci,m = 4 Ci,m = 4 Active Active M A U 4 + + + + – 10 10 20 80 40 2 Fig. 3 The same model recapitulates memory of activation. a, b Experimental test of epigenetic memory of activation6 (see main text for details). a Heat- shock GA-4 and UAS PRE/TRE reporter constructs. Hsp: heat shock promoter. mw: miniwhite. b Summary of results in ref. 6. Active LacZ: blue; active mw: red. c–e Simulated time courses of Drosophila development showing model promoter activity (c, d) and PRE/TRE activity (e) over time for the ﬁrst 7.5 h of development and for the eye disc up to mid third instar (22–100 h). Division cycles are indicated (right). Plots show results from model 1 (see also Supplementary Figs. 1 and 2). The simulation was run continuously from 0 to 100 h, but only the time windows indicated are shown. 22 h–100 h: dark red: active; yellow: silenced, to represent miniwhite activity. For each condition, data from 50 independent simulations are shown, each of which is continuous throughout the vertical scale. The input value of the parameter p1 was varied as indicated on the plots in (c). The same p1 values were used for the corresponding time segments in (d, e). All other parameter values and initial conditions were as in Fig. 2b, c (middle). Ce and Ci,m values are indicated. f–i Boxplots of promoter state averaged over 90–100 h for 400 independent simulations. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 In simulations, the promoter level at end of third instar (100 h) did not change over further developmental time, thus it is used here as a proxy for adult eye colour. NB: number of promoter sites in B conﬁguration, s: total number of sites. Boxplot parameters as in Fig. 2d, e. f Effect of heatshock timing. g–i Under conditions in which a heatshock was given early, and the PRE/ TRE was coupled to the promoter at the onset of initiation phase, different additional parameters were varied. g Initial state of all nucleosomes in the PRE/ TRE set to A, U or M. h Ce, was varied as shown. Later coupling was not varied (Ci,m = 4). i Length of cycles 1–13 was varied as indicated. See also Supplementary Figs. 1 and 3. Time (h:m) 2:10 4:50 7:30 none Heat shock: Model promoter without PRE/TRE c 22:00 100:00 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.6 p1 = 0.6 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 Ce = 0 early late Ci,m = 0 Time (h:m) 2:10 4:50 7:30 none Heat shock: Model promoter without PRE/TRE Model promoter with PRE/TRE Model PRE/TRE with promoter c e d F B A M Promoter state PRE/TRE state Embryo (LacZ) F B Promoter state A M PRE/TRE state Adult (mw) Eye disc cycles 22:00 100:00 Silent Silent Cycle no. Initiation Maintenance 1 4 7 10 13 14 15 Cycles 1 13 2 4 6 8 p1 = 10–3 p1 = 10–3 p1 = 10–3 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.6 p1 = 0.6 p1 = 0.06 p1 = 0.06 p1 = 0.06 p1 = 0.06 Ce = 0 Ce = 0 Ce = 0 ) early late none early late none early late Ci,m = 0 Ci,m = 4 Ci,m = 4 Active Active c Model promoter without PRE/TRE d Model promoter with PRE/TRE Model PRE/TRE with promoter e F B Promoter state A M PRE/TRE state Adult (mw) Eye disc cycles Silent Cycle no. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 We next asked whether the model can recapitulate the epigenetic memory of activation shown by a PRE/TRE in a transgenic Drosophila reporter assay6, and if so, with which parameters. In the activa- tion assay, the yeast GAL4 transcription activator protein is expressed under control of a heat shock promoter (Fig. 3a). The reporter construct carries an upstream activating sequence (UAS), which is activated by GAL4, upstream of a PRE/TRE and two reporter genes. The LacZ reporter enables detection of expression in embryos and larvae, and the miniwhite (mw) reporter gives a red pigment in adult eyes (Fig. 3b). In the absence of heat shock, both reporters are silenced by the PRE/TRE, and the embryo and the adult eyes are mostly light in colour with some variegation (Fig. 3b, left). If a 1-h heat shock is given during late developmental stages, e.g., in the larva, then the reporter is transiently activated but is re-silenced before the adult stage (Fig. 3b, middle). In contrast, if a 1-h heat shock is given during embryonic stages then the reporter is activated and remains active The system gave optimum memory over a range of values for (p3, p4) and p5. However, these parameters also showed upper and lower limits, due to the stringent criteria that must be fulﬁlled for memory to be effective (Supplementary Fig. 2). The feedback- dependent parameters p3 and p4 determine the speed with which the PRE/TRE can be converted to one state or the other. This is critical in the model, as the PRE/TRE has to be fully converted to 5 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 f Effect of heatshock timing. g–i Under conditions in which a heatshock was given early, and the PRE/ TRE was coupled to the promoter at the onset of initiation phase, different additional parameters were varied. g Initial state of all nucleosomes in the PRE/ TRE set to A, U or M. h Ce, was varied as shown. Later coupling was not varied (Ci,m = 4). i Length of cycles 1–13 was varied as indicated. See also Supplementary Figs. 1 and 3. Fig. 3 The same model recapitulates memory of activation. a, b Experimental test of epigenetic memory of activation6 (see main text for details). a Heat- shock GA-4 and UAS PRE/TRE reporter constructs. Hsp: heat shock promoter. mw: miniwhite. b Summary of results in ref. 6. Active LacZ: blue; active mw: red. c–e Simulated time courses of Drosophila development showing model promoter activity (c, d) and PRE/TRE activity (e) over time for the ﬁrst 7.5 h of development and for the eye disc up to mid third instar (22–100 h). Division cycles are indicated (right). Plots show results from model 1 (see also Supplementary Figs. 1 and 2). The simulation was run continuously from 0 to 100 h, but only the time windows indicated are shown. 22 h–100 h: dark red: active; yellow: silenced, to represent miniwhite activity. For each condition, data from 50 independent simulations are shown, each of which is continuous throughout the vertical scale. The input value of the parameter p1 was varied as indicated on the plots in (c). The same p1 values were used for the corresponding time segments in (d, e). All other parameter values and initial conditions were as in Fig. 2b, c (middle). Ce and Ci,m values are indicated. f–i Boxplots of promoter state averaged over 90–100 h for 400 independent simulations. In simulations, the promoter level at end of third instar (100 h) did not change over further developmental time, thus it is used here as a proxy for adult eye colour. NB: number of promoter sites in B conﬁguration, s: total number of sites. Boxplot parameters as in Fig. 2d, e. f Effect of heatshock timing. g–i Under conditions in which a heatshock was given early, and the PRE/ TRE was coupled to the promoter at the onset of initiation phase, different additional parameters were varied. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Initiation Maintenance 1 4 7 10 13 14 15 Cycles 1–13 2 4 6 8 Ce = 0 none early late Ci,m = 4 Active e Model PRE/TRE with promoter Time (h:m) 0 + + + + + – – – – – – – 0.2 0.4 0.6 0.8 1.2 f PRE/TRE Heatshock late Heatshock early 1.0 Promoter (NB/s Eye colour (NB/s) 0 0.2 0.4 0.6 0.8 1.2 1.0 Ce 0 h Eye colour (NB/s) 4 2 0 0.2 0.4 0.6 0.8 1.0 1.2 Length of cycle 1–13 (mins) PRE/ TRE i Eye colour (NB/s) + + + + – 10 10 20 80 40 0 0.2 0.4 0.6 0.8 1.2 g 1.0 Initial PRE/TRE state Eye colour (NB/s) M A U i h Early heat shock Fig. 3 The same model recapitulates memory of activation. a, b Experimental test of epigenetic memory of activation6 (see main text for details). a Heat- shock GA-4 and UAS PRE/TRE reporter constructs. Hsp: heat shock promoter. mw: miniwhite. b Summary of results in ref. 6. Active LacZ: blue; active mw: red. c–e Simulated time courses of Drosophila development showing model promoter activity (c, d) and PRE/TRE activity (e) over time for the ﬁrst 7.5 h of development and for the eye disc up to mid third instar (22–100 h). Division cycles are indicated (right). Plots show results from model 1 (see also Supplementary Figs. 1 and 2). The simulation was run continuously from 0 to 100 h, but only the time windows indicated are shown. 22 h–100 h: dark red: active; yellow: silenced, to represent miniwhite activity. For each condition, data from 50 independent simulations are shown, each of which is continuous throughout the vertical scale. The input value of the parameter p1 was varied as indicated on the plots in (c). The same p1 values were used for the corresponding time segments in (d, e). All other parameter values and initial conditions were as in Fig. 2b, c (middle). Ce and Ci,m values are indicated. f–i Boxplots of promoter state averaged over 90–100 h for 400 independent simulations. In simulations, the promoter level at end of third instar (100 h) did not change over further developmental time, thus it is used here as a proxy for adult eye colour. NB: number of promoter sites in B conﬁguration, s: total number of sites. Boxplot parameters as in Fig. 2d, e. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 In contrast, a 1-h heat shock given during the initiation phase (2h10 to 4h50) gave stable activation and memory, even after the removal of the heat shock stimulus (Fig. 3d, early). Parameter space analysis revealed a similar optimum range of values for the three free parameters as those deﬁned for memory of silencing (Supplementary Fig. 3). In somatic cells, whilst PC, H3K4me1 and H3K4me3 were detectable in interphase nuclei at stages 5–6, the other modiﬁca- tions (H3K36me and H3K27me3) ﬁrst became detectable in a subset of interphase nuclei at stage 9 (Fig. 4d). The same modiﬁcations became visible earlier at stages 6–7 in mitotic nuclei (ref. 37, marked ‘m’ in Fig. 4 and Supplementary Fig. 5). The appearance in mitotic nuclei before detection in interphase is likely due to the condensation of mitotic chromatin, making the epitope more readily detectable. We note that38 also detected H3K27me3 in early mitotic nuclei and in pole cell nuclei, however, immunoﬂuorescence data on accumulation in inter- phase nuclei were not presented, and time points beyond stage 5 were not analysed in that study. y y We did not detect histone modiﬁcations by immunoﬂuores- cence during cycles 1–13. Several authors have observed histone modiﬁcations present in chromatin during these early cycles by ChIP34,38. However, these studies did not address time points later than stage 5 (2h50, see Supplementary Fig. 6), and a recent study performing mass spectrometry did not address the early cycles 1–335. To evaluate PC and histone modiﬁcations across a wider time window, we performed ChIP followed by whole- genome qPCR on staged embryo collections (Fig. 4e, see “Methods”). This analysis showed that PC and all histone modiﬁcations were detectable on chromatin during the ﬁrst 2 h of development (stages 1–4, corresponding to cycles 1–13), and subsequently accumulated at different rates, reaching maximal levels at 8–10 h. The H3K4me1 and me3 modiﬁcations were detectable at earlier stages than H3K27me and H3K36me, consistent with our immunoﬂuorescence data and with a recent ChIP study using more highly resolved time windows34 (Supplementary Fig. 6). We note that different histone modiﬁca- tions and PC show different kinetics of detectable accumulation in both the immunoﬂuorescence and ChIP assays, whereas the model predicts a smooth increase of all states during cycles 5–9. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 g Initial state of all nucleosomes in the PRE/ TRE set to A, U or M. h Ce, was varied as shown. Later coupling was not varied (Ci,m = 4). i Length of cycles 1–13 was varied as indicated. See also Supplementary Figs. 1 and 3. NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications 6 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 H3K4me1 and H3K4me3). Unmodiﬁed histone H3 served as an internal control. The results are summarised in Fig. 4 and Supplementary Fig. 5. throughout development, giving a proportion of adults with red eyes and some variegation (Fig. 3b, right)6. Several different PRE/TREs have been shown to give memory of activation in this assay6,10,17. The effectiveness of heat shock during embryonic but not larval stages has been proposed to indicate a qualitative change in the mechanism of PRE/TRE action during development6. Neither histone modiﬁcations nor the PC protein showed any detectable signal in interphase nuclei during cycles 1–13 (Fig. 4b, c and Supplementary Fig. 5). All accumulated with different kinetics over the subsequent developmental stages. At 11 h (late maintenance phase), all nuclei contained visible signals for all marks. Interestingly, the earliest detectable signal for several modiﬁcations was in the pole cells, typically becoming visible at stage 5–6 (early cycle 14), before the somatic nuclei showed a signal (Fig. 4b-d and Supplementary Fig. 5). This observation is consistent with the prediction of the model (Fig. 4a), and suggests that early cell cycle exit may contribute to accumulation of chromatin modiﬁcations. p To ask whether the model can recapitulate these observations we adapted the model promoter without the PRE/TRE to the experimental heat shock conditions (without heatshock, with late or with early heatshock; Fig. 3c). We then evaluated the effect of coupling a PRE/TRE to this model promoter in time courses simulated for the whole of development, including the adult eye (Fig. 3c-e and Supplementary Tables 1 and 2, Supplementary Methods). Remarkably, the same values of the four parameters (p3, p4) (feedback) p5 (feedback independent transitions) Ce (early coupling) and Cim (late coupling) as used above (Fig. 2b, c (middle)) were also able to accurately recapitulate the experi- mentally observed memory of heat shock-induced activated states (Fig. 3d). In particular, a 1-h heat shock given after the end of the initiation phase (4h50) allowed transient activation of the reporter but failed to induce stable switching and memory (Fig. 3d, late). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 However, in the model, the ‘A’ and ‘M’ system states comprise all modiﬁcations and bound proteins that contribute to that state, and thus in its current form the model does not enable predictions about separate molecular events. The important similarities of our own and other published data to the model prediction are that the PC protein and speciﬁc histone modiﬁcations are present on chromatin at low levels during the early cycles 1–13, accumulate during development, and reach maximum levels during the maintenance phase. Similarly to the silencing experiment described above, the strength of memory was independent of initial PRE/TRE state (Fig. 3g), but was dependent on coupling and length of the ﬁrst 13 cycles (Fig. 3h, i). We conclude that the same model, with the same parameter values and identical cell cycle constraints are sufﬁcient to recapitulate memory of both silencing and activation. Chromatin modiﬁcations increase during development. Several previous theoretical studies of chromatin-based epigenetic memory have shown that stability of memory in bistable model systems increases with increasing cell cycle length21,31. However, these models have not been applied to the observed real changes in cell cycle length that occur during development. To examine the effect of developmental changes in Drosophila division cycle length on the rate of global accumulation of A and M states in the model, we ran simulations over a developmental time course. We ran 1000 simulations on a nucleosomal array that was not cou- pled to a promoter, over 11 h of development and scored the average levels of A and M states at different time points (Fig. 4a). This showed that in the model, A and M states are present at low levels from the onset of development and accumulate slowly during development, reaching maximum levels only during late embryonic development. This is consistent with the observation that heterochromatin features (HP1 and H3K9 methylation) accumulate slowly during cycles 11–13 and increase substantially during cycle 1432,33, and with ChIP34 and mass spectrometry analysis35 showing that several other histone modiﬁcations also accumulate during Drosophila development. The model further predicts that histone modiﬁcations will accumulate earlier in pole cells, which exit the cell cycle at cycle 10 and commit to the germ cell fate (Fig. 4a)36. Two PRE/TREs respond differently to dynamic input. We have established model conditions that recapitulate two classical paradigms of PRE/TRE mediated regulation, namely epigenetic memory of silent and of active promoter states. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 stage Phase Stage 4:20 NM/S NA/S Germline cells Somatic cells Somatic cells p p p p B’ C’ Germline cells 2:50 2:50 Phase Model cycles Dev. stage Early cycles Initiation Mainte- nance Early cycles (1–13) Initiation (cycle 14) Maintenance (cycle 15 & above) e ChIP 0 0.1 αH3K36me2 αH3K36me3 Early cycles 1– 4 5 – 9 10–14 αPC 0 0.15 0.1 0.05 Phase Stage Embryo age (h) 0–2 αH3K4me3 αH3K4me1 8–10 2–4 4–6 0 0.4 0.8 0 0.1 0.2 Init- iation Maintenance Enrichment relative to Histone H3 ChIP 0 0.2 0.4 0 0.1 0.2 6–8 0.2 αH3K27me3 0.3 0.3 p p m m m m p p p p p p m a Model predictions A state (active) 0 Early cycles (1–13) Initiation (cycle 14) Maintenance (cycle 15 & above) 0 2:10 4:20 11:20 1 – 4 5 – 9 10 – 14 0.4 M state (silent) Phase Model cycles Dev. stage NM/S Somatic cells Germline cells 2:50 Stage 2 cycles 3–8 (0 h:25-1 h:05 ) Stage 3 cycle 9 (1 h:05-1 h:20) Stage 4 cycles 10–13 (1 h:20-2 h:10) Stage 5 cellularisation (2 h:10-2 h:50) Stage 6 gastrulation (I) (2 h:50-3 h:00) Stage 7 gastrulation (II) (3 h:00-3 h:10) Stage 9 germ band extension (3 h:40-4 h:20) Stage 14 Segmentation (10 h:20-11 h:20) Early cycles 1–13 Initiation Maintenance b αH3 αH3K27me3 Merge c B’ p p m m m m p p a Model predictions Embryo age (h:m) 0.4 0 0 2:10 11:20 1 – 4 5 – 9 10 – 14 A state (active) 0 Early cycles (1–13) Initiation (cycle 14) Maintenance (cycle 15 & above) 0 2:10 4:20 11:20 1 – 4 5 – 9 10 – 14 0.4 M state (silent) Phase Model cycles Dev. stage 4:20 NM/S NA/S Germline cells Somatic cells Somatic cells Germline cells 2:50 2:50 Phase Model cycles Dev. stage Early cycles (1–13) Initiation (cycle 14) Maintenance (cycle 15 & above) b a Early cycles 1–13 Embryo age (h:m) 0.4 0 0 2:10 11:20 1 – 4 5 – 9 10 – 14 A state (active) 4:20 NA/S Germline cells Somatic cells 2:50 Phase Model cycles Dev. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 stage Early cycles (1–13) Initiation (cycle 14) Maintenance (cycle 15 & above) B’ Stage 5 cellularisation (2 h:10-2 h:50) p p Stage 6 gastrulation (I) (2 h:50-3 h:00) p m p Interhpase nuclei with detectable modification All 0 αH3K27me3 αH3K36me2 αH3K36me3 Partial All 0 Partial All Partial 1– 4 14 d 5 6 7 9 p p p p p All 0 Partial αPC Summary of image analysis (panel D, E and Figure S7) Embryo age (h:m) 0 2:10 11:20 4:20 αH3K4me3 αH3K4me1 All 0 Partial All 0 Partial Phase Stage p p p p Early cycles Initiation Mainte- nance d m Stage 7 gastrulation (II) (3 h:00-3 h:10) m Stage 9 germ band extension (3 h:40-4 h:20) m m Stage 14 Segmentation (10 h:20-11 h:20) ally different in its eciﬁc cells from a t does not appear repression because 0, and it displays a Fig. 5b (for more ethods). The eya 5′ region contains two potential PRE/TREs that are enriched for PcG and/or TrxG proteins in ChIP-seq data sets (Fig. 5a). One putative PRE/TRE is at the promoter and a second is in the ﬁrst intron. The intronic PRE/TRE is well-characterised in reporter assays and contains a high density of binding sites for the PHO, GAF and Zeste proteins12,15 (Fig. 5a). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 To evaluate whether the model is applicable to a more dynamic mode of regulation we challenged it with an extreme case: the eyes absent To test the model predictions in individual animals, we analysed the accumulation of the Polycomb (PC) protein and of histone modiﬁcations in stage 2–14 Drosophila embryos by immunoﬂuorescence (Fig. 4 and Supplementary Figs. 4–6). We analysed the PC protein and histone modiﬁcations associated with silencing (H3K27me3), and activation (H3K36me2, H3K36me3, TURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications 7 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 We chose to experimentally analyse the function of the intronic PRE/TRE in M Stage 2 cycles 3–8 (0 h:25-1 h:05 ) Stage 3 cycle 9 (1 h:05-1 h:20) Stage 4 cycles 10–13 (1 h:20-2 h:10) Stage 5 cellularisation (2 h:10-2 h:50) Stage 6 gastrulation (I) (2 h:50-3 h:00) Stage 7 gastrulation (II) (3 h:00-3 h:10) Stage 9 germ band extension (3 h:40-4 h:20) Stage 14 Segmentation (10 h:20-11 h:20) Early cycles 1–13 Initiation Maintenance αH3 αH3K36me2 Merge c C’ p p p p m Stage 2 cycles 3–8 (0 h:25-1 h:05 ) Stage 3 cycle 9 (1 h:05-1 h:20) Stage 4 cycles 10–13 (1 h:20-2 h:10) Stage 5 cellularisation (2 h:10-2 h:50) Stage 6 gastrulation (I) (2 h:50-3 h:00) Stage 7 gastrulation (II) (3 h:00-3 h:10) Stage 9 germ band extension (3 h:40-4 h:20) Early cycles 1–13 Initiation ce αH3 αH3K36me2 Merge c C’ p p p p m c e ChIP 0 0.1 αH3K36me2 αH3K36me3 Early cycles 1– 4 5 – 9 10–14 αPC 0 0.15 0.1 0.05 Phase Stage Embryo age (h) 0–2 αH3K4me3 αH3K4me1 8–10 2–4 4–6 0 0.4 0.8 0 0.1 0.2 Init- iation Maintenance Enrichment relative to Histone H3 ChIP 0 0.2 0.4 0 0.1 0.2 6–8 0.2 αH3K27me3 0.3 0.3 e C’ Initiation Embryo age (h) The eya 5′ region contains two potential PRE/TREs that are enriched for PcG and/or TrxG proteins in ChIP-seq data sets (Fig. 5a). One putative PRE/TRE is at the promoter and a second is in the ﬁrst intron. The intronic PRE/TRE is well-characterised in reporter assays and contains a high density of binding sites for the PHO, GAF and Zeste proteins12,15 (Fig. 5a). We chose to experimentally analyse the function of the intronic PRE/TRE in (eya) gene (Fig. 5). The eya gene is fundamentally different in its regulation to the Hox genes: it switches in speciﬁc cells from a silent to an active state late in development39, it does not appear to require a classical memory of activation and repression because its activators and repressors remain present13,40, and it displays a gradient rather than an all-or-none pattern39, Fig. 5b (for more details on eya regulation see Supplementary Methods). (eya) gene (Fig. 5). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 (eya) gene (Fig. 5). The eya gene is fundamentally different in its regulation to the Hox genes: it switches in speciﬁc cells from a silent to an active state late in development39, it does not appear to require a classical memory of activation and repression because its activators and repressors remain present13,40, and it displays a di t th th ll tt 39 Fi 5b (f The eya 5′ region contains two potential PRE/TREs that are enriched for PcG and/or TrxG proteins in ChIP-seq data sets (Fig. 5a). One putative PRE/TRE is at the promoter and a second is in the ﬁrst intron. The intronic PRE/TRE is well-characterised in reporter assays and contains a high density of binding sites for th PHO GAF d Z t t i 12 15 (Fi 5 ) W h t a Interhpase nuclei with detectable modification All 0 αH3K27me3 αH3K36me2 αH3K36me3 Partial All 0 Partial All Partial 1– 4 14 d 5 6 7 9 p p p p p All 0 Partial αPC Stage 2 cycles 3–8 (0 h:25-1 h:05 ) Stage 3 cycle 9 (1 h:05-1 h:20) Stage 4 cycles 10–13 (1 h:20-2 h:10) Stage 5 cellularisation (2 h:10-2 h:50) Stage 6 gastrulation (I) (2 h:50-3 h:00) Stage 7 gastrulation (II) (3 h:00-3 h:10) Stage 9 germ band extension (3 h:40-4 h:20) Stage 14 Segmentation (10 h:20-11 h:20) Early cycles 1–13 Initiation Maintenance b αH3 αH3K27me3 Merge Stage 2 cycles 3–8 (0 h:25-1 h:05 ) Stage 3 cycle 9 (1 h:05-1 h:20) Stage 4 cycles 10–13 (1 h:20-2 h:10) Stage 5 cellularisation (2 h:10-2 h:50) Stage 6 gastrulation (I) (2 h:50-3 h:00) Stage 7 gastrulation (II) (3 h:00-3 h:10) Stage 9 germ band extension (3 h:40-4 h:20) Stage 14 Segmentation (10 h:20-11 h:20) Early cycles 1–13 Initiation Maintenance αH3 αH3K36me2 Merge c Summary of image analysis (panel D, E and Figure S7) Model predictions Embryo age (h:m) 0 2:10 11:20 4:20 αH3K4me3 αH3K4me1 All 0 Partial All 0 Partial Embryo age (h:m) 0.4 0 0 2:10 11:20 1 – 4 5 – 9 10 – 14 A state (active) 0 Early cycles (1–13) Initiation (cycle 14) Maintenance (cycle 15 & above) 0 2:10 4:20 11:20 1 – 4 5 – 9 10 – 14 0.4 M state (silent) Phase Model cycles Dev. ARTICLE ARTICLE Fig. 4 Chromatin modiﬁcations increase during Drosophila development. a Simulated time course of accumulation of M (top) and A (bottom) nucleosomes as a proportion of total nucleosomes (S) in the array over developmental time, averaged over 1000 individual simulations. Solid grey: somatic cells, red line: germline cells, modelled by implementing early cycles 1–9, followed by cycle 10 of 2 h, and subsequent G2 arrest for the rest of the simulation36. Parameters as in Fig. 2b, c (no coupling between PRE/TRE and promoter). b, c Embryos were ﬁxed and double stained with antibodies against unmodiﬁed histone H3 (green, left panels) and H3K27me3 (b) or H3K36me2 (c) (magenta, middle panels). Merge: right panels. Embryos were staged according to morphology, as indicated. Three slides per antibody were prepared, typically containing 50–100 embryos, of which 5–10 were at the required stage and all showed similar staining for a given stage. White boxes at stages 5 and 6 indicate pole cells (p), which ﬁrst become apparent at stage 5. B’ and C’ show 3× zoom of boxed area. Mitotic cells are indicated at stages 6–9 where visible (m). Scale bar represents 75 μm and is the same for all panels except B’ and C’. d Summary of immunoﬂuorescence analysis shown in (b, c) and Supplementary Fig. 5. Red circles, p, indicate the stage at which staining of each modiﬁcation was visible in pole cell nuclei. Stages at which histone modiﬁcations became visible in interphase nuclei are indicated (white or grey bars). “Partial” indicates that a proportion of nuclei showed detectable histone modiﬁcations. “All” indicates that all nuclei contained signal for the modiﬁcation or protein. e ChIP analysis of PC and histone modiﬁcations in staged embryos as indicated. Global levels are shown as a proportion of histone H3 ChIP for each stage (see “Methods”). Data are presented as mean values ± SD of two IPs each made from two independent chromatin preparations (= four independent IPs in total). Individual data values are shown as black circles. To compare the properties of different PRE/TREs in this assay, we replaced the 1.5 kb eya intronic PRE/TRE in the reporter construct with a 1.5 kb fragment of the bithoraxoid (bxd) PRE/ TRE, which shows memory of both silencing and activation in the assays described in Figs. 2 and 32,3,10. Surprisingly, the bxd PRE/ TRE caused strong variegation of GFP expression (Fig. 6g, h). ARTICLE The same model recapitulates the behaviour of both PRE/ TREs. We next asked whether our simple model can capture these differences, and if so, with which parameters? We imple- mented two versions of the model, one in which the model reporter locus contains two PRE/TREs (one at the TSS and one intronic, Fig. 5a), and a second in which it contains only one intronic PRE/TRE. Both models reached similar conclusions regarding the properties of the eya and bxd PRE/TREs (Supple- mentary Fig. 10 and Supplementary Methods). Thus we describe the results of the simpler, one-PRE/TRE model here. The two- PRE/TRE model is described in detail in Supplementary Methods. We ﬁrst adapted the model to recapitulate the events involved in eye disc proliferation and differentiation (see Supplementary Methods). We then modelled the observed gradient shape of the eya1::GFP reporter gene without the intronic PRE/TRE in the 3rd instar larval eye disc. To this end, we dynamically adjusted the parameter p1 to create a gradient across the simulated eye disc, and ﬁtted this gradient to the experimentally observed shape (Fig. 6 and Supplementary Fig. 8, Supplementary Methods). To examine the effect of the model PRE/TRE on this gradient, we coupled the PRE/TRE to the promoter and ran simulations for the whole of development from embryonic cycle 1 up to mid 3rd instar. We searched for PRE/TRE parameter combinations that would recapitulate the experimentally observed effect of the eya PRE/TRE on the promoter, namely to sharpen the anterior- posterior gradient in a smooth manner (Fig. 6f and Supplemen- tary Fig. 8). Deletion and mutation analysis of the eya1::GFP transgene conﬁrmed that the intronic PRE/TRE is required for PcG- dependent repression of the mw reporter. In the absence of this intronic PRE/TRE (Fig. 5d, l), or in a version in which binding sites for the PcG protein PHO were mutated, the mw reporter was strongly derepressed (Fig. 5s, t and Supplementary Fig. 7D). To evaluate genetic interactions, we introduced each variant of the eya1::GFP transgene into a heterozygous mutant background for the PcG gene polymoeotic (ph410). Constructs containing the enhancer were highly expressed and showed no further activation in a ph410 mutant background (Fig. 5o, p). The construct lacking both the enhancer and the intronic PRE/TRE was expressed at lower levels but also showed no response to the ph410 mutation (Fig. 5r). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 The eya gene is fundamentally different in its regulation to the Hox genes: it switches in speciﬁc cells from a silent to an active state late in development39, it does not appear to require a classical memory of activation and repression because its activators and repressors remain present13,40, and it displays a gradient rather than an all-or-none pattern39, Fig. 5b (for more details on eya regulation see Supplementary Methods). NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications 8 ARTICLE Interestingly, this variegation was also graded across the eye disc, with cells that express the reporter gene ahead of the morphogenetic furrow doing so at a higher level than those behind it (Fig. 6g). We conclude that the bxd PRE/TRE acts as a bistable element in this assay, but is nevertheless receptive to gene expression level differences across the gradient. Thus the two PRE/TREs we have tested behave very differently when coupled to the dynamically regulated eya promoter and enhancer. this study, as the putative promoter PRE/TRE spans the transcription start site, which is required for reporter expression, and thus its deletion would disrupt reporter gene transcription. An upstream enhancer drives eya expression in the developing larval eye disc39 (Figs. 5a, and 6a–c). y g Before implementing the model, we ﬁrst dissected the contributions of the enhancer and the intronic PRE/TRE to the eya expression pattern experimentally. To this end, we placed a green ﬂuorescent protein (GFP) reporter gene under control of eya regulatory sequences by generating the eya1::GFP reporter construct. The reporter contains 9.2 kb of genomic eya sequence, including the eya enhancer, promoter, putative TSS PRE/TRE, exon 1 and the intronic PRE/TRE. A TurboGFP reporter gene is fused in frame to the ﬁrst three codons of eya exon 2, and a second mw reporter enables analysis in adults (Figs. 5c and 6c). To exclude genomic position effects, transgenic ﬂies were generated in which the eya1::GFP construct and variants lacking the PRE/TRE or the enhancer were integrated into the same genomic location as described in41 (Figs. 5c–f and 6c). The eya1:: GFP reporter construct expressed GFP in a similar pattern to wild type eya mRNA, in a manner dependent on the presence of the eya enhancer (Fig. 5b–j). The same model recapitulates the behaviour of both PRE/ TREs. We next asked whether our simple model can capture these differences, and if so, with which parameters? We imple- mented two versions of the model, one in which the model reporter locus contains two PRE/TREs (one at the TSS and one intronic, Fig. 5a), and a second in which it contains only one intronic PRE/TRE. Both models reached similar conclusions regarding the properties of the eya and bxd PRE/TREs (Supple- mentary Fig. 10 and Supplementary Methods). Thus we describe the results of the simpler, one-PRE/TRE model here. The two- PRE/TRE model is described in detail in Supplementary Methods. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 eya1::GFPΔPRE/TRE eya1::GFPΔen eya1::GFP ΔenΔPRE/TRE c d e f l m n mw het mw hom k h i j g GFP eya1::GFP MF A P MF A P MF A P MF A P GAF/PSQ GAGAG PHO/PHOL GCCAT ZESTE YGAGYG 0 180 180 0 exon 1 Enhancer eya exon 1 MF MF RNA In situ hybridisation 1kb Key to elements in C - F eya enhancer eya 5′ UTR eya coding sequence eya PRE/TRE gfp reporter miniwhite reporter A P A P p q r o ph410 ph+ 0 210 eya1::GFP eya1::GFP ΔPHO het hom eya locus Chr2L:6548067-6544683 GAF H3K27me3 180 0 PRE/TRE score PSC 351 bp deleted in Δenhancer 1559 bp deleted in ΔPRE/TRE DAPI eya exon 1 s t a b eporter constructs for eya PRE/TRE and enhancer. a 3.4 kb of eyes absent (eya)_ locus included in the eya 1 transgene. BDGP Drosop gaster genome version R5/dm3, Chr2L:6548067–6544683. Coloured bars show motif occurrence as indicated. PRE/TRE score (blue tra ed according to ref. 12. Grey: ChIP-seq tracks for the proteins or histone modiﬁcations indicated, obtained from ModEncode (http://comp edu/modencode/webpage/Chromatin.v0.6.html#ChIP-seq%20and%20ChIP-chip%20data). ChIP-seq 14–16 h Oregon R embryos, Karp me3 (track ID 3955), PSC, Posterior Sex Combs (track ID 3960), GAF, GAGA Factor (track ID 4119). Below the plots the regions that are d nic reporters are indicated: 1559 bp deleted in ΔPRE/TRE: Chr 2L:6546243–6544711. 351 bp deleted in Δenhancer lines Chr 2L: 6547869–6 n situ hybridisation against exon 1 of eya on wild type 3rd instar eye-antenna imaginal disc. Anterior (A) posterior (P). Morphogenetic furro r, 100 μm, same for (g–j). c eya1::GFP reporter construct. Sequence encoding turboGFP (green) was fused to exon 2 of the eya gene (dark enic reporter construct carrying the eya enhancer (yellow), exon 1 (blue) and the intronic eya PRE/TRE (black). The sequence shown in onds to the enhancer, ﬁrst exon and intronic PRE/TRE. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 The construct contains a mw reporter (red) for detection of expression in adult ants of the reporter construct lacking the intronic PRE/TRE (d), lacking the enhancer (e), or lacking both (f), were integrated by ΦC31 int i i f GFP t i i t l di f 3 d i t l h f th t i di t d G GFP d SYTO R d l GAF/PSQ GAGAG PHO/PHOL GCCAT ZESTE YGAGYG 0 180 180 0 exon 1 Enhancer 1kb 0 210 eya locus Chr2L:6548067-6544683 GAF H3K27me3 180 0 PRE/TRE score PSC 351 bp deleted in Δenhancer 1559 bp deleted in ΔPRE/TRE a b a 1559 bp deleted in ΔPRE/TRE eya1::GFPΔPRE/TRE eya1::GFPΔen eya1::GFP ΔenΔPRE/TRE c d e f l m n mw het mw hom k h i j g GFP eya1::GFP MF A P MF A P MF A P MF A P Key to elements in C - F eya enhancer eya 5′ UTR eya coding sequence eya PRE/TRE gfp reporter miniwhite reporter p q r o ph410 ph+ eya1::GFP eya1::GFP ΔPHO het hom e a ce / s t eya1::GFP eya1::GFP ΔPHO c d h e Key to elements in C - F eya enhancer eya 5′ UTR eya coding sequence eya PRE/TRE gfp reporter miniwhite reporter Fig. 5 Reporter constructs for eya PRE/TRE and enhancer. a 3.4 kb of eyes absent (eya)_ locus included in the eya 1 transgene. BDGP Drosophila melanogaster genome version R5/dm3, Chr2L:6548067–6544683. Coloured bars show motif occurrence as indicated. PRE/TRE score (blue track) was calculated according to ref. 12. Grey: ChIP-seq tracks for the proteins or histone modiﬁcations indicated, obtained from ModEncode (http://compbio.med. harvard.edu/modencode/webpage/Chromatin.v0.6.html#ChIP-seq%20and%20ChIP-chip%20data). ChIP-seq 14–16 h Oregon R embryos, Karpen lab. H3K27me3 (track ID 3955), PSC, Posterior Sex Combs (track ID 3960), GAF, GAGA Factor (track ID 4119). Below the plots the regions that are deleted in transgenic reporters are indicated: 1559 bp deleted in ΔPRE/TRE: Chr 2L:6546243–6544711. 351 bp deleted in Δenhancer lines Chr 2L: 6547869–6547519. b RNA in situ hybridisation against exon 1 of eya on wild type 3rd instar eye-antenna imaginal disc. Anterior (A) posterior (P). Morphogenetic furrow (MF). Scale bar, 100 μm, same for (g–j). c eya1::GFP reporter construct. Sequence encoding turboGFP (green) was fused to exon 2 of the eya gene (dark blue) in a transgenic reporter construct carrying the eya enhancer (yellow), exon 1 (blue) and the intronic eya PRE/TRE (black). ARTICLE In contrast, the construct lacking the enhancer but containing the intronic PRE/TRE showed derepression in a ph410 mutant background (Fig. 5q). Taken together these results indicate that the intronic PRE/TRE plays a role in PcG mediated regulation of the eya1::GFP reporter. We ﬁrst adapted the model to recapitulate the events involved in eye disc proliferation and differentiation (see Supplementary Methods). We then modelled the observed gradient shape of the eya1::GFP reporter gene without the intronic PRE/TRE in the 3rd instar larval eye disc. To this end, we dynamically adjusted the parameter p1 to create a gradient across the simulated eye disc, and ﬁtted this gradient to the experimentally observed shape (Fig. 6 and Supplementary Fig. 8, Supplementary Methods). To examine the effect of the model PRE/TRE on this gradient, we coupled the PRE/TRE to the promoter and ran simulations for the whole of development from embryonic cycle 1 up to mid 3rd instar. We searched for PRE/TRE parameter combinations that would recapitulate the experimentally observed effect of the eya PRE/TRE on the promoter, namely to sharpen the anterior- posterior gradient in a smooth manner (Fig. 6f and Supplemen- tary Fig. 8). To obtain quantitative information, we performed in situ hybridisations against the GFP mRNA and compared the patterns with and without the intronic PRE/TRE. Interestingly, the PRE/ TRE had the effect of sharpening the gradient pattern established by the enhancer: the level of activation ahead of the morphoge- netic furrow was higher, and the expression level posterior to the furrow was lower than in the construct without the PRE/TRE (Fig. 6d–f). We conclude that the intronic PRE/TRE ﬁne-tunes the expression pattern of GFP in the eya1::GFP reporter construct. y g This analysis showed that the model can indeed recapitulate the smooth and sharpened gradient with speciﬁc parameters (Fig. 6m, n and Supplementary Fig. 9). Interestingly the optimum parameter range differed from that deﬁned previously for the memory assays (Figs. 2 and 3) in several respects. First, the timing and strength of coupling between the PRE/TRE and the promoter was critical. The best results were obtained when coupling was 9 NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications 9 ARTICLE NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 The sequence shown in (a) corresponds to the enhancer, ﬁrst exon and intronic PRE/TRE. The construct contains a mw reporter (red) for detection of expression in adult eyes. d–f Variants of the reporter construct lacking the intronic PRE/TRE (d), lacking the enhancer (e), or lacking both (f), were integrated by ΦC31 integration. g–j Live imaging of GFP protein in eye antennal discs of 3rd instar larvae homozygous for the transgenes as indicated. Green: GFP; red: SYTO Red live DNA stain. Six to 10 discs per genotype were analysed, giving similar results. k–n Eye colours of 4 day-old adult female ﬂies heterozygous (left) or homozygous (right) for the transgenes as indicated. o–r Females homozygous for the transgene variants in a heterozygous mutant background for the Polycomb group gene polyhomeotic (ph). s, t eya1::GFP ΔPHO: all eight PHO binding sites of the eya1::GFP transgene were mutated (see Supplementary Fig. 7D). Four-day old females heterozygote (S) and homozygote (T) for the transgenes are shown. introduced only at the onset of eya activation, and was 3-fold weaker than that required for memory in the previous simulations (Supplementary Fig. 9). Earlier and stronger coupling resulted in silencing of the PRE/TRE in response to the early silent state of the promoter (Supplementary Fig. 9). Second, the best ﬁt was obtained when the values of p3 and p4 (feedback) were reduced from 0.25 to 0.2, and the feedback-independent transition parameter, p5, was increased approximately four-fold to 0.17. The large increase in p5 and concomitant reduction in (p3, p4) renders the PRE/TRE more ﬂexible, allowing it to respond continuously to the promoter state. Thus with speciﬁc parameter changes the simple model is able to quantitatively NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunicatio 10 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 We note that in disc, the model predicted a higher aver a 2 3 1 GFP bxd PRE/TRE en g MF MF c No PRE/TRE MF MF A P With PRE/TRE MF P A Anterior Posterior Antenna disc Eye disc Model promoter output 1.0 (active) 0.0 (silent) MF P A No PRE/TRE GFP PRE/TRE en eya1::GFP 1kb Relative GFP intensity GFP mRNA GFP mRNA eya1::bxd::GFP Model promoter output p3, p4 = 0.25 p5 = 0.04 o 2 3 1 Anterior Posterior Model eye disc Disc zone Disc zone 2 3 1 2 3 1 MF (morphogenetic furrow) eya expression Model promoter output A Model promoter - PRE/TRE p1, p2 (p3, p4), p5 C C b f i k j d n l Model promoter - bxd PRE/TRE Experiment Model eya exon 1 exon 2 DAPI; GFP mRNA DAPI; GFP mRNA 2 3 1 2 3 1 MF P A 0 1 2 3 1 Disc zone Disc zone With PRE/TRE No PRE/TRE MF P A 0 1 2 3 1 Disc zone Model promoter output p With PRE/TRE No PRE/TRE MF P A 0 1 2 3 1 h With PRE/TRE No PRE/TRE MF P A 0 1 2 3 1 Disc zone Disc zone Relative GFP intensity Disc zone Disc zone MF (morphogenetic furrow) With PRE/TRE MF MF A P p3, p4 = 0.2 p5 = 0.17 MF P A GFP mRNA e m With PRE/TRE 2 3 1 DAPI; GFP mRNA Disc zone With PRE/TRE No PRE/TRE A P A P A P A P Model promoter output 1.0 (active) 0.0 (silent) P MF A P MF 2 3 1 Anterior Posterior Model eye disc Disc zone i Model MF (morphogenetic furrow) i a a 2 3 1 Anterior Posterior Antenna disc Eye disc Disc zone MF (morphogenetic furrow) Experiment c eya1::GFP 2 3 1 MF (morphogenetic furrow) eya expression b Disc zone A P MF (morphogenetic furrow) 2 3 1 Model promoter output A Model promoter - PRE/TRE k j Disc zone (morphogenetic furrow) P MF j b k c c GFP PRE/TRE en eya1::GFP 1kb eya exon 1 exon 2 l l Model promoter output 1.0 (active) 0.0 (silent) MF P A No PRE/TRE l 2 3 1 Disc zone m Disc No PRE/TRE MF MF A P GFP mRNA d DAPI; GFP mRNA A P A P Model promoter output 1.0 (active) 0.0 (silent) MF P A No PRE/TRE n l 2 3 1 Disc Disc zone p3, p4 = 0.2 p5 = 0.17 MF P A m With PRE/TRE 2 3 1 Disc zone d With PRE/TRE MF MF A P GFP mRNA e DAPI; GFP mRNA A P A P m e Relative GFP intensity f With PRE/TRE No PRE/TRE MF P A 0 1 2 3 1 Disc zone f n Model promoter output o n Model promoter - bxd PRE/TRE MF P A 0 1 2 3 1 Disc zone With PRE/TRE No PRE/TRE GFP bxd PRE/TRE en g MF MF GFP mRNA eya1::bxd::GFP DAPI; GFP mRNA h With PRE/TRE No PRE/TRE MF P A 0 1 2 3 1 Disc zone Relative GFP intensity o g GFP bxd PRE/TRE en g MF MF GFP mRNA eya1::bxd::GFP DAPI; GFP mRNA Model promoter - bxd PRE/TRE Model promoter - bxd PRE/TRE With PRE/TRE MF P A p3, p4 = 0.25 p5 = 0.04 2 3 1 Model promoter output 1.0 (active) 0.0 (silent) h ; h With PRE/TRE No PRE/TRE MF P A 0 1 2 3 1 Disc zone Relative GFP intensity p Model promoter output p With PRE/TRE No PRE/TRE MF P A 0 1 2 3 1 Disc zone eya, and for all other parameters (p3, p4 and p5), identical values to those determined for the memory assay (Fig. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 bserved behaviour of the eya PRE/TRE in the sc. hen searching for parameters that best reﬂected he bxd PRE/TRE in this assay we found that an eya, and for all other parameters (p3, p4 to those determined for the memory Supplementary Fig. 9). NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Fig. 6 Two PRE/TREs respond differently to a dynamically regulated promoter. a Third instar larval eye-antenna disc. b eya expression levels in mid-3rd instar larval eye disc across zones 1–3 (A, anterior; P, posterior,). c eya1::GFP reporter construct. Yellow (en), eya enhancer; blue, eya exons (light blue, 5′ UTR; dark blue, coding region); black, eya PRE/TRE. Grey bar: RNA in situ probe used in (d) and (e). d, e RNA in situ hybridisation against GFP on mid 3rd instar eye-antenna disc of larvae homozygous for the eya1::GFP transgene (e), or the same transgene lacking the PRE/TRE (d). Scale bar, 100 μm, same for (g). Six to ten discs for each transgene gave similar results. f Signal intensity proﬁles along the A-P axis (dotted lines shown on the discs in d, e). Multiple line scans perpendicular to the morphogenetic furrow were averaged for six - 10 discs for each transgene and aligned to the morphogenetic furrow, the anterior limit of the expression domain and the posterior edge of the disc. Vertical scale represents signal intensity relative to the maximum of GFP mRNA. g Top: eya1::bxd::GFP reporter construct. 1.5 kb of the eya PRE/TRE was replaced in the reporter construct with 1.5 kb of the bxd PRE/TRE (see “Methods”). Bottom: RNA in situ hybridization against GFP mRNA shows variegation. Similar results were obtained for eight discs. h Average signal intensity proﬁles generated as in (f), except that the whole disc area was scanned to reduce noise due to variegation. i Model eye disc. The eye disc was modeled as described in Supplementary Methods and Supplementary Fig. 8. j The model promoter output was quantiﬁed by averaging the anterior-posterior proﬁles of 100 simulations. k–m Model promoter and PRE/TRE. l No PRE/TRE. m With PRE/TRE (Model 1, Ci,m = 0 (zone 1); Ceya = 2.5 (zones 2 and 3)). n Average model promoter output for 100 independent simulations. o Simulated eye disc showing best ﬁt to eya1::bxd::GFP data shown in (h). p Average model promoter output for 100 independent simulations. See also Supplementary Figs. 1, 7 and 8. maintenance phase, PRE/TRE states become ﬁxed and are imposed on a coupled promoter, giving epigenetic memory (Figs. 2 and 3). In the model, these three phases depend only on cell cycle length and timing of coupling, and no other develop- mental change in PRE/TRE properties is required to recapitulate experimental data for epigenetic memory. Discussion We have combined mathematical modelling and quantitative experimental analysis to examine different modes of PcG/TrxG- mediated gene regulation in Drosophila and how they are modulated during development. The unique feature of the model described here in comparison to previous models of chromatin- mediated epigenetic regulation21,22,31,42 is that chromatin states and transcriptional states are considered explicitly and indepen- dently from each other, with regulated coupling between them. These features were essential for the model to recapitulate the observed data, thus identifying the simplest model required to describe the phenomena of interest43. We identify speciﬁc roles for cell cycle length, transcriptional input, PRE/TRE-promoter coupling and PRE/TRE identity in determining system behaviour. This simple model system is able to generate a rich repertoire of outputs including memory, variegation and ﬁne-tuning and thus provides a unifying theoretical framework for profoundly differ- ent modes of PcG/TrxG-mediated gene regulation (Fig. 7). In the model, regulated coupling between the PRE/TRE and promoter, and the absence of coupling during cycles 1–13, was essential to recapitulate all three data sets (memory of silencing, memory of activation, and the eya gradient). This suggests that in the living animal, coupling and decoupling between PRE/TREs and promoters or enhancers may be devel- opmentally regulated, locus-speciﬁc and biologically important. Several candidate mechanisms for coupling exist, including spreading of marks from a nucleation site49, speciﬁc PRE/TRE- promoter communication19,50, regulation by non-coding RNAs20,51, long-range chromatin interactions52 and trans-regulation by homo- logous pairing53,54 (see Supplementary Discussion). Interestingly, several of these phenomena are absent during the early cycles and accumulate during cycle 14, the time window in which we predict that coupling is required19,50,52. We have shown both experimentally and theoretically that the identity of the PRE/TRE itself is decisive in determining the regulatory properties of the system. We show experimentally that when two different PRE/TREs are placed in an otherwise identical regulatory setting, the bxd PRE/TRE causes variegation whilst the eya PRE/TRE gives a smooth gradient. Since the reporters are placed in identical genomic locations, these differences in An important ﬁnding of this study is that the model system becomes increasingly stable as development proceeds (Fig. 7a). During the ﬁrst 13 rapid division cycles the frequent disruptions of replication result in a naive PRE/TRE state in the model. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 may be due to the large variation in degree of variegation in the imaged eye discs. Nevertheless, the criteria for a good ﬁt, namely the shape of the gradient and the variegating pattern, were fulﬁlled by a range of values for the parameters p3, p4 and p5 that are mutually exclusive to those that ﬁt the smooth eya PRE/TRE gradient (Supplementary Fig. 9). The parameter p5 (feedback-independent transitions) for the best ﬁt of the bxd PRE/TRE model to the data (p5 = 0.04) was four-fold lower than for the eya PRE/TRE model (p5 = 0.17). The best- ﬁt values of the feedback parameters (p3, p4) were higher for the bxd PRE/TRE ((p3, p4) = 0.25) than for the eya PRE/TRE ((p3, p4) = 0.2). Low values of p5 and high values of (p3, p4) tend to make the system more bistable and inﬂexible, leading to variegation, caused by the perpetuation of random active and silent PRE/TRE states that were adopted before coupling. The gradient imposed by the promoter was nevertheless discernable in the bxd model at intermediate coupling strengths (Supplementary Fig. 9D). At higher coupling strengths, the variegation imposed by the PRE/TRE dominated completely and the gradient was lost (Supplementary Fig. 9F, G). Taken together these results demonstrate that deﬁned parameter changes enable the simple model to capture both the late switching, ﬁne-tuning behaviour of the eya PRE/TRE and the variegation of the bxd PRE. The dependence of stability of epigenetic memory on cell cycle length has been observed in several theoretical studies21,31. Here, we extend those observations to model the entire lifespan of a developing animal. We propose that developmentally regulated changes in cell cycle length may contribute to the observed developmental increase in stability of chromatin states in Drosophila3,6,44–47 and to the progressive accumulation of histone modiﬁcations during development34,35. In ref. 31 this idea was proposed but not explicitely tested. Interestingly, a similar effect has been observed for heterochromatin in Drosophila. HP1 and H3K9 methylation progressively accumulate with the progressive lengthening of S-phase during cycles 11–1332,33. This accumu- lation is in increased upon arrest of cycle 1333, and abrogated in grp mutants, which have shortened cycles 11–1348. The fact that the system becomes globally more stable as development proceeds raises the question of how individual genes locally escape this constraint to allow late switching and dynamic responses to promoter input (Fig. 7b). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 We show here theoretically that the two most important factors enabling this escape are the coupling between PRE/TRE and promoter, and the inherent properties of the PRE/TRE itself. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 6o, p and Supplementary Fig. 9). We note that in the posterior part of the disc, the model predicted a higher average expression level than was observed in the experiment (compare Fig. 5h and p). This eya, and for all other parameters (p3, p4 and p5), identical values to those determined for the memory assay (Fig. 6o, p and Supplementary Fig. 9). We note that in the posterior part of the disc, the model predicted a higher average expression level than was observed in the experiment (compare Fig. 5h and p). This recapitulate the observed behaviour of the eya PRE/TRE in the developing eye disc. recapitulate the observed behaviour of the eya PRE/TRE in the developing eye disc. p g y Remarkably, when searching for parameters that best reﬂected the behaviour of the bxd PRE/TRE in this assay we found that an optimal ﬁt was obtained with the same coupling parameters as for Remarkably, when searching for parameters that best reﬂected the behaviour of the bxd PRE/TRE in this assay we found that an optimal ﬁt was obtained with the same coupling parameters as for 11 NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Inputs (global) Cell cycle length TrxG/PcG levels Coupling Promoter strength Inputs (local) t p3 p5 p4 C p1 p2 b p3 p5 p4 Histone turnover TrxG/PcG recruitment Recruitment-independent conversions b Promoter PRE/TRE PcG TrxG TFs Output depends on PRE/TRE state AND promoter state Promoter PRE/TRE PcG TrxG TFs Output depends on PRE/TRE state AND promoter state Fig. 7 Summary of model and main conclusions. a Summary of developmental changes in stability of memory in the model. b Summary of the model and its parameters. PcG: Polycomb group proteins; TrxG: Trithorax group proteins; TFs: transcription factors. Global inputs are those that are expected to affect all PRE/TRE regulated genes in the genome. Local inputs are those that may be regulated speciﬁcally for different loci. The output of the model in terms of transcriptional activity depends on the interplay between PRE/TRE state and promoter state. See main text for details. differences in known motifs between the bxd and eya PRE/TREs is the conﬁguration of binding sites for the TrxG protein GAGA factor (GAF) (Supplementary Fig. 7A–C). Whereas the bxd PRE/ TRE contains more single instances of the GAF binding motif (GAGAG), the eya PRE/TRE contains a run of ﬁve consecutive GAF motifs (Supplementary Fig. 7A). The GAF protein binds its target sites cooperatively and has nucleosome remodelling activity56,57. We propose that the conﬁguration of GAF binding sites in the eya PRE/TRE may contribute to increased nucleosome remodelling and thus to reduced stability of chromatin states. The identiﬁcation of model parameters and deﬁned DNA sequences that cause large differences in output will allow the design of precise perturbation experiments in future to address mechanistic questions linking regulatory output to DNA sequence features. PRE/TRE sequences are complex, diverse, and evolve rapidly12,24,58. To date, research has focused on the role of PRE/TRE sequence motifs in recruiting PcG and TrxG proteins24,25. We propose that PRE/TRE sequence may also inﬂuence the stability of PRE/TRE states, bringing a fresh per- spective to the PRE/TRE code. regulatory output must depend on nucleic acid sequence differ- ences between the two PRE/TREs. In our simple model, the dif- ference between the two PRE/TREs depends on differences in only two parameters: (p3, p4) (feedback) and p5 (feedback- independent transitions). In the model, the eya PRE/TRE with high p5 and low (p3, p4) is ﬂexible and able to respond continuously to promoter inputs, but unable to maintain memory. Discussion In the initiation phase (cycle 14), in the absence of replication, this naive PRE/TRE stabilises into active and silent states, and can still receive information from a coupled promoter. Finally, in the URE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications 12 ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Promoter PRE/TRE PcG TrxG TFs Inputs (global) Cell cycle length TrxG/PcG levels Coupling Promoter strength Inputs (local) Output depends on PRE/TRE state AND promoter state t p3 p5 p4 C p1 p2 Initiation Cycles 1–13 Maintenance Increasing stability of memory PRE/TRE naive: Memory destabilised by cell cycles PRE/TRE switchable: Can be instructed by promoter PRE/TRE stable: Can instruct promoter; Switching to new state requires local disruption of memory a b p3 p5 p4 Histone turnover TrxG/PcG recruitment Recruitment-independent conversions Fig. 7 Summary of model and main conclusions. a Summary of developmental changes in stability of memory in the model. b Summary of the model and its parameters. PcG: Polycomb group proteins; TrxG: Trithorax group proteins; TFs: transcription factors. Global inputs are those that are expected to affect all PRE/TRE regulated genes in the genome. Local inputs are those that may be regulated speciﬁcally for different loci. The output of the model in terms of transcriptional activity depends on the interplay between PRE/TRE state and promoter state. See main text for details. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 ARTICLE Initiation Cycles 1–13 Maintenance a a Initiation Cycles 1–13 Maintenance Increasing stability of memory PRE/TRE naive: Memory destabilised by cell cycles PRE/TRE switchable: Can be instructed by promoter PRE/TRE stable: Can instruct promoter; Switching to new state requires local disruption of memory a a PRE/TRE stable: Can instruct promoter; Switching to new state requires local disruption of memory PRE/TRE stable: Can instruct promoter; Switching to new state requires local disruption of memory Promoter PRE/TRE PcG TrxG TFs Inputs (global) Cell cycle length TrxG/PcG levels Coupling Promoter strength Inputs (local) Output depends on PRE/TRE state AND promoter state t p3 p5 p4 C p1 p2 b p3 p5 p4 Histone turnover TrxG/PcG recruitment Recruitment-independent conversions Fig. 7 Summary of model and main conclusions. a Summary of developmental changes in stability of memory in the model. b Summary of the model and its parameters. PcG: Polycomb group proteins; TrxG: Trithorax group proteins; TFs: transcription factors. Global inputs are those that are expected to affect all PRE/TRE regulated genes in the genome. Local inputs are those that may be regulated speciﬁcally for different loci. The output of the model in terms of transcriptional activity depends on the interplay between PRE/TRE state and promoter state. See main text for details. NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications Mathematical modelling. See Supplementary Methods. The H3 ChIP samples were not diluted, and thus a relative enrichment of 1 would be obtained if the quantity of IP material from ChIP x were equivalent to 100% of the IP material from the H3 ChIP. We note that such a result would not imply that 100% of H3 molecules carry the modiﬁcation in question, because the antibodies used are very unlikely to have identical afﬁnities for their epitopes. In ref. 35 the absolute numbers of modiﬁcations are calculated as a percentage of total H3 molecules, which is a different calculation than the one we perform here. To avoid confusion, we use the term “enrichment relative to Histone H3 ChIP” on the axis of Fig. 4e. The vertical scale indicates the result of the calculation 2(Ct ChIP H3 – Ct ChIP x). Generation of eya1::GFP reporter construct and variants. The eya1::GFP reporter construct was generated using primers and templates as shown in Sup- plementary Table 6. All PCR steps were performed using Phusion polymerase (NEB). Two genomic fragments A and B of the eya locus (see Fig. 5) were each cloned separately into pCRII TOPO (Invitrogen) and sequenced. Fragment A (2L:6544449–6548972, ﬂybase version FB2010_05; D.mel genome version R5.28) contains the eye speciﬁc enhancer, the ﬁrst exon and the eya PRE/TRE (see Fig. 6). Fragment B (2L:6530964–6535677) consists of intronic sequence including a branch site and the ﬁrst three codons of exon 2 in order to enable splicing. Fragment B was fused to a fragment encoding TurboGFP (fragment C) using overlap extension PCR62 (Supplementary Table 6). The fused product B,C was cloned into pCRII TOPO and sequenced. Fragment A was cloned with EcoRI into pCRII TOPO BC, and the fused product was cloned into the EcoRI/ XhoI cut pKC27_mw vector41, resulting in the ﬁnal eya1::GFP construct. The eya1::GFP deletion constructs were generated by overlap extension PCR62 using the eya1::GFP construct as a template, and primers as shown in Supplementary Table 6. Details of cloning are given in the legend to Supplementary Table 6. For the bxd replacement construct, eya PRE/TRE coordinates Chr2L:6546243–6544711, were replaced with bxd PRE/TRE coordinates Chr3R:12590916–12589364, D.mel genome BDGP R5/ dm3. eya1::GFP was modiﬁed to replace the eya PRE/TRE with four unique cloning sites by replacing a 1982bp FspAI/Bsu36I fragment containing the PRE/TRE with a synthetic 569 bp FspAI/Bsu36I fragment (Mr.Gene http://mrgene.com), carrying NheI, AvrII, KpnI and SacI sites in place of the PRE/TRE sequence. Mathematical modelling. See Supplementary Methods. As a control for linker ligation, a gel-puriﬁed 448 bp blunt-ended PvuII fragment of pBluescriptIIKS+ was ligated to linkers under the same conditions as the ChIP ligations, and a dilution series of known quantities was subjected to qPCR, showing that linker ligation is efﬁcient and ampliﬁcation is linear with fragment concentration. Primer efﬁciency was calculated on the basis of plasmid and input dilution series to be 2.0 ± 0.3 (meaning that the product is ampliﬁed 2.0 ± 0.3 times per cycle). To further control for uniform ampliﬁcation by qPCR we compared input chromatin samples before and after blunting, ligation and qPCR by agarose gel electrophoresis. This showed a similar size distribution of fragments around the mean of 500 bp, thus the linker-mediated qPCR does not appear to introduce bias for a particular fragment size. Dot blot. Peptides were purchased from Jpt Innovative Peptide Solutions in unbiotinylated form. Dot blots were performed using 2 μl of the different histone peptides (Supplementary Table 3) spotted in different quantities (300 pmol, 150 pmol, 75 pmol and 37 pmol in PBS) onto a nitrocellulose membrane (Thermo Fisher Scientiﬁc). The membrane was dried for 1 h at room temperature before blocking with 5% BSA in TBST (20 mM Tris, pH 7.5, 150 mM NaCl, 0.05% Tween20) and incubation with primary antibody in the same buffer, each for 30 min at room temperature with gentle shaking. The membrane was extensively washed in TBST before incubation for 30 min with secondary antibody in TBST, 5% BSA, followed by extensive TBST and ﬁnally, TBS washes. See also59. Primary and secondary antibodies (Supplementary Tables 4 and 5) were used at 1:1000 dilution. Signals were visualized by enhanced chemiluminescence (ECL, Thermo Fisher Scientiﬁc) and quantiﬁed using the gel analysis function in ImageJ (NIH). ChIP and input samples from two independent chromatin preparations were subjected to qPCR in the same plate, each in a 5-fold dilution series (1 μl, 0.2 μl and 0.04 μl of the ligation reaction or of a 1:100 dilution of the input chromatin). We compared technical replicates by evaluating the correlation between data series from ChIP and inputs from independent chromatin preparations from each time point. We observed a high correlation between independent ChIP assays performed with the same antibody from the same time point (Supplementary Fig. 6). Immunoﬂuorescence on Drosophila embryos. Embryos were collected, ﬁxed and stained according to ref. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 In contrast, the bxd PRE/TRE with a low p5 and high (p3, p4) is more bistable, less ﬂexible and more able to maintain memory of initial states. In the model, if such a strongly bistable PRE/TRE is coupled to a promoter during the initiation phase, it responds to the promoter and later gives stable memory. However, if this same strongly bistable PRE/TRE is coupled to a promoter at later stages, it causes the promoter to variegate. Thus the model predicts that variegation (frequently observed in PRE/TRE reporter assays3,55, but rarely for endo- genous genes in the developing ﬂy) is a result of the late coupling of a strongly bistable PRE/TRE to a promoter. These observations provide a coherent hypothesis for the mechanistic and develop- mental relationship between variegation and epigenetic memory. In conclusion, this work provides a general theoretical fra- mework for PcG/TrxG regulation and systematically extends What might be the molecular basis of the difference between the eya and bxd PRE/TREs? Interestingly, one of the largest 13 TURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 concepts of regulation beyond epigenetic memory. The model allows a quantitative dissection of the interplay between chro- matin states and transcriptional states, the relationship between PRE/TRE sequence and regulatory output, and the capacity of the system to respond to developmental and environmental signals. Thus this work has broad implications for understanding genome-wide PcG/TrxG regulation throughout development, and the consequences of its disruption in disease. (Roche). 7 μL of the blunted IP material was ligated to a molar excess of double- stranded linkers (1 μM ﬁnal concentration linkers in 10 μL reaction volume). The linkers are designed so that the annealed 20 mer and 24 mer can only ligate via their blunt end to a blunt-ended fragment. The blunt end of the annealed linkers carries a 5′ phosphate (because the 24 mer was phosphorylated). The overhanging 4 bp do not allow self-annealing. Thus each blunt-ended ChIP fragment can only be ligated to one linker on each end. g Linker-mediated ChIP ampliﬁcation is normally used with 10–15 PCR cycles to amplify material prior to site-speciﬁc qPCR. Mathematical modelling. See Supplementary Methods. We note that such a result would not imply that 100% of H3 molecules carry the modiﬁcation in question, because the antibodies used are very unlikely to have identical afﬁnities for their epitopes. In ref. 35 the absolute numbers of modiﬁcations are calculated as a percentage of total H3 molecules, which is a different calculation than the one we perform here. To avoid confusion, we use the term “enrichment relative to Histone H3 ChIP” on the axis of Fig. 4e. The vertical scale indicates the result of the calculation 2(Ct ChIP H3 – Ct ChIP x). We ca cu ated e c e ts o eac od cat o a d o u od ed sto e H3 relative to (input/100) for each dilution using the following formula: enrichment relative to input = 2(Ct input – Ct ChIP). (Ct = the cycle number at which the signal crosses the threshold). This gives enrichments as a proportion of input, where an enrichment of 1 would be obtained if the quantity of IP material were equivalent to 1% of input (because the input samples were diluted 100-fold prior to qPCR). This dilution step was necessary to give input Ct values in a similar range to those of the ChIP samples). The ChIP vs. input calculation gave a measure of the reproducibility of ChIP assays (see also Supplementary Fig. 6). To calculate the enrichment of each modiﬁcation relative to Histone H3 ChIP one could use (2(Ct input – Ct ChIPx))/(2(Ct input – Ct H3 ChIP)). However, this expression simpliﬁes to 2(Ct ChIP H3 – Ct ChIPx), and thus the input is not required in this case. Indeed we found that including the input in the former calculation led to ampliﬁcation of errors Chromatin immunoprecipitation (ChIP). ChIP and chromatin preparation on whole embryos was performed as in refs. 58,61 and as follows: Before embryo collection, a pre-laying step of several hours was included, to ensure that only fresh embryos were collected. Embryos of different ages (0–2 h, 2–4 h, 4–6 h, 6–8 h and 8–10 h) were collected, by allowing ﬂies to lay for 2 h and ageing collections for the appropriate time period at 25 °C. For each collection, 1 g of embryos was dechorionated in 3% bleach (2.8% hypochloride) in eggwash (0.7% NaCl, 0.03% Triton X-100) and washed extensively with eggwash, then once with PBS, 0.01% Triton. Mathematical modelling. See Supplementary Methods. 60, and stored in methanol at −20 °C. Primary and sec- ondary antibodies (Supplementary Tables 4 and 5) were diluted 1:5000, with the exception of αH3, which was diluted 1:1000. After washing, the embryos were incubated overnight in ProLong Gold containing DAPI (Life Technologies). Embryos were imaged using a Leica DMi8 inverted microscope. Images were acquired with a HC PL Fluotar ×20 dry objective with numerical aperture of 0.55. For deconvolution of the images, the standard express, unsupervised deconvolution option in the Huygens Essential Software 16.05 was used. Image analysis was carried out using Fiji (ImageJ) v2.0.0.-rc_41, and Adobe Photoshop 2015.01. We calculated enrichments for each modiﬁcation and for unmodiﬁed histone H3 relative to (input/100) for each dilution using the following formula: We calculated enrichments for each modiﬁcation and for unmodiﬁed histone H3 relative to (input/100) for each dilution using the following formula: enrichment relative to input = 2(Ct input – Ct ChIP). (Ct = the cycle number at which the signal crosses the threshold). This gives enrichments as a proportion of input, where an enrichment of 1 would be obtained if the quantity of IP material were equivalent to 1% of input (because the input samples were diluted 100-fold prior to qPCR). This dilution step was necessary to give input Ct values in a similar range to those of the ChIP samples). The ChIP vs. input calculation gave a measure of the reproducibility of ChIP assays (see also Supplementary Fig. 6). To calculate the enrichment of each modiﬁcation relative to Histone H3 ChIP one could use (2(Ct input – Ct ChIPx))/(2(Ct input – Ct H3 ChIP)). However, this expression simpliﬁes to 2(Ct ChIP H3 – Ct ChIPx), and thus the input is not required in this case. Indeed we found that including the input in the former calculation led to ampliﬁcation of errors. Thus for the data shown in Fig. 4e and Supplementary Fig. 6C, the enrichment of each modiﬁed histone or PC was calculated as a proportion of the H3 ChIP using the latter calculation. The H3 ChIP samples were not diluted, and thus a relative enrichment of 1 would be obtained if the quantity of IP material from ChIP x were equivalent to 100% of the IP material from the H3 ChIP. NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 In constrast, for whole-genome qPCR, samples were not ampliﬁed after linker ligation but were subjected directly to real- time PCR using Sso Advanced SYBR Green Supermix (BioRad) and the 20 mer oligo as primer at 150 nM ﬁnal concentration, with the following program parameters: 2 min 50 °C, 2 min 94 °C, 40 × [15 s 94 °C,1 min 60 °C, 1 min 72 °C]. Methods Mathematical modelling. See Supplementary Methods. Mathematical modelling. See Supplementary Methods. Embryos were transferred to crosslinking solution (9.5 ml of [50 mM Hepes pH8, 1 mM EDTA, 0.5 mM EGTA, 100 mM NaCl], 30 ml of n-heptane, 0.5 ml of 37% formaldehyde) and incubated with vigorous horizontal shaking for 15 minutes at room temperature. Embryos were transferred to stop solution (0.01% Triton X- 100, 125 mM glycine in PBS), then subsequently washed for 10 min with rotation in wash solution A (10 mM Hepes, pH7.6, 10 mM EDTA, 0.5 mM EGTA, 0.25% Triton X-100), and wash solution B (10 mM Hepes, pH7.6, 200 mM NaCl, 1 mM EDTA, 0.5 mM EGTA, 0.01% Triton X-100). Crosslinked embryos were resus- pended in 5.5 ml sonication buffer (4 mM EDTA, 50 mM Hepes, pH 7.6, 300 mM NaCl, with addition of Protease inhibitors (Roche Complete, 1 tablet per 30 ml buffer), and 1 ml aliquots were sheared using a Covaris S220 ultra sonicator with the following settings: 65 Watt, 20% DF, 200 cycles per burst. The time of soni- cation was optimised to obtain chromatin fragments of ca. 500 bp as follows: 0–2 h or 2–4 h old embryos: 45 min; 4–6 h old embryos: 30 min; 6–8 h old embryos: 25 min; 8–10 h old embryos: 20 min. The DNA content of a 50 μL aliquot of each chromatin preparation was measured by spectrophotometry (Denovix) after reverse crosslink and DNA extraction. The chromatin preparations were diluted to equivalent DNA concentrations. Two independent chromatin preparations and two to three independent ChIP assays were performed for each time point. IPs were performed using 50 μg DNA in 300 μl volume with 4.5 μg of each primary antibody as listed in Supplementary Table 4. IPs were performed in 300 μl volume overnight at 4 °C in RIPA buffer (10 mM Tris, pH8.0, 140 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.1% SDS, 0.1% Na deoxycholate, 1 mM PMSF). Immunocomplexes were puriﬁed using Protein A Sepharose beads (CL4B, Amersham, cat # 17–0780–01) (60 μl per IP). ChIP H3 Ct ChIPx), and thus the input is not required in this case. Indeed we found that including the input in the former calculation led to ampliﬁcation of errors. Thus for the data shown in Fig. 4e and Supplementary Fig. 6C, the enrichment of each modiﬁed histone or PC was calculated as a proportion of the H3 ChIP using the latter calculation. Methods M h i Methods Mathematical modelling. See Supplementary Methods. ARTICLE A silencer is required for maintenance of transcriptional repression throughout Drosophila development. Development 124, 4343–4350 (1997). development. Development 124, 4343–4350 (1997). 9. Americo, J. et al. A complex array of DNA-binding proteins required for pairing-sensitive silencing by a polycomb group response element from the Drosophila engrailed gene. Genetics 160, 1561–1571 (2002). Live GFP imaging. 3rd instar larval eye discs were stained in 5 µM SYTO Red Fluorescent Nucleic Acid Stain (Invitrogen) and imaged on a confocal microscope LSM 700/Axioimager (Zeiss). 10. Rank, G., Prestel, M. & Paro, R. Transcription through intergenic chromosomal memory elements of the Drosophila bithorax complex correlates with an epigenetic switch. Mol. Cell Biol. 22, 8026–8034 (2002). In situ hybridisation. PCR products were generated for two in situ probes (eya exon 1 and GFP) using the primers and templates shown in Supplementary Table 6. PCR products were cloned into pCRII vector (Invitrogen), sequenced, and checked for the correct orientation. For in situ hybridization, probe sequences were PCR ampliﬁed from pCRII using M13 primers such that the PCR products con- tained the T7 promoter. Probes were then in vitro transcribed using T7 polymerase (Roche). The probe was labelled with ﬂuorescein (Roche), detected with primary antibody: Mouse anti-ﬂuorescein (Roche), secondary antibody: goat anti-mouse- HRP (Invitrogen) and visualised with Alexa Fluor 488 Tyramide (Invitrogen) using Tyramide Signal Ampliﬁcation kit (TSATM, Invitrogen) according to manu- facturers instructions. Tissues were mounted in Vectashield with DAPI (Vector Laboratories) on microscope slides. Images were taken using confocal microscopy, with LSM 700 Axioimager (Zeiss). 11. Ringrose, L. & Paro, R. Epigenetic regulation of cellular memory by the Polycomb and Trithorax group proteins. Annu Rev. Genet 38, 413–443 (2004 12. Ringrose, L., Rehmsmeier, M., Dura, J. & Paro, R. Genome-wide prediction of Polycomb/Trithorax response elements in Drosophila melanogaster. Dev. Cell 5, 759–771 (2003). 13. Ringrose, L. Polycomb comes of age: genome-wide proﬁling of target sites. Curr. Opin. Cell Biol. 19, 290–297 (2007). 14. Oktaba, K. et al. Dynamic regulation by polycomb group protein complexes controls pattern formation and the cell cycle in Drosophila. Dev. Cell 15, 877–889 (2008). 15. Schuettengruber, B. et al. Functional anatomy of polycomb and trithorax chromatin landscapes in Drosophila embryos. PLoS Biol. 7, e13 (2009). Signal intensity proﬁles were generated by evaluation of maximum intensity projections using the line scan function of the imaging software MetaMorph (Version 7.1.1.0). ARTICLE ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 2. Simon, J., Chiang, A., Bender, W., Shimell, M. J. & O’Connor, M. Elements of the Drosophila bithorax complex that mediate repression by Polycomb group products. Dev. Biol. 158, 131–144 (1993). TRE (see Fig. 6, Supplementary Fig. 7) was ampliﬁed from genomic DNA using primers shown in Supplementary Table 6, and was cloned with NheI/KpnI into this modiﬁed eya1::GFP vector. The eya PRE/TRE with mutated PHO sites was syn- thesised as a 1982bp fragment (Mr.Gene http://mrgene.com) with mutations as shown in Supplementary Fig. 7D and legend. All constructs were sequenced prior to injection. p 3. Chan, C. S., Rastelli, L. & Pirrotta, V. A Polycomb response element in the Ubx gene that determines an epigenetically inherited state of repression. EMBO J. 13, 2553–2564 (1994). 4. Steffen, P. A. & Ringrose, L. What are memories made of? How Polycomb and Trithorax proteins mediate epigenetic memory. Nat. Rev. Mol. Cell Biol. 15, 340–356 (2014). Generation and genetic analysis of transgenic ﬂies. Transgenic ﬂy lines carrying eya1::GFP or variants were generated by ΦC31 mediated integration by the IMBA ﬂy injection service. The landing site used here corresponds to site 2 at position 46E1 in ref. 41 (chr. 2R) at genomic location 5,965,083 (ﬂybase version FB2010_05; D.mel genome version R5.28). All ﬂy stocks are available on request. To test the effects of the ph410 mutation on miniwhite expression, ﬂies of the following gen- otypes were crossed for each transgene: females: w -, ph410/ w -, ph410;Transgene/ CyO x males: w -; Transgene/CyO. Female progeny lacking the CyO balancer were heterozygous for the ph410 mutation and homozygous for the transgene. These ﬂies were aged for four days before taking photographs. The ph410 allele (FBal0013768) is a hypomorphic allele of ph. It disrupts the proximal ph transcription unit but leaves the closely related distal transcription unit intact. 5. Woo, C. J., Kharchenko, P. V., Daheron, L., Park, P. J. & Kingston, R. E. A region of the human HOXD cluster that confers polycomb-group responsiveness. Cell 140, 99–110 (2010). 6. Cavalli, G. & Paro, R. The Drosophila Fab-7 chromosomal element conveys epigenetic inheritance during mitosis and meiosis. Cell 93, 505–518 (1998). 7. Hagstrom, K., Muller, M. & Schedl, P. A Polycomb and GAGA dependent silencer adjoins the Fab-7 boundary in the Drosophila bithorax complex. Genetics 146, 1365–1380 (1997). 8. Busturia, A., Wightman, C. D. & Sakonju, S. ARTICLE The line scans were placed in anterior-posterior orientation at 90° to the morphogenetic furrow as shown in Fig. 6d, e. Three to four scans were performed for each disc. Average Y-values of each channel (Red/Green/Blue) were obtained by applying a scan width of 50 pixels. Line scans from six to ten individual eye discs of similar size from each experiment were aligned with respect to the position of the morphogenetic furrow and the posterior disc edge on the X-axis, and averaged for the plots shown in Fig. 6f. For vertical scaling the average line scans were normalised by setting the maximum measured intensity of each disc to 1. 16. Schwartz, Y. B. et al. Alternative epigenetic chromatin states of polycomb target genes. PLoS Genet. 6, e1000805 (2010). 17. Maurange, C. & Paro, R. A cellular memory module conveys epigenetic inheritance of hedgehog expression during Drosophila wing imaginal disc development. Genes Dev. 16, 2672–2683 (2002). p 18. Martinez, A. M., Colomb, S., Dejardin, J., Bantignies, F. & Cavalli, G. Polycomb group-dependent Cyclin A repression in Drosophila. Genes Dev. 20, 501–513 (2006). 19. Perez, L. et al. Enhancer-PRE communication contributes to the expansion of gene expression domains in proliferating primordia. Development 138, 3125–3134 (2011). 20. Herzog, V. A. et al. A strand-speciﬁc switch in noncoding transcription switches the function of a Polycomb/Trithorax response element. Nat. Genet. 46, 973–981 (2014). Reporting summary. Further information on research design is available in the Nature Research Reporting Summary linked to this article. 21. Dodd, I. B., Micheelsen, M. A., Sneppen, K. & Thon, G. Theoretical analysis of epigenetic cell memory by nucleosome modiﬁcation. Cell 129, 813–822 (2007). Data availability The ChIP-seq data shown in Fig. 5a are publically available at [http://compbio.med. harvard.edu/modencode/webpage/Chromatin.v0.6.html#ChIP-seq%20and%20ChIP-chip %20data]. H3K27me3 (track ID 3955), PSC, Posterior Sex Combs (track ID 3960), GAF, GAGA Factor (track ID 4119). All other relevant data supporting the key ﬁndings of this study are available within the article and its Supplementary Information ﬁles or from the corresponding author upon reasonable request. Source data for Fig. 4e are provided with this paper. A reporting summary for this Article is available as a Supplementary Information ﬁle. Source data are provided with this paper. 22. Angel, A., Song, J., Dean, C. & Howard, M. A Polycomb-based switch underlying quantitative epigenetic memory. Nature 476, 105–108 (2011). 22. Angel, A., Song, J., Dean, C. & Howard, M. A Polycomb-based switch underlying quantitative epigenetic memory. Nature 476, 105–108 (2011). 23. Sneppen, K. & Ringrose, L. Theoretical analysis of Polycomb-Trithorax systems predicts that poised chromatin is bistable and not bivalent. Nat. Commun. 10, 2133 (2019). in press. 23. Sneppen, K. & Ringrose, L. Theoretical analysis of Polycomb-Trithorax systems predicts that poised chromatin is bistable and not bivalent. Nat. Commun. 10, 2133 (2019). in press. 24. Kassis, J. A. & Brown, J. L. Polycomb group response elements in Drosophila and vertebrates. Adv. Genet. 81, 83–118 (2013). 25. Bauer, M., Trupke, J. & Ringrose, L. The quest for mammalian Polycomb response elements: are we there yet? Chromosoma 125, 471–496 (2016). Code availability 26. Martínez-Balbás, M. A., Dey, A., Rabindran, S. K., Ozato, K. & Wu, C. Displacement of sequence-speciﬁc transcription factors from mitotic chromatin. Cell 83, 29–38 (1995). The computer code used in this study is available as supplementary software and has been deposited at [https://github.com/Ringrose546/Beyond_memory_V2]. 27. Parsons, G. G. & Spencer, C. A. Mitotic repression of RNA polymerase II transcription is accompanied by release of transcription elongation complexes. Mol. Cell. Biol. 17, 5791–5802 (1997). Received: 10 May 2019; Accepted: 24 August 2020; Received: 10 May 2019; Accepted: 24 August 2020; 28. Spencer, C. A., Kruhlak, M. J., Jenkins, H. L., Sun, X. & Bazett-Jones, D. P. Mitotic transcription repression in vivo in the absence of nucleosomal chromatin condensation. J. Cell Biol. 150, 13–26 (2000). 29. Alabert, C. et al. Two distinct modes for propagation of histone PTMs across the cell cycle. Genes Dev. 29, 585–590 (2015). Mathematical modelling. See Supplementary Methods. The bxd PRE/ Linker ligation and qPCR to quantify ChIP enrichments. Global ChIP enrich- ments were quantiﬁed by linker ligation followed by qPCR as follows (see also refs. 58,61): 7 μL IP or input material was rendered blunt-ended in 25 μL total volume using the Quick Blunting Kit (NEB). Double-stranded linkers were pre- pared by annealing 20 mer and 24 mer oligos (see Supplementary Table 6). Prior to annealing, the 24 mer oligo was 5′ phosphorylated using T4 Polynucleotide Kinase 14 URE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications Acknowledgements y p 38. Zenk, F. et al. Germ line-inherited H3K27me3 restricts enhancer function during maternal-to-zygotic transition. Science 357, 212–216 (2017). This work was supported by Austrian Academy of Sciences http://www.oeaw.ac.at/ (L.R. and F.R.) the EU FP7 Network of Excellence Epigenesys (FP7-HEALTH-F4-2010- 257082) http://www.epigenesys.eu/ (L.R. and F.R.), the DFG Exzellenzinitiativ IRI- Lifesciences https://www.iri-ls.hu-berlin.de/en (L.R. and J.R.), the EU H2020-MSCA-ITN PEP-NET (Project number 813282) and by the BBSRC Institute Strategic Programmes GRO (BB/J004588/1) and GEN (BB/P013511/1) (M.H.). We acknowledge support by the Open Access Publication Fund of Humboldt-Universität zu Berlin. during maternal-to-zygotic transition. Science 357, 212–216 (20 39. Bui, Q. T., Zimmerman, J. E., Liu, H., Gray-Board, G. L. & Bonini, N. M. Functional analysis of an eye enhancer of the Drosophila eyes absent gene: differential regulation by eye speciﬁcation genes. Dev. Biol. 221, 355–364 (2000). 40. Kumar, J. P. The molecular circuitry governing retinal determination. Biochim Biophys. Acta 1789, 306–314 (2009). 41. Okulski, H., Druck, B., Bhalerao, S. & Ringrose, L. Quantitative analysis of polycomb response elements (PREs) at identical genomic locations distinguishes contributions of PRE sequence and genomic environment. Epigenetics Chromatin 4, 4 (2011). NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 31. Zerihun, M. B., Vaillant, C. & Jost, D. Effect of replication on epigenetic memory and consequences on gene transcription. Phys. Biol. 12, 026007 (2015). 58. Hauenschild, A., Ringrose, L., Altmutter, C., Paro, R. & Rehmsmeier, M. Evolutionary plasticity of polycomb/trithorax response elements in Drosophila species. PLoS Biol. 6, e261 (2008). ( ) 32. Shermoen, A. W., McCleland, M. L. & O’Farrell, P. H. Developmental control of late replication and S phase length. Curr. Biol. 20, 2067–2077 (2010). p 59. Ringrose, L., Ehret, H. & Paro, R. Distinct contributions of histone H3 lysine 9 and 27 methylation to locus-speciﬁc stability of polycomb complexes. Mol. Cell 16, 641–653 (2004). 33. Yuan, K. & O’Farrell, P. H. TALE-light imaging reveals maternally guided, H3K9me2/3-independent emergence of functional heterochromatin in Drosophila embryos. Genes Dev. 30, 579–593 (2016). 60. Rothwell, W. F., Sullivan, W. Fluorescent analysis of Drosophila embryos in Drosophila protocols (eds. Sullivan, W., Ashburner, M. & Scott Hawley, R.) 141–157 (CSHL Press, 2000). Drosophila embryos. Genes Dev. 30, 579–593 (2016). 34. Li, X. Y., Harrison, M. M., Villalta, J. E., Kaplan, T. & Eisen, M. B. Establishment of regions of genomic activity during the Drosophila maternal to zygotic transition. eLife 3, e03737 (2014). 61. Orlando, V., Jane, E. P., Chinwalla, V., Harte, P. J. & Paro, R. Binding of trithorax and Polycomb proteins to the bithorax complex: dynamic changes during early Drosophila embryogenesis. Embo J. 17, 5141–5150 (1998). yg f 35. Bonnet, J. et al. Quantiﬁcation of proteins and histone marks in Drosophila embryos reveals stoichiometric relationships impacting chromatin regulation. Dev. Cell 51, 632–644 e636 (2019). 62. Horton, R. M., Hunt, H. D., Ho, S. N., Pullen, J. K. & Pease, L. R. Engineering hybrid genes without the use of restriction enzymes: gene splicing by overlap extension. Gene 77, 61–68 (1989). 36. Williamson, A. & Lehmann, R. Germ cell development in Drosophila. Annu. Rev. cell Dev. Biol. 12, 365–391 (1996). 37. Foe, V. E. Mitotic domains reveal early commitment of cells in Drosophila embryos. Development 107, 1–22 (1989). Correspondence and requests for materials should be addressed to L.R. 49. Yang, H. et al. Distinct phases of Polycomb silencing to hold epigenetic memory of cold in Arabidopsis. Science 357, 1142–1145 (2017). Peer review information Nature Communications thanks Ian Dodd and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available. y p 50. Kwon, D. et al. Enhancer-promoter communication at the Drosophila engrailed locus. Development 136, 3067–3075 (2009). 51. Ringrose, L. Noncoding RNAs in Polycomb and Trithorax regulation: a quantitative perspective. Annu Rev. Genet. 51, 385–411 (2017). Author contributions L.R. conceived the study, performed modelling and wrote the computer code and the paper. M.H. hosted L.R. during the initial stages of the project, provided advice on modelling and contributed to the paper. J.R. performed experiments shown in Fig. 4 and Supplementary Figs. 4, 5 and 6. F.R. performed experiments shown in Figs. 5, 6 and Supplementary Fig. 7. L.R. supervised J.R. and F.R. Epigenetics Chromatin 4, 4 (2011). 42. Berry, S., Dean, C. & Howard, M. Slow chromatin dynamics allow polycomb target genes to ﬁlter ﬂuctuations in transcription factor activity. Cell Syst. 4, 445–457 e448 (2017). 43. Sneppen, K. Models of life: epigenetics, diversity and cycles. Reports on progress in physics. Phys. Soc. 80, 042601 (2017). p g The authors declare no competing interests. g 46. Meshorer, E. et al. Hyperdynamic plasticity of chromatin proteins in pluripotent embryonic stem cells. Dev. Cell 10, 105–116 (2006). pluripotent embryonic stem cells. Dev. Cell 10, 105–116 (2006). References 30. Blythe, S. A. & Wieschaus, E. F. Coordinating cell cycle remodeling with transcriptional activation at the Drosophila MBT. Curr. Top. Dev. Biol. 113, 113–148 (2015). References 1. Ptashne, M. & Gann, A. Genes and Signals. (Cold Spring Harbor Laboratory Press, 2002). 15 TURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunications Funding O A g y y 44. Ficz, G., Heintzmann, R. & Arndt-Jovin, D. J. Polycomb group protein complexes exchange rapidly in living Drosophila. Development 132, 3963–3976 (2005). g Open Access funding provided by Projekt DEAL. Open Access funding provided by Projekt DEAL. 45. Fonseca, J. P. et al. In vivo Polycomb kinetics and mitotic chromatin binding distinguish stem cells from differentiated cells. Genes Dev. 26, 857–871 (2012). Additional information 47. Perino, M. & Veenstra, G. J. Chromatin control of developmental dynamics and plasticity. Dev. Cell 38, 610–620 (2016). Supplementary information is available for this paper at https://doi.org/10.1038/s41467- 020-18507-4. Supplementary information is available for this paper at https://doi.org/10.1038/s41467- 020-18507-4. 48. Seller, C. A., Cho, C. Y. & O’Farrell, P. H. Rapid embryonic cell cycles defer the establishment of heterochromatin by Eggless/SetDB1 in Drosophila. Genes Dev. 33, 403–417 (2019). 48. Seller, C. A., Cho, C. Y. & O’Farrell, P. H. Rapid embryonic cell cycles defer the establishment of heterochromatin by Eggless/SetDB1 in Drosophila. Genes Dev. 33, 403–417 (2019). Correspondence and requests for materials should be addressed to L.R. ARTICLE NATURE COMMUNICATIONS \| https://doi.org/10.1038/s41467-020-18507-4 Reprints and permission information is available at http://www.nature.com/reprints 52. Hug, C. B., Grimaldi, A. G., Kruse, K. & Vaquerizas, J. M. Chromatin architecture emerges during zygotic genome activation independent of transcription. Cell 169, 216–228 e219 (2017). 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. 53. Gemkow, M. J., Verveer, P. J. & Arndt-Jovin, D. J. Homologous association of the Bithorax-Complex during embryogenesis: consequences for transvection in Drosophila melanogaster. Development 125, 4541–4552 (1998). 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 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 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 license, visit http://creativecommons.org/ licenses/by/4.0/. 54. Kassis, J. A. Unusual properties of regulatory DNA from the Drosophila engrailed gene: three “pairing-sensitive” sites within a 1.6-kb region. Genetics 136, 1025–1038 (1994). 55. Fauvarque, M. O. & Dura, J. M. polyhomeotic regulatory sequences induce developmental regulator-dependent variegation and targeted P-element insertions in Drosophila. Genes Dev. 7, 1508–1520 (1993). 56. Wall, G., Varga-Weisz, P. D., Sandaltzopoulos, R. & Becker, P. B. Chromatin remodeling by GAGA factor and heat shock factor at the hypersensitive Drosophila hsp26 promoter in vitro. EMBO J. 14, 1727–1736 (1995). 57. Katsani, K. R., Hajibagheri, M. A. & Verrijzer, C. P. Co-operative DNA binding by GAGA transcription factor requires the conserved BTB/ POZ domain and reorganizes promoter topology. EMBO J. 18, 698–708 (1999). © The Author(s) 2020 16 NATURE COMMUNICATIONS \| (2020) 11:4782 \| https://doi.org/10.1038/s41467-020-18507-4 \| www.nature.com/naturecommunicatio
https://openalex.org/W2134813969	https://www.scielo.br/j/mr/a/Zrk8LjwfxSFgZXpcwLZKk8s/?lang=en&format=pdf	English	null	Tailoring activated carbon by surface chemical modification with O, S, and N containing molecules	Materials research	2,003	cc-by	4,054	*e-mail: rochel@dedalus.lcc.ufmg.br Trabalho apresentado no I Simpósio Mineiro de Ciências dos Materiais, Ouro Preto, Novembro de 2001. 129 © 2003 129 © 2003 Vol. 6, No. 2, 2003 Tailoring Activated C Materials Research, Vol. 6, No. 2, 129-135, 2003. Received: November 11, 2001; Revised: March 24, 2003 In this work the surface of activated carbon was chemically modified in order to introduce O, S and N containing groups. The activated carbon surface was selectively oxidized with concen- trated HNO3 under controlled conditions. Characterization by thermogravimetric analyses, infra- red spectroscopy and NaOH titration suggested the formation of mainly –COOH and small amounts of –OH groups, with concentration of approximately 4.1021 groups/g of carbon. These -COOH functionalized carbons showed high adsorption capacity for metal cations in aqueous solution in the following order: Pb+2>Cu+2>Ni+2>Cd+2~Co+2>Ca+2, suggesting a cation exchange mechanism via a surface complex [–COO-M+2]. These –COOHsurf groups can be reacted with SOCl2 to produce a surface acylchloride group, -COCl. This surface -COCl group proved to be a very reactive and versatile intermediate for the grafting of different S and N containing molecules onto the carbon surface, such as 1,2-ethaneditiol (EDT-, HSCH2CH2SH) 1,7-dimercapto-4-thioheptane (DMTH- HSCH2CH2CH2SCH2CH2CH2SH) or 1,2-ethylenediamine (EDA- NH2CH2CH2NH2) and triethyltetraamine, TEA (H2NCH2CH2NHCH2CH2NHCH2CH2NH2). The characterization of these materials was carried out by TG, IR and TPDMS (Temperature Programmed Decomposition Mass Spectrometry) experiments suggesting the formation of thioesther and amide surface groups, i.e. –COSR and –CONHR, with yields of approximately 50 and 75% for the reaction with DME and EDA, respectively. Preliminary adsorption experiments showed that these materials can efficiently remove metals such as Pb+2, Cu+2 and Ni+2 from aqueous medium. Keywords: activated carbon, surface modification, functionalization, cation exchange, heavy metals high surface area and porous structure9 it can adsorb gases and compounds dispersed or dissolved in liquids10,11. The type of contaminant to be adsorbed and the adsorption/ remotion efficiency of these carbons is strongly dependent on their surface chemical features. Therefore, the surface chemical modification of carbon is of great interest in order to produce materials for specific applications. This modifi- cation has been mainly carried out by oxidative methods, producing a more hydrophilic structure with a large number of oxygen-containing groups. Various reagents have been used as oxidants: concentrated nitric or sulfuric acid, so- dium hypochlorite, permanganate, bichromate, hydrogen peroxide, transition metals and ozone-based gas mixtures12. It was found that the type of surface structures and the ex- tent of their formation depends on the oxidizing agent, the concentration and the pH of the oxidizing solution13-15. Al- though the creation of oxygen groups on the carbon surface 1. Introduction Surface modification1 has a fundamental role on the application of organic and inorganic supports in industrial and environmental processes, such as selective purification processes, gas separation, solvent recovery, drinking water purification, adsorption of taste, odor and other micro-pol- lutants2, ion exchange properties for metal adsorption3, cata- lyst preparation4, adhesion phenomenon4,6, electrode modi- fication7 and polymer technologies8. Activated carbon is one of the most important industrial adsorbent/support and can be prepared by carbonization and activation of a large number of raw materials such as coco- nut shells, wood, peat and coal. These carbons show high- developed porous structure and a large internal specific sur- face area, which is generally greater than 400 m2/g but of- ten exceeds this value, reaching 1000 m2/g 3-5. Due to this Rios et al. 130 Materials Research MS chamber (P = 6.10-6 Torr). The probe was heated at 5 °C/min up to 275 °C and all the volatile decomposition products were analyzed by the mass spectrometer. is relatively well known, the functionalization with S and N containing molecules has been much less investigated. In this work the activated carbon surface was chemi- cally modified in order to introduce oxygen, nitrogen and sulfur containing groups. 2.6 Potentiometric Titration of the Carbon C/HNO3 with NaOH 2.1 HNO3 Treatment (C/HNO3) The activated carbon (20 g, Aldrich Norit, 930 m2/g) was treated with 100 ml of concentrated HNO3 under re- flux for 2, 4, 8, 16 or 38 h. After reflux, the resulting mate- rial was filtered and extensively washed with hot distilled water until the cleansing water pH was approximately 7. The carbons were dried in vacuum at 60 °C for 24 h. (H2NCH2CH2NHCH2CH2NHCH2CH2NH2) After treatment with SOCl2 the material (0.80 g) was reacted with of the amine (2.6 ml) in 6 ml of chloroform under reflux for 48 h. The product was extensively washed with CHCl3 and separated by centrifugation. 3 Thermogravimetric analyses were carried out in a Mettler TA 4000 System under N2 or air flow with a heating rate of 10 °C/min. Infrared spectroscopic analyses were obtained in a FTIR Mattson Instrument using KBr. BET surface areas were obtained in a Quantachrome Nova 1200 equipment with the carbon samples treated at 200 °C under vacuum for 3 h. 2. Experimental The carbon (0.1 g) suspended in 25 ml of distilled water was titrated with a 0.02 M NaOH solution with continuous stirring. The titration was carried out under nitrogen atmos- phere to avoid interference of CO2 from the air. NaOH so- lution was added stepwise in 0.5 ml in intervals of 5-10 min to allow the reaction to take place and the pH to stabilize. 3.1 Treatment with HNO3 The controlled oxidation of the activated carbon was carried out with concentrated HNO3 under reflux for differ- ent periods. After this treatment the obtained material was extensively washed with hot water and carefully dried in vacuum at 60 °C. For comparison, the untreated original activated carbon was also washed with water and dried un- der vacuum. The mass balance with the weight losses ob- served after different reflux times is shown in Fig. 1. The HNO3 treated carbon (0.80 g) was dried under vacuum at 80 °C to remove water and other substance which could interfere on the reaction. The reaction was carried out with 5 ml benzene and 5 ml of SOCl2 under reflux for 24 h. The mixture was dried in a rotatory-evaporator and extensively washed with benzene to eliminate residual SOCl2. 2.3 Reaction with Ethaneditiol (EDT, HSCH2CH2SH) or 1,7-dimercapto-4-thioheptane (DMTH, HSCH2CH2CH2SCH2CH2CH2SH) 2.3 Reaction with Ethaneditiol (EDT, HSCH2CH2SH) or 1,7-dimercapto-4-thioheptane (DMTH, HSCH2CH2CH2SCH2CH2CH2SH) It can be observed that the weight loss increases with the reflux time, probably due to the oxidation of the acti- vated carbon to gaseous products, e.g. CO2, and water solu- ble derivatives such as mellitic acid16. After treatment with SOCl2 the material C/HNO3/SOCl2 (0.80 g) was reacted with 9 mmol of 1,2-ethaneditiol or 1,7-dimercapto-4-thioheptane in 5 ml of chloroform under reflux for 48 h. The product was washed extensively with CHCl3 and separated by centrifugation. Infrared spectroscopic analyses of the HNO3 treated car- bons (Fig. 2) showed strong absorption at approximately 1720 cm-1 (-C=O), 1550 cm-1 (-COO-), 1250 cm-1 (-C-O) and 3. Results and Discussion 2.2 Reaction of the Activated Carbon with Thionyl Chloride (C/HNO3/SOCl2) 2.4 Reaction with Etilenodiamina (EDAH2NCH2CH2NH2) or Triethyltetraamine (H2NCH2CH2NHCH2CH2NHCH2CH2NH2) Figure 1. Weight losses after refluxing activated carbon with con- centrated HNO3 for different periods. 2.5 Temperature Programmed Decomposition Coupled with a Mass Spectrometer (TPDMS) The TPDMS experiments were carried out in a HP 5989 II mass spectrometer using 5 mg sample in a direct probe equipped with a heating system which was inserted in the Figure 1. Weight losses after refluxing activated carbon with con- centrated HNO3 for different periods. 131 Vol. 6, No. 2, 2003 Tailoring Activated Carbon by Surface Chemical Modification with O, S, and N Containi Tailoring Activated Carbon by Surface Chemical Modification with O, S, and N Containing Molecules Vol. 6, No. 2, 2003 Tailoring Activated Carbon by Surface Chemical Modification with O, S, and N Containing Molecules 3450 cm-1 (-OH) which can be assigned to carboxylic acid groups16. To estimate the surface concentration of carboxylic groups, TG experiments were carried out keeping the tem- perature at 750 °C until no weight loss was observed. The total weight loss obtained from 200 °C was considered to be related to the amount of CO2 formed. It was also carried out potentiometric titrations of the carbons with NaOH to obtain the number of acid sites. The HNO3 treated carbons were also analyzed by thermogravimetry. Figure 3 shows the TG profiles for the different HNO3 treated carbons. It can be observed that the untreated activated carbon showed very small weight decrease during TG analysis. On the other hand, the treated carbons showed significant weight loss after 200 °C. These weight losses observed for HNO3 treated carbons have been assigned mainly to the decompo- sition of carboxylic surface groups and in a lesser exten- sion to –OH groups according to the following processes2,16: The number of acid (COOH) sites for the different re- flux time obtained by both TG and NaOH titration is dis- played in Fig. 4. It can be observed that up to 8 h reflux with HNO3 the –COOH concentration increases to ca. 2-4 . 1021 sites/g, but it remains approximately constant even if the carbon is -COOHsurf → CO2 -OHsurf → CO -COOHsurf → CO2 -OHsurf → CO Figure 4. Acid sites concentration (as –COOH) on carbon surface for the different reflux time. Figure 4. Acid sites concentration (as –COOH) on carbon surface for the different reflux time Figure 3. TG profiles for the different HNO3 treated carbons. Figure 4. Acid sites concentration (as –COOH) on carbon surface for the different reflux time. Figure 5. Potentiometric titration of the carbons C/HNO3 with NaOH (0.02 M). Figure 2. 2.5 Temperature Programmed Decomposition Coupled with a Mass Spectrometer (TPDMS) IR spectra of the HNO3 treated carbons. Figure 2. IR spectra of the HNO3 treated carbons. Figure 2. IR spectra of the HNO3 treated carbons. Figure 4. Acid sites concentration (as –COOH) on carbon surface for the different reflux time. Figure 2. IR spectra of the HNO3 treated carbons. Figure 2. IR spectra of the HNO3 treated carbons. Figure 5. Potentiometric titration of the carbons C/HNO3 with NaOH (0.02 M). Figure 5. Potentiometric titration of the carbons C/HNO3 with NaOH (0.02 M). Figure 5. Potentiometric titration of the carbons C/HNO3 with NaOH (0.02 M). Figure 3. TG profiles for the different HNO3 treated carbons. 132 Rios et al. Materials Research further refluxed with HNO3. further refluxed with HNO3. Pb+2 > Cu+2 > Ni+2 > Cd+2 ~ Co+2 > Ca+2, suggesting that the stability of the surface complex formed is important for the adsorption of the cation. In fact, the adsorption capacity is very well correlated with the stability constant of cation 3 Figure 5 shows the NaOH titration curves. The obtained titration curves suggested that the number and the strength of surface acid sites tends to increase as the reflux time in HNO3 increased. For example, the carbon treated for 38 h with concentrated HNO3 showed the strongest acid sites whereas the treatment for 2-16 h did show much difference. Moreover, it can be observed the presence of two or more inflection points in the titration curve, which suggests the presence of sites with different acid strength. Scheme 1. Cation exchange mechanism with the carbon surface carboxylic groups. It was also investigated the effect of the HNO3 treat- ment on the carbon surface area (Table 1). The oxidation with concentrated HNO3 has a strong ef- fect on the carbon surface area, decreasing from 912 for the untreated carbon to 605 m2/g after 38 h reflux. This de- crease in surface area is related to the destruction of the porous structure caused by the severe oxidation with con- centrated nitric acid. Scheme 1. Cation exchange mechanism with the carbon surface carboxylic groups. Figure 6. Chlorobenzene adsorption capacity on carbon treated with HNO3 for different reflux time. 2.5 Temperature Programmed Decomposition Coupled with a Mass Spectrometer (TPDMS) These results suggest that a treatment with reflux time of 2 up to 8 h, will not strongly affect the activated carbon properties, since it produces a high concentration of sur- face acid sites, causes low weight losses and a relatively low decrease on surface area. 3.3 Carbon Surface Functionalization. Figure 8 shows the correlation of the stability constants with the adsorption capacities. Functionalization of carbon surfaces with S and N con- taining groups have been carried out by different methods such as: treatment with sulfur17 and CS2 18 at high tempera- tures, direct reaction of surface COOH groups with H2NR compounds19 and direct surface deposition of pyrazolone derivatives20. However, these proposed methods show in general complex procedures with low efficiency. Hereon, it is described a simple and versatile method for the introduc- tion of heteroatoms such as S and N containing molecules onto carbon surface. It can be observed that as the stability (log K) increases the adsorption capacity of the carbons increases linearly. This excellent cation exchange properties shown by these Figure 8. Stability constants versus adsorption capacity for the different metal cations onto 16 h reflux treated carbon. The following reaction scheme was used to introduce S and N groups on the carbon surface starting from the –COOH groups. The S and N molecules used in this work are described in Table 2. The IR spectra obtained for the –COOH, -COCl inter- mediates and the materials obtained by the reaction with EDT and EDA are shown in Fig. 9. After the reaction with SOCl2, the IR spectrum (Fig. 9b) showed new and well defined bands at approximately 1700, 1350, 1210 and 790 cm-1 which are clearly related to the formation of the group –COCl. The IR spectrum for C6H5COCl shows exactly the same bands. On the other hand, it can also be noticed the presence of the original bands, which might be indicating that not all the –COOH groups present on the carbon surface reacted with SOCl2 21. Figure 8. Stability constants versus adsorption capacity for the different metal cations onto 16 h reflux treated carbon. 2 The IR spectrum of the carbon after reaction with EDT (Fig. 9c) suggests the occurrence of the reaction of –COCl with the HSR compound to form –COSR. Typical bands of –COCl disappeared and well defined absorptions at 1400, 1200, 750 and 700 cm-1 related to the EDT molecule are observed. The IR spectrum for the material obtained by the reaction with EDA (Fig. 9d) also suggested the reaction of HNHR with the surface group –COCl, with the presence of Figure 9. 3.2 Adsorption Properties of –COOH Functionalized Carbons The adsorption of chlorobenzene from water onto the carbons treated with HNO3 for different reflux time is shown in Fig. 6. The treatment with HNO3 almost inhibit the adsorption of chlorobenzene. This is likely related mainly to presence of –COOH groups, which gives to the hydrophobic carbon surface a strong hydrophilic character. Figure 6. Chlorobenzene adsorption capacity on carbon treated with HNO3 for different reflux time. On the other hand, the presence of these –COOH sur- face groups strongly increases the adsorption capacity for metal cations in water (Fig. 7). It can be observed that as the –COOH surface concen- tration increases the adsorption capacity also increases, sug- gesting that the adsorption is probably taking place at the carboxylic groups via a cation exchange mechanism. Figure 7. Metal cations adsorption on functionalized activated carbons. The following cation adsorption order was observed Table 1. BET N2 surface area for the HNO3 treated carbons. HNO3 treatment time (h) BET surface area (m2/g) 0 912 2 831 4 821 8 792 16 722 38 605 Table 1. BET N2 surface area for the HNO3 treated carbons. HNO3 treatment time (h) BET surface area (m2/g) Table 1. BET N2 surface area for the HNO3 treated carbons. Figure 7. Metal cations adsorption on functionalized activated carbons. Vol. 6, No. 2, 2003 Tailoring Activated Carbon b complex formation with acetate (a model –COO- groups) in solution: M+2 + CH3COO-↔ [M(CH3COO)]+ K = Figure 8 shows the correlation of the stab with the adsorption capacities. It can be observed that as the stability (lo the adsorption capacity of the carbons incre This excellent cation exchange properties Figure 8. Stability constants versus adsorption different metal cations onto 16 h reflux treated c [M [M+ Tailoring Activated Carbon by Surface Chemical Modification with O, S, and N Containing Molecules Vol. 6, No. 2, 2003 Tailoring Activated Carbon by Surface Chemical Modification with O, S, and N Containing Molecules 133 complex formation with acetate (a model for the surface –COO- groups) in solution: carbon containing –COOH surface groups makes them promising adsorbents to remove metal contaminants in aque- ous effluents. M+2 + CH3COO-↔ [M(CH3COO)]+ K = [M(CH3COO)]+ [M+2].[CH3COO-] 3.3 Carbon Surface Functionalization. IR spectra for the (a) carbon treated with HNO3 (8 h); (b) after reaction with SOCl2, followed by reaction with; (c) HSCH2CH2SH (EDT) or (d) H2NCH2CH2NH2 (EDA). Scheme 2. Reactions used to functionalize the carbon surface with S and N groups. Figure 9. IR spectra for the (a) carbon treated with HNO3 (8 h); (b) after reaction with SOCl2, followed by reaction with; (c) HSCH2CH2SH (EDT) or (d) H2NCH2CH2NH2 (EDA). Scheme 2. Reactions used to functionalize the carbon surface with S and N groups. 134 Rios et al. Materials Research Table 2. S and N molecules grafted on carbon surface. Molecule Formulae 1,2-ethaneditiol (EDT) HSCH2CH2SH 1,6-dimercapto-4-thioheptane (DMTH) HSCH2CH2CH2SCH2CH2CH2SH 1,2-diaminoethane (EDA) NH2CH2CH2NH2 Triethyltetraamine (TEA) H2NCH2CH2NHCH2CH2NHCH2CH2NH2 Table 2. S and N molecules grafted on carbon surface. Formulae HSCH2CH2SH HSCH2CH2CH2SCH2CH2CH2SH NH2CH2CH2NH2 H2NCH2CH2NHCH2CH2NHCH2CH2NH2 Figure 10. TG analyses of the (a) HNO3; (b) EDT; (c) EDA functionalyzed carbons in air. Molecule 1,2-ethaneditiol (EDT) 1,6-dimercapto-4-thioheptane (DMTH) 1,2-diaminoethane (EDA) Triethyltetraamine (TEA) HSCH2CH2SH HSCH2CH2CH2SCH2CH2CH2SH NH2CH2CH2NH2 2 2 2 2 H2NCH2CH2NHCH2CH2NHCH2CH2NH2 the following bands: ~1720 cm-1 (-C=O), 1675 and 1600 cm-1 (-N-H), 1520 cm-1 (-RCONHR), 950-1000 cm-1 (DTE characteristic bands) and 1200 cm-1 (-C-N) and 3000-3500 cm-1 (N-H). Similar results have been observed for the functionalization with DMTH and TEA, with the presence of bands typical of –COSR and –CONHR groups and also characteristic absorption presented by the R groups. 3.4 Thermogravimetric Analyses of the Functionalysed Carbons TG analyses of the DMTH and TEA functionalysed car- bons are shown in Fig. 10 It can be observed for the EDT functionalyzed material in Fig. 9b a weight loss of ca. 46% centered at 250 °C. From this weight loss it is possible to estimate the concentration of the EDT groups grafted on the carbon surface of approxi- mately 2 mmol/g of carbon. Therefore, considering the –COOH concentration of ca. 4.1021 sites/g this weight loss suggests that approximately 50% of these sites have been actually converted to the thio surface group -COSCH2CH2SH. From the EDA analysis (Fig. 9c), the weight loss of 62% at the temperature 300 °C, it can be estimated a concentration of surface groups of 3.4 mmol/g of carbon corresponding to a yield of approximately 75%. For both analyses the weigh losses observed at temperatures higher than 500 °C are re- lated to the oxidation of the carbon by air. Figure 10. TG analyses of the (a) HNO3; (b) EDT; (c) EDA functionalyzed carbons in air. Figure 11. 3.5 Temperature Programmed Decomposition/Mass Spectrometry Experiments (TPDMS) 3.5 Temperature Programmed Decomposition/Mass Spectrometry Experiments (TPDMS) In these TPDMS experiments the functionalized carbons were heated in a He flow at 10 °C/min and upon the ther- mal decomposition the volatile products were analyzed by a mass spectrometer. The results are displayed in Fig. 11. The TPDMS profile show two peaks in the total ion sig- nal (TIC) centered at 150 and 225 °C. It was observed that the peak at 150 °C is composed by several m/z fragments such as 182 (HSCH2CH2CH2SCH2CH2CH2SH+) (Fig. 11) and 74 (CH2CH2CH2SH+) suggesting that these fragments originate from the thermal decomposition of the surface DMTH group, which is released to the gas phase and de- tected by the mass spectrometer. The TPDMS peak at 225 °C is composed by a m/z signal 44, CO2, formed by the thermal decomposition of –COOH surface groups. Similar TPD profile was obtained for the HNO3 treated carbon. Figure 11. TPDMS experiment with the DMTH functionalised carbon. 3.3 Carbon Surface Functionalization. TPDMS experiment with the DMTH functionalised carbon. 5. Conclusion 15. Bradley, R.H.; Sutherland, I.; Sheng, E. J. Chem. Soc. Faraday Trans., v. 91, p. 3201, 1995. Activated carbon can be tailored by surface chemical modification to produce materials with unique properties and for specific applications. These modifications can be carried out via the key surface intermediate: -COOH. The carbon containing –COOH groups shows excellent cation exchange properties and can be used as adsorbent for metal contaminants in water. These –COOHsurf groups can be con- verted to -COCl surf which is a very reactive and versatile group for the grafting of other molecules onto the carbon surface. From this acylchloride intermediate several S and N molecules, for example HSR and HNR2, can be easily bounded to the carbon surface. These S and N functionalized carbon show great potential as adsorbent of metals, spe- cially heavy metals which represent a serious environmen- tal problem nowadays. 16. Boehm, H.P. Adv. Catal., v. 16, p. 179, 1966. 17. Boehm, H.P.; Diehl, E.; Heck, W. J. Chem. Soc. Ind., v. 369, 1966. 18. Boehm, H.P.; Tereczik, B.; Schanz, K. Proc. Colloque Intern. Adsorption Interfaces gaz/solide, Aix-en-Pro- vence, 1981, Elsevier, p. 395, 1982. 19. Puri, B.P.; Hazra, R.S. Carbon, v. 9, p. 123, 1971. 20. Pittman, C.U.; He, G.R.; Wu, B.; Gardner, S.D. Carbon, v. 35, p. 317, 1997. 21. Lakov, L.; Vassileva, P.; Peshev, O. Carbon, v. 37, p. 1655, 1999. 22. Zawadzki, J. Infrared spectroscopy in surface chemis- try of carbons. In: Chemistry and Physics of Carbon. P. A. Thrower (Ed.), vol. 21, Marcel Dekker, New York, p. 147-386, 1989. 4. Metal Adsorption These sulfur and nitrogen functionalyzed carbons show great potential for the adsorption of heavy metals such as Hg2+, Zn2+, Cd2+ and Pb2+ ions due to the presence of surface –SH and –NH groups22. The materials obtained in this work 135 Vol. 6, No. 2, 2003 Vol. 6, No. 2, 2003 Tailoring Activated Carbon by Surface Chemical Modification with O, S, and N Containing Molecules Figure 12. Adsorption of Pb+2, Cu+2 and Ni+2 in aqueous solution on untreated carbon (ST) and on the EDT (C/S) and EDA (N/C) functionalyzed carbons. References 1. Spencer, N.D. Advanced Surfaces: Their Tailoring and Analysis, Chimia, v. 52, p. 598, 1998. 2. Bansal, R.C.; Donnet, J.B.; Stoeckli, F. Active Carbon, Marcel Dekker Inc., New York, p. 27-35, 1988. 3. Shim, J.W.; Lee, S.M.; Rhee, B.S.; Ryu, S.K. Adsorption of Ni(II), Cu(II), Cr(VI) from multicomponent aqueous solution by pitch-based ACF. In: Extended abstracts, European Carbon Conference, Newcastle, UK, , British Carbon Group, p. 242-243, 1996. 4. Rodriguez-Reinoso, F. The role of carbon materials in heterogeneous catalysis. Carbon, v. 36, p. 159, 1998. 5. Peebles, L.H.Jr. In: Carbon fibers: formation, structure, and properties, CRC, Ann Arbor, p. 125 chapter 8, 1994. 6. Park, S.J.; Kim, M.H. J. Mater. Sci., v. 35, p. 1901, 2000. 7. E.C. Peters, M. Petro, F.Svec and J.M.J. Frechet, Anal. Chem., v. 70, p. 2288, 1998. Figure 12. Adsorption of Pb+2, Cu+2 and Ni+2 in aqueous solution on untreated carbon (ST) and on the EDT (C/S) and EDA (N/C) functionalyzed carbons. 8. M.V. Michel, C. Marzin, G. Tarrago and J. Durand, J. Appl. Polym. Sci., v. 70, p. 359, 1998. 9. Gregg, S.J.; Sing, K.S.W. In: Adsorption, Surface Area and Porosity, Academic Press, London, p. 121-189, 1982. 10. Matson, P.; Mark, H.B. Activated carbon: Surface Chem- istry and Adsorption from Solution. Marcel Dekker, New York, 1971. were also tested for the adsorption of Pb+2, Cu+2 and Ni+2 in aqueous solution. The comparison of the adsorptions on the untreated carbon, the EDT (C/S) and EDA (N/C) functionalyzed carbons are shown in Fig. 12. 11. Ruthven, D.M. Principles of Adsorption and Adsorp- tion Processes. John Wiley & Sons Inc. New York, 1984. It can be seen that the functionalization of the carbon strongly increases its adsorption for the different metals stud- ied. The results suggested that the EDT funtionalyzed ma- terial can adsorb more efficiently Ni+2 whereas the EDA carbon showed a stronger interaction with Pb+2 cations. 12. Lisovskii, A.; Semiat, R.; Aharoni, C. Carbon, v. 35, p. 1639, 1997. 13. Ramos, M.A.; Serrano, V.G.; Calahlorro, C.V.; Peinado, A.J.L. Spec. Lett., v. 26, p. 1117, 1993 14. Ogino, K. Studies Surface Sci. Catalysis, v. 80, p. 491, 1992. Acknowledgements 23. Gonzales, J.D.L.; Castilla, C.M.; Guerrero, A.; Reinoso, F.R. 15th Bienn. Conf. Carbon, Philadelphia, p. 250, 1981. The authors are grateful for funding from PRPq/UFMG, CAPES, FAPEMIG and CNPq.
https://openalex.org/W2966694263	https://periodicos.unemat.br/index.php/reps/article/download/10189/6731	Portuguese	null	Educação para o trânsito nos primeiros anos do ensino fundamental	Eventos Pedagógicos	2,019	cc-by	6,077	TRAFFIC EDUCATION IN THE FIRST YEARS TRAFFIC EDUCATION IN THE FIRST YEARS OF ELEMENTARYSCHOOL TRAFFIC EDUCATION IN THE FIRST YEARS Sandra Regina Mendes Sandra Regina Mendes RESUMO Este artigo apresenta um estudo sobre a educação para o trânsito como tema interdisciplinar nos primeiros anos do ensino fundamental nas escolas municipais de Sinop, Mato Grosso. O objetivo foi identificar como esta educação está contribuindo para um melhor comportamento no trânsito e com este propósito foram realizadas pesquisas bibliográficas com base em autores como Eduardo Alcântara Vasconcelos e João Pedro Martins. A pesquisa foi de cunho descritivo e qualitativo e envolveu entrevistas com professoras e gestores públicos. Posteriormente à análise dos dados obtidos, concluiu-se que é necessário um trabalho de caráter permanente nas escolas juntamente com o apoio fam Este artigo apresenta um estudo sobre a educação para o trânsito como tema interdisciplinar nos primeiros anos do ensino fundamental nas escolas municipais de Mato Grosso. O objetivo foi identificar como esta educação está contribuindo para um melhor comportamento no trânsito e com este propósito foram realizadas pesquisas bibliográficas com base em autores como Eduardo Alcântara Vasconcelos . A pesquisa foi de cunho descritivo e qualitativo e envolveu entrevistas com professoras e gestores públicos. Posteriormente à análise dos se que é necessário um trabalho de caráter permanente nas escolas juntamente com o apoio familiar para a obtenção de resultados relevantes. Este artigo apresenta um estudo sobre a educação para o trânsito como tema interdisciplinar nos primeiros anos do ensino fundamental nas escolas municipais de Mato Grosso. O objetivo foi identificar como esta educação está contribuindo para um melhor comportamento no trânsito e com este propósito foram realizadas pesquisas bibliográficas com base em autores como Eduardo Alcântara Vasconcelos . A pesquisa foi de cunho descritivo e qualitativo e envolveu entrevistas com professoras e gestores públicos. Posteriormente à análise dos se que é necessário um trabalho de caráter permanente nas enção de resultados relevantes. Palavras-chave: Ensino Fundamental. Educação. Trânsito. Abordagem qualitativa. Ensino Fundamental. Educação. Trânsito. Professores e gestores. Professores e gestores. Abordagem qualitativa. EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS TRAFFIC EDUCATION IN THE FIRST YEARS REP’s REP’s REP’s REP’s Número Regular Sinop, v. ISSN http://sinop.unemat.br/projetos/revista/index.php/eventos/index DOI: EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL1 TRAFFIC EDUCATION IN THE FIRST YEARS OF ELEMENTARYSCHOOL REP’s REP’s REP’s REP’s ---- Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 ISSN 2236-3165 http://sinop.unemat.br/projetos/revista/index.php/eventos/index DOI: 10.30681/2236-3165 EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS TRAFFIC EDUCATION IN THE FIRST YEARS Educação e Literatura: saberes, cultura e leitura http://sinop.unemat.br/projetos/revista/index.php/eventos/index EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS TRAFFIC EDUCATION IN THE FIRST YEARS REP’s REP’s REP’s REP’s Número Regular Sinop, v. ISSN http://sinop.unemat.br/projetos/revista/index.php/eventos/index DOI: EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL1 TRAFFIC EDUCATION IN THE FIRST YEARS OF ELEMENTARYSCHOOL REP’s REP’s REP’s REP’s ---- Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 ISSN 2236-3165 http://sinop.unemat.br/projetos/revista/index.php/eventos/index DOI: 10.30681/2236-3165 EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS TRAFFIC EDUCATION IN THE FIRST YEARS Educação e Literatura: saberes, cultura e leitura http://sinop.unemat.br/projetos/revista/index.php/eventos/index REP’s REP’s REP’s REP’s Número Regular Sinop, v. ISSN http://sinop.unemat.br/projetos/revista/index.php/eventos/index DOI: REP’s REP’s REP’s REP’s ---- Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 ISSN 2236-3165 http://sinop.unemat.br/projetos/revista/index.php/eventos/index DOI: 10.30681/2236-3165 Educação e Literatura: saberes, cultura e leitura http://sinop.unemat.br/projetos/revista/index.php/eventos/index Palavras-chave: Ensino Fundamental. Educação. Trânsito. Abordagem qualitativa. Ensino Fundamental. Educação. Trânsito. Professores e gestores. Professores e gestores. Recebido em: 09 de maio de 2019. Aprovado em: 04 de junho de 2019. Link: http://sinop.unemat.br/projetos/revista/index.php/eventos/article/view/3515/2480 Recebido em: 09 de maio de 2019. Aprovado em: 04 de junho de 2019. Link: http://sinop.unemat.br/projetos/revista/index.php/eventos/article/view/3515/2480 Recebido em: 09 de maio de 2019. Aprovado em: 04 de junho de 2019. Recebido em: 09 de maio de 2019. Aprovado em: 04 de junho de 2019. Link: http://sinop.unemat.br/projetos/revista/index.php/eventos/article/view/3515/2480 Aprovado em: 04 de junho de 2019. p j Link: http://sinop.unemat.br/projetos/revista/index.php/eventos/article/view/351 Página 452 – Sandra Regina Mendes ABSTRACT2 1 Este artigo é um recorte do Trabalho de Conclusão de Curso intitulado TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL Antônio de Souza, Curso de Pedagogia, Faculdade de Educação e Linguagem (FAEL) da Universidade do Estado de Mato Grosso (UNEMAT), Câm 2 Resumo traduzido pela professora Mestra Betsemens Barbosa de Souza Marcelino. Professora interina do curso de Letras da UNEMAT/Sinop. Mestra em Estudos de Linguagem pe 2015.Graduada em Licenciatura Plena em Letras, Português/Inglês pela UNEMAT/Sinop, 2013. ste artigo é um recorte do Trabalho de Conclusão de Curso intitulado EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL, sob a orientação do Dr. Edison Curso de Pedagogia, Faculdade de Educação e Linguagem (FAEL) da Universidade do Estado de Mato Grosso (UNEMAT), Câmpus Universitário de Sinop, 2018/2. Resumo traduzido pela professora Mestra Betsemens Barbosa de Souza Marcelino. Professora interina do curso de Letras da UNEMAT/Sinop. Mestra em Estudos de Linguagem pe 2015.Graduada em Licenciatura Plena em Letras, Português/Inglês pela UNEMAT/Sinop, 2013. EDUCAÇÃO PARA O , sob a orientação do Dr. Edison Curso de Pedagogia, Faculdade de Educação e Linguagem (FAEL) da pus Universitário de Sinop, 2018/2. Resumo traduzido pela professora Mestra Betsemens Barbosa de Souza Marcelino. Professora interina do curso de Letras da UNEMAT/Sinop. Mestra em Estudos de Linguagem pela UFMT/Cuiabá, 2015.Graduada em Licenciatura Plena em Letras, Português/Inglês pela UNEMAT/Sinop, 2013. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. d g g d g g d g g d g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 This article presents a study on traffic education as an interdisciplinary theme addressed in the first years of elementary school in schools of municipal Department in Sinop city, Mato Grosso. The objective was to identify how does school education is contributing for a better behave in traffic and, on this purpose, it were carried out bibliographic research based on authors such as Eduardo Alcântara Vasconcelos and João Pedro Martins. The research had a qualitative approach and involved interviews with teachers and public managers. After analyzing the data obtained, it was concluded that regarding the studied subject Correspondência: Sandra Regina Mendes. Graduanda em Pedagogia pela Universidade do Estado de Mato Grosso (UNEMAT), Faculdade de Educação e Linguagem (FAEL). Sinop, Mato Grosso, Brasil. E-mail: sandrajjmendes@hotmail.com 1 INTRODUÇÃO O presente artigo apresenta um dos maiores desafios do governo federal, estadual e municipal que é assegurar e proteger a população nas vias públicas. Esta proposta de pesquisa pretende discutir o que está sendo desenvolvido para mostrar e amenizar o alto índice de acidentes de trânsito em nossa cidade, no qual muitas pessoas perdem a vida, e outras convivem com sequelas dos acidentes pelo resto de suas vidas, na maior parte das vezes por falha humana. Neste artigo a problemática é o alto índice de acidentes em Sinop e a importância da discussão sobre o mesmo. Está incluída ainda uma proposta de pesquisa que pretende debater a necessidade de desenvolver um trabalho nas escolas com crianças a partir dos primeiros anos do ensino fundamental, para se pensar num futuro diferente para a nossa sociedade onde o professor é uma das figuras determinantes, pois a educação influencia uma nação e conduz um país ao pleno desenvolvimento que seu povo precisa e merece. O objetivo deste artigo é mostrar a necessidade de inserir as crianças num contexto escolar, para que entendam que pedestres e motoristas são os indivíduos Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. R d g g R P d g g R P d g g R d g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 d g g d g g d g g d g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 essenciais no trânsito e que a criança passa a ser vista como um multiplicador de educação no trânsito. 1 INTRODUÇÃO Fica claro a necessidade de introduzir uma educação interdisciplinar para o trânsito ainda que nos primeiros anos do ensino fundamental. A pesquisa foi realizada com a contribuição de autores como: Eduardo Alcântara Vasconcelos (2010) e João Pedro Martins (2007). Para o êxito desta pesquisa é indispensável o apoio da instituição e dos professores para a concretização e realizações das etapas desse projeto. A obtenção de dados por meio dos questionários, foram para uma análise mais detalhada sobre a ação pedagógica em relação a educação para o trânsito nos primeiros anos do ensino fundamental. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. g g d g g d g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Para que haja uma mudança segundo Martins (2007, p. 83), “É necessário conscientizar o cidadão que a reeducação, a se iniciar nos bancos escolares, já nas primeiras séries, não pode se limitar à situação escolar. Ela precisa mobilizar as crianças, os familiares, a comunidade, o estado e a nação”. Faz-se necessário que esta educação comece nos primeiros anos escolares e de continuidade até sua formação acadêmica. Com propriedade Vasconcelos (2010, p. 11) afirma que: “trânsito é um lugar de uso coletivo, onde todas as pessoas têm direitos e deveres, logo é necessário respeito e compreensão de todas as pessoas, para evitar que aconteça os terríveis acidentes”. Uma intervenção educacional é necessária para que haja diversas mudanças de comportamento no trânsito. O trabalho da educação é importante para que mudanças ocorram de forma permanente, a escola tem papel fundamental nesse processo. Pois, segundo Marin e Queiroz (2000, p. 12): [...] é necessário um conhecimento maior das culturas e das condições de vida locais para que as atitudes dos motoristas possam ser compreendidas, aproveitando esse conhecimento em programas de capacitação, reabilitação e educação, que promovam um comportamento mais adequado. Educação para o trânsito se faz necessária no atual cenário em que vivemos, a busca pelo progresso, o grande desejo de facilitar e simplificar a locomoção humana em tempo recorde, levam as pessoas a disputarem espaços físicos muitas vezes de maneira irresponsáveis, não respeitando a própria vida nem a vida alheia. TRÂNSITO E EDUCAÇÃO 2 Segundo Pascarelli Filho (2012, p. 02), “[..] é preciso capacitar docentes para trabalhar o tema trânsito em sala de aula atendendo à força de lei e praticar a responsabilidade social de modo a mudar, através da educação sistemática” Considerando o número de acidentes fatais e com sequelas no trânsito em nosso país, e acreditando na relevância e a necessidade de uma intervenção pedagógica contextualizada, para que no futuro possamos ter mudanças realmente significativas no comportamento dos jovens e motoristas de todo o país, defendendo a necessidade da educação para o trânsito como tema interdisciplinar nas escolas como necessidade de urgência. Vasconcelos (2010, p. 9) descreve: “O trânsito é, assim, o conjunto de todos os deslocamentos diários, feitos pelas calçadas e vias da cidade, e que aparece na rua na forma de movimentação geral de pedestres e veículos”. Logo, imagina-se que há: adultos, crianças e idosos a pé e de bicicleta, motos, carros, ônibus, caminhões fazendo entregas etc., animais domésticos que circulam normalmente pelas vias. Martins (2007, p. 19), discorre que: “É preciso humanizar a realidade do trânsito, corrigindo os erros com campanhas educativas bem conduzidas e direcionadas pelos diversos meios de comunicação, valendo-se de estratégias diversificadas”. Espera-se que a educação para o trânsito venha desempenhar e resgatar valores e atitudes para que possamos ver em nossa sociedade uma verdadeira transformação de conduta e princípios. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. g g d g g d g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 g g g g g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 O local escolhido para a pesquisa foi o município de Sinop-MT, a coleta de dados ocorreu nos meses de agosto a outubro de 2018, com a visita às entidades. Para a elaboração da pesquisa optou-se pela abordagem qualitativa.de forma que Minayo (2002, p. 21-22) descreve como: A pesquisa qualitativa responde à questão muito particulares. Ela se preocupa, nas ciências sociais, com um nível de realidade que não pode ser quantificado. Ou seja, ela trabalha com o universo de significados, motivos, aspirações, crenças, valores e atitudes, o que corresponde ao um espaço mais profundo das relações, dos processos e dos fenômenos que não podem ser reduzidos a operações de variáveis. A coleta de dados foi obtida através de questionários objetivos, realizados com duas professoras em uma determinada escola municipal, naSecretaria de Educação Esportes e Cultura e na Secretaria de Trânsito Urbano (STU), município de Sinop, Mato Grosso. Para obtenção dos resultados, primeiramente realizamos a pesquisa com a Secretaria de Educação Esporte e Cultura. A Secretaria Municipal de Educação busca desenvolver projetos na Educação para o trânsito? (01) Secretaria Municipal de Educação, Esporte e Cultura: A secretaria Municipal de Educação, Esporte e Cultura é parceira ativa com entidades que já desempenham atividades voltadas a esta temática. Esta é uma temática inserida nos Componentes Curriculares trabalhados em todas unidades educativas da Rede Municipal. 3 EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL NA CONCEPÇÃO DOS PARTICIPANTES DESTA PESQUISA Para a realização da pesquisa será utilizada a metodologia de cunho qualitativo através de entrevistas semiestruturadas com duas professoras de uma escola municipal, funcionários da Secretária de Trânsito Urbano e a Secretária de Educação Esporte e Cultura. Página 454 – Sandra Regina Mendes Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Quais os projetos que são desenvolvidos? Quais os projetos que são desenvolvidos? (02) Secretaria Municipal de Educação, Esporte e Cultura: Evidenciando a parceria, Projeto Agente Mirim Escolar – AME com a Secretaria de Trânsito que tem como objetivo formar cidadãos comprometidos com a segurança no trânsito. FETHAB – Trânsito/Consciência Cidadã através da Arte Cênica. (02) Secretaria Municipal de Educação, Esporte e Cultura: Evidenciando a parceria, Projeto Agente Mirim Escolar – AME com a Secretaria de Trânsito que tem como objetivo formar cidadãos comprometidos com a segurança no trânsito. FETHAB – Trânsito/Consciência Cidadã através da Arte Cênica. “Associar educação a mudança não é novidade. Tem sido um costume desde pelo menos as primeiras décadas do século” (BRANDÃO, 1981, p. 83). EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL – Página 455 Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Há recursos destinados para estes projetos? Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. d g g d g g d g g d g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 g g g g g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 (06) Professora B: Sim. Tem um projeto que é desenvolvido nas escolas por um/a agente de trânsito nas escolas. O agente vem até as escolas uma vez por semana e trabalha com os alunos a conscientização e prevenção de acidentes com as crianças. O material é elaborado pela Secretaria Nacional do Trânsito. Detran, CONTRAN. Segundo Paulo Freire (2011, p. 97) “Toda vez, porém, que a conjuntura o exige, a educação dominante é progressiva à sua maneira, progressiva pela metade”. (03) Secretaria Municipal de Educação, Esporte e Cultura: Os recursos utilizados são oriundos de várias fontes que subsidiam a Educação. (03) Secretaria Municipal de Educação, Esporte e Cultura: Os recursos utilizados são oriundos de várias fontes que subsidiam a Educação. A Secretaria Municipal de Educação, Esporte e Cultura oferece formação para os professores na área de educação para o Trânsito? (04) Secretaria Municipal de Educação, Esporte e Cultura: No formato direcionado ao Professor especificamente não ofertamos. Contemplamos esta temática dentro das diversas formações ofertadas, abordando orientações sobre como desenvolver as práticas pedagógicas para a Educação do Trânsito. Participamos de campanhas que abordam o tema, com alunos das Unidades Escolares Municipais participantes do Projeto AME. O objetivo do Projeto Agente Mirim Escolare-AME é ter multiplicadores de informação e formar cidadãos mais conscientes. De acordo Com o Código de Transito Brasileiro-CTB, Lei nº 9.503 de 23 de setembro de 1997 (BRASIL, 2013, p. 39-40), no art. 76 diz que: A educação para o trânsito será promovida na pré-escola e nas escolas de 1º, 2º e 3º graus, por meio de planejamento e ações coordenadas entre os órgãos e entidades do Sistema Nacional de Trânsito e de Educação, da União, dos Estados, do Distrito Federal e dos Municípios, nas respectivas áreas de atuação. II - A adoção de conteúdos relativos à educação para o trânsito nas escolas de formação para o magistério e o treinamento de professores e multiplicadores. A Secretaria Municipal de Educação busca desenvolver projetos na Educação para o trânsito? (05) Professora A: A escola em si não tem projetos, porém realiza atividades voltadas ao trânsito com os alunos quando a Secretaria de Educação em parceria com a Secretaria de trânsito traz até a escola palestras, teatro e panfletos com orientações sobre o respeito e educação para o trânsito. Página 456 – Sandra Regina Mendes Página 456 – Sandra Regina Mendes Quais os projetos que são desenvolvidos? Quais os projetos que são desenvolvidos? (07) Professora A: Não realizo projetos sobre Educação para o trânsito com meus alunos, somente trabalho o conteúdo proposto na proposta pedagógica. (08) Professora B: Na escola, além do Projeto da Secretaria do Trânsito, nas salas o assunto é abordado como tema transversal. De acordo com Vasconcelos (2010, p. 64) “A educação deve ser vista como um processo continuo, para que tenha efetividade real”. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. d g g R d g g R d g g d g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 g g g g g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 (11) Professora A: Já participei de uma formação sobre Educação para trânsito realizada pela Secretaria de Educação no ano de 1999, quando tínhamos como prefeito Adenir Alves Barbosa. Lembro-me que foi uma parceria com o DETRAN de Cuiabá e ganhamos até o material pedagógico para trabalhar com os alunos. Depois disso não tivemos mais esse tipo de formação. E o trânsito, infelizmente só é lembrado pelas secretarias quando está na semana nacional do trânsito. Página 458 – Sandra Regina Mendes Há recursos destinados para estes projetos? Há recursos destinados para estes projetos? (09) Professora A: A escola não recebe recursos para trabalhar projetos voltados a Educação para o Trânsito e nem para outros projetos que possa vir a desenvolver. (10) Professora B: Do Projeto da Secretaria Municipal do Trânsito deve ter sim, mas é gerido pela própria secretaria. Quanto à escola, tem os repasses da Prefeitura e do FNDE/PDDE que uma parte é destinada às ações pedagógicas. A Secretaria Municipal de Educação, Esporte e Cultura oferece formação aos professores na área de educação para o trânsito? EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL – Página 457 Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. g g g g g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 g g g g g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Existe uma continuidade por parte das escolas após o término desses projetos? (15) Secretaria de Trânsito Urbano: A cada ano contemplamos outras escolas com o projeto AME. E, até o momento não há uma continuidade por parte das escolas. De acordo com Biavati e Martins (2008, p. 10) “O alerta global demarca a urgência de ações que se desdobram em muitas dimensões, mas apontam para um eixo fundamental: o processo educacional do jovem cidadão” Existe algum programa de capacitação em Educação para o Trânsito para os professores? Qual seria? (16) Secretaria de Trânsito Urbano: Não temos um programa de capacitação para os professores. Quais são os projetos desenvolvidos pela Secretaria de Trânsito Urbano(STU) nas escolas? Quais são os projetos desenvolvidos pela Secretaria de Trânsito Urbano(STU) nas escolas? (13) Secretaria de Trânsito Urbano: A STU juntamente com o Departamento de Educação para o Trânsito realiza palestras sobre Educação para o Trânsito nas escolas, faculdades, empresas públicas e privadas, realiza também blitz educativas em frente às escolas e demais pontos da cidade. Têm também o Projeto Agente Mirim Escolar - AME e este ano contempla duas escolas (EMEB Rodrigo Damasceno e EMEB Armando Dias) sendo todos os alunos dos 5º anos, num total de 330 alunos participando do Projeto semanalmente durante seis meses. Segundo Vasconcelos, (2010, p. 65) “Se nós ensinarmos às crianças algumas regras de bom comportamento no trânsito e eles virem os adultos agindo erradamente a toda hora, o resultado do processo poderá ser muito prejudicado” (VASCONCELOS, 2010, p. 65). Faz se necessário que todos os pais e comunidade se envolva para uma real mudança de comportamento. Há recursos destinados para estes projetos? Há parcerias? (14) Secretaria de Trânsito Urbano: Os recursos para a realização dos projetos educativos para o trânsito são da própria Prefeitura. A Secretaria de Trânsito Urbano vem confirmar o que a secretaria de educação disse com respeito aos trabalhos que ambas realizam juntas como parceiras. (16) Secretaria de Trânsito Urbano: Não temos um programa de capacitação para os professores. Segundo Vasconcelos (2010, p. 65): “Em todos esses níveis, o processo educacional no Brasil pode ser considerado muito deficiente: não há atividade escolar adequada, o exame de habilitação é falho e as campanhas são esporádicas, sem planejamento global”. A Prefeitura Municipal de Sinop tem algum projeto para implantar semáforos na cidade? (17) Secretaria de Trânsito Urbano: Está em discussão a implantação de mais semáforos, bem como a implantação de radares de velocidade. Seria possível desenvolver juntamente com a Unemat um projeto permanente de Educação para o Trânsito em Sinop? (18) Secretaria de Trânsito Urbano: Com o objetivo de melhorar o trânsito de nossa cidade, a STU está sempre aberta a discussões e a novas parcerias. (18) Secretaria de Trânsito Urbano: Com o objetivo de melhorar o trânsito de nossa cidade, a STU está sempre aberta a discussões e a novas parcerias. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. g g d g g d g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 g g g g g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Vasconcelos, (2010, p. 62). “Mas será que após ver tudo isto devemos continuar a usar a palavra “acidente”? Essa palavra se aplica a algo inesperado e inevitável, mas pudemos constatar que esses “acidentes” são produto de uma série de causas evitáveis”. Se todos os setores educacionais se empenharem em prol do bem comum, com certeza os resultados aparecerão como decorrência da prática. Tabela 1 - Total de acidentes no município de Sinop-MT de 2010 a 2018 2010 2011 2012 2013 2014 2015 2016 2017 2018 JAN 104 108 122 160 148 131 134 210 153 FEV 117 117 144 153 158 176 186 180 159 MAR 137 138 169 153 195 176 200 223 176 ABR 135 153 159 219 157 182 161 212 194 MAI 173 164 185 222 198 194 173 234 178 JUN 151 141 155 185 150 217 155 180 203 JUL 131 147 140 217 179 176 175 144 156 AGO 165 154 174 183 189 160 166 243 227 SET 148 147 143 199 176 148 168 197 159 OUT 150 163 180 184 196 160 177 181 213 NOV 151 163 160 200 179 177 183 179 DEZ 143 149 181 164 175 209 187 139 TOTAIS 1705 1744 1912 2239 2100 2106 2065 2322 1818 Fonte: Secretaria de trânsito urbano, Sinop (2018) Gráfico 1: Total de óbitos no Município de 2010 a 2017. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Fonte: Secretaria de trânsito urbano, Sinop, (2018) 43 52 56 42 26 23 23 16 19 20 29 31 6 43 52 56 61 46 52 54 22 2010 2011 2012 2013 2014 2015 2016 2017 ÓBITOS (IML) STU ROD TOTAL Tabela 1 - Total de acidentes no município de Sinop-MT de 2010 a 2018 2010 2011 2012 2013 2014 2015 2016 2017 2018 JAN 104 108 122 160 148 131 134 210 153 FEV 117 117 144 153 158 176 186 180 159 MAR 137 138 169 153 195 176 200 223 176 ABR 135 153 159 219 157 182 161 212 194 MAI 173 164 185 222 198 194 173 234 178 JUN 151 141 155 185 150 217 155 180 203 JUL 131 147 140 217 179 176 175 144 156 AGO 165 154 174 183 189 160 166 243 227 SET 148 147 143 199 176 148 168 197 159 OUT 150 163 180 184 196 160 177 181 213 NOV 151 163 160 200 179 177 183 179 DEZ 143 149 181 164 175 209 187 139 TOTAIS 1705 1744 1912 2239 2100 2106 2065 2322 1818 Fonte: Secretaria de trânsito urbano, Sinop (2018) Gráfico 1: Total de óbitos no Município de 2010 a 2017. Gráfico 1: Total de óbitos no Município de 2010 a 2017. Página 460 – Sandra Regina Mendes Página 460 – Sandra Regina Mendes EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL – Página 461 4 CONSIDERAÇÕES FINAIS O presente trabalho buscou informação de como é tratado o tema educação para o trânsito nos primeiros anos do ensino fundamental no município de Sinop, Mato Grosso. Dentro desta perspectiva e apesar da lei ser clara em dizer que a educação para o trânsito deve ser adotada em todos os níveis de ensino, desde a pré-escola ao ensino superior, descobriu-se que na realidade isso não acontece na prática, pois, toda a informação sobre o trânsito praticada nas escolas fica limitado a semana nacional do trânsito, em palestras e distribuições de folhetos. Não tendo uma continuidade significativa. Espera-se que a educação para o trânsito venha desempenhar e resgatar valores e atitudes para que possamos ver em nossa sociedade uma verdadeira transformação de conduta e princípios, pois nossas crianças serão os futuros condutores de veículos. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. 451-463, jan./jul. 2019 solidariedade e responsabilidade constituem os eixos determinantes da transformação do comportamento do homem no trânsito”. REFERÊNCIAS BIAVATI, Eduardo; MARTINS, Heloisa. Trânsito e Mobilidade: em trânsito consultoria 2008. BIAVATI, Eduardo; MARTINS, Heloisa. Trânsito e Mobilidade: em trânsito consultoria 2008. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. g g g g g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 Tabela 2 - População de município de Sinop e frota de carros de 2010 a 2017 PERIODO 2010 2011 2012 2013 2014 2015 2016 2017 POPULAÇÃ O (IBGE) 113.099 116.013 117.448 118.883 123.634 126.817 129.916 132.945 FROTA (DETRAN - Julho/17) 61.800 69.681 77.400 86.025 93.164 101.323 105.761 106.777 Fonte: Secretaria de trânsito urbano, Sinop (2018) Tabela 2 - População de município de Sinop e frota de carros de 2010 a 2017 Tabela 3 - Total de óbitos dentro do município de Sinop. p p ÓBITOS ÁREA DE ATUAÇÃO STU (IML) 2010 2011 2012 2013 2014 2015 2016 2017 2018 JAN PERÍODO NÃO DISTRIBUÍDO 3 3 1 1 4 FEV 3 2 4 3 0 MAR 4 2 2 3 2 5 ABR 2 1 3 3 4 MAI 2 2 1 3 2 1 JUN 5 4 5 5 JUL 4 1 4 4 2 3 AGO 6 7 2 4 SET 5 5 1 3 2 4 OUT 3 3 3 1 3 5 NOV 1 3 2 5 DEZ 4 2 2 7 TOTAIS 43 52 56 42 26 23 23 37 35 Fonte: Secretaria de trânsito urbano, Sinop (2018) ÓBITOS ÁREA DE ATUAÇÃO STU (IML) ÓBITOS ÁREA DE ATUAÇÃO STU (IML) Tabela 4 - Totais de óbitos urbanos e rodoviários. Tabela 4 - Totais de óbitos urbanos e rodoviários. Tabela 4 - Totais de óbitos urbanos e rodoviários. ÓBITO EM ACIDENTES DE TRÂNSITO (IML) ÓBITO 2013 2014 2015 2016 2017 2018 Atuação STU Masculino 33 19 18 20 22 25 Feminino 9 7 5 3 12 9 Não informado 3 1 Sub-total STU 42 26 23 23 37 35 Rodovias Masculino 15 16 23 26 13 16 Feminino 4 4 6 5 9 10 Sub-total ROD 19 20 29 31 22 26 TOTAL 61 46 52 54 59 61 Fonte: Secretaria de trânsito urbano, Sinop (2018) De acordo com Ecco e Banaszeski (2009, p. 04) “Educar para o trânsito é preservar a vida, evitar acidentes, exercer a cidadania, no qual respeito, cortesia, Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. g g g g g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. Revista Even. Pedagóg. g g g g Número Regular: Educação e Literatura: saberes, cultura e leitura Sinop, v. 10, n. 1 (26. ed.), p. 451-463, jan./jul. 2019 FREIRE, Paulo. Pedagogia da autonomia: saberes necessários à práticas educativas. São Paulo: Paz e Terra, 2011. FREIRE, Paulo. Pedagogia da autonomia: saberes necessários à práticas educativas. São Paulo: Paz e Terra, 2011. MARTINS, João Pedro. A Educação de Trânsito: campanhas educativas nas escolas. Belo Horizonte: Autêntica, 2007. MINAYO, Maria Cecília de Souza. O desafio da pesquisa social. In: MINAYO, Maria Cecília de Souza; GOMES, Suely Ferreira Deslandes Romeu (org.). Pesquisa social: teoria, método e criatividade. 21. ed. Petrópolis: Editora Vozes, 2002, p. 09- 29. MINAYO, Maria Cecília de Souza. O desafio da pesquisa social. In: MINAYO, Maria Cecília de Souza; GOMES, Suely Ferreira Deslandes Romeu (org.). Pesquisa social: teoria, método e criatividade. 21. ed. Petrópolis: Editora Vozes, 2002, p. 09- 29. PASCARELLI FILHO, Nelson. A Triste Realidade do Trânsito Brasileiro. Modulo 2. GPECE: Educação a distância. In: Curso: Gestão Educacional e Prática Pedagógica, 2012. p.1-7. Disponível em: h //i / d/d / i lid d d â i b il i PASCARELLI FILHO, Nelson. A Triste Realidade do Trânsito Brasileiro. Modulo 2. GPECE: Educação a distância. In: Curso: Gestão Educacional e Prática Pedagógica, 2012. p.1-7. Disponível em: PASCARELLI FILHO, Nelson. A Triste Realidade do Trânsito Brasileiro. Modulo 2. GPECE: Educação a distância. In: Curso: Gestão Educacional e Prática Pedagógica, 2012. p.1-7. Disponível em: h //i / d/d / i lid d d â i b il i ttps://issuu.com/cursosgpecead/docs/a_triste_realidade_do_trânsito_brasileiro. cesso em: 15 set. 2018. https://issuu.com/cursosgpecead/docs/a_triste_realidade_do_trânsito_brasileiro. Acesso em: 15 set. 2018. PROFESSORA A. Dados de Pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. PROFESSORA A. Dados de Pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. PROFESSORA A. Dados de Pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. PROFESSORA B. Dados de Pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. PROFESSORA B. Dados de Pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. Secretaria Municipal De Educação, Esporte E Cultura. Dados de Pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. Secretaria Municipal De Educação, Esporte E Cultura. Dados de Pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. BIAVATI, Eduardo; MARTINS, Heloisa. Trânsito e Mobilidade: em trânsito consultoria 2008. BRANDÃO, Carlos Rodrigues. O que é Educação. São Paulo: Editora Brasiliense, 1981. BRASIL. [Código de trânsito brasileiro (1997)]. Código de trânsito brasileiro [recurso eletrônico]: Lei nº. 9.503, de 23 de setembro de 1997, e legislação correlata. 5. ed. Brasília, DF: Câmara dos Deputados, Edições Câmara, 2013. Disponível em: http://bd.camara.gov.br/bd/bitstream/handle/bdcamara/18141/codigo_transito_5ed.p df?sequence=8. Acesso em: 26 nov. 2018. ECCO, Idanir; BANASZESKI, Alexandra Auziliero. Educação para o Trânsito: um Olhar para o Contexto Escolar. WebArtigos. Publicado em 06 de março de 2009. Disponível em: https://www.webartigos.com/artigos/educacao-para-o-transito-um- olhar-para-o-contexto-escolar/15180. Acesso em: 27 jan. 2018. Página 462 – Sandra Regina Mendes FREIRE, Paulo. Pedagogia da autonomia: saberes necessários à práticas educativas. São Paulo: Paz e Terra, 2011. MARIN, Letícia; QUEIROZ, Marcos Souza. A atualidade dos acidentes de trânsito na era da velocidade: uma visão geral. Cadernos de Saúde Pública, v.16, n.1, p.7-21, 2000. Disponível em: http://www.scielo.br/scielo.php?pid=S0102- 311X2000000100002&script=sci_abstract&tlng=pt. Acesso em: 18 jul. 2018. MARIN, Letícia; QUEIROZ, Marcos Souza. A atualidade dos acidentes de trânsito na era da velocidade: uma visão geral. Cadernos de Saúde Pública, v.16, n.1, p.7-21, 2000. Disponível em: http://www.scielo.br/scielo.php?pid=S0102- 311X2000000100002&script=sci_abstract&tlng=pt. Acesso em: 18 jul. 2018. MARTINS, João Pedro. A Educação de Trânsito: campanhas educativas nas escolas. Belo Horizonte: Autêntica, 2007. MARTINS, João Pedro. A Educação de Trânsito: campanhas educativas nas escolas. Belo Horizonte: Autêntica, 2007. MARTINS, João Pedro. A Educação de Trânsito: campanhas educativas nas escolas. Belo Horizonte: Autêntica, 2007. SECRETARIA DE TRÂNSITO URBANO. Dados de pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. SECRETARIA DE TRÂNSITO URBANO. Dados de pesquisa. [Entrevista cedida à]: Sandra Regina Mendes. Educação para o trânsito nos primeiros anos do ensino Fundamental. Sinop, UNEMAT, Curso de Pedagogia, ago/dez 2018. INOP. Secretaria de Trânsito Urbano – STU: estatísticas de acidentes, 2018. SINOP. Secretaria de Trânsito Urbano – STU: estatísticas de acidentes, 201 VASCONCELOS, Eduardo Alcântara. O Que é Trânsito. São Paulo: Editora Brasiliense, 2010. VASCONCELOS, Eduardo Alcântara. O Que é Trânsito. São Paulo: Editora Brasiliense, 2010. VASCONCELOS, Eduardo Alcântara. O Que é Trânsito. São Paulo: Editora Brasiliense, 2010. EDUCAÇÃO PARA O TRÂNSITO NOS PRIMEIROS ANOS DO ENSINO FUNDAMENTAL – Página 463
https://openalex.org/W3126032471	https://www.research.ed.ac.uk/files/196307608/journal.ppat.1009224_1_.pdf	English	null	Reliable, scalable functional genetics in bloodstream-form Trypanosoma congolense in vitro and in vivo	PLOS pathogens	2,021	cc-by	18,068	Edinburgh Research Explorer General rights C i h f h General rights Copyright for the publications made accessible via the Edinburgh Research Explorer is retained by the author(s) and / or other copyright owners and it is a condition of accessing these publications that users recognise and abide by the legal requirements associated with these rights. Published In: PLoS Pathogens Publisher Rights Statement: Publisher Rights Statement: Copyright: © 2021 Awuah-Mensah 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. Digital Object Identifier (DOI): 10.1371/journal.ppat.1009224 Link: Link to publication record in Edinburgh Research Explorer Document Version: Publisher's PDF, also known as Version of record Edinburgh Research Explorer Reliable, scalable functional genetics in bloodstream-form Trypanosoma congolense in vitro and in vivo Citation for published version: Awuah-Mensah, G, McDonald, J, Steketee, P, Autheman, D, Whipple, S, D'Archivio, S, Brandt, C, Clare, S, Harcourt, K, Wright, GJ, Morrison, L, Gadelha, C & Wickstead, B 2021, 'Reliable, scalable functional genetics in bloodstream-form Trypanosoma congolense in vitro and in vivo', PLoS Pathogens. https://doi.org/10.1371/journal.ppat.1009224 Citation for published version: Awuah-Mensah, G, McDonald, J, Steketee, P, Autheman, D, Whipple, S, D'Archivio, S, Brandt, C, Clare, S, Harcourt, K, Wright, GJ, Morrison, L, Gadelha, C & Wickstead, B 2021, 'Reliable, scalable functional genetics in bloodstream-form Trypanosoma congolense in vitro and in vivo', PLoS Pathogens. https://doi.org/10.1371/journal.ppat.1009224 Take down policy Take down policy The University of Edinburgh has made every reasonable effort to ensure that Edinburgh Research Explorer content complies with UK legislation. If you believe that the public display of this file breaches copyright please contact openaccess@ed.ac.uk providing details, and we will remove access to the work immediately and investigate your claim. Download date: 24. Oct. 2024 PLOS PATHOGENS PLOS PATHOGENS Abstract Citation: Awuah-Mensah G, McDonald J, Steketee PC, Autheman D, Whipple S, D’Archivio S, et al. (2021) Reliable, scalable functional genetics in bloodstream-form Trypanosoma congolense in vitro and in vivo. PLoS Pathog 17(1): e1009224. https://doi.org/10.1371/journal.ppat.1009224 Animal African trypanosomiasis (AAT) is a severe, wasting disease of domestic livestock and diverse wildlife species. The disease in cattle kills millions of animals each year and inflicts a major economic cost on agriculture in sub-Saharan Africa. Cattle AAT is caused predominantly by the protozoan parasites Trypanosoma congolense and T. vivax, but labo- ratory research on the pathogenic stages of these organisms is severely inhibited by difficul- ties in making even minor genetic modifications. As a result, many of the important basic questions about the biology of these parasites cannot be addressed. Here we demonstrate that an in vitro culture of the T. congolense genomic reference strain can be modified directly in the bloodstream form reliably and at high efficiency. We describe a parental single marker line that expresses T. congolense-optimized T7 RNA polymerase and Tet repressor and show that minichromosome loci can be used as sites for stable, regulatable transgene expression with low background in non-induced cells. Using these tools, we describe organ- ism-specific constructs for inducible RNA-interference (RNAi) and demonstrate knockdown of multiple essential and non-essential genes. We also show that a minichromosomal site can be exploited to create a stable bloodstream-form line that robustly provides >40,000 independent stable clones per transfection–enabling the production of high-complexity libraries of genome-scale. Finally, we show that modified forms of T. congolense are still infectious, create stable high-bioluminescence lines that can be used in models of AAT, and follow the course of infections in mice by in vivo imaging. These experiments establish a base set of tools to change T. congolense from a technically challenging organism to a rou- tine model for functional genetics and allow us to begin to address some of the fundamental questions about the biology of this important parasite. Editor: Christine Clayton, Heidelberg University Center for Molecular Biology (ZMBH), GERMANY Received: October 5, 2020 Accepted: December 7, 2020 Published: January 22, 2021 Received: October 5, 2020 Accepted: December 7, 2020 Published: January 22, 2021 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. RESEARCH ARTICLE Reliable, scalable functional genetics in bloodstream-form Trypanosoma congolense in vitro and in vivo Georgina Awuah-MensahID1, Jennifer McDonaldID1¤a, Pieter C. SteketeeID2, Delphine Autheman3, Sarah Whipple1, Simon D’ArchivioID1¤b, Cordelia Brandt4, Simon Clare4, Katherine Harcourt4, Gavin J. WrightID3,5, Liam J. MorrisonID2, Catarina GadelhaID1, Bill WicksteadID1 1 School of Life Sciences, University of Nottingham, Nottingham, United Kingdom, 2 The Roslin Institute, Royal (Dick) School of Veterinary Studies, University of Edinburgh, Edinburgh, United Kingdom, 3 Cell Surface Signalling Laboratory, Wellcome Sanger Institute, Cambridge, United Kingdom, 4 Pathogen Support Team, Wellcome Sanger Institute, Cambridge, United Kingdom, 5 Department of Biology, Hull York Medical School, York Biomedical Research Institute, University of York, York, United Kingdom a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 ¤a Current address: Department of Pathology, University of Cambridge, United Kingdom ¤b Current address: Sygnature Discovery, Pennyfoot Street, Nottingham, United Kingdom * catarina.gadelha@nottingham.ac.uk (CG); bill.wickstead@nottingham.ac.uk (BW) PRJEB41578. All other relevant data are within the manuscript and its Supporting Information files. PRJEB41578. All other relevant data are within the manuscript and its Supporting Information files. Author summary Funding: This work was supported by University of Nottingham/Wellcome Trust Institutional Strategic Support Fund 204843/Z/16/Z awards to BW and to CG; Sir Halley Stewart Medical Research Grant R410 to CG and GAM; BBSRC studentship 1364116 to JM; BBSRC grants BB/N007492/1, BB/ S00243X/1, and BBS/E/D/20002173 to LJM and PCS; and Wellcome Trust grant 206194 and BBSRC grant BB/S001980/1 to DA, CB, SC, KH and GJW. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The parasites Trypanosoma congolense and T. vivax are the most significant causative agents of Animal African trypanosomiasis (AAT). AAT kills an estimated 3 million cattle each year and represents a huge financial burden on food production in sub-Saharan Africa. A critical tool for understanding pathogen biology is the ability to make genetic modifications, especially creating specific mutants of target genes that can be used to investigate the locations of gene products, the effects of changes in expression, or conse- quence of complete gene removal. However, work on AAT is severely limited by difficul- ties in making even small genetic modifications and lack of tools for many functional genetics applications. Here, we design, test and validate a set of tools for T. congolense that brings for the first time: routine high-efficiency gene tagging and knockout, regulatable transgene expression from silent loci, a species-specific system for inducible gene knock- down, bioluminescent lines for in vivo disease models, and a means to generate highly complex libraries of mutants that will enable genome-scale work. These data and the tools around them will greatly aid research into AAT and T. congolense biology. Competing interests: The authors have declared that no competing interests exist. Introduction Animal African trypanosomiasis (AAT) is a parasitic disease associated with anaemia, loss of condition and death in sub-Saharan livestock. The impact of the disease on cattle farming is particularly severe. Each year, AAT causes ~3 million cattle deaths, with economic losses in cattle production alone estimated to be US$ 1–1.2 billion [1]. The disease in cattle is caused by trypanosomes of the species Trypanosoma brucei (sub-species of which also cause human dis- ease), T. vivax, and T. congolense. Of these species, T. congolense may cause the majority of dis- ease in sub-Saharan cattle [2,3], as well as making up a substantial proportion of infections [4]. In spite of the importance of T. congolense and T. vivax for AAT, the vast majority of labo- ratory research in African trypanosomes uses T. brucei, which is responsible for only a minor proportion of infections. However, there are known to be significant differences in the biology of these species, including in genomic content [5,6], antigenic variation [6–8], developmental progression [9], and disease symptoms and tropism [10,11], in addition to the differences in host specificity. There are also substantial differences in the differential regulation of genes involved in specific biological processes during infection [12] and between metabolomes (Ste- ketee et al., in preparation). Such differences create a pressing need for species-specific models of AAT that can be used and modified in axenic culture and also transmitted through animals. The focus of molecular research on T. brucei was historically driven by its association with human disease, but recent bias has been heavily influenced by the paucity of means for per- forming functional genetics in T. congolense. This is true for all lifecycle stages of T. congolense, but is particularly acute for bloodstream-form cells, where even generation of stable transfor- mants has been difficult–necessitating laborious strategies for genetic modification involving multiple rounds of transfection and selection in procyclic cells followed by in vitro differentia- tion first to epimastigotes, then to metacyclic cells and finally to bloodstream forms (some- times requiring passage through animals; [13,14]). Where transfections have been performed, they have been through application of plasmids designed and tested in T. brucei with little/no modification. Moreover, there has been no establishment of silent loci suitable for integration of inducible constructs, so although tetracycline-regulated gene expression has been demon- strated in T. Abstract The editorial history of this article is available here: https://doi.org/10.1371/journal.ppat.1009224 Copyright: © 2021 Awuah-Mensah 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: RNA-seq read data are available from the European Nucleotide Archive database (accession number ERP125374) link: https://www.ebi.ac.uk/ena/browser/view/ 1 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Introduction congolense [14], the level of regulation was very poor (~4-fold), precluding their PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 2 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense use for applications such as expression of toxic gene products. Finally, work in T. brucei has been fundamentally changed by the development of methods that enable the generation of many thousands of transfectants from a single transfection [15], enabling the production of high-complexity libraries, such as in the RIT-Seq approach for high-throughput testing of RNA-interference mutants [16]. The lack of such methods in T. congolense severely restricts our ability to probe the mechanisms of pathogenicity, lifecycle control and resistance to drugs, amongst other critical questions. Limitations in functional genetics tools for T. congolense not only restrict experimental work to species that may not be good models for AAT infections, but prevent the addressing of fundamental questions about the biology of these important species–including decoding the genetic basis for the differences seen in their pathology, virulence and resistance to drugs. In addition, since T. congolense is the only African trypanosome for which the whole lifecycle can be recapitulated in vitro by means of chemical cues alone [14], this organism has unique advantages as a model for studying differentiation across the African trypanosome clade. Here, taking as a start point a stable axenic bloodstream culture of the genome reference strain of T. congolense (IL3000), we have designed and implemented tools that allow direct, reliable transgenesis of multiple types in the pathogenic stage of this species. We demonstrate efficient modification of endogenous loci and gene knockout without the need for heterologous DNA modifiers, such as recombinases, restriction endonucleases or Cas9. We describe the design and use of a T. congolense-optimized construct to allow for inducible T7RNAP-driven expres- sion and investigation of loci on the silent minichromosomes as sites for regulatable transgene expression. We also build and test new T. congolense-specific constructs for RNA-interference (RNAi) and describe a freeze/thaw-stable cell line enabling transfection efficiencies compatible with the production of genome-scale high-complexity libraries. Finally, we produce a stable, luminescent line that can be used in models of AAT in mice to increase efficiency of vaccinol- ogy and drug development studies. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Routine high-efficiency transfection in bloodstream-form T. congolense Although transfection of procyclic (insect midgut) forms of T. congolense is generally routine– albeit at low efficiency (~10 independent transfectants from a single transfection of ~108 pro- cyclic cells; [17])–transfection of cultures of bloodstream-form T. congolense has been seen as challenging, with the majority of transfections giving no stable transfectants at all [14]. How- ever, very few loci have been targeted to date, and published transfection attempts in blood- stream forms have used heterologous sequence from the T. brucei tubulin locus to direct integration (~400 bp homology at each end with ~10% mismatch to T. congolense), the impact of which on efficacy is unclear. To test the efficiency of routine transfection directed by homologous targeting sequences in bloodstream-form T. congolense, we took the stable axenic bloodstream culture of the IL3000 strain of T. congolense originally derived by Coustou et al. ([14]; kindly provided by Michael Barrett, University of Glasgow) and targeted modifications to 9 different endogenous loci, either as modifications to the 5’- or 3’-end of the CDS, or as complete knockout of a gene. Unlike previous descriptions using heterologous sequence, 10 constructs with homologous sequence, covering 4 different plasmid architectures (see Materials and Methods), could be transfected at efficiencies more than sufficient for routine modification (100–4000 indepen- dent clones per transfection; Fig 1A). These efficiencies are at or slightly above those we rou- tinely achieve targeting endogenous loci in T. brucei Lister 427 cells (typically ~1000 independent clones). The same preparations of DNA gave very similar numbers of clones PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 3 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Fig 1. High-efficiency modification of endogenous loci in bloodstream-form T. congolense. A) Numbers of independent stable clones produced from electroporation of ~2.5×107 cells with 10–15 μg linearised DNA targeting CDS encoding fluorescent proteins (sfG, superfolder GFP; Y, eYFP; mSt, mStrawberry; mSc, mScarlet-I) to endogenous loci in T. congolense. Modifications are either N-terminal tags (RPA2, CITFA2, Nopp140, Vex1), C-terminal tags (ESPs), or full replacement of a gene (Δtubb::GFP). Selection of integrants with genes conferring resistance to hygromycin (HYG) or G418 (NEO) give 100–5000 independent clones from a single transfection in both bloodstream-form (BSF) and procyclic-form (PCF) cells. Bars: 90% confidence intervals for estimates of clone numbers resulting from individual transfections (reflecting the total number of wells counted). B) Examples of localisation of expressed tagged proteins to expected cellular compartments in bloodstream-form T. Routine high-efficiency transfection in bloodstream-form T. congolense congolense. T. congolense ESP10 homologue localises to the flagellar pocket region, visualised with fluorophore-conjugated tomato lectin (TL), while T. congolense Nopp140 localises to the nucleolus. Representative interphase cells are shown. See S1 Fig for larger fields of cells. https://doi org/10 1371/journal ppat 1009224 g001 Fig 1. High-efficiency modification of endogenous loci in bloodstream-form T. congolense. A) Numbers of independent stable clones produced from electroporation of ~2.5×107 cells with 10–15 μg linearised DNA targeting CDS encoding fluorescent proteins (sfG, superfolder GFP; Y, eYFP; mSt, mStrawberry; mSc, mScarlet-I) to endogenous loci in T. congolense. Modifications are either N-terminal tags (RPA2, CITFA2, Nopp140, Vex1), C-terminal tags (ESPs), or full replacement of a gene (Δtubb::GFP). Selection of integrants with genes conferring resistance to hygromycin (HYG) or G418 (NEO) give 100–5000 independent clones from a single transfection in both bloodstream-form (BSF) and procyclic-form (PCF) cells. Bars: 90% confidence intervals for estimates of clone numbers resulting from individual transfections (reflecting the total number of wells counted). B) Examples of localisation of expressed tagged proteins to expected cellular compartments in bloodstream-form T. congolense. T. congolense ESP10 homologue localises to the flagellar pocket region, visualised with fluorophore-conjugated tomato lectin (TL), while T. congolense Nopp140 localises to the nucleolus. Representative interphase cells are shown. See S1 Fig for larger fields of cells. https://doi.org/10.1371/journal.ppat.1009224.g001 https://doi.org/10.1371/journal.ppat.1009224.g001 when transformed into procyclic-form IL3000 cells (Fig 1A). Moreover, in all clones tested (n = 2–8, depending on transfection), correct integration of DNA could be confirmed by detection of tagged protein or PCR, and sub-cellular localisations were in line with predictions based on putative function (see Fig 1B for examples). These data demonstrate that there is no intrinsic barrier to transgenesis in bloodstream-form T. congolense IL3000 and that specific loci can be selectively targeted with good efficiency. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 A ‘single-marker’ T. congolense line for regulated ectopic expression Although regulatable systems based on IPTG [18], vanillic acid [19] and cumate [20] have been used in T. brucei, the tetracycline-based system is to date the best characterised and best regulated [21]. A bloodstream-form T. congolense cell line expressing bacterial tetracycline- repressor protein (TetR) and T7 bacteriophage RNA polymerase (T7RNAP) has been gener- ated previously by in vitro differentiation of a procyclic IL3000 cells [14], but this line: i) uses 2 selection markers that would otherwise be available for further modification, ii) depends on heterologous sequences for processing of transgenes, and iii) does not include the codon-opti- mization used in the T. brucei SmOx lines [22] that has been shown to increase expression in this organism [23]. As a basis for regulated ectopic expression in T. congolense, we designed and built a single-marker construct (pTcoSM) that modifies the T. congolense tubulin locus by full replacement of a TUBB CDS with codon-optimized T7RNAP and TetR. Similarly to T. bru- cei pSmOx [22], pTcoSM integrates into the T. congolense genome without bacterial DNA to PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 4 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Fig 2. Production of a ‘single marker’ bloodstream-form T. congolense line (TcoSM) for inducible expression of transgenes. A) Schematic representation of genetic modification in TcoSM cells. PAC confers resistance to puromycin. TcoPFR1 and TcoPFR2 are the intergenic regions between copies of the native T. congolense IL3000 PFR1 and PFR2 arrays, respectively. B) Quantitation of mRNA levels for transgenes in TcoSM against common modifications of T. brucei (SMB, Wirtz et al. ‘single-marker bloodstream’ cells [21]; AnTat-SMOx, pleomorphic EATRO1125 bloodstream cells transformed with the Poon et al. pSmOx plasmid [22]; see Figs 8 and 10 for modifications producing TcoSM-LUC2 and TcoTTS cells). Read density is normalised to total reads resulting from TUBA. C) Read density across the modified regions in SMB, AnTat-SMOx and TcoSM cells. Peaks in intergenic regions in untransformed parental lines shows reads mapping to UTRs shared between integrated DNA and endogenous loci. D) Levels of transgenic TetR produced in modified cells assessed by immunoblotting of whole cell lysates with an anti-TetR monoclonal antibody. A section (encompassing the VSG region) of the same membrane stained with Ponceau S is shown as a control for loading. ‘wt’ indicates unmodified parental cells. A ‘single-marker’ T. congolense line for regulated ectopic expression The TetR protein in SMB cells lacks the N-terminal NLS used in AnTat-SMOx and TcoSM, so has a lower molecular mass. Data from one of two biological replicates are shown. See S2 Fig for full view of immunoblot membrane. https://doi.org/10.1371/journal.ppat.1009224.g002 Fig 2. Production of a ‘single marker’ bloodstream-form T. congolense line (TcoSM) for inducible expression of transgenes. A) Schematic representation of genetic modification in TcoSM cells. PAC confers resistance to puromycin. TcoPFR1 and TcoPFR2 are the intergenic regions between copies of the native T. congolense IL3000 PFR1 and PFR2 arrays, respectively. B) Quantitation of mRNA levels for transgenes in TcoSM against common modifications of T. brucei (SMB, Wirtz et al. ‘single-marker bloodstream’ cells [21]; AnTat-SMOx, pleomorphic EATRO1125 bloodstream cells transformed with the Poon et al. pSmOx plasmid [22]; see Figs 8 and 10 for modifications producing TcoSM-LUC2 and TcoTTS cells). Read density is normalised to total of a ‘single marker’ bloodstream-form T. congolense line (TcoSM) for inducible expression of transgenes. A) Schematic Fig 2. Production of a ‘single marker’ bloodstream-form T. congolense line (TcoSM) for inducible expression of transgenes. A) Schematic representation of genetic modification in TcoSM cells. PAC confers resistance to puromycin. TcoPFR1 and TcoPFR2 are the intergenic regions between copies of the native T. congolense IL3000 PFR1 and PFR2 arrays, respectively. B) Quantitation of mRNA levels for transgenes in TcoSM against common modifications of T. brucei (SMB, Wirtz et al. ‘single-marker bloodstream’ cells [21]; AnTat-SMOx, pleomorphic EATRO1125 bloodstream cells transformed with the Poon et al. pSmOx plasmid [22]; see Figs 8 and 10 for modifications producing TcoSM-LUC2 and TcoTTS cells). Read density is normalised to total reads resulting from TUBA. C) Read density across the modified regions in SMB, AnTat-SMOx and TcoSM cells. Peaks in intergenic regions in untransformed parental lines shows reads mapping to UTRs shared between integrated DNA and endogenous loci. D) Levels of transgenic TetR produced in modified cells assessed by immunoblotting of whole cell lysates with an anti-TetR monoclonal antibody. A section (encompassing the VSG region) of the same membrane stained with Ponceau S is shown as a control for loading. ‘wt’ indicates unmodified parental cells. The TetR protein in SMB cells lacks the N-terminal NLS used in AnTat-SMOx and TcoSM, so has a lower molecular mass. Data from one of two biological replicates are shown. See S2 Fig for full view of immunoblot membrane. Fig 2. Production of a ‘single marker’ bloodstream-form T. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Inducible transgene expression from stable T. congolense minichromosomal loci Essential for the development of well-regulated systems for inducible expression in T. brucei has been the identification of loci that are normally transcriptionally silent, but can support high-level transcription on induction. Lack of such identified loci is likely the major cause of the poor regulation (only 2.8- to 4.5-fold) seen in previous attempts at inducible transgene expression in T. congolense, as tetracycline-responsive elements were integrated into the active tubulin locus resulting in substantial expression in the absence of induction [14]. The most common integration target in T. brucei is a spacer region found between arrayed copies of the 18S/28S ribosomal RNA polycistron. Targeting this sequence in the opposite ori- entation to rRNA transcription can provide >1000-fold transgene regulation [21,24], although regulation appears to be highly dependent on the precise spacer at which integration occurs [24–26]. Unlike in T. brucei, the 18S/28S rRNA polycistrons in T. congolense do not appear to be arrayed [5], so there is no site homologous to the T. brucei rRNA spacer. Integration has been successfully directed to a non-homologous sequence upstream of the T. congolense rRNA promoter [27,28], but it is unclear if this site offers the same level of transcriptional regulation as the T. brucei rRNA spacer. As an alternative to sites on the housekeeping chromosomes, loci on the transcriptionally silent minichromosomes have been shown to provide sites for consis- tent, well-regulated expression in T. brucei [26,29]. As minichromosomes are even more abun- dant in T. congolense than in T. brucei [5] and transcriptomic data suggest them to be silent [12], we reasoned they might make good integration sites for regulated expression. To test the utility of T. congolense minichromosomes as sites for regulated expression, we targeted a GFP and hygromycin resistance marker polycistron transcribed by a single tetracy- cline-responsive T7 promoter to 3 sequences present on minichromosomal contigs in the recent long-read assembly of the IL3000 genome [5]: i) the 369 bp repeat found abundantly on minichromosomes [30,31], ii) loci encoding an expanded family of putative DEAH-box RNA helicases (DBRH) related to Tb927.6.740 but present on minichromosomes in T. congolense (e.g. TcIL3000.A.H_000093500), and iii) a silent telomeric VSG gene (TcIL3000.A. H_000093600). 369 bp repeat and silent VSG were targeted with an efficiency indistinguishable from modifications made at pol II transcribed loci (Figs 1A and 3B). These efficiencies are con- sistent with integration into the T. A ‘single-marker’ T. congolense line for regulated ectopic expression congolense line (TcoSM) for inducible expression of transgenes. A) Schematic representation of genetic modification in TcoSM cells. PAC confers resistance to puromycin. TcoPFR1 and TcoPFR2 are the intergenic regions between copies of the native T. congolense IL3000 PFR1 and PFR2 arrays, respectively. B) Quantitation of mRNA levels for transgenes in TcoSM against common modifications of T. brucei (SMB, Wirtz et al. ‘single-marker bloodstream’ cells [21]; AnTat-SMOx, pleomorphic EATRO1125 bloodstream cells transformed with the Poon et al. pSmOx plasmid [22]; see Figs 8 and 10 for modifications producing TcoSM-LUC2 and TcoTTS cells). Read density is normalised to total reads resulting from TUBA. C) Read density across the modified regions in SMB, AnTat-SMOx and TcoSM cells. Peaks in intergenic regions in untransformed parental lines shows reads mapping to UTRs shared between integrated DNA and endogenous loci. D) Levels of transgenic TetR produced in modified cells assessed by immunoblotting of whole cell lysates with an anti-TetR monoclonal antibody. A section (encompassing the VSG region) of the same membrane stained with Ponceau S is shown as a control for loading. ‘wt’ indicates unmodified parental cells. The TetR protein in SMB cells lacks the N-terminal NLS used in AnTat-SMOx and TcoSM, so has a lower molecular mass. Data from one of two biological replicates are shown. See S2 Fig for full view of immunoblot membrane. https://doi.org/10.1371/journal.ppat.1009224.g002 avoid potential mis-integration when using additional constructs containing common plasmid sequence, but in pTcoSM all introduced CDS are flanked by intergenic sequences associated with highly-expressed T. congolense loci (tubulin, PFR1 and PFR2; Fig 2A). Transfection of pTcoSM into bloodstream-form IL3000 cells generated the T. congolense single marker line, TcoSM. Transcriptome analysis showed levels of T7RNAP and TetR mRNA in TcoSM are higher than in T. brucei ‘single-marker bloodstream’ cells [21], which have been very widely used in this organism, and similar to T. brucei lines carrying the pSmOx modification (Fig 2B and 2C), demonstrating both pol II read-through and processing of transgenes by the homologous intergenic sequences as expected. The discontinuation of com- mercial antibodies against T7RNAP precluded the direct testing of levels for this protein, but levels of TetR were also substantially above those in SMB cells (Fig 2D) which, together with the high mRNA levels, suggested levels of protein suitable for regulated expression. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 5 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Inducible transgene expression from stable T. congolense minichromosomal loci congolense genome being independent of target copy num- ber, as is also seen in T. brucei [32], as targets present in the long-read assemblies at 1, 30 and 13 000 copies per haploid genome direct integration with the same efficiency (Fig 3B). In con- trast, transfection targeting DBRH produced ~20-fold fewer clones. This was independent of whether transgene transcription ran in the same or opposite orientation to the DBRH coding direction, showing that the effect was not a product of a specific DNA preparation, but likely an intrinsic property of either the targeting sequence (e.g. due to heterogeneity in sequence across the family) or inability of some of these loci to support transgene expression. All successfully selected transfectants supported T7-driven expression at levels greatly above those from pol II read-through at the tubulin locus (Figs 3C and 4A). Mean protein lev- els were ~20-times higher when T7RNAP transcribed compared to pol II, which is at least as good as those from SMB or 90-13 T. brucei cells (~12 and 15-times, respectively; [21]), con- firming that there is good production of T7RNAP in TcoSM cells. Removal of tetracycline reduced mean fluorescence of transfectants to background levels (>500-fold regulation at the protein level) for all except cells containing transgenes integrated into DBRH in the coding ori- entation, where there was substantial fluorescence in the absence of tetracycline (Figs 4A and S3). Interestingly, this does not appear to be due to direct basal transcription of GFP by 6 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Fig 3. Integration of inducible constructs at T. congolense minichromosomal loci. A) Schematic representation of integrated construct and loci targeted. VSG, silent minichromosomal VSG TcIL3000.A.H_000093600; DBRH, gene encoding putative ATP-dependent DEAH- box RNA helicase (e.g. TcIL3000.A.H_000093500); 369bp repeat, minichromosome-specific satellite repeat [5,31]. Constructs targeting DBRH are identical other than integrating with inducible transcription running in the forward (F) or reverse (R) direction relative to DBRH CDS. B) Transfection efficiency when directing integration to minichromosomal or TUBB loci. ‘Copy number’ refers to number in the TriTrypDB T. congolense IL3000_2019 assemblies (v46) based on the work of [5]. Bars: 90% confidence intervals for estimates of clone numbers resulting from individual transfections. C) Expression of GFP (green) in cells in the presence of tetracycline. Counter-staining of cells with 40,6-diamidino-2-phenylindole (DAPI; magenta) is also shown. https://doi.org/10.1371/journal.ppat.1009224.g003 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Inducible transgene expression from stable T. congolense minichromosomal loci All lines were captured and processed equally, except that GFP signal from Δtubb::GFP has been increased 6-fold for visualisation. Representative images from one clone for each modification are shown. https://doi.org/10.1371/journal.ppat.1009224.g003 NS Tractable functional genetics in bloodstream-form Trypanosoma congolense PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Fig 3. Integration of inducible constructs at T. congolense minichromosomal loci. A) Schematic representation of integrated construct and loci targeted. VSG, silent minichromosomal VSG TcIL3000.A.H_000093600; DBRH, gene encoding putative ATP-dependent DEAH- box RNA helicase (e.g. TcIL3000.A.H_000093500); 369bp repeat, minichromosome-specific satellite repeat [5,31]. Constructs targeting DBRH are identical other than integrating with inducible transcription running in the forward (F) or reverse (R) direction relative to DBRH CDS. B) Transfection efficiency when directing integration to minichromosomal or TUBB loci. ‘Copy number’ refers to number in the TriTrypDB T congolense IL3000 2019 assemblies (v46) based on the work of [5] Bars: 90% confidence intervals for estimates of clone Fig 3. Integration of inducible constructs at T. congolense minichromosomal loci. A) Schematic representation of integrated construct and loci targeted. VSG, silent minichromosomal VSG TcIL3000.A.H_000093600; DBRH, gene encoding putative ATP-dependent DEAH- box RNA helicase (e.g. TcIL3000.A.H_000093500); 369bp repeat, minichromosome-specific satellite repeat [5,31]. Constructs targeting DBRH are identical other than integrating with inducible transcription running in the forward (F) or reverse (R) direction relative to DBRH CDS. B) Transfection efficiency when directing integration to minichromosomal or TUBB loci. ‘Copy number’ refers to number in the TriTrypDB T. congolense IL3000_2019 assemblies (v46) based on the work of [5]. Bars: 90% confidence intervals for estimates of clone numbers resulting from individual transfections. C) Expression of GFP (green) in cells in the presence of tetracycline. Counter-staining of cells with 40,6-diamidino-2-phenylindole (DAPI; magenta) is also shown. All lines were captured and processed equally, except that GFP signal from Δtubb::GFP has been increased 6-fold for visualisation. Representative images from one clone for each modification are shown. https://doi.org/10.1371/journal.ppat.1009224.g003 7 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Tractable functional genetics in bloodstream-form Trypanosoma congolense PLOS PATHOGENS Fig 4. Regulated expression from T. congolense minichromosomal loci. Constructs and target sequences are as in Fig 3. A) Quantification of GFP regulation. Mean fluorescence in 0 (non-induced) or 1 μg ml-1 (induced) tetracycline was calculated by flow cytometry and is shown as levels above mean background (untagged TcoSM cells). For comparison, range of triplicate measurements from 2 independent clones of Δtubb::GFP are also shown. Inducible transgene expression from stable T. congolense minichromosomal loci Clone number is shown below x-axis. ‘pop’ indicates data from uncloned populations. Distributions of fluorescence for each cell line are shown in S3 Fig. B) Change in mean induced GFP levels following maintenance of cells for 21 days (~50 generations) without selection. Individual points represent same clones/populations as shown in Fig 3D and are from a single experimental replicate. Fig 4. Regulated expression from T. congolense minichromosomal loci. Constructs and target sequences are as in Fig 3. A) Quantification of GFP regulation. Mean fluorescence in 0 (non-induced) or 1 μg ml-1 (induced) tetracycline was calculated by flow cytometry and is shown as levels above mean background (untagged TcoSM cells). For comparison, range of triplicate measurements from 2 independent clones of Δtubb::GFP are also shown. Clone number is shown below x-axis. ‘pop’ indicates data from uncloned populations. Distributions of fluorescence for each cell line are shown in S3 Fig. B) Change in mean induced GFP levels following maintenance of cells for 21 days (~50 generations) without selection. Individual points represent same clones/populations as shown in Fig 3D and are from a single experimental replicate. https://doi.org/10.1371/journal.ppat.1009224.g004 endogenous pol II, but to high-level GFP production in a sub-population of cells–presumably due to transient displacement of TetR (S3 Fig). Induced GFP levels for 14 of 15 lines were essentially unchanged by growth for 21 days in the absence of transgene selection (Fig 4B), showing that most T. congolense minichromosomal loci have sufficient stability with respect to mitosis to also be good sites for transgenetic modifi- cations where it may not be possible to constantly maintain selective drug (e.g. in vivo). PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Direct inducible RNA interference in T. congolense bloodstream forms Due to difficulties in performing genetic modifications, no previous work has performed RNAi by direct modification of bloodstream-form T. congolense–instead making modifica- tions (at actively transcribed loci) in procyclic cells and then differentiating to bloodstream forms, either with [13] or without [14] passage through animals. To take advantage of the abil- ity to stably modify minichromosomal loci in bloodstream-form T. congolense, we designed 3 constructs for inducible RNAi that integrate at the silent minichromosomal VSG, TcIL3000.A. H_000093600. These constructs contain a tetracycline-responsive RNAi cassette slightly modi- fied from p2T7Ti [33] that is widely used in T. brucei, a constitutively-expressed selection cas- sette derived from those providing the greatest transfection efficiency for endogenous-locus modification (Fig 1A), and targeting sequence to integrate at the silent VSG without bacterial PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 8 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense DNA (Fig 5A). Three different arrangements of promoter and terminator around the selection cassette were tested in the plasmids: p2T7-TcoV (NEO-expression driven by the T. congolense rRNA promoter), p3T7-TcoV (T7-driven NEO expression), and p3T7-TcoVTT (T7-driven expression terminated at end of cassette). To test the efficiency of knockdown without the confounding influence of gene essentiality, we directed RNAi against a heterologous gene (GFP) integrated into the tubulin locus by inducible expression of ~600 bp GFP double-stranded RNA. Herein, this RNAi is indicated δgfp by analogy to nomenclature used for gene knock-outs. All three T. congolense-specific RNAi plasmids reduced the abundance of GFP on induction (Figs 5B and S3), although knockdown was weaker in lines with pol I-driven selection than those with three T7 promoters (presumably as a result of transcriptional interference). There was no evidence at the protein level for leakiness of RNAi in the absence of tetracycline, but greatest knockdown was only to 30–40% of non-induced levels for the best performing construct (p3T7-TcoV). Similar levels of knockdown were seen when targeting an endogenous gene encoding flagellar pocket pro- tein ESP14 [34]; although in T. brucei, RNAi against ESP14 reduces proteins levels to an esti- mated <10% of parental, knockdown of the T. congolense orthologue (TcIL3000_7_180) is only to ~30% (Fig 5C and 5D). This effect is manifest at the RNA level, as RNAi targetted against a further 5 endogenous genes (PPDK, TcIL3000.A.H_000922100; CHC, TcIL3000.A. Direct inducible RNA interference in T. congolense bloodstream forms H_000768200; FBPase, TcIL3000.A.H_000671500; FH1, TcIL3000.A.H_000909500; PEPCK, TcIL3000.A.H_000300300) by dsRNAs of 493–707 bp consistently reduced mRNA levels, but only to 30–60% of parental or non-induced levels (Fig 5E and 5F). Given that T7-driven transcription in TcoSM cells from minichromosomal loci is at least as strong (in comparison to pol II transcription) as seen in T. brucei bloodstream forms [21], the lower penetrance of RNAi in IL3000 T. congolense does not appear to be a feature of integra- tion site. Indeed, levels of knockdown with p3T7-TcoV are considerably stronger than for inducible constructs integrated at the tubulin locus (without the apparent reduction of levels in non-induced cells), and as strong as for constitutively expressed hairpins introduced into the genome [13]. Instead, it is likely that there is an intrinsically lower effect size resulting from expression of double-strand RNA in T. congolense, at least for the genome reference strain. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Inducible knockdown of essential genes in T. congolense bloodstream forms To test if levels of RNAi knockdown in T. congolense could produce defects expected due to loss of essential proteins, we used p3T7-TcoV to knockdown genes encoding α-tubulin (e.g. TcIL3000.A.H_000560300; δtuba) and clathrin heavy chain (TcIL3000.A.H_000768200; δchc). Knockdown of these genes produce distinctive ‘FAT’ and ‘BigEye’ phenotypes when targeted in T. brucei [35,36]. Consistent with RNAi against non-essential genes, RNAi reduced CHC mRNA levels to ~30% of parental (Fig 5E). Both δtuba and δchc produced the expected mor- phological change in ~50% of cells in independent clonal populations by 48 h post-induction (56% and 54% for δtuba and δchc, respectively; Fig 6A and 6B), similar to proportions of cells with morphological defects seen previously in inducible knockdown of α-tubulin in procyclic- form T. congolense [28]. Again, these data are compatible with an intrinsically lower pene- trance of RNAi in IL3000 compared to T. brucei strains. They also suggest that there may be some attenuation of RNAi effect at later time points–as seen in a reduction of δchc cells with an detectable morphological defect at 72 h versus 48 h (Fig 6B; mean 35% and 54%, respec- tively) and a slight increase in mRNA levels for all genes followed over 72 h induction (Fig 5F). Induction of RNAi against either TUBA or CHC caused a growth defect in all clones ana- lysed from 24 h post-induction (Fig 6C). However, even though defects were still apparent at later time points, in neither case did populations stop dividing completely–doubling every PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 9 / 29 Tractable functional genetics in bloodstream-form Trypanosoma congolense PLOS PATHOGENS Fig 5. Inducible RNA-interference from a silent minichromosomal locus in bloodstream-form T. congolense. Numbers in parentheses following knockdown name refer to the total length of the gene fragment inserted between promoters in the RNAi construct. A) Schematic representation of RNAi construct architectures and integration. B) Quantification of knockdown of GFP expressed from the tubulin locus by RNAi targeting GFP (δgfp). Numbers on the x-axis represent independent clones. C) Localisation of the T. congolense orthologue of ESP14 (TcIL3000_7_180) by endogenous-locus tagging. Counter-staining of cells with DAPI (magenta) is also shown. D) Knockdown of endogenous locus-tagged ESP14 in bloodstream-form T. brucei and two independent clones of T. congolense. E) Quantification of target transcript levels following RNAi targeting pyruvate phosphate dikinase (TcIL3000.A.H_000922100, δppdk) or clathrin heavy chain (δchc) genes. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Inducible knockdown of essential genes in T. congolense bloodstream forms Mean and SEM from 3 (δppdk) or 4 (δchc) biological replicates are shown along with individual measurements (dots). F) Time courses for target transcript levels following RNAi targeting 4 genes with putative metabolic roles. Bars: SEM from 3 biological replicates. https://doi.org/10.1371/journal.ppat.1009224.g005 Fig 5. Inducible RNA-interference from a silent minichromosomal locus in bloodstream-form T. congolense. Numbers in parentheses following knockdown name refer to the total length of the gene fragment inserted between promoters in the RNAi construct. A) Schematic representation of RNAi construct architectures and integration. B) Quantification of knockdown of GFP expressed from the tubulin locus by RNAi targeting GFP (δgfp). Numbers on the x-axis represent independent clones. C) Localisation of the T. congolense orthologue of ESP14 (TcIL3000_7_180) by endogenous-locus tagging. Counter-staining of cells with DAPI (magenta) is also shown. D) Knockdown of endogenous locus-tagged ESP14 in bloodstream-form T. brucei and two independent clones of T. congolense. E) Quantification of target transcript levels following RNAi targeting pyruvate phosphate dikinase (TcIL3000.A.H_000922100, δppdk) or clathrin heavy chain (δchc) genes. Mean and SEM from 3 (δppdk) or 4 (δchc) biological replicates are shown along with individual measurements (dots). F) Time courses for target transcript levels following RNAi targeting 4 genes with putative metabolic roles. Bars: SEM from 3 biological replicates. 20.0 ± 1.8 h and 19.3 ± 2.0 h after 24 h induction for δtuba and δchc, respectively, compared to 10.3 ± 0.3 h for non-induced cells. Cells not producing a detectable morphological defect early in induction are not due to a substantial proportion of cells refractory to RNAi in the non- 20.0 ± 1.8 h and 19.3 ± 2.0 h after 24 h induction for δtuba and δchc, respectively, compared to 10.3 ± 0.3 h for non-induced cells. Cells not producing a detectable morphological defect early in induction are not due to a substantial proportion of cells refractory to RNAi in the non- PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 10 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Fig 6. RNAi targeting essential genes in T. congolense causes growth defects and morphological changes similar to those seen in T. brucei. A) Phenotypic changes seen following knockdown of α-tubulin (δtuba) or clathrin heavy chain (δchc) in T. congolense. Images from 2 technical replicates of induction of a single clone (Clone 1) are shown. B) Quantification of gross morphological changes. Representative data from 1 of 2 biological repeats are shown. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Inducible knockdown of essential genes in T. congolense bloodstream forms Number below bars are time post-induction (h); numbers above are cells analysed; P, parental (no RNAi construct). C) Growth of cells containing RNAi constructs in 0 (non-induced) or 1 μg ml-1 (induced) tetracycline. htt //d i /10 1371/j l t 1009224 006 Fig 6. RNAi targeting essential genes in T. congolense causes growth defects and morphological changes similar to those seen in T. brucei. A) Phenotypic changes seen following knockdown of α-tubulin (δtuba) or clathrin heavy chain (δchc) in T. congolense. Images from 2 technical replicates of induction of a single clone (Clone 1) are shown. B) Quantification of gross morphological changes. Representative data from 1 of 2 biological repeats are shown. Numbers below bars are time post-induction (h); numbers above are cells analysed; P, parental (no RNAi construct). C) Growth of cells containing RNAi constructs in 0 (non-induced) or 1 μg ml-1 (induced) tetracycline. https://doi org/10 1371/journal ppat 1009224 g006 https://doi.org/10.1371/journal.ppat.1009224.g006 https://doi.org/10.1371/journal.ppat.1009224.g006 induced population, as they produce progeny that can be affected by knockdown. Instead, there appears to be a relatively stable proportion of the population where levels fall below a threshold necessary for apparent defect, at least in the short-term following induction. Importantly, this proportion was unchanged by freeze-thawing of clones, meaning that RNAi lines can be stored and re-analysed in future experiments (S4 Fig). However, there was some attenuation of RNAi defects after growth in culture without induction for 8 weeks (>120 generations; S4 Fig), PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 11 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense suggesting that there is some long-term pressure imposed by the presence of the RNAi con- struct–even without induction. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 https://doi.org/10.1371/journal.ppat.1009224.g007 Inducible double-strand break creation in T. congolense minichromosomes A key technical development in the evolution of T. brucei genetic modification was the demon- stration that double-strand breaks at specific loci could hugely increase the efficiency and spec- ificity of transgene integration [15]. This is a critical component of high-complexity library approaches such as genome-scale RNA-interference target sequencing (RIT-Seq; [16]). The technology as currently applied in T. brucei involves introducing an I-SceI meganuclease rec- ognition sequence at the target locus followed by transient expression of I-SceI to induce cut- ting. To implement a similar approach at T. congolense minichromosomes, we designed and built constructs to introduce an inducible, self-excising gene encoding I-SceI (with a monopar- tite nuclear localisation sequence from SV40 Large T-antigen) at the silent minichromosomal VSG (Fig 7A). Fig 7. Production of inducible, self-excising locus on T. congolense minichromosome. A) Schematic representation of construct architectures and integration. B) Experimental set-up for estimation of locus loss. Tet, tetracycline; Hyg, hygromycin; P(TbrRNA), T. brucei rRNA gene promoter. C) Number of positive wells on input plate and replicas in the absence (-Hyg) or presence (+Hyg) of hygromycin. D) Inferred proportion of hygromycin-sensitive cells following induction. Three independent clones are shown for pTcoScetiT7 (1–3). Bars: 90% confidence intervals for estimates from individual biological replicates. https://doi.org/10.1371/journal.ppat.1009224.g007 Fig 7. Production of inducible, self-excising locus on T. congolense minichromosome. A) Schematic representation of construct architectures and integration. B) Experimental set-up for estimation of locus loss. Tet, tetracycline; Hyg, hygromycin; P(TbrRNA), T. brucei rRNA gene promoter. C) Number of positive wells on input plate and replicas in the absence (-Hyg) or presence (+Hyg) of hygromycin. D) Inferred proportion of hygromycin-sensitive cells following induction. Three independent clones are shown for pTcoScetiT7 (1–3). Bars: 90% confidence intervals for estimates from individual biological replicates. Fig 7. Production of inducible, self-excising locus on T. congolense minichromosome. A) Schematic representation of construct architectures and integration. B) Experimental set-up for estimation of locus loss. Tet, tetracycline; Hyg, hygromycin; P(TbrRNA), T. brucei rRNA gene promoter. C) Number of positive wells on input plate and replicas in the absence (-Hyg) or presence (+Hyg) of hygromycin. D) Inferred proportion of hygromycin-sensitive cells following induction. Three independent clones are shown for pTcoScetiT7 (1–3). Bars: 90% confidence intervals for estimates from individual biological replicates. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 12 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense In T. Inducible double-strand break creation in T. congolense minichromosomes brucei, the published approach targeting a rDNA spacer has the disadvantage that leakiness in I-SceI expression means that cells containing the I-SceI construct (known as Sce cells) are unstable on freeze-thaw and also during medium-term culture [37]. Although our data suggest levels of transgene expression from the targeted locus are very low in the absence of induction (Fig 4), it is not known what level of protein is necessary to produce cutting in a significant proportion of cells. To accommodate differences in I-SceI threshold, constructs were designed with the gene under the control of 2 alternative promoters (both controlled by two TetO elements): a T7 promoter (pTcoScetiT7) expected to produce very strong expression, and a heterologous promoter for the rRNA genes from T. brucei (pTcoSceTbrRNA). The latter is highly dissimilar to the endogenous rRNA promoter and our data suggest it results in very low transgene expression when used in T. congolense, but is known to be well regulated in T. brucei [25]. Cutting of I-SceI in cells transfected with either pTcoScetiT7 or pTcoSceTbrRNA would cause loss of telomere-proximal SceI and HYG (Fig 7A). To test for the efficiency of double-strand break induction, we grew cells for 48 h in the presence or absence of tetracycline, plated with- out selection and then applied drug selection to test for loss of HYG (Fig 7B). Cells containing I-SceI controlled by heterologous TbrRNA promoter presented no evidence of HYG loss in the absence of tetracycline but also few hygromycin-sensitive wells on induction (0/71 and 6/76, respectively, equating to a probability of loss in individual cells of 0.08 and 0.1–0.3; p = 0.03, Fisher’s exact test; Fig 7C). In contrast, induction of T7-driven I-SceI expression resulted in production of hygromycin-sensitive wells at a rate equivalent to 65–92% of cells losing HYG across 3 independent clones (Fig 7C and 7D). Although the number of sensitive wells pro- duced by non-induced lines was also significantly higher with this strong promoter compared to heterologous rRNA promoter (11/196 and 0/71, respectively; p = 0.04), the estimated loss for even the highest clone is equivalent to a loss in ~3% of cells per generation (17% over 48 h), which is sufficiently low for short-term growth without selection. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Inducible double-strand break creation greatly increases transfection efficiency To test if double-strand break induction increases transfection efficiency in T. congolense, we took TcoSM cells transfected with pTcoScetiT7 or pTcoSceTbrRNA, induced I-SceI expression for 16 h and then introduced a T. congolense RNAi plasmid that targets the Sce-modified locus (p3T7-TcoV containing a fragment of TUBA; Fig 8A). I-SceI induction time was chosen as being the approximate peak in specific ssDNA production seen in T. brucei [15] and also for simple convenience in experimental set-up. It is possible that other lengths of induction would further change efficiency, but this was not investigated here. Induction of I-SceI controlled by heterologous rRNA promoter had no impact on transfec- tion efficiency, which was indistinguishable from transfection of construct into TcoSM cells (~250 independent clones per transfection; Fig 8B). Induction of T7-driven expression, how- ever, increased efficiencies to >10,000 clones per transfection (~50-fold above efficiency with- out cutting). For 2 of 3 independent clones there was evidence of considerable cutting even without tetracycline (Fig 8B). Given the very low background transcription for tetracycline- inducible transgenes at this minichromosomal locus, this seems anomalous–particularly as there was no corresponding loss of HYG in these clones during normal growth (Fig 7D). Our current hypothesis is that this reflects stochastic displacement of TetR in a proportion of cells as a result of electroporation stress. This would explain the high variability in the effect (100-fold between experiments and clones; Fig 8B). Fortunately, it is of no practical disadvan- tage, as induced efficiencies are consistent and high across clones. Significantly, all clones were (100-fold between experiments and clones; Fig 8B). Fortunately, it is of no practical disadvan- tage, as induced efficiencies are consistent and high across clones. Significantly, all clones were PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 13 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Fig 8. Increased efficiency of integration at induced double-strand break locus. A) Schematic representation of integration events and resultant unique amplicon sizes. The targetted locus, TcIL3000.A.H_000093600, is first modified by integration of pTcoScetiT7 or pTcoSceTbrDNA and may be subsequently replaced by p3T7-TcoV-TUBA, which has the same targeting sequences. B) Efficiency of stable transfection following induction of double-strand break in 3 independent clones carrying self-excising locus at minichromosome VSG. Independent replicates of transfection are shown for each clone. Bars: 90% confidence intervals for estimates in individual transfections. Clone 1 was selected as the progenitor of the TcoTTS cell line. https://doi.org/10.1371/journal.ppat.1009224.g008 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Inducible double-strand break creation greatly increases transfection efficiency A single purified preparation of p3T7-TcoV-TUBA was used for all transfections. Bars: 90% confidence intervals for estimates in individual transfections. https://doi.org/10.1371/journal.ppat.1009224.g009 https://doi.org/10.1371/journal.ppat.1009224.g009 also stable to freeze-thawing of cultures (Fig 8B) and could be maintained in culture for at least 3 weeks without loss of high-efficiency. We selected the clone with greatest fold-change in transfection efficiency as the basis for a stable library-recipient cell line. This cell line, based on IL3000 TcoSM cells, is named TcoTTS for the presence of T7RNAP, TetR and inducible I-SceI. TcoTTS cells are compatible with both our RNAi and over-expression constructs targeting VSG TcIL3000.A.H_000093600, and are suitable for generation of high-complexity libraries. Optimization of transfection parame- ters in TcoTTS cells following frozen storage showed that this line can routinely and consis- tently produce >40,000 independent clones per transfection (using 20 μg input DNA and 5x107 induced cells; Fig 9). Although the targeted VSG (TcIL3000.A.H_000093600) is present in the long-read IL3000 assemblies as a single copy [5], analysis of integration events shows at least 3 copies in IL3000 cells, since TcoTTS cells contain both modified and unmodified TcIL3000.A.H_000093600 loci and introduction of p3T7-TcoV without induction of I-SceI can either replace pTcoScetiT7 (e.g. clone 1 in Fig 8C) or modify a VSG copy (e.g. clone 3 in Fig 8C) and still leave 1 unmod- ified copy. Copies of TcIL3000.A.H_000093600 likely represent multiple copies of the mini- chromosome itself that could not be distinguished during assembly. As expected, induction of I-SceI pushes all integration events to the single modified locus (Fig 8C), meaning that all inte- gration events in induced TcoTTS are modified at precisely the same genomic location. Importantly, phenotypes associated with δtuba could be induced equally in cell lines derived by I-SceI induction as those derived without induction (Figs 6B and 8D), showing that the pro- cess does not interfere with generating RNAi against essential genes. Inducible double-strand break creation greatly increases transfection efficiency Freeze/thaw indicates samples that have been stored in liquid nitrogen and defrosted prior to transfection. C) Multiplex PCR showing modification of minichromosome VSG loci following introduction of a construct targeting the VSG (p3T7-TcoV-TUBA) after 0 h (-Tet) or 16 h (+Tet) induction of double-strand break at the target locus in TcoTTS cells. Distinguishable amplicons are produced from unmodified loci (wt), and loci modified by either pTcoScetiT7/pTcoSceTbrDNA (SceI) or p3T7-TcoV-TUBA (δtuba). D) Production of phenotypic changes due to knockdown of α-tubulin (δtuba) in 2 independent clones produced from TcoTTS cells either without (-) or with (+) induction of double-strand break during transfection. htt //d i /10 1371/j l t 1009224 008 Fig 8. Increased efficiency of integration at induced double-strand break locus. A) Schematic representation of integration events and resultant unique amplicon sizes. The targetted locus, TcIL3000.A.H_000093600, is first modified by integration of pTcoScetiT7 or pTcoSceTbrDNA and may be subsequently replaced by p3T7-TcoV-TUBA, which has the same targeting sequences. B) Efficiency of stable transfection following induction of double-strand break in 3 independent clones carrying self-excising locus at minichromosome VSG. Independent replicates of transfection are shown for each clone. Bars: 90% confidence intervals for estimates in individual transfections. Clone 1 was selected as the progenitor of the TcoTTS cell line. Freeze/thaw indicates samples that have been stored in liquid nitrogen and defrosted prior to transfection. C) Multiplex PCR showing modification of minichromosome VSG loci following introduction of a construct targeting the VSG (p3T7-TcoV-TUBA) after 0 h (-Tet) or 16 h (+Tet) induction of double-strand break at the target locus in TcoTTS cells. Distinguishable amplicons are produced from unmodified loci (wt), and loci modified by either pTcoScetiT7/pTcoSceTbrDNA (SceI) or p3T7-TcoV-TUBA (δtuba). D) Production of phenotypic changes due to knockdown of α-tubulin (δtuba) in 2 independent clones produced from TcoTTS cells either without (-) or with (+) induction of double-strand break during transfection. Fig 8. Increased efficiency of integration at induced double-strand break locus. A) Schematic representation of integrati https://doi.org/10.1371/journal.ppat.1009224.g008 14 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Fig 9. Optimization of transfection efficiency in TcoTTS cells following induction of double-strand break. A single purified preparation of p3T7-TcoV-TUBA was used for all transfections. Bars: 90% confidence intervals for estimates in individual transfections. https://doi.org/10.1371/journal.ppat.1009224.g009 Fig 9. Optimization of transfection efficiency in TcoTTS cells following induction of double-strand break. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Genetic modified IL3000 cells for in vivo disease models In order to use directly modified bloodstream-form T. congolense in animal models of disease, it is necessary that cells following selection still establish infections with reasonable dynamics. It would also be beneficial to generate forms of IL3000 that can be followed through an infec- tion by in vivo imaging. To this end, we designed and built constructs to stably express firefly luciferase using an endogenous rRNA promoter (Fig 10A), in a manner analogous to those PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 15 / 29 Tractable functional genetics in bloodstream-form Trypanosoma congolense PLOS PATHOGENS Fig 10. Transgenic T. congolense bloodstream lines for visualisation of in vivo infection. A) Schematic representation of construct architectures and integration. Constructs differ only in T. congolense intergenic region between CDSs in transgenic bicistron. B) Luminescence in wild-type (wt) or TcoSM cells stably transfected with pTcoLUC1 or pTcoLUC2. Technical duplicates from 3 independent clones of each line are shown along with mean and SEM across clones. C) Linearity of luminescence in the highest-expressing wild-type (wt) or TcoSM LUC1/2 clones. D) Infection dynamics of T. congolense LUC2 cells in a mouse model of AAT monitored by bioluminescence and parasitaemia in peripheral blood (5 animals for each inoculum). Inoculum refers to total dose of parasites used to infect animal at time 0. E) In vivo imaging of infection density in mice inoculated with T. congolense LUC2 cells. Numbers below images of individual mice are estimates of the peripheral parasitaemia (cell ml-1) made by haemocytometry. Data from 2 representative animals for each inoculum are shown. https://doi.org/10.1371/journal.ppat.1009224.g010 Fig 10. Transgenic T. congolense bloodstream lines for visualisation of in vivo infection. A) Schematic representation of construct architectures and integration. Constructs differ only in T. congolense intergenic region between CDSs in transgenic bicistron. B) Luminescence in wild-type (wt) or TcoSM cells stably transfected with pTcoLUC1 or pTcoLUC2. Technical duplicates from 3 independent clones of each line are shown along with mean and SEM across clones. C) Linearity of luminescence in the highest-expressing wild-type (wt) or TcoSM LUC1/2 clones. D) Infection dynamics of T. congolense LUC2 cells in a mouse model of AAT monitored by bioluminescence and parasitaemia in peripheral blood (5 animals for each inoculum). Inoculum refers to total dose of parasites used to infect animal at time 0. E) In vivo imaging of infection density in mice inoculated with T. congolense LUC2 cells. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Genetic modified IL3000 cells for in vivo disease models Numbers below images of individual mice are estimates of the peripheral parasitaemia (cell ml-1) made by haemocytometry. Data from 2 representative animals for each inoculum are shown. htt //d i /10 1371/j l t 1009224 010 16 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense used in T. brucei [38] and T. vivax [39]. Integration is duplicative, such that transcription of the downstream RRNA polycistron is not disrupted and the modifications carry no bacterial DNA to the integration site. Two alternative endogenous intergenic regions from highly expressed T. congolense genes (between the genes encoding L5 and L15 ribosomal proteins, and TUBA-TUBB) were placed at the 3’ end of the luciferase CDS. Integration of either con- struct in IL3000 wild-type or TcoSM cells produced detectable luciferase at levels >1000-fold above background (Fig 10B), but levels associated with the tubulin intergenic (LUC2) were sig- nificantly higher (p < 0.001; Student’s t-test). Luminescence in vitro is linear across >5-orders of magnitude and as few as 10 cells in total are detectable above background (Fig 10C). g g g To create a model of acute trypanosomiasis, 106 modified LUC2 cells were injected into BALB/c mice via the intraperitoneal route. First parasitaemic peak was seen at day 4 post- infection, demonstrating that modified cells are still infectious directly from culture. Cells from the second parasitaemic wave were further passaged twice in BALB/c mice, each to the first parasitaemic peak. Luciferase expression was maintained across this in vivo growth in the absence of selection and infections with this line result in a predictable dose-dependent time to first peak parasitaemia across at least 2-orders of magnitude (Fig 10D). Infection dynamics are also very similar in male and female mice and in an albino strain derived from C57BL/6 (S5 Fig). Monitoring parasites by in vivo bioluminescence imaging shows linear doubling of para- site numbers during exponential growth proportional to the initial dose and with an estimated in vivo doubling time of 8.4 h. However, counts of parasites in peripheral blood appear to sub- stantially underestimate total T. congolense abundance at low parasitaemia (Fig 10D)–probably due to the vascular adherence of parasites that is a feature of T. congolense infections [40]. Dis- tribution of T. Genetic modified IL3000 cells for in vivo disease models congolense in this acute model extends widely through the animal (Fig 10E) and there is no strong evidence for tissue-specific accumulation as is seen at later stages of infection with T. brucei [38] or T. vivax [41]. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Transfection of bloodstream-form IL3000 cells Here we have described, for the first time routine, direct genetic modification of bloodstream- form T. congolense IL3000. Transfection efficiencies when targeting endogenous loci are good (100–4000 independent clones per transfection)–at least as high as for Lister 427 bloodstream- form cells, which are the most commonly modified T. brucei strain–and do not require heter- ologous DNA modifiers, such as recombinases, restriction endonucleases or Cas9. While this is a great benefit for T. congolense research, it does raise the question as to why previous attempts at stable transfection proved so challenging. The IL3000 line used herein is a descendant of the original in vitro bloodstream-form line [14] and is still infective in animals. Our transfection procedure is covered in detail in Materi- als and Methods, but is also only modestly different from standard short-pulse electroporation (‘nucleofection’) procedures used previously–most of which gave no stable transfectants. Although not systematically addressed, our experience suggests that bloodstream-form T. con- golense cells are much more sensitive to bacterial endotoxin than T. brucei and high-purity DNA appears to be an absolute requirement for high efficiency stable transfection. We only perform transfection with DNA purified by anion exchange and transfection buffer made from high-purity stocks. Notwithstanding influence from the above, the major determinant of previous difficulties in genetic modification of T. congolense is likely to be simply that the constructs were not designed for this organism, but were heterologous application of T. brucei constructs. Many genomic elements definitely cannot be transferred in this manner: the T. brucei procyclin PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 17 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense promoter is not functional in T. congolense [27], and our data suggest that the T. brucei rRNA promoter has very limited activity. Moreover, while it is clear that at least some T. brucei inter- genic regions are processed sufficiently in T. congolense to produce detectable levels of protein [14,17,28], the efficiency of processing compared to native sequences is unknown and likely to be much lower in many cases. More significantly, the majority of reported transfection attempts have also used heterologous T. brucei sequence for targeting integration [13,14,17]. It is unknown how transfection efficiency scales with identity of targeting sequence in T. congo- lense, but in T. brucei comparable levels of mismatch (10% for TUBB) would cause a 50-fold decrease in efficiency [42]. RNAi effectiveness in T. congolense IL3000 Tetracycline-regulated T7RNAP-driven transcription from minichromosomal loci in TcoSM cells is at least as strong with respect to pol II transcription as in T. brucei SMB and 90-13 lines [21], but T7-based RNAi in T. congolense had consistently lower effect size compared to typical T. brucei knockdowns. To our knowledge, there are no published systematic tests of the effect of dsRNA length on knock-down effectiveness in T. brucei, although fragments as small as 50– 60 bp have been reported to reduce endogenous mRNA levels when expressed in cells [36,43]. In the absence of strong evidence for length effects, most knockdowns using head-to-head pro- moter arrangements in T. brucei employ fragments in the range 250–800 bp. In our T. congo- lense system, gene fragments in the range ~500–800 bp reduced mRNA and protein levels to ~30–50% of parental levels with no evidence for length effects. It is possible that much longer dsRNA could increase RNAi effectiveness in T. congolense, but current data suggest this is unlikely. Moreover, similar levels of knockdown were seen previously when expressing RNA hairpins from constitutively active loci [13] and our attempts to improve RNAi effect size using inducible single-promoter hairpins gave no greater phenotype penetrance than head-to- head constructs. These data imply that T. congolense and/or the IL3000 strain have an intrinsi- cally lower RNAi effect size than commonly tested T. brucei strains. This does not appear to be due to differences in expression of the core RNAi machinery, as levels of mRNA encoding the trypanosome homologues of Argonaute [44,45], Dicer [46,47] and the RNA Interference Fac- tor 4 (RIF4) are slightly higher in bloodstream-form T. congolense compared to T. brucei (Fig 11A). The encoded proteins are also well conserved within African trypanosomes and no more divergent in T. congolense than would be expected from lineage history (Fig 11B). Importantly, although RNAi effectiveness–both in terms of proportion of cells exhibiting a defect and the defect severity–is lower than in T. brucei, knockdown is sufficient to produce gross morphological defects in ~50% of cells when targeting two essential genes. This effect on only part of a clonal population does not appear to be due to a high proportion of refractory cells. It appears instead to be a threshold effect, and progeny of cells that had no detectable morphological change can go on to develop defects. Transfection of bloodstream-form IL3000 cells This may not explain all the improvement in efficiency we have achieved (perhaps as much as 1000-fold), but likely provides a substantial contribution. Impor- tantly, this implies that, although the work here is focused on bloodstream-form cells, all the tools developed will also greatly facilitate work using other replicative stages. None of our T. congolense constructs contain stage-specific elements and we have shown that procyclic-form cells can be modified with similar ease to bloodstream forms (Fig 1)–with approximately 500-fold greater efficiency than previous studies (>50 and ~0.1 independent clones per million cells, respectively; [17]). promoter is not functional in T. congolense [27], and our data suggest that the T. brucei rRNA promoter has very limited activity. Moreover, while it is clear that at least some T. brucei inter- genic regions are processed sufficiently in T. congolense to produce detectable levels of protein [14,17,28], the efficiency of processing compared to native sequences is unknown and likely to be much lower in many cases. More significantly, the majority of reported transfection attempts have also used heterologous T. brucei sequence for targeting integration [13,14,17]. It is unknown how transfection efficiency scales with identity of targeting sequence in T. congo- lense, but in T. brucei comparable levels of mismatch (10% for TUBB) would cause a 50-fold decrease in efficiency [42]. This may not explain all the improvement in efficiency we have achieved (perhaps as much as 1000-fold), but likely provides a substantial contribution. Impor- tantly this implies that although the work here is focused on bloodstream form cells all the promoter is not functional in T. congolense [27], and our data suggest that the T. brucei rRNA promoter has very limited activity. Moreover, while it is clear that at least some T. brucei inter- genic regions are processed sufficiently in T. congolense to produce detectable levels of protein [14,17,28], the efficiency of processing compared to native sequences is unknown and likely to be much lower in many cases. More significantly, the majority of reported transfection PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 RNAi effectiveness in T. congolense IL3000 Sequence for both Lister 427 (Tb427) and TREU927/4 (Tb927) strains are shown for T. brucei. T. congolense gene models suffixed “_full” have been corrected to start at the unannotated in-frame ATG, which in all cases produces an uninterrupted CDS >99% identical to either TcIL3000_10_9140 or TcIL3000_10_9150 in the short-read assemblies. Phylogenies were inferred from aligned protein sequences using FastTree 2.1.9 ([60]; LG substitution matrix, gamma-distributed variation in substitution rate approximated to 12 discrete categories). Node support is derived from 100 bootstrap replicates. https://doi.org/10.1371/journal.ppat.1009224.g011 https://doi.org/10.1371/journal.ppat.1009224.g011 induced changes are readily amenable to identification by high-throughput methods relying on differential cell growth such as RIT-Seq (for example, the fold-change in TUBA fragments by day 6 of RIT-Seq in T. brucei is also ~20-fold; [16]). This is particularly relevant given that we have also developed here a system for production of high-complexity libraries in T. congo- lense analogous to Sce cells used in T. brucei [15]. RNAi effectiveness in T. congolense IL3000 As a result, the corresponding halving of the population growth rate after 24 h RNAi induction is persistent, at least over a few days, and a large difference in total cell numbers can thus accumulate (~20-fold difference by 5 days post-induction). This means that, even in the absence of dramatic cell death, such RNAi- PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 18 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Fig 11. Expression and conservation of components of the RNAi machinery in T. brucei and T. congolense. A) Total mRNA levels (fragments per million; FPM) for AGO1, DCL1, DCL2 and RIF4 homologues in bloodstream-form cells. For comparison, the distribution of FPM for all detected genes is also shown (grey). RNA levels for the 3 near- identical AGO1 genes in T. congolense (>99% identical at the nucleotide level and indistinguishable across most of their length in RNA-Seq) have been summed to reflect the total levels of mRNA for the protein. B) Maximum likelihood phylogenies for homologues of components in Leishmania braziliensis (LbrM), Trypanosoma grayi (DQ4), T. vivax (TvY486), T. congolense (TcIL3000) and T. brucei. Sequence for both Lister 427 (Tb427) and TREU927/4 (Tb927) strains are shown for T. brucei. T. congolense gene models suffixed “_full” have been corrected to start at the unannotated in-frame ATG, which in all cases produces an uninterrupted CDS >99% identical to either TcIL3000_10_9140 or TcIL3000_10_9150 in the short-read assemblies. Phylogenies were inferred from aligned protein sequences using FastTree 2.1.9 ([60]; LG substitution matrix, gamma-distributed variation in substitution rate approximated to 12 discrete categories). Node support is derived from 100 bootstrap replicates. https://doi.org/10.1371/journal.ppat.1009224.g011 Fig 11. Expression and conservation of components of the RNAi machinery in T. brucei and T. congolense. A) Fig 11. Expression and conservation of components of the RNAi machinery in T. brucei and T. congolense. A) Total mRNA levels (fragments per million; FPM) for AGO1, DCL1, DCL2 and RIF4 homologues in bloodstream-form cells. For comparison, the distribution of FPM for all detected genes is also shown (grey). RNA levels for the 3 near- identical AGO1 genes in T. congolense (>99% identical at the nucleotide level and indistinguishable across most of their length in RNA-Seq) have been summed to reflect the total levels of mRNA for the protein. B) Maximum likelihood phylogenies for homologues of components in Leishmania braziliensis (LbrM), Trypanosoma grayi (DQ4), T. vivax (TvY486), T. congolense (TcIL3000) and T. brucei. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 A toolkit for functional genetics in Trypanosoma congolense Work in T. brucei (and then only in a limited set of strains) dominates functional genetics in African trypanosomes. This is in spite of T. congolense and T. vivax being the more important parasites for AAT and it being known that the species differ in significant aspects of their biol- ogy. The dominance of T. brucei work is due in large part to the wealth of tools for functional genetics built up across more than 3 decades of work. Here we have attempted to substantially accelerate work in T. congolense by developing a basic set of tools and methods that open the organism up to functional genetics–designing each tool from first principles to suit the T. con- golense genome. With homologous targeting sequences, we have shown that modification of endogenous loci for tagging or gene knockout is routine in T. congolense, even for bloodstream stages. To date, we have only used 3 selective drugs (hygromycin, G418 and puromycin), but bleomycin has been used elsewhere [14,28] and additional selective drugs used in T. brucei are highly PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 19 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense likely to be usable. Development of a well-regulated system for inducible ectopic (over-)expres- sion was a key early tool in T. brucei. We have described a T. congolense single marker line (TcoSM) that expresses codon-optimized TetR and T7RNAP using T. congolense-specific pro- cessing sequences. Combining this line with the use of minichromosomal loci as stable, silent targets for inducible transgenes brings levels of expression in the absence of induction from only ~4-fold below induced as previously achieved [14] to below the level of detection (>700-fold regulation), creating a system suitable for toxic gene expression or conditional knockouts. We have also designed a T. congolense-specific RNAi construct, p3T7-TcoV, that integrates at a silent minichromosomal VSG and validated it against multiple genes. This is by no means the first application of RNAi in T. congolense, which goes back nearly two decades [28], but should greatly increase the utility of a technology that has been so important in analysis of gene function in T. brucei. To date, we have seen no evidence of significant clonal variation in knockdown at the mRNA, protein or phenotypic levels, suggesting that we can anticipate good reproducibility of RNAi using p3T7-TcoV. A toolkit for functional genetics in Trypanosoma congolense In addition to this, genetically modified IL3000 bloodstream-form cells are still infectious and the production of a stable, highly-lumi- nescent line provides a tool for vaccinology or other work requiring in vivo disease models. Since modified T. congolense minichromosomes have good stability in the absence of selection, this system also allows regulation of transgene expression or RNAi during an infection. The tools above enable work analogous to the majority of studies in T. brucei to be per- formed directly in bloodstream-form T. congolense. Since specific modification of endogenous loci in bloodstream-form T. congolense is now routine, we have not prioritized testing of CRISPR/Cas systems in the first instance, but such a system would be a useful future extension to allow modification of multiple alleles/loci. However, some of the most transformative tech- nologies in T. brucei in recent years depend on the production of not individual mutants, but libraries containing many thousand mutants (e.g. RIT-Seq [16,37]). The TcoTTS line (T7RNAP, TetR and inducible I-SceI) allows high-complexity library production in T. congo- lense for the first time, robustly giving >40,000 independent clones per transfection with p3T7-TcoV. Importantly, this is sufficient efficiency for genome-scale applications encompass- ing several mutants for each gene in the genome. The stability of TcoTTS with respect to stor- age and at least medium-term growth is an additional convenience in comparison to the current technology in T. brucei as it removes the need for fresh derivation before each experi- ment. Moreover, since all copies of the locus appear to behave indistinguishably with respect to transgene expression/RNAi, this technology can be trivially recreated in other parental lines without the necessity of screening multiple clones. The work here is focused on bloodstream-form T. congolense, as the lifecycle stage of highest experimental interest. However, all the constructs developed are applicable without modifica- tion in other stages–an application for which T. congolense is well-suited due to the ability to reproduce its whole lifecycle in vitro [14]. It also provides an example of the application of genome and transcriptomic sequence information, cost-effective DNA engineering tools, and information from decades of work in T. brucei for the rapid development of functional genetics in other important disease-causing trypanosomes. Application of this to other lines or species is relatively easy, but can only be applied to cells that can be grown as stable in vitro cultures. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Cell culture and transformation Bloodstream-form T. congolense IL3000 strain, originally isolated at the International Live- stock Research Institute (Nairobi, Kenya) and derived from infected mouse blood by [14], was a kind gift of Prof. Mike Barrett (University of Glasgow, UK). Cells were cultured at 34˚C and 5% CO2 in TcBSF1 without red blood cells [14], with the modification that only 15% goat serum (GIBCO) was used and no serum-plus. Procyclic-form T. congolense IL3000 were a kind gift of Prof. Wendy Gibson (University of Bristol, UK) and were cultured in unvented flasks at 28˚C in TcPCF3 [14]. Bloodstream-form T. brucei was grown in HMI-9 medium sup- plemented with 15% fetal bovine serum at 37˚C and 5% CO2 [48]. 2-3x107 cells and 10 μg of cut plasmid were used unless otherwise stated in the text. DNA was concentrated for transfection by ethanol precipitation and resuspension in 30 μl Tb-BSF (90 mM sodium phosphate, 5 mM KCl, 0.15 mM CaCl2, 50 mM HEPES, pH 7.3; [49]). Actively-dividing, healthy cells (at ~2x106 and ~2x107 cells ml-1, for bloodstream- and procyc- lic-form trypanosomes, respectively) were harvested by centrifugation at 1200 g for 10 min, washed in a small volume of Tb-BSF, re-pelleted at 1200 g for 2 min, and resuspended in 90 μl Tb-BSF. Cut DNA in Tb-BSF was added to these suspensions (final volume of 120 μl) and cells were electroporated in 2 mm cuvettes using an Amaxa Nucleofector 2b device (Lonza) with program ‘Z-001’. Cells were quickly transferred to flasks containing pre-warmed medium and allowed to recover for 8 h. After recovery, selection was applied by addition of antibiotics to final concentrations of 0.5 μg ml-1 puromycin, 0.2 μg ml-1 G418, or 0.5 μg ml-1 of hygromycin B. For pGad constructs (containing a single GFP and drug polycistron), selection was carried out in the presence of induction with 1 μg ml-1 tetracycline. For transfection of TcoTTS cells with induction of I-SceI expression, the same procedure was followed as above, except that selective pressure for background (hygromycin B) was removed from cells 48 h prior to electroporation, and 1 μg ml-1 tetracycline added 16 h prior to electroporation. To estimate transfection efficiency, samples of cells immediately following recovery after electroporation were diluted in fresh, selective medium and distributed across 96 well plates. Ethics statement Animal experiments were performed under UK Home Office regulations (project licences PD3DA8D1F and P98FFE489) and European directive 2010/63/EU. Research was ethically approved by the Sanger Institute Animal Welfare and Ethical Review Board. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 20 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Cell culture and transformation The proportion of total population plated was selected based on expected transfection effi- ciency, but for routine transfections 20% and 4% of each transfection was plated on 2 indepen- dent plates, and 2% and 0.4% for induced TcoTTS cells (providing accurate estimates of efficiency in ranges 100–6000 and 1000–60000 independent clones, respectively). Numbers of independent transfectants were derived from the number of positive wells based on the expected Poisson-distributed filling of wells. Confidence intervals of estimates were derived from the distribution of quantiles of 106 simulated platings at 0.001 to 10 cells well-1. Induction of RNAi or ectopic transgene expression was by addition of 1 μg ml-1 tetracycline to culture medium. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Design and construction of plasmids Sequences and graphical maps for all new constructs used in this work are available at www. wicksteadlab.co.uk and www.catarinagadelha.com, and also in S1 Dataset. All primers used for amplification of T. congolense targeting sequences are provided in S1 Table. For tagging of proteins by modification of endogenous loci, constructs were derived from pEnNY0, pEnNmSt0-N [50] and pEnNmSc0-N [51] in the case of N-terminal tagging, or pSiS-HHsfG [34] for C-terminal tagging. In all cases, 300–800 bp of targeting sequence from each side of the integration site was amplified from T. congolense IL3000 genomic DNA and incorporated into the constructs using standard methods (see S1 Table). pGad10-TcoTUB, for knockout of a TUBB allele in T. congolense, was derived from pGad9 [26] by replacement of 5’- Sequences and graphical maps for all new constructs used in this work are available at www. wicksteadlab.co.uk and www.catarinagadelha.com, and also in S1 Dataset. All primers used for amplification of T. congolense targeting sequences are provided in S1 Table. For tagging of proteins by modification of endogenous loci, constructs were derived from pEnNY0, pEnNmSt0-N [50] and pEnNmSc0-N [51] in the case of N-terminal tagging, or pSiS-HHsfG [34] for C-terminal tagging. In all cases, 300–800 bp of targeting sequence from each side of the integration site was amplified from T. congolense IL3000 genomic DNA and incorporated into the constructs using standard methods (see S1 Table). pGad10-TcoTUB, for knockout of a TUBB allele in T. congolense, was derived from pGad9 [26] by replacement of 5’- PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 21 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense and 3’- end of the GFP-HYG polycistron (without the promoter) by TUBA-TUBB and TUBB-- TUBA intergenic regions, respectively (see S1 Dataset). Since codon usage bias is very similar between T. brucei and T. congolense, CDS encoding TetR and T7RNAP for pTcoSM were taken from pSmOx ([22]; kindly provided by Steve Kelly, University of Oxford). All intergenic and targeting regions were amplified from T. congolense IL3000 genomic DNA, and the construct built by Gibson assembly in a pBluescriptII-SK (+)-derived backbone (see S1 Dataset). Constructs for ectopic inducible expression of GFP from minichromosomal loci were derived from pGad9 [26] by incorporating 300–800 bp targeting sequences from each side of the integration sites amplified from T. congolense IL3000 genomic DNA. Design and construction of plasmids Amplicons of 369 bp repeat were size-selected for 2 tandem repeats (1 repeat at each end of integration site) by exci- sion from agarose gel. p2T7-TcoV was derived from p2T7Ti [33], pEnNmSt0-N [50] and amplicons from T. congolense IL3000 genomic DNA by Gibson assembly (see S1 Dataset). This was subsequently modified to create p3T7-TcoV and p3T7-TcoVTT by standard means. To generate a self-excising construct that creates double-strand breaks at minichromosomal VSG, sequence encoding I-SceI with an N-terminal nuclear localisation signal and C-terminal HA tag controlled by an inducible T. brucei rRNA promoter was obtained from pLew100:: NLS-ISceI-HA ([52]; kindly provided by Nina Papavasiliou, German Cancer Research Center, Heidelberg). This was combined by Gibson assembly with sequence elements from pGad9 and p3T7-TcoV to create pTcoSceTbrRNA, which was subsequently modified by replacement of the inducible promoter by standard means to create pTcoScetiT7. For stable expression of luciferase, CDS encoding firefly luciferase was taken from pLEW100 [21] and neomycin resistance cassette from p3T7-TcoV. These were assembled with amplicons from T. congolense IL3000 genomic DNA by Gibson assembly in a pBluescriptII-SK (+)-derived backbone. All constructs were amplified in XL1 Blue Escherichia coli and purified using an anion- exchange column of the appropriate size (QIAGEN plasmid kits). PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Quantitative analysis of RNA All RNA preparations were made from 6-10x107 actively growing cells using High Pure RNA isolation kit (Roche) and integrity of RNA assessed by denaturing agarose gel electrophoresis. g y y g g g p For RNA-Seq, sample preparation and sequencing was performed by University of Notting- ham Deep Seq facility. RNA concentrations were measured using a Qubit Fluorometer and Qubit RNA BR Assay Kit (ThermoFisher Scientific) and integrity assessed using a TapeStation 4200 and RNA ScreenTape Assay Kit (Agilent). mRNA was purified from 1 μg of total RNA using the NEBNext Poly(A) mRNA Magnetic Isolation Module (New England BioLabs). Indexed sequencing libraries were then prepared using the NEBNext Ultra Directional RNA Library Preparation Kit for Illumina (New England BioLabs) and NEBNext Multiplex Oligos for Illumina, Index Primers Set 2 and Set 3 (New England BioLabs). Libraries were quantified using a Qubit Fluorometer and Qubit dsDNA HS Kit (ThermoFisher Scientific). Fragment- length distributions were analysed using a TapeStation 4200 and High Sensitivity D1000 ScreenTape Assay (Agilent). Libraries were pooled in equimolar amounts and final library quantification performed using the KAPA Library Quantification Kit for Illumina (Roche). The pool was sequenced using a NextSeq 500 System (Illumina) and mid-output 150 cycle kit v2.5 (Illumina), providing >12 million pairs of 75-bp paired-end reads passing filter per sam- ple. http://www.ebi.ac.uk/ena The pool was sequenced using a NextSeq 500 System (Illumina) and mid-output 150 cycle kit v2.5 (Illumina), providing >12 million pairs of 75-bp paired-end reads passing filter per sam- ple. http://www.ebi.ac.uk/ena Reads were trimmed for quality and adapter using Trim Galore v0.4.4 (www. bioinformatics.babraham.ac.uk; -q 28—illumina—stringency 3—length 50) and aligned the PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 22 / 29 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense TriTrypDB T. congolense IL3000_2019 or T. brucei Lister 427_2018 assemblies (v46; based on the work of [5,53]) using bowtie2 v2.3.4 ([54];—no-mixed—no-discordant -I 50 -X 500). Total transcript abundance was assessed using HTSeq v0.5.4 [55]. Read depth analysis was per- formed using bedtools v2.25.0 [56]. For quantitative RT-PCR of transcripts in δchc cells, total RNA was reverse transcribed using avian myeloblastosis virus reverse transcriptase (New England BioLabs) primed by T15N. cDNA (or no-RT control) generated from 20 ng total RNA was used as template in each PCR reaction replicate. qPCR was performed on an Mx3000 system (Agilent) with a stan- dard Taq polymerase mix supplemented with 0.5x EvaGreen dsDNA dye (Biotium). Quantitative analysis of RNA Target mRNA levels were normalised using levels of PFR1 (TcIL3000.A.H_000346600) transcript. For quantitative RT-PCR in other RNAi cell lines, total RNA was reverse transcribed using a High Capacity cDNA RT kit (Applied Biosystems). qPCR reactions were carried out with SensiFAST SYBR Hi-ROX mix (Bioline), in a Rotor-Gene 3000 system (Corbett Research). Target mRNA levels were normalised using levels of TERT (TcIL3000.A.H_000960200, for other RNAi) tran- script. Primers for TERT were as in [57], other primers used in quantitative PCR are provided in S2 Table. Samples were biological triplicates (for individual clones) or duplicates of 2 inde- pendent clones (δchc). Microscopy For analysis of localisation of tagged proteins or soluble fluorescent protein by native fluores- cence, cells were harvested from mid-log phase cultures, washed twice in phosphate-buffered saline (PBS; 137 mM NaCl, 3 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4) and allowed to settle for 5 min onto glass slides at ~2x107 cell ml-1 density. Cells were fixed for 5 min in 2% (w/v) formaldehyde, permeabilised in -20˚C methanol for 10 min, re-hydrated in PBS and incubated with 15 ng ml-1 40,6-diamidino-2-phenylindole for 5 min, before mounting in 1% (w/v) 1,4-diazabicyclo[2.2.2]octane, 90% (v/v) glycerol, 50 mM sodium phosphate, pH 8.0. Counter-staining of cells expressing ESP14-sfGFP was by incubation with 25 μg ml-1 tomato lectin conjugated to AlexaFluor 594 (Invitrogen) for 20 min. Images were captured on an Olympus BX51 microscope equipped with a 100x UPlanApo objective (1.35 NA; Olympus) and Retiga R1 CCD camera (Qimaging) without binning. All images of fluorescent proteins were captured at equal exposure settings without prior illumination. Images for level compari- son were also processed in parallel with the same alterations to minimum and maximum dis- play levels, except where stated. Image acquisition was controlled by μManager open source software [58]. Processing and analysis were performed in ImageJ [59]. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Infections with transgenic T. congolense and bioluminescence measurements In vitro luciferase activity was detected using the Luciferase Reporter Gene Assay, high sensi- tivity (Roche) and luminescence detected using a Glomax 96 Microplate Luminometer (Pro- mega). For infections, female BALB/c strain animals were used except where indicated. Bloodstream-form T. congolense LUC2 cells were obtained from an infected donor mouse at the peak of parasitaemia, diluted in PBS supplemented with 20 mM glucose and used to inocu- late recipient mice by intravenous injection at total doses of 100, 1000, or 10,000 parasites. D- luciferin was administered to animals at a dose of 200 mg kg-1, by intraperitoneal injection 10 minutes before imaging. The mice were allowed 3 minutes of movement before being anaes- thetized and placed in the imaging chamber where anaesthesia was maintained during acquisi- tion. In vivo bioluminescence was captured on an IVIS Spectrum Imaging System (Perkin Elmer) and quantified using Living Image software (Xenogen). Flow cytometry For quantitative analysis of fluorescence by flow cytometry, ~3x106 cells were harvested by centrifugation at 1200 g, 5 min and resuspended in 120 μl PBS containing 1% (w/v) formalde- hyde, 0.2% (w/v) glutaraldehyde and 10 μg ml-1 propidium iodide. This suspension was incu- bated at room temperature for 10 minutes, after which time the suspension was diluted 5-fold with PBS and analysed on a FC500 flow cytometer (Beckman Coulter), collecting a >20,000 gated events for each sample. Events were gated on the basis of forward- and side-scatter to remove clumps of cells and debris, and only cells with intact plasma membrane (negative for propidium) were included for analysis. Testing of double-strand break induction To test for loss of HYG due to induction of I-SceI endonuclease expression, cells were removed from hygromycin selection and cultured with or without 1 μg ml-1 tetracycline for 48h. These cultures were then diluted in fresh culture medium without antibiotic, and distributed across 96 well plates at an estimated 1.5 cells well-1. After growth for 7 days, plates were replicated by 5-fold dilution into plates containing fresh medium with or without 2 μg ml-1 hygromycin B and cultured for a further 1.5 days. The number of wells that had become hygromycin-sensi- tive was used to derive the proportion of HYG-negative cells based on Poisson-distributed fill- ing of wells (λ estimated from populated wells on initial plating) and Binomially-distributed loss of HYG. Confidence intervals of estimates were derived from the distribution of quantiles from 106 simulated platings with resampling λ (Normally-distributed around each estimate) and a range of probability of loss from 0–1. Immunoblotting To test either tagged protein ablation by RNAi or ectopic gene expression, lysates from 2x107 cells were separated by reducing SDS-polyacrylamide gel electrophoresis and electro-trans- ferred onto nitrocellulose membrane (GE healthcare) in 25 mM Tris, 192 mM glycine, 0.02% SDS, 10% methanol. Membranes were blocked with 5% (w/v) skimmed milk in TBS-T (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.1% (v/v) Tween-20) and protein detected by either 800 ng ml-1 mixture of two anti-GFP monoclonal antibodies (7.1 and 13.1; Roche), or 500 ng ml-1 anti-tetracycline repressor protein monoclonal antibody (9G9; Clontech), followed by 80 ng ml-1 goat anti-mouse immunoglobulins conjugated to horseradish peroxidase (Sigma) in TBS-T containing 1% (w/v) skimmed milk. Antibodies were detected using Western Lightning enhanced chemiluminescence reagent (GE healthcare) captured on a Fusion FX Spectra (Vil- ber Lourmat). 23 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Genotypic analysis of modified cell lines For testing of genotypes by PCR, ~105 cells from clonal lines or populations were harvested, washed in PBS and subjected to hot alkaline lysis in 50 mM NaOH, 2 mM EDTA for 10 min at 95˚C, followed by neutralisation by addition of Tris-HCl pH 6.8 to a final concentration of 75 mM. Small samples of these preparations were used as templates in multiplex PCR using prim- ers for specific bait regions to screen transformants. Supporting information Median and range of 5% and 95% quantiles are shown by dot and bars, respec- tively. (TIF) S4 Fig. Stability of phenotypes associated with knockdown of clathrin heavy chain to con- tinuous passage or freeze-thaw of cell lines. Proportions of cells with clear morphological defect (see Fig 6) are shown for cells with no RNAi construct (parental) and two independent clones of δchc cells following 0 or 48 h induction with tetracycline. Clones had previously been grown in culture for 1 week or 8 weeks following transfection and selection, or frozen after 1 week in culture, stored under liquid nitrogen and then brought back into growth (freeze- thaw). Bonferroni adjusted p-values from proportions test are shown above columns (n/s: not significant; : p<0.05; : p<0.001). ( ) S5 Fig. Infection dynamics of T. congolense LUC2 cells in different albino mouse strains. Male or female BALB/c and male B6 Albino (C57BL/6N-TyrcWTSI) were infected with 1000 T. congolense LUC2 cells at time 0 and dynamics monitored by whole-animal bioluminescence. Data from 15 independent infections using male BALB/c and B6 Albino animals, and 4 infec- tions with female BALB/c animals are shown. (TIF) S5 Fig. Infection dynamics of T. congolense LUC2 cells in different albino mouse strains. Male or female BALB/c and male B6 Albino (C57BL/6N-TyrcWTSI) were infected with 1000 T. congolense LUC2 cells at time 0 and dynamics monitored by whole-animal bioluminescence. Data from 15 independent infections using male BALB/c and B6 Albino animals, and 4 infec- tions with female BALB/c animals are shown. (TIF) Supporting information S1 Fig. Localisation of proteins by modification of endogenous loci in bloodstream-form T. congolense. Representative fields of view are shown for cells expressing T. congolense ESP10 24 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense C-terminally tagged with superfolder GFP (ESP10-sfG), or Nopp140 N-terminally tagged with mStrawberry (mSt-Nopp140). Native fluorescence from tagged proteins is shown, alongside counter-staining with 40,6-diamidino-2-phenylindole (DAPI; cyan). ESP10-sfG cells have been additionally stained with AlexaFluor 594-conjugated tomato lectin (TL), to highlight the flagel- lar pocket and endosomal machinery. (TIF) C-terminally tagged with superfolder GFP (ESP10-sfG), or Nopp140 N-terminally tagged with mStrawberry (mSt-Nopp140). Native fluorescence from tagged proteins is shown, alongside counter-staining with 40,6-diamidino-2-phenylindole (DAPI; cyan). ESP10-sfG cells have been additionally stained with AlexaFluor 594-conjugated tomato lectin (TL), to highlight the flagel- lar pocket and endosomal machinery. (TIF) S2 Fig. Levels of transgenic TetR produced in modified cells. Figure shows full view of immunoblot membrane excerpted in Fig 2D containing transferred whole cell lysates. Ponceau S staining is shown as a control for loading, plus two exposures of the membrane immuno- blotted with an anti-TetR monoclonal antibody. ‘wt’ indicates unmodified parental cells. (TIF) S3 Fig. Distribution of cellular fluorescence in cell lines expressing transgenic GFP from endogenous loci or ectopic inducible loci, and in the presence of RNAi against GFP. A) Violin plots of green fluorescence of cells expressing GFP from the tubulin locus (Δtubb::GFP) or from inducible minichromosomal loci in 0 (non-induced) or 1 μg ml-1 (induced) tetracy- cline (see Figs 3A and 4A). B) Violin plots of green fluorescence of cells expressing GFP from the tubulin locus (Δtubb::GFP) in which RNA-interference against GFP had been induced (see Fig 5A and 5B). Median and range of 5% and 95% quantiles are shown by dot and bars, respec- tively. (TIF) S3 Fig. Distribution of cellular fluorescence in cell lines expressing transgenic GFP from endogenous loci or ectopic inducible loci, and in the presence of RNAi against GFP. A) Violin plots of green fluorescence of cells expressing GFP from the tubulin locus (Δtubb::GFP) or from inducible minichromosomal loci in 0 (non-induced) or 1 μg ml-1 (induced) tetracy- cline (see Figs 3A and 4A). B) Violin plots of green fluorescence of cells expressing GFP from the tubulin locus (Δtubb::GFP) in which RNA-interference against GFP had been induced (see Fig 5A and 5B). PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 S1 Dataset. Sequences and graphical maps for all new constructs used in this work. Anno- tated sequences are provided in GenBank format and were used to generate graphical maps. (ZIP) S1 Dataset. Sequences and graphical maps for all new constructs used in this work. Anno- tated sequences are provided in GenBank format and were used to generate graphical maps. (ZIP) S1 Table. All primers used for amplification of T. congolense targeting sequences. (XLSX) S2 Table. Primers used in quantitative RT-PCR. (XLSX) S1 Supporting Information. Data supporting figures in main document and supplement. (XLSX) 25 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense Acknowledgments We are grateful to Mike Barrett (University of Glasgow) and Wendy Gibson (University of Bristol) for providing bloodstream-form and procyclic IL3000 cells, respectively. pSmOx and pLew100::NLS-ISceI-HA were gifts of Steve Kelly (University of Oxford) and Nina Papavasi- liou (German Cancer Research Center, Heidelberg), respectively. pLew100::NLS-ISceI-HA was obtained via AddGene (#21299). We thank Tom Miller (University of Nottingham) for assistance with immunoblotting. RNA sequencing and library preparation were performed by Nadine Holmes in the Deep Seq facility, University of Nottingham. Project administration: Catarina Gadelha, Bill Wickstead. Project administration: Catarina Gadelha, Bill Wickstead. Project administration: Catarina Gadelha, Bill Wickstead. Resources: Simon D’Archivio, Cordelia Brandt, Simon Clare, Katherine Harcourt. Supervision: Gavin J. Wright, Liam J. Morrison, Catarina Gadelha, Bill Wickstead. Validation: Georgina Awuah-Mensah, Jennifer McDonald, Pieter C. Steketee, Sarah Whipple. Visualization: Georgina Awuah-Mensah, Pieter C. Steketee, Delphine Autheman, Catarina Gadelha, Bill Wickstead. Writing – original draft: Catarina Gadelha, Bill Wickstead. Writing – review & editing: Georgina Awuah-Mensah, Jennifer McDonald, Pieter C. Steketee, Delphine Autheman, Sarah Whipple, Simon D’Archivio, Cordelia Brandt, Simon Clare, Katherine Harcourt, Gavin J. Wright, Liam J. Morrison, Catarina Gadelha, Bill Wickstead. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 Author Contributions Conceptualization: Catarina Gadelha, Bill Wickstead. Conceptualization: Catarina Gadelha, Bill Wickstead. Data curation: Catarina Gadelha, Bill Wickstead. Data curation: Catarina Gadelha, Bill Wickstead. Formal analysis: Georgina Awuah-Mensah, Jennifer McDonald, Pieter C. Steketee, Delphine Autheman, Sarah Whipple, Simon D’Archivio, Gavin J. Wright, Liam J. Morrison, Catarina Gadelha, Bill Wickstead. Formal analysis: Georgina Awuah-Mensah, Jennifer McDonald, Pieter C. Steketee, Delphine Autheman, Sarah Whipple, Simon D’Archivio, Gavin J. Wright, Liam J. Morrison, Catarina Gadelha, Bill Wickstead. Funding acquisition: Gavin J. Wright, Liam J. Morrison, Catarina Gadelha, Bill Wickstead. Investigation: Georgina Awuah-Mensah, Jennifer McDonald, Pieter C. Steketee, Delphine Autheman, Sarah Whipple. References References 1. United Nations. http://www.fao.org/paat/the-programme/the-disease/en/. 2014. 2. Sheferaw D, Abebe R, Fekadu A, Kassaye S, Amenu K, Data D, et al. Prevalence of bovine trypanoso- mosis and vector density in a dry season in Gamo-Gofa and Dawuro Zones, Southern Ethiopia. Vet Parasitol Reg Stud Reports. 2019; 18:100343. https://doi.org/10.1016/j.vprsr.2019.100343 PMID: 31796171 3. Tchamdja E, Kulo AE, Vitouley HS, Batawui K, Bankole´ AA, Adomefa K, et al. Cattle breeding, trypano- somosis prevalence and drug resistance in Northern Togo. Vet Parasitol. 2017; 236:86–92. https://doi. org/10.1016/j.vetpar.2017.02.008 PMID: 28288771 4. Muhanguzi D, Mugenyi A, Bigirwa G, Kamusiime M, Kitibwa A, Akurut GG, et al. African animal trypano- somiasis as a constraint to livestock health and production in Karamoja region: a detailed qualitative and quantitative assessment. BMC Vet Res. 2017; 13:355. https://doi.org/10.1186/s12917-017-1285-z PMID: 29178951 5. Abbas AH, Silva Pereira S, D’Archivio S, Wickstead B, Morrison LJ, Hall N, et al. The structure of a con- served telomeric region associated with variant antigen loci in the blood parasite Trypanosoma congo- lense. Genome Biol Evol. 2018; 10:2458–73. https://doi.org/10.1093/gbe/evy186 PMID: 30165630 1. United Nations. http://www.fao.org/paat/the-programme/the-disease/en/. 2014. 1. United Nations. http://www.fao.org/paat/the-programme/the-disease/en/. 2014. 2. Sheferaw D, Abebe R, Fekadu A, Kassaye S, Amenu K, Data D, et al. Prevalence of bovine trypanoso- mosis and vector density in a dry season in Gamo-Gofa and Dawuro Zones, Southern Ethiopia. Vet Parasitol Reg Stud Reports. 2019; 18:100343. https://doi.org/10.1016/j.vprsr.2019.100343 PMID: 31796171 3. Tchamdja E, Kulo AE, Vitouley HS, Batawui K, Bankole´ AA, Adomefa K, et al. Cattle breeding, trypano- somosis prevalence and drug resistance in Northern Togo. Vet Parasitol. 2017; 236:86–92. https://doi. org/10.1016/j.vetpar.2017.02.008 PMID: 28288771 4. Muhanguzi D, Mugenyi A, Bigirwa G, Kamusiime M, Kitibwa A, Akurut GG, et al. African animal trypano- somiasis as a constraint to livestock health and production in Karamoja region: a detailed qualitative and quantitative assessment. BMC Vet Res. 2017; 13:355. https://doi.org/10.1186/s12917-017-1285-z PMID: 29178951 5. Abbas AH, Silva Pereira S, D’Archivio S, Wickstead B, Morrison LJ, Hall N, et al. The structure of a con- served telomeric region associated with variant antigen loci in the blood parasite Trypanosoma congo- lense. Genome Biol Evol. 2018; 10:2458–73. https://doi.org/10.1093/gbe/evy186 PMID: 30165630 26 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense 6. Jackson AP, Berry A, Aslett M, Allison HC, Burton P, Vavrova-Anderson J, et al. Antigenic diversity is generated by distinct evolutionary mechanisms in African trypanosome species. Proc Natl Acad Sci USA. References 2012; 109:3416–21. https://doi.org/10.1073/pnas.1117313109 PMID: 22331916 7. Fish WR, Muriuki CW, Muthiani AM, Grab DJ, Lonsdale-Eccles JD. Disulfide bond involvement in the maintenance of the cryptic nature of the cross-reacting determinant of metacyclic forms of Trypano- soma congolense. Biochemistry. 1989; 28:5415–21. https://doi.org/10.1021/bi00439a015 PMID: 2476173 8. Silva Pereira S, Casas-Sa´nchez A, Haines LR, Ogugo M, Absolomon K, Sanders M, et al. Variant anti- gen repertoires in Trypanosoma congolense populations and experimental infections can be profiled from deep sequence data using universal protein motifs. Genome Res. 2018; 28:1383–94. https://doi. org/10.1101/gr.234146.118 PMID: 30006414 9. Rotureau B, Van Den Abbeele J. Through the dark continent: African trypanosome development in the tsetse fly. Front Cell Infect Microbiol. 2013; 3:53. https://doi.org/10.3389/fcimb.2013.00053 PMID: 24066283 10. Gitonga PK, Ndung’u K, Murilla GA, Thande PC, Wamwiri FN, Auma JE, et al. Differential virulence and tsetse fly transmissibility of Trypanosoma congolense and Trypanosoma brucei strains. Onderstepoort J Vet Res. 2017; 84:e1–10. 11. Hemphill A, Frame I, Ross CA. The interaction of Trypanosoma congolense with endothelial cells. Para- sitology. 1994; 109:631–41. https://doi.org/10.1017/s0031182000076514 PMID: 7831098 12. Silvester E, Ivens A, Matthews KR. A gene expression comparison of Trypanosoma brucei and Trypa- nosoma congolense in the bloodstream of the mammalian host reveals species-specific adaptations to density-dependent development. PLoS Negl Trop Dis. 2018; 12:e0006863. https://doi.org/10.1371/ journal.pntd.0006863 PMID: 30307943 13. Coustou V, Plazolles N, Guegan F, Baltz T. Sialidases play a key role in infection and anaemia in Trypa- nosoma congolense animal trypanosomiasis. Cell Microbiol. 2012; 14:431–45. https://doi.org/10.1111/ j.1462-5822.2011.01730.x PMID: 22136727 14. Coustou V, Guegan F, Plazolles N, Baltz T. Complete in vitro life cycle of Trypanosoma congolense: development of genetic tools. PLoS Negl Trop Dis. 2010; 4:e618. https://doi.org/10.1371/journal.pntd. 0000618 PMID: 20209144 15. Glover L, Horn D. Site-specific DNA double-strand breaks greatly increase stable transformation effi- ciency in Trypanosoma brucei. Mol Biochem Parasitol. 2009; 166:194–7. https://doi.org/10.1016/j. molbiopara.2009.03.010 PMID: 19459229 16. Alsford S, Turner DJ, Obado SO, Sanchez-Flores A, Glover L, Berriman M, et al. High-throughput phe- notyping using parallel sequencing of RNA interference targets in the African trypanosome. Genome Res. 2011; 21:915–24. https://doi.org/10.1101/gr.115089.110 PMID: 21363968 17. Kay C, Peacock L, Gibson W. Trypanosoma congolense: In vitro culture and transfection. Curr Protoc Microbiol. 2019; 53:e77. https://doi.org/10.1002/cpmc.77 PMID: 30707507 18. Tschopp F, Charrière F, Schneider A. In vivo study in Trypanosoma brucei links mitochondrial transfer RNA import to mitochondrial protein import. EMBO Rep. 2011; 12:825–32. https://doi.org/10.1038/ embor.2011.111 PMID: 21720389 19. Sunter JD. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 References A vanillic acid inducible expression system for Trypanosoma brucei. Mol Biochem Parasitol. 2016; 207:45–8. https://doi.org/10.1016/j.molbiopara.2016.04.001 PMID: 27063979 20. Li F-J, Xu Z-S, Aye HM, Brasseur A, Lun Z-R, Tan KSW, et al. An efficient cumate-inducible system for procyclic and bloodstream form Trypanosoma brucei. Mol Biochem Parasitol. 2017; 214:101–4. https:// doi.org/10.1016/j.molbiopara.2017.04.007 PMID: 28438458 21. Wirtz E, Leal S, Ochatt C, Cross GA. A tightly regulated inducible expression system for conditional gene knock-outs and dominant-negative genetics in Trypanosoma brucei. Mol Biochem Parasitol. 1999; 99:89–101. https://doi.org/10.1016/s0166-6851(99)00002-x PMID: 10215027 22. Poon SK, Peacock L, Gibson W, Gull K, Kelly S. A modular and optimized single marker system for gen- erating Trypanosoma brucei cell lines expressing T7 RNA polymerase and the tetracycline repressor. Open Biol. 2012; 2:110037. https://doi.org/10.1098/rsob.110037 PMID: 22645659 23. de Freitas Nascimento J, Kelly S, Sunter J, Carrington M. Codon choice directs constitutive mRNA lev- els in trypanosomes. Elife. 2018; 7:e32467. https://doi.org/10.7554/eLife.32467 PMID: 29543152 24. Biebinger S, Wirtz LE, Lorenz P, Clayton C. Vectors for inducible expression of toxic gene products in bloodstream and procyclic Trypanosoma brucei. Mol Biochem Parasitol. 1997; 85:99–112. https://doi. org/10.1016/s0166-6851(96)02815-0 PMID: 9108552 25. Alsford S, Kawahara T, Glover L, Horn D. Tagging a T. brucei RRNA locus improves stable transfection efficiency and circumvents inducible expression position effects. Mol Biochem Parasitol. 2005; 144:142–8. https://doi.org/10.1016/j.molbiopara.2005.08.009 PMID: 16182389 27 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense 26. Wickstead B, Ersfeld K, Gull K. Targeting of a tetracycline-inducible expression system to the transcrip- tionally silent minichromosomes of Trypanosoma brucei. Mol Biochem Parasitol. 2002; 125:211–6. https://doi.org/10.1016/s0166-6851(02)00238-4 PMID: 12467990 27. Downey N, Donelson JE. Search for promoters for the GARP and rRNA genes of Trypanosoma congo- lense. Mol Biochem Parasitol. 1999; 104:25–38. https://doi.org/10.1016/s0166-6851(99)00135-8 PMID: 10589979 28. Inoue N, Otsu K, Ferraro DM, Donelson JE. Tetracycline-regulated RNA interference in Trypanosoma congolense. Mol Biochem Parasitol. 2002; 120:309–13. https://doi.org/10.1016/s0166-6851(02)00015- 4 PMID: 11897138 29. Wickstead B, Ersfeld K, Gull K. The mitotic stability of the minichromosomes of Trypanosoma brucei. Mol Biochem Parasitol. 2003; 132:97–100. https://doi.org/10.1016/j.molbiopara.2003.08.007 PMID: 14599670 30. Gibson WC, Dukes P, Gashumba JK. Species-specific DNA probes for the identification of African try- panosomes in tsetse flies. Parasitology. 1988; 97:63–73. https://doi.org/10.1017/s0031182000066749 PMID: 3174239 31. Moser DR, Cook GA, Ochs DE, Bailey CP, McKane MR, Donelson JE. Detection of Trypanosoma con- golense and Trypanosoma brucei subspecies by DNA amplification using the polymerase chain reac- tion. Parasitology. 1989; 99:57–66. References https://doi.org/10.1017/s0031182000061023 PMID: 2797872 32. Wickstead B, Ersfeld K, Gull K. The frequency of gene targeting in Trypanosoma brucei is independent of target site copy number. Nucleic Acids Res. 2003; 31:3993–4000. https://doi.org/10.1093/nar/ gkg445 PMID: 12853615 33. LaCount DJ, Bruse S, Hill KL, Donelson JE. Double-stranded RNA interference in Trypanosoma brucei using head-to-head promoters. Mol Biochem Parasitol. 2000; 111:67–76. https://doi.org/10.1016/ s0166-6851(00)00300-5 PMID: 11087917 34. Gadelha C, Zhang W, Chamberlain JW, Chait BT, Wickstead B, Field MC. Architecture of a host-para- site interface: complex targeting mechanisms revealed through proteomics. Mol Cell Proteomics. 2015; 14:1911–26. https://doi.org/10.1074/mcp.M114.047647 PMID: 25931509 35. Allen CL, Goulding D, Field MC. Clathrin-mediated endocytosis is essential in Trypanosoma brucei. EMBO J. 2003; 22:4991–5002. https://doi.org/10.1093/emboj/cdg481 PMID: 14517238 36. Ngoˆ H, Tschudi C, Gull K, Ullu E. Double-stranded RNA induces mRNA degradation in Trypanosoma brucei. Proc Natl Acad Sci USA. 1998; 95:14687–92. https://doi.org/10.1073/pnas.95.25.14687 PMID: 9843950 37. Glover L, Alsford S, Baker N, Turner DJ, Sanchez-Flores A, Hutchinson S, et al. Genome-scale RNAi screens for high-throughput phenotyping in bloodstream-form African trypanosomes. Nat Protoc. 2015; 10:106–33. https://doi.org/10.1038/nprot.2015.005 PMID: 25502887 38. McLatchie AP, Burrell-Saward H, Myburgh E, Lewis MD, Ward TH, Mottram JC, et al. Highly sensitive in vivo imaging of Trypanosoma brucei expressing ‘red-shifted’ luciferase. PLoS Negl Trop Dis. 2013; 7: e2571. https://doi.org/10.1371/journal.pntd.0002571 PMID: 24278497 39. D’Archivio S, Medina M, Cosson A, Chamond N, Rotureau B, Minoprio P, et al. Genetic engineering of Trypanosoma (Dutonella) vivax and in vitro differentiation under axenic conditions. PLoS Negl Trop Dis. 2011; 5:e1461. https://doi.org/10.1371/journal.pntd.0001461 PMID: 22216367 40. Banks KL. Binding of Trypanosoma congolense to the walls of small blood vessels. J Protozool. 1978; 25:241–5. https://doi.org/10.1111/j.1550-7408.1978.tb04405.x PMID: 690955 41. D’Archivio S, Cosson A, Medina M, Lang T, Minoprio P, Goyard S. Non-invasive in vivo study of the Try- panosoma vivax infectious process consolidates the brain commitment in late infections. PLoS Negl Trop Dis. 2013; 7:e1976. https://doi.org/10.1371/journal.pntd.0001976 PMID: 23301112 42. Bell JS, McCulloch R. Mismatch repair regulates homologous recombination, but has little influence on antigenic variation, in Trypanosoma brucei. J Biol Chem. 2003; 278:45182–8. https://doi.org/10.1074/ jbc.M308123200 PMID: 12933800 43. Ullu E, Djikeng A, Shi H, Tschudi C. RNA interference: advances and questions. Philos Trans R Soc Lond B Biol Sci. 2002; 357:65–70. https://doi.org/10.1098/rstb.2001.0952 PMID: 11839183 44. Durand-Dubief M, Absalon S, Menzer L, Ngwabyt S, Ersfeld K, Bastin P. The Argonaute protein TbAGO1 contributes to large and mini-chromosome segregation and is required for control of RIME ret- roposons and RHS pseudogene-associated transcripts. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 References Mol Biochem Parasitol. 2007; 156:144–53. https://doi.org/10.1016/j.molbiopara.2007.07.016 PMID: 17822785 45. Shi H, Djikeng A, Tschudi C, Ullu E. Argonaute protein in the early divergent eukaryote Trypanosoma brucei: control of small interfering RNA accumulation and retroposon transcript abundance. Mol Cell Biol. 2004; 24:420–7. https://doi.org/10.1128/mcb.24.1.420-427.2004 PMID: 14673174 28 / 29 PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 PLOS PATHOGENS Tractable functional genetics in bloodstream-form Trypanosoma congolense 46. Shi H, Tschudi C, Ullu E. An unusual Dicer-like1 protein fuels the RNA interference pathway in Trypano- soma brucei. RNA. 2006; 12:2063–72. https://doi.org/10.1261/rna.246906 PMID: 17053086 47. Patrick KL, Shi H, Kolev NG, Ersfeld K, Tschudi C, Ullu E. Distinct and overlapping roles for two Dicer- like proteins in the RNA interference pathways of the ancient eukaryote Trypanosoma brucei. Proc Natl Acad Sci U S A. 2009; 106:17933–8. https://doi.org/10.1073/pnas.0907766106 PMID: 19815526 48. Hirumi H, Hirumi K. Continuous cultivation of Trypanosoma brucei blood stream forms in a medium con- taining a low concentration of serum protein without feeder cell layers. J Parasitol. 1989; 75:985–9. PMID: 2614608 49. Schumann Burkard G, Jutzi P, Roditi I. Genome-wide RNAi screens in bloodstream form trypanosomes identify drug transporters. Mol Biochem Parasitol. 2011; 175:91–4. https://doi.org/10.1016/j. molbiopara.2010.09.002 PMID: 20851719 50. D’Archivio S, Wickstead B. Trypanosome outer kinetochore proteins suggest conservation of chromo- some segregation machinery across eukaryotes. J Cell Biol. 2017; 216:379–91. https://doi.org/10. 1083/jcb.201608043 PMID: 28034897 51. Brusini L, D’Archivio S, McDonald J, Wickstead B. Ndc80/Nuf2-like protein KKIP1 connects a stable kinetoplastid outer kinetochore complex to the inner kinetochore and responds to metaphase tension. bioRxiv. 2019;764829: https://doi.org/10.1101/764829 52. Boothroyd CE, Dreesen O, Leonova T, Ly KI, Figueiredo LM, Cross GAM, et al. A yeast-endonuclease- generated DNA break induces antigenic switching in Trypanosoma brucei. Nature. 2009; 459:278–81. https://doi.org/10.1038/nature07982 PMID: 19369939 53. Mu¨ller LSM, Cosentino RO, Fo¨rstner KU, Guizetti J, Wedel C, Kaplan N, et al. Genome organization and DNA accessibility control antigenic variation in trypanosomes. Nature. 2018; 563:121–5. https://doi. org/10.1038/s41586-018-0619-8 PMID: 30333624 54. Langmead B, Salzberg SL. Fast gapped-read alignment with Bowtie 2. Nat Methods. 2012; 9:357–9. https://doi.org/10.1038/nmeth.1923 PMID: 22388286 55. Anders S, Pyl PT, Huber W. HTSeq: a Python framework to work with high-throughput sequencing data. Bioinformatics. 2015; 31:166–9. https://doi.org/10.1093/bioinformatics/btu638 PMID: 25260700 56. Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformat- ics. 2010; 26:841–2. https://doi.org/10.1093/bioinformatics/btq033 PMID: 20110278 57. Tihon E, Imamura H, Van den Broeck F, Vermeiren L, Dujardin J-C, Van Den Abbeele J. PLOS Pathogens \| https://doi.org/10.1371/journal.ppat.1009224 January 22, 2021 References Genomic anal- ysis of isometamidium chloride resistance in Trypanosoma congolense. Int J Parasitol Drugs Drug Resist. 2017; 7:350–61. https://doi.org/10.1016/j.ijpddr.2017.10.002 PMID: 29032180 58. Edelstein AD, Tsuchida MA, Amodaj N, Pinkard H, Vale RD, Stuurman N. Advanced methods of micro- scope control using μManager software. J Biol Methods. 2014; 1:e10. https://doi.org/10.14440/jbm. 2014.36 PMID: 25606571 59. Schneider CA, Rasband WS, Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat Meth- ods. 2012; 9:671–5. https://doi.org/10.1038/nmeth.2089 PMID: 22930834 60. Price MN, Dehal PS, Arkin AP. FastTree 2—approximately maximum-likelihood trees for large align- ments. PLoS One. 2010; 5:e9490. https://doi.org/10.1371/journal.pone.0009490 PMID: 20224823 29 / 29
https://openalex.org/W2074049645	https://bmccancer.biomedcentral.com/counter/pdf/10.1186/1471-2407-9-130	English	null	Eligibility of patients with advanced non-small cell lung cancer for phase III chemotherapy trials	BMC cancer	2,009	cc-by	4,839	BioMed Central BioMed Central Received: 5 December 2008 Accepted: 29 April 2009 BMC Cancer 2009, 9:130 doi:10.1186/1471-2407-9-130 This article is available from: http://www.biomedcentral.com/1471-2407/9/130 © 2009 Vardy 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: Evidence that chemotherapy improves survival and quality of life in patients with stage IIIB & IV non small cell lung cancer (NSCLC) is based on large randomized controlled trials. The purpose of this study was to determine eligibility of patients with advanced NSCLC for major chemotherapy trials. Methods: Physicians treating stage IIIB/IV NSCLC at Sydney Cancer Centre assessed patient eligibility for the E1594, SWOG9509 and TAX326 trials for patients presenting from October 2001 to December 2002. A review of the centre's registry was used to obtain missing data. Results: 199 patients with advanced NSCLC were registered during the 14-month period. Characteristics of 100 patients were defined prospectively, 85 retrospectively: 77% males, median age 68 (range 32–88), 64% stage IV disease. Only 35% met trial eligibility for E1594 and 28% for SWOG9509 and TAX326. Common reasons for ineligibility were: co-morbidities 75(40%); ECOG Performance Status ≥2 72(39%); symptomatic brain metastasis 15(8%); and previous cancers 21(11%). Many patients were ineligible by more than one criterion. Conclusion: The majority of patients with advanced NSCLC were ineligible for the large chemotherapy trials. The applicability of trial results to advanced lung cancer populations may be limited. Future trials should be conducted in a more representative population. and chemotherapy reported a 27% reduction in the risk of death (p < 0.0001), with an increase in median survival of 1.5 months, in patients receiving cisplatin-based chemo- therapy[1]. A phase III study randomising patients with stage IIIB (pleural effusion) or IV NSCLC to BSC or BSC and chemotherapy with mitomycin C, ifosfamide and cis- platin (MIC2), found a median survival of 6.7 months in Open Ac Research article Eligibility of patients with advanced non-small cell lung cancer for phase III chemotherapy trials Janette Vardy1,2,3, Ryan Dadasovich1,2,3, Philip Beale1,2,3, Michael Boyer1,2,3 and Stephen J Clarke1,2,3 Open Access Address: 1Sydney Cancer Centre, Concord Repatriation General Hospital and Royal Prince Alfred Hospital, Sydney, NSW, Australia, 2Sydney Cancer Centre, Concord Repatriation General Hospital, Hospital Rd, Concord, NSW, 2139 Australia and 3Sydney Cancer Centre, Royal Prince Alfred Hospital, Missenden Rd, Camperdown, 2050, NSW, Australia Email: Janette Vardy - jvardy@med.usyd.edu.au; Ryan Dadasovich - ryan.dadasovitch@gmp.med.usyd.edu.au; Philip Beale - Philip.beale@sswahs.nsw.gov.au; Michael Boyer - Michael.boyer@sswahs.nsw.gov.au; Stephen J Clarke - sclarke@med.usyd.edu.au mail: Janette Vardy* - jvardy@med.usyd.edu.au; Ryan Dadasovich - ryan.dadasovitch@gmp.med.usyd.edu.au; hilip Beale - Philip.beale@sswahs.nsw.gov.au; Michael Boyer - Michael.boyer@sswahs.nsw.gov.au; tephen J Clarke - sclarke@med.usyd.edu.au * Corresponding author Published: 29 April 2009 BMC Cancer 2009, 9:130 doi:10.1186/1471-2407-9-130 Received: 5 December 2008 Accepted: 29 April 2009 This article is available from: http://www.biomedcentral.com/1471-2407/9/130 © 2009 Vardy 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/licen which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Received: 5 December 2008 Accepted: 29 April 2009 Received: 5 December 2008 Accepted: 29 April 2009 Page 1 of 6 (page number not for citation purposes) Methods Patients were recruited from the Royal Prince Alfred Hos- pital and Concord Repatriation General Hospital, Sydney, Australia from October 2001 until December 2002. Ethics committee approval was obtained from both participat- ing hospitals. All physicians from Medical Oncology, Radiation Oncology, Respiratory Medicine and Thoracic Surgery were requested to complete a questionnaire at the time of the initial consultation for each new patient with Stage IIIB (pleural effusion) and Stage IV NSCLC present- ing to their service. Basic demographic information was collected and histological subtype and staging was recorded as well as the presence (or not) of measurable or evaluable disease. Information on co-morbidities, prior history of malignancy and laboratory tests, including full blood count, liver function tests and creatinine, were recorded. Any previous treatment for lung cancer was doc- umented and Eastern Co-operative Oncology Group (ECOG) performance status (PS) was evaluated. Physi- cians were asked to estimate patient's life expectancy as less than or greater than 12 weeks, to document weight loss of greater than 10% in the past 6 weeks, and to assess whether the patient was capable of giving informed con- sent. Patients seen by more than one service were only counted once. Although the large chemotherapy versus BSC trials, and the more recent combination chemotherapy trials, reported a significant benefit from chemotherapy in both survival and quality of life in advanced NSCLC, all were performed in relatively young, highly selected groups of patients. Some of the more influential recent studies performed for patients with NSCLC have included (i) European Co- operative Oncology Group (ECOG) E1594[4], (ii) South Western Oncology Group SWOG 9509[5] and (iii) TAX 326[6]. The ECOG 1594 trial compared paclitaxel and cis- platin with the following regimens: gemcitabine and cis- platin; docetaxel and cisplatin; or paclitaxel and carboplatin in 1155 patients. There was no difference in response rate (19%), median survival (7.9 months) or 1- year survival (33%) between the four arms. Time to pro- gression was significantly better in the gemcitabine/cispl- atin arm but there was greater renal toxicity. Initially patients with ECOG PS 2 were eligible, but the criteria were changed to exclude them after it became evident that patients with PS 2 had a significantly higher rate of adverse events than PS 0–1. Median survival in patients with PS 0 was 10.8 months, PS 1 7.1 months and in PS 2 3.9 months (p < 0.001)[4]. Methods All patients referred to the Sydney Cancer Centre are reg- istered in a data base. This cancer registry was reviewed to determine how many patients with advanced NSCLC had been referred to the service during this period. A retrospec- tive review of the medical records of patients not evalu- ated prospectively was undertaken to obtain as much information as possible: the hospital and oncology medi- cal records were reviewed, the individual specialist was contacted and the hospital electronic pathology, labora- tory and imaging results were examined. SWOG 9509 enrolled 406 patients with advanced NSCLC, all with ECOG performance status 0–1. Patients were ran- domised to receive either vinorelbine and cisplatin, or paclitaxel and carboplatin. There was no difference in median survival, 1 or 2 year survival or quality of life[5]. TAX326 randomised 1218 patients to receive either docetaxel and cisplatin, docetaxel and carboplatin or vinorelbine and cisplatin. Patients receiving docetaxel and cisplatin had a median survival of 11.3 months compared to 10.1 months for vinorelbine/cisplatin (p = 0.044), with no significant difference in median survival between vinorelbine/cisplatin and docetaxel/carboplatin, but poorer quality of life in the vinorelbine arm [6]. Each patient's data were then reviewed to determine whether or not they met the eligibility criteria for each of the following trials: (i) ECOG E1594, (ii) SWOG 9509 and (iii) TAX 326. Background g There is level one evidence that patients with stage IIIB and IV non-small cell lung cancer (NSCLC) and good per- formance status benefit from chemotherapy, both in terms of survival and improved quality of life. A meta- analysis by the Non-Small Cell Lung Cancer Collaborative Group, comparing best supportive care (BSC) with BSC Page 1 of 6 (page number not for citation purposes) Page 1 of 6 (page number not for citation purposes) BMC Cancer 2009, 9:130 http://www.biomedcentral.com/1471-2407/9/130 http://www.biomedcentral.com/1471-2407/9/130 Cancer Centre (excluding the Palliative Care department) would have been eligible for these recent randomised tri- als of chemotherapy and to document the reasons for ineligibility. the chemotherapy arm compared with 4.8 months in the BSC arm (p = 0.03)[2]. Quality of life improved from baseline to six weeks in the chemotherapy arm and dete- riorated in the BSC arm (p = 0.007)[2]. A systematic review by Clegg found that the chemotherapeutic agents paclitaxel, docetaxel, gemicitabine and virorelbine improved survival in advanced NSCLC by 2–4 months compared to BSC, without compromising quality of life and that it was cost effective. [3] the chemotherapy arm compared with 4.8 months in the BSC arm (p = 0.03)[2]. Quality of life improved from baseline to six weeks in the chemotherapy arm and dete- riorated in the BSC arm (p = 0.007)[2]. A systematic review by Clegg found that the chemotherapeutic agents paclitaxel, docetaxel, gemicitabine and virorelbine improved survival in advanced NSCLC by 2–4 months compared to BSC, without compromising quality of life and that it was cost effective. [3] Page 2 of 6 (page number not for citation purposes) Results A total of 199 patients with advanced NSCLC were regis- tered in the Sydney Cancer Centre data base between Octo- ber 2001 and December 2002. Prospective questionnaires were obtained for 100 patients and data from retrospective chart review for an additional 85 patients. There were a fur- ther 14 patients who were registered but for whom no evi- dence could be found of their attending Sydney Cancer Centre or of having investigations during this period. The major reasons for ineligibility are outlined in Table 4. Seventy two (39%) were ineligible for all three trials due to a performance status of 2 or worse. This included 39 patients who were ECOG performance status 2; of whom only 15/39 (38%) met all other eligibility criteria for SWOG 9509 or TAX326. Patient characteristics are shown in table 1. The median age of the entire cohort was 68 years (age range 32–88 years) with 77% males and 23% females. Sixty-six patients (36%) had Stage IIIB disease, and 119 (64%) had stage IV disease. The most common histological diagnoses were: large cell 78 (42%), adenocarcinoma 67 (36%) and squa- mous cell carcinoma 23 (12%). Seventy-five patients (40.5%) would have been excluded from SWOG 9509 and TAX326 for co-morbidities, most commonly severe or uncontrolled cardiac or pulmonary disease. gations, it was assumed that they had been a non-attendee and they were excluded from the data set. gations, it was assumed that they had been a non-attendee and they were excluded from the data set. Using the combined prospective and retrospective data set 64/185 (34%) patients were eligible for E1594, 53 (28%) for SWOG 9509 and 52 (28%) for TAX326: this included 3 patients for whom there were inadequate data to deter- mine eligibility. (Table 2) Many patients were ineligible on more than one criterion. In the E1594 study 71 (38%) patients, and in the SWOG 9509 study 75 (40.5%), were ineligible on at least two criteria. (Table 3). Statistical methods Descriptive methods were used to summarise demo- graphic characteristics. Analysis involved simple summary statistics, performed on Excel. Analysis of the prospective and retrospective data are presented separately as well as combined. Where information was unobtainable retro- spectively (e.g. life expectancy) a conservative estimate was assumed. Where no evidence could be found that a patient had attended any appointment or had any investi- The highly selected patients treated in these clinical trials raises doubts about the applicability of their results to other patients with lung cancer. We therefore designed a questionnaire to determine how many of the total pool of patients with advanced NSCLC presenting to the Sydney Page 2 of 6 (page number not for citation purposes) Page 2 of 6 (page number not for citation purposes) BMC Cancer 2009, 9:130 http://www.biomedcentral.com/1471-2407/9/130 http://www.biomedcentral.com/1471-2407/9/130 http://www.biomedcentral.com/1471-2407/9/130 gations, it was assumed that they had been a non-attendee and they were excluded from the data set. * retrospective patients were added to the Medical Oncology department referral base # Blood abnormalities leading to exclusion from SWOG lung trial * retrospective patients were added to the Medical Oncology department referral base Bl d b l l d l f SWOG l l * retrospective patients were added to the Medical Oncology departmen # Blood abnormalities leading to exclusion from SWOG lung trial Results (Table 4) increasingly complex and expensive treatment regimens, and it is important that such therapies be evaluated in the l i C l d i d h h Table 2: Eligibility for E1594 and SWOG9509 Prospective Data N = 100 (%) Retrospective Data N = 85 (%) Entire Cohort N = 185 (%) E1594: Ineligible 61 (61%) 60 (70.5%) 121 (65%) Eligible 39 (39%) 22 (26%) 61 (33%) Not assessable 0 (0%) 3 (3.5%) 3 (2%) SWOG 9509 Ineligible 66 (66%) 66 (78%) 132 (71%) Eligible 34 (34%) 16 (19%) 50 (27%) Not assessable 0 (0%) 3 (3.5%) 3 (2%) Table 2: Eligibility for E1594 and SWOG9509 ria and would have excluded 50 (27%) patients based on their co-morbidities. (Table 4) ria and would have excluded 50 (27%) patients based on their co-morbidities. (Table 4) increasingly complex and expensive treatment regimens, and it is important that such therapies be evaluated in the target population. Consequently, we determined whether the patients enrolled in large chemotherapy trials were typ- ical of the lung cancer patients presenting to cancer centres. Eligibility for the newer molecular targeted therapy trials were not included as these were not widely available at our centre at the time of this study; however a second study is in progress to assess this. Across all studies 21 (11%) patients were excluded due to second malignancies. Fifteen patients were not eligible because of unstable brain metastasis. Some of these patients might have subsequently qualified for inclusion in E1594 after cerebral radiotherapy. Eleven patients would have been excluded from SWOG 9509 on the basis of abnormal blood test results: most commonly elevated creatinine and raised bilirubin. In 16 patients, blood results were not available, as they either hadn't been performed (generally because patients were too unwell to consider chemotherapy) or they were done in private laboratories and not available. For the purposes of determining eligibility these were regarded as being normal. Of the 3 trials whose inclusion/exclusion criteria were evalu- ated, the ECOG-1594 study had the least restrictive criteria and would have excluded 121 (65%) of our NSCLC patients, whilst a further 3 had insufficient data available to determine their eligibility. Results Of the 61 (33%) that would have met E1594 eligibility criteria, 7 patients were estimated to have a life expectancy of less than 12 weeks, 4 had severe concomitant pulmonary disease and eight had weight loss of greater than 10% in the 6 weeks prior to assessment. Ten patients were ineligible due to non-measurable dis- ease and twelve patients because they had received prior chemotherapy. Life expectancy was estimated to be less than 12 weeks in 54 (29%) of patients. This is likely a conservative estimate, as physicians are known to consistently overestimate life expectancy in people with advanced cancer[7], and where life expectancy was not documented we assumed it was greater than 12 weeks. Physicians estimated a life expect- ancy of <12 weeks in 6 patients who met all eligibility crite- ria for E1594, SWOG 9509 and TAX 326. Results The E1594 study has more liberal inclusion crite- Table 1: Patient Characteristics Patient Characteristics Prospective Data n = 100 (%) Retrospective Data n = 85 (%) Total Cohort n = 185 (%) Sex: Males 75 (75%) 68 (80%) 143 (77%) Females 25 (25%) 17 (20%) 42 (23%) Median age (years) 66(32–88) 68 (45–84) 68 (32–88) Mean age 66 67 67 Staging: Stage IIIB (effusion) 37 (37%) 29 (34%) 66 (36%) Stage IV 63 (63%) 56 (66%) 119 (64%) Histological Subtype: Large cell carcinoma 50 (50%) 28 (33%) 78 (42%) Adenocarcinoma 36 (36%) 31 (36%) 67 (36%) Squamous cell carcinoma 10 (10%) 13 (15%) 23 (12%) Bronchoalveolar 2 (2%) 4 (5%) 6 (3%) Undifferentiated carcinoma 2 (2%) 4 (5%) 6 (3%) Mixed 0 (0%) 2 (2%) 2 (1%) Not available 0 (0%) 3 (3.5%) 3 (2%) Measurable disease 88 (88%) 82 (96%) 170 (92%) Evaluable disease 4 (4%) 2 (2%) 6 (3%) Non-measurable disease 8 (8%) 1 (1%) 9 (5%) ECOG Performance Status: ECOG 0 28 (28%) 17 (20%) 45 (24%) ECOG 1 35 (35%) 26 (30.5%) 61 (33%) ECOG 2 23 (23%) 15 (18%) 38 (20.5%) ECOG 3 & 4 14 (14%) 20 (23%) 34 (19%) Not available 0 (0%) 7 (8%) 7 (4%) Estimated life expectancy: <12 weeks 34 (34%) 20 (23.5%) 54 (29%) >12 weeks 66 (66%) 58 (68%) 124 (67%) Not available 0 (0%) 7 (8%) 7 (4%) Weight loss > 10% in 6 weeks 26 (26%) 26 (30.5%) 52 (28%) Weight loss Not available 0 (0%) 10 (12%) 10 (5%) Incapable of giving informed consent 1 (1%) 3 (3.5%) 4 (2%) * retrospective patients were added to the Medical Oncology department referral base # Blood abnormalities leading to exclusion from SWOG lung trial Table 1: Patient Characteristics http://www.biomedcentral.com/1471-2407/9/130 BMC Cancer 2009, 9:130 ria and would have excluded 50 (27%) patients based on their co-morbidities. Discussion The last decade has seen a shift from studies comparing BSC with chemotherapy in patients with advanced NSCLC, to studies comparing different chemotherapy combina- tions, administered either alone or in combination with molecular targeted agents, most commonly targeted against the epidermal growth factor receptor (EGFR) or vas- cular endothelial growth factor (VEGF). Discussion Exami- nation of our lung cancer data base reveals that it is likely that our study obtained data on 93% of patients who had presented to the Sydney Cancer Centre with advanced NSCLC during the study period; although almost half of the data were obtained retrospectively. However, our study likely underestimates the number of patients that would be excluded from advanced NSCLC trials, as the sickest patients and patients with other co-morbidities are often not referred to a Cancer Centre, but instead are man- aged by their family physician, geriatrician or palliative care services with symptomatic treatment only. It is likely that a true population based study would result in an even smaller percent of patients meeting inclusion criteria. mented weight loss of greater than 10% within the preced- ing six weeks: of these patients 12 (6%) were eligible for the E1594 trial and 13 (7%) met eligibility criteria for both the SWOG 9509 and TAX 326 trials. Excluding patients with ECOG PS of 2–4 immediately excluded 72 (39%) of our patients with advanced NSCLC. Of those with a performance status of 2, only 15/39 (38%) would have met all the other eligibility criteria for SWOG 9509 or TAX326. Co-morbidities were a major cause of trial ineligibility: 40.5% in SWOG 9509 and 27% in E1594. With approxi- mately 86% of lung cancers in men and 49% in women being current smokers, or having a significant past history of smoking[10], it is not surprising that there is a high rate of co-morbidities, particularly pulmonary (e.g. chronic obstructive pulmonary disease) and cardiovascular dis- ease. In addition 11% of our patients had a past history of a previous malignancy, excluding non-melanoma skin lesions and cervical cancer in situ. A total of 35 (19%) patients were excluded for having non-measurable dis- ease, previous malignancy and/or prior receipt of chemo- therapy. Of these, 24 (69%) remained ineligible on other grounds, and 11 patients were rendered ineligible on these criteria only, and would not necessarily have been restricted from receiving chemotherapy off trial. Whilst cisplatin-based chemotherapy regimens have been shown to increase survival and quality of life in patients who are eligible for large clinical trials compared to those treated with best supportive care, we need to recognise limitations in generalisability of results of those trials to the broader population with advanced lung cancer. Discussion This has resulted in Weight loss is a poor prognostic factor for patients with advanced NSCLC[8,9] Fifty two (28%) patients had docu- Table 3: Number (%) of Criteria Resulting in Ineligibility for Trials No of Criteria No of Patients E1594 No of Patients SWOG9509 Resulting in Ineligibility Prospective Data N = 100 (%) Retrospective Data N = 85 (%) Entire Cohort N = 185(%) Prospective Data N = 100 (%) Retrospective Data N = 85 (%) Entire Cohort N = 185 (%) 1 32 (32%) 38 (45%) 71 (38%) 34 (34%) 41 (48%) 75 (40.5%) 2 22 (22%) 11 (13%) 33 (18%) 23 (23%) 12 (14%) 35 (19%) 3 6 (6%) 9 (11%) 15 (8%) 7 (7%) 11 (13%) 18 (10%) 4 1 (1%) 2 (2%) 3 (2%) 2 (2%) 2 (2%) 4 (2%) * Eligibility for TAX326 is almost identical to SWOG9509 and so is not listed separately Table 3: Number (%) of Criteria Resulting in Ineligibility for Trials http://www.biomedcentral.com/1471-2407/9/130 BMC Cancer 2009, 9:130 Table 4: Number (%) of Ineligible Patients Based on Trials Exclusion Criteria* E1594 SWOG 9509 Prospective Data N = 100 (%) Retro-spective Data N = 85 (%) Entire Cohort N = 185 (%) Prospective Data N = 100 (%) Retro-spective Data N = 85 (%) Entire Cohort N = 185 (%) Performance Status 37 (37%) 35 (41%) 72 (39%) 37 (37%) 35 (41%) 72 (39%) Co-morbidities 19 (19%) 31 (36%) 50 (27%) 34 (34%) 41 (48%) 75 (40%) Previous History of Cancer 12 (12%) 9 (11%) 21 (11%) 12 (12%) 9 (10.5%) 21 (11%) Brain Metastases 14 (14%) 1 (1%) 15 (8%) 14 (14%) 1 (1%) 15 (8%) Non Evaluable or Non Measurable Disease 9 (9%) 1 (1%) 10 (5%) 9 (9%) 1 (1%) 10 (5%) Abnormal Blood Parameter 2 (2%) 6 (7%) 8 (4%) 2 (2%) 7 (8%) 9 (5%) Previous Chemotherapy 2 (2%) 10 (12%) 12 (6.5%) 2 (2%) 10 (12%) 12 (6.5%) Unable to give Informed Consent 1 (1%) 3 (3.5%) 4 (2%) 1 (1%) 3 (3.5%) 4 (2%) * Does not equal 100% due to many patients being excluded on more than one criterion – see Table 3 Table 4: Number (%) of Ineligible Patients Based on Trials Exclusion Criteria* * Does not equal 100% due to many patients being excluded on more than one criterion – see Table 3 able to prospectively survey consecutive patients. Discussion The median age of our cohort was 68 years, consistent with the median age of presentation of patients with advanced NSCLC, whereas the median age in E1594, SWOG9509 and TAX326 was 61–63 years. It is only recently that we have seen the emergence of studies investigating the effects of chemotherapy in the elderly[11,12], as well as ongoing studies in those with poorer performance status. The major limitation of this study is that it is not popula- tion based. Due to failure of physicians to complete the questionnaire on every new patient presenting to the Syd- ney Cancer Centre with advanced NSCLC, we were not Page 5 of 6 (page number not for citation purposes) Competing interests p g The authors declare that they have no competing interests. References 1. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 ran- domised clinical trials. Non-small Cell Lung Cancer Collabo- rative Group. BMJ (Clinical research ed) 1995, 311(7010):899-909. 1. Chemotherapy in non-small cell lung cancer: a meta-analysis using updated data on individual patients from 52 ran- domised clinical trials. Non-small Cell Lung Cancer Collabo- rative Group. BMJ (Clinical research ed) 1995, 311(7010):899-909. 2. Cullen MH, Billingham LJ, Woodroffe CM, Chetiyawardana AD, Gower NH, Joshi R, Ferry DR, Rudd RM, Spiro SG, Cook JE, et al.: Mitomycin, ifosfamide, and cisplatin in unresectable non- small-cell lung cancer: effects on survival and quality of life. J Clin Oncol 1999, 17(10):3188-3194. ( ) 3. Clegg A, Scott DA, Hewitson P, Sidhu M, Waugh N: Clinical and cost effectiveness of paclitaxel, docetaxel, gemcitabine, and vinorelbine in non-small cell lung cancer: a systematic review. Thorax 2002, 57(1):20-28. ( ) 4. Schiller JH, Harrington D, Belani CP, Langer C, Sandler A, Krook J, Zhu J, Johnson DH: Comparison of four chemotherapy regi- mens for advanced non-small-cell lung cancer. The New Eng- land journal of medicine 2002, 346(2):92-98. 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 6 of 6 (page number not for citation purposes) 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 Publish with BioMed Central and every scientist can read your work free of charge 5. Acknowledgements We would like to thank all the Medical Oncology, Respiratory, Cardiotho- racic surgery and Radiation Oncology physicians and registrars at Sydney Cancer Centre who completed the questionnaires for this study: particu- larly Drs Rohini Sharma and Anthony Joshua. We would like to acknowl- edge the assistance of Dr Ian Tannock in providing an editorial review of the manuscript. The study was presented in part at the ASCO Annual Meeting in 2003 Authors' contributions The pre-publication history for this paper can be accessed here: The pre-publication history for this paper can be accessed here: JV and SJC were responsible for concept and design, acquisition of data, analysis and interpretation of data and writing of the manuscript. RD contributed to acquisi- tion of data. PB and MB contributed to study design and patient recruitment. All authors reviewed the manuscript. http://www.biomedcentral.com/1471-2407/9/130/pre pub http://www.biomedcentral.com/1471-2407/9/130/pre pub Conclusion Our study demonstrates that 65–71% of ANSCLC patients presenting to Sydney Cancer Centre, would not have been Page 5 of 6 (page number not for citation purposes) http://www.biomedcentral.com/1471-2407/9/130 http://www.biomedcentral.com/1471-2407/9/130 BMC Cancer 2009, 9:130 eligible for the major lung cancer trials E1594, SWOG 9509 and TAX356. There is a lack of evidence-based data on which to base treatment decisions in the majority of patients with advanced NSCLC. The optimal treatment for these patients is not known. non-small-cell lung cancer treated with third-generation chemotherapy regimens based on eastern cooperative oncology group data. J Clin Oncol 2005, 23(1):175-183. non-small-cell lung cancer treated with third-generation chemotherapy regimens based on eastern cooperative oncology group data. J Clin Oncol 2005, 23(1):175-183. 9. 9. Dewys WD, Begg C, Lavin PT, Band PR, Bennett JM, Bertino JR, Cohen MH, Douglass HO Jr, Engstrom PF, Ezdinli EZ, et al.: Prognos- tic effect of weight loss prior to chemotherapy in cancer patients. Eastern Cooperative Oncology Group. The American journal of medicine 1980, 69(4):491-497. j f ( ) 10. Parkin DM, Bray FI, Devesa SS: Cancer burden in the year 2000. The global picture. Eur J Cancer 2001, 37(Suppl 8):S4-66. Our recommendations are that future trials need to be developed with more liberal eligibility and/or to target specific subsets of patients within the population with advanced NSCLC to ensure that results are applicable to more patients. 11. Gridelli C: The ELVIS trial: a phase III study of single-agent vinorelbine as first-line treatment in elderly patients with advanced non-small cell lung cancer. Elderly Lung Cancer Vinorelbine Italian Study. The oncologist 2001, 6(Suppl 1):4-7. y g ( pp ) 12. Gridelli C, Maione P, Illiano A, Piantedosi FV, Favaretto A, Bearz A, Robbiati SF, Filipazzi V, Lorusso V, Carrozza F, et al.: Cisplatin plus gemcitabine or vinorelbine for elderly patients with advanced non small-cell lung cancer: the MILES-2P studies. J Clin Oncol 2007, 25(29):4663-4669. y g ( pp ) 12. Gridelli C, Maione P, Illiano A, Piantedosi FV, Favaretto A, Bearz A, Robbiati SF, Filipazzi V, Lorusso V, Carrozza F, et al.: Cisplatin plus gemcitabine or vinorelbine for elderly patients with advanced non small-cell lung cancer: the MILES-2P studies. J Clin Oncol 2007, 25(29):4663-4669. References Kelly K, Crowley J, Bunn PA Jr, Presant CA, Grevstad PK, Moinpour CM, Ramsey SD, Wozniak AJ, Weiss GR, Moore DF, et al.: Rand- omized phase III trial of paclitaxel plus carboplatin versus vinorelbine plus cisplatin in the treatment of patients with advanced non – small-cell lung cancer: a Southwest Oncol- ogy Group trial. J Clin Oncol 2001, 19(13):3210-3218. gy p J ( ) 6. Fossella F, Pereira JR, von Pawel J, Pluzanska A, Gorbounova V, Kaukel E, Mattson KV, Ramlau R, Szczesna A, Fidias P, et al.: Rand- omized, multinational, phase III study of docetaxel plus plat- inum combinations versus vinorelbine plus cisplatin for advanced non-small-cell lung cancer: the TAX 326 study group. J Clin Oncol 2003, 21(16):3016-3024. g p J ( ) 7. Glare P: Clinical predictors of survival in advanced cancer. The journal of supportive oncology 2005, 3(5):331-339. g p J ( ) 7. Glare P: Clinical predictors of survival in advanced cancer. The journal of supportive oncology 2005, 3(5):331-339. j pp gy ( ) 8. Hoang T, Xu R, Schiller JH, Bonomi P, Johnson DH: Clinical model to predict survival in chemonaive patients with advanced j pp gy ( ) 8. Hoang T, Xu R, Schiller JH, Bonomi P, Johnson DH: Clinical model to predict survival in chemonaive patients with advanced
https://openalex.org/W1496801626	https://uwe-repository.worktribe.com/preview/813542/Parry%20and%20Tasker.pdf	English	null	Value and Servitization: Creating Complex Deployed Responsive Services	Strategic change	2,014	cc-by	13,705	1 J.E.L. classification codes: D21 (Firm Behavior); D83 (Search; Learning; Information and Knowledge; Communication; Belief);M21 (Business economics); Z10 Cultural Economics; Economic Sociology; Economic Anthropology: General; B41 (Economic Methodology); E.F.M. classification codes: 760 (Methodological issues) Value and servitization; creating complex deployed 1 responsive services1 2 Glenn Parry 3 Faculty of Business & Law, University of the West of England, Bristol, United Kingdom 4 Paul Tasker 5 School of Applied Sciences, Cranfield University, Cranfield, United Kingdom 6 Correspondence to: 7 Dr Glenn Parry 8 Bristol Business School, 9 University of the West of England, 10 Frenchay Campus 11 Bristol, BS16 1QY 12 United Kingdom 13 e-mail: glenn.parry@uwe.ac.uk 14 One sentence summary: 15 This paper presents a value framework which captures how manufacturing 16 engineering firms are transforming from product to complex service provision 17 where the service is delivered within the customer’s dynamic environment and their 18 ability to capture worth is determined by the success of their customer. 19 Value and servitization; creating complex deployed 1 responsive services1 2 Glenn Parry 3 Faculty of Business & Law, University of the West of England, Bristol, United Kingdom 4 Paul Tasker 5 School of Applied Sciences, Cranfield University, Cranfield, United Kingdom 6 Correspondence to: 7 Dr Glenn Parry 8 Bristol Business School, 9 University of the West of England, 10 Frenchay Campus 11 Bristol, BS16 1QY 12 United Kingdom 13 e-mail: glenn.parry@uwe.ac.uk 14 One sentence summary: 15 This paper presents a value framework which captures how manufacturing 16 engineering firms are transforming from product to complex service provision 17 where the service is delivered within the customer’s dynamic environment and their 18 ability to capture worth is determined by the success of their customer. 19 Value and servitization; creating complex deployed 1 responsive services1 2 Glenn Parry 3 Faculty of Business & Law, University of the West of England, Bristol, United Kingdom 4 Paul Tasker 5 School of Applied Sciences, Cranfield University, Cranfield, United Kingdom 6 Correspondence to: 7 Dr Glenn Parry 8 Bristol Business School, 9 University of the West of England, 10 Frenchay Campus 11 Bristol, BS16 1QY 12 United Kingdom 13 e-mail: glenn.parry@uwe.ac.uk 14 One sentence summary: 15 This paper presents a value framework which captures how manufacturing 16 engineering firms are transforming from product to complex service provision 17 where the service is delivered within the customer’s dynamic environment and their 18 ability to capture worth is determined by the success of their customer. 19 Key points: 20 1. A value framework presents the business models for service transformation 21 which requires managers to consider and capture their value proposition, 22 value realisation and worth capture processes 23 2. A characterisation of a particular form of service, named complex deployed 24 responsive services [CDRS] 25 3. CDRS are delivered in partnership with customers, realised off-site and in the 26 customer’s environment and must be responsive to their demands such that 27 their success determines the success of the provider. 28 1. A value framework presents the business models for service transformation 21 which requires managers to consider and capture their value proposition, 22 value realisation and worth capture processes 23 3. CDRS are delivered in partnership with customers, realised off-site and in the 26 customer’s environment and must be responsive to their demands such that 27 their success determines the success of the provider. 1 Introduction 36 Servitization highlights the trend in which firms seek to gain revenue by offering fuller market 37 packages or bundles of customer-focused combinations of products and services. Many 38 product offers have become commoditised in the eyes of the end user which has led 39 traditional manufacturing firms in particular to pursue extra revenue downstream through 40 services. For many manufacturers the provision of service, previously seen as additional 41 activity (Ren, 2009), would now appear to be a necessity to maintain financial viability (Neely, 42 2008). This change in business focus and strategy brings about new challenges and 43 opportunities. 44 As manufacturers are ‘adding service’ there is a tendency in both literature and practice to 45 treat service as an extension of the manufacturing and engineering knowledge base (Ng et al., 46 2012). However, service and service provision is a very different form of business to 47 manufacture. Manufacturing firms produce a unit and the transformation of materials and 48 equipment undertaken in the production process is normally considered as the value creating 49 activity and the unit of analysis (Slack et al., 2013). The focus of value realisation is at the point 50 of exchange where the unit is sold and worth is captured for the manufacturing firm, usually as 51 money. The customer’s use or consumption activity is frequently seen as separate from the 52 manufacturer’s value creation activity. A focus on exchange as the point of value realisation is 53 reflected in theory as a goods dominant logic (Vargo and Lusch 2004, 2008). 54 Service has proven difficult to define but has been characterised as different to product 55 manufacture (Zeithaml et al., 1985) and the realisation of service value is often presented as 56 simultaneous with its production. A service provider can only create a proposition for a 57 Servitization highlights the trend in which firms seek to gain revenue by offering fuller market 37 packages or bundles of customer-focused combinations of products and services. Many 38 product offers have become commoditised in the eyes of the end user which has led 39 traditional manufacturing firms in particular to pursue extra revenue downstream through 40 services. For many manufacturers the provision of service, previously seen as additional 41 activity (Ren, 2009), would now appear to be a necessity to maintain financial viability (Neely, 42 2008). Value and servitization; creating complex deployed 1 responsive services1 2 Glenn Parry 3 Faculty of Business & Law, University of the West of England, Bristol, United Kingdom 4 Paul Tasker 5 School of Applied Sciences, Cranfield University, Cranfield, United Kingdom 6 Correspondence to: 7 Dr Glenn Parry 8 Bristol Business School, 9 University of the West of England, 10 Frenchay Campus 11 Bristol, BS16 1QY 12 United Kingdom 13 e-mail: glenn.parry@uwe.ac.uk 14 One sentence summary: 15 This paper presents a value framework which captures how manufacturing 16 engineering firms are transforming from product to complex service provision 17 where the service is delivered within the customer’s dynamic environment and their 18 ability to capture worth is determined by the success of their customer. 19 28 Page 1 of 25 29-Aug-14 4. Application of the value framework to a number of business-to-business CDRS 29 has demonstrated its utility in identification and understanding of 30 opportunities for worth capture 31 5. The proposed framework helps firms consider how to avoid value slippage, 32 which is the process where the value creator is unable to capture the worth 33 from their effort. 34 35 4. Application of the value framework to a number of business-to-business CDRS 29 has demonstrated its utility in identification and understanding of 30 opportunities for worth capture 31 5. The proposed framework helps firms consider how to avoid value slippage, 32 which is the process where the value creator is unable to capture the worth 33 from their effort. 34 35 Page 2 of 25 Page 2 of 25 29-Aug-14 1 Introduction 36 This change in business focus and strategy brings about new challenges and 43 opportunities. 44 As manufacturers are ‘adding service’ there is a tendency in both literature and practice to 45 treat service as an extension of the manufacturing and engineering knowledge base (Ng et al., 46 2012). However, service and service provision is a very different form of business to 47 manufacture. Manufacturing firms produce a unit and the transformation of materials and 48 equipment undertaken in the production process is normally considered as the value creating 49 activity and the unit of analysis (Slack et al., 2013). The focus of value realisation is at the point 50 of exchange where the unit is sold and worth is captured for the manufacturing firm, usually as 51 money. The customer’s use or consumption activity is frequently seen as separate from the 52 manufacturer’s value creation activity. A focus on exchange as the point of value realisation is 53 reflected in theory as a goods dominant logic (Vargo and Lusch 2004, 2008). 54 Service has proven difficult to define but has been characterised as different to product 55 manufacture (Zeithaml et al., 1985) and the realisation of service value is often presented as 56 simultaneous with its production. A service provider can only create a proposition for a 57 customer which has potential value as value is only realised when the service is enacted. As 58 service production is simultaneous with its consumption by the customer, customer and 59 supplier firms are proactively involved in the realization of value, a construct described as 60 29-Aug-14 Page 3 of 25 Page 3 of 25 29-Aug-14 being ‘co-opted’ into the design and delivery of services (Prahalad and Ramaswamy, 2000 61 and2003). The competence to create value from service comes from skilful co-ordination of 62 complex resource combinations of products, providers, suppliers and often the customer 63 (Vargo and Lusch 2008; Daliwal et al., 2011; Angelis et al., 2011). Worth may be captured 64 through a fee but payment may be contingent upon the customer realising value from the 65 offer. Therefore the notions of value proposition, realisation and worth capture are different 66 to those of traditional manufacture. 1 Introduction 36 These are the elements of the business model (Baden- 67 Fuller and Morgan, 2010), and past work has suggested that servitization requires a paradigm 68 shift in both the perspective taken by managers and the business model they employ (Barnett 69 et al., 2013). 70 This paper takes a business model perspective and examines the new business models 71 employed by manufacturers following servitization. Through case study analysis this paper 72 identifies and describes three manufacturer engineering business-to-business services using a 73 framework of value proposition, realisation and worth capture. The three examples are for 74 business-to-business services providing engine support services for civil and military aerospace 75 and military ships. They are provided at the global scale and require multiple organisational 76 resources for the service to operate. They illustrate a particular business model as an outcome 77 of servitization as firms transform from sale of an asset to an offer of a use service based on 78 the assured availability of assets. 79 The paper will proceed as follows. First theory to support the case analysis includes the nature 80 of servitization, the issues of unit of analysis, service complexity and a model for value 81 creation. A brief methodology is followed by the three case studies. Discussion of the case 82 studies in light of theory then leads to the conclusion and future work. 83 being ‘co-opted’ into the design and delivery of services (Prahalad and Ramaswamy, 2000 61 and2003). The competence to create value from service comes from skilful co-ordination of 62 complex resource combinations of products, providers, suppliers and often the customer 63 (Vargo and Lusch 2008; Daliwal et al., 2011; Angelis et al., 2011). Worth may be captured 64 through a fee but payment may be contingent upon the customer realising value from the 65 offer. Therefore the notions of value proposition, realisation and worth capture are different 66 to those of traditional manufacture. These are the elements of the business model (Baden- 67 Fuller and Morgan, 2010), and past work has suggested that servitization requires a paradigm 68 shift in both the perspective taken by managers and the business model they employ (Barnett 69 et al., 2013). 70 This paper takes a business model perspective and examines the new business models 71 employed by manufacturers following servitization. 1 Introduction 36 Through case study analysis this paper 72 identifies and describes three manufacturer engineering business-to-business services using a 73 framework of value proposition, realisation and worth capture. The three examples are for 74 business-to-business services providing engine support services for civil and military aerospace 75 and military ships. They are provided at the global scale and require multiple organisational 76 resources for the service to operate. They illustrate a particular business model as an outcome 77 of servitization as firms transform from sale of an asset to an offer of a use service based on 78 the assured availability of assets. 79 The paper will proceed as follows. First theory to support the case analysis includes the nature 80 of servitization, the issues of unit of analysis, service complexity and a model for value 81 creation. A brief methodology is followed by the three case studies. Discussion of the case 82 studies in light of theory then leads to the conclusion and future work. 83 The paper will proceed as follows. First theory to support the case analysis includes the nature 80 of servitization, the issues of unit of analysis, service complexity and a model for value 81 creation. A brief methodology is followed by the three case studies. Discussion of the case 82 studies in light of theory then leads to the conclusion and future work. 83 Page 4 of 25 29-Aug-14 2 Servitization 84 2.1 The Unit of analysis 85 The transition from product manufacture to a focus upon service activity has been named 86 “servitization” (Vandermerwe and Rada, 1988; Matthyssens and Vandembempt, 1988; 87 Anderson & Narus, 1995). There is an issue with regards the unit of analysis when servitization 88 is discussed as although there is a long standing agreement over the definition of 89 products/goods, their characteristics and their production through manufacture, the definition 90 of services has never reached consensus (Parry et al., 2011a). Whilst ‘manufacturer’ frequently 91 forms the start point for a firm’s servitization journey, the end point is varied. 92 The extent of servitization may be conceptualised as reflecting the spectrum of potential 93 service offerings, beginning with a base service offering products and on-going supply of spare 94 parts; intermediate services offering scheduled maintenance and in-field service; and 95 advanced complex services such as customer support or rental type agreements (Baines et al., 96 2009; Baines, et al., 2011a). Neely (2008) identifies five categories of product and service 97 offerings which may result from servitization: Product oriented Product-Service System [PSS[ 98 where ownership of the product is transferred to the customer and product related services 99 are provided; use oriented service systems where ownership of the product is retrained by the 100 provider and the customer purchases use, as in lease arrangements; results oriented PSS 101 where the product may disappear entirely and the customer pays for the result, such as voice 102 messaging; Integration oriented PSS where firms seek to add services by going downstream 103 and vertically integrate, such as when an oil company also sells fuel to customers by operating 104 petrol stations; and service oriented PSS which occur when firms build services into their 105 products, such as intelligent health monitoring systems and their associated services. The 106 ‘direction’ of servitization has further been conceptualised as forwards integration where the 107 focal firm takes over operations of a customer and backwards where they take over operations 108 29-Aug-14 Page 5 of 25 29-Aug-14 of a supplier (Baines et al., 2011b). 2 Servitization 84 Neely notes that these services are conceptualised in the 109 language of goods dominant logic (Vargo and Lusch, 2004) where the focus of value is in the 110 exchange relationship as opposed to on a broader understanding of value as co-created with, 111 and for, the parties engaging in the activity (Vargo and Lusch, 2008). 112 of a supplier (Baines et al., 2011b). Neely notes that these services are conceptualised in the 109 language of goods dominant logic (Vargo and Lusch, 2004) where the focus of value is in the 110 exchange relationship as opposed to on a broader understanding of value as co-created with, 111 and for, the parties engaging in the activity (Vargo and Lusch, 2008). 112 of a supplier (Baines et al., 2011b). Neely notes that these services are conceptualised in the 109 language of goods dominant logic (Vargo and Lusch, 2004) where the focus of value is in the 110 exchange relationship as opposed to on a broader understanding of value as co-created with, 111 and for, the parties engaging in the activity (Vargo and Lusch, 2008). 112 2.2 Complex deployed responsive services 113 As firms have specialised and focussed on development of their own core competences to 114 create and deliver services they must collaborate with partner firms (Mills et al, 2012). This 115 adds to the complexity of multi-organisational service and raises a particular challenge for 116 managers attempting to co-ordinate the resources employed to deliver the outcome of a 117 service, as they must take a holistic approach, seeing beyond the individual business units and 118 company structures and manage the whole system. The lead provider organization must 119 impose a holistic management perspective on a complex system of interconnected and 120 interdependent activities undertaken by a diverse network of stakeholders (Purchase et al., 121 2011a). It is this enterprise that in the end delivers the service experience. 122 Complex deployed responsive services [CDRS] are a particular form of engineering service 123 where the service is primarily based not in the provider firm, but out in the customers 124 operating environment (Parry et al., 2011). CDRS have been characterised by recognition of 125 three core interrelated business challenges: geographic coverage, customer demand, meeting 126 demand. These three characteristics were identified during analysis of business to consumer 127 services and a single, relatively simple, global aviation field repair service. 2 Servitization 84 128 The first challenge relates to the provision of geographic coverage such that the service is able 129 to be in the correct location when required. Depending upon the service offered this may be 130 local, national, regional or global Organisations typically divide their geographic area into 131 zones depending upon the scale of the second challenge, customer demand (Parry et al., 132 2011). Customer demand is challenging for firms new to this service provision as to predict 133 p p y p As firms have specialised and focussed on development of their own core competences to 114 create and deliver services they must collaborate with partner firms (Mills et al, 2012). This 115 adds to the complexity of multi-organisational service and raises a particular challenge for 116 managers attempting to co-ordinate the resources employed to deliver the outcome of a 117 service, as they must take a holistic approach, seeing beyond the individual business units and 118 company structures and manage the whole system. The lead provider organization must 119 impose a holistic management perspective on a complex system of interconnected and 120 interdependent activities undertaken by a diverse network of stakeholders (Purchase et al., 121 2011a). It is this enterprise that in the end delivers the service experience. 122 Complex deployed responsive services [CDRS] are a particular form of engineering service 123 where the service is primarily based not in the provider firm, but out in the customers 124 operating environment (Parry et al., 2011). CDRS have been characterised by recognition of 125 three core interrelated business challenges: geographic coverage, customer demand, meeting 126 demand. These three characteristics were identified during analysis of business to consumer 127 services and a single, relatively simple, global aviation field repair service. 128 Complex deployed responsive services [CDRS] are a particular form of engineering service 123 where the service is primarily based not in the provider firm, but out in the customers 124 operating environment (Parry et al., 2011). CDRS have been characterised by recognition of 125 three core interrelated business challenges: geographic coverage, customer demand, meeting 126 demand. These three characteristics were identified during analysis of business to consumer 127 services and a single, relatively simple, global aviation field repair service. 128 The first challenge relates to the provision of geographic coverage such that the service is able 129 to be in the correct location when required. 2 Servitization 84 Depending upon the service offered this may be 130 local, national, regional or global Organisations typically divide their geographic area into 131 zones depending upon the scale of the second challenge, customer demand (Parry et al., 132 2011). Customer demand is challenging for firms new to this service provision as to predict 133 The first challenge relates to the provision of geographic coverage such that the service is able 129 to be in the correct location when required. Depending upon the service offered this may be 130 local, national, regional or global Organisations typically divide their geographic area into 131 zones depending upon the scale of the second challenge, customer demand (Parry et al., 132 2011). Customer demand is challenging for firms new to this service provision as to predict 133 Page 6 of 25 29-Aug-14 29-Aug-14 Page 6 of 25 likely demands require knowledge of the variables which drive demand. The third challenge, 134 meeting demand, requires processes of communication such that the specific service 135 requirement of the customer can be forecast and captured efficiently. Having captured the 136 requirement the most appropriate resources must be deployed to address that specific 137 demand. Managing customer demand becomes easier with time as a record of likely demand 138 linked to environmental factors becomes established. For example, in the UK, the Royal 139 Automobile Club (RAC) provides a national breakdown recovery service for cars. Through 140 analysis of data they recognise that factors such as sporting events, national holidays, time of 141 day and particularly weather are key drivers of demand. By establishing variables for analysis 142 allows prediction of likely demand that enables better demand planning. Further, common 143 failure modes may be captured along with the likely way customers experience and 144 communicate that failure. This knowledge allows for appropriate resources to meet demand 145 are deployed. Over time, if complex services can be learning organisations, they are able to 146 exploit their knowledge to become efficient and increasingly cost effective and competitive. 147 2.3 Challenges of Complexity 148 One of the key challenges identified involves understanding and managing the complexity 149 experienced in multi-organisational service enterprises (Purchase et al., 2011b). 2 Servitization 84 The term 150 complexity is frequently used but is resistant to clear definition and measurement (Foley, 151 1996; Murmann, 1994; Pighin, 1998; Kim and Wilemon, 2003; Schlick et al., 2007) and there is 152 resistance to clarification of the term if it involves simplification of the concept (Elliot and Kiel, 153 1997; Cilliers, 1998). Complex systems are non-linear, they do not necessarily act in a 154 mechanical way and give outcomes that are sensitive to the initial conditions (Kao 1997). 155 Typically there is a disconnect between the behaviour observed locally and the whole system 156 level behaviour which can lead to system level outcomes which can be counterintuitive, 157 named emergence (Bonabeau, 2003). 158 likely demands require knowledge of the variables which drive demand. The third challenge, 134 meeting demand, requires processes of communication such that the specific service 135 requirement of the customer can be forecast and captured efficiently. Having captured the 136 requirement the most appropriate resources must be deployed to address that specific 137 demand. Managing customer demand becomes easier with time as a record of likely demand 138 linked to environmental factors becomes established. For example, in the UK, the Royal 139 Automobile Club (RAC) provides a national breakdown recovery service for cars. Through 140 analysis of data they recognise that factors such as sporting events, national holidays, time of 141 day and particularly weather are key drivers of demand. By establishing variables for analysis 142 allows prediction of likely demand that enables better demand planning. Further, common 143 failure modes may be captured along with the likely way customers experience and 144 communicate that failure. This knowledge allows for appropriate resources to meet demand 145 are deployed. Over time, if complex services can be learning organisations, they are able to 146 exploit their knowledge to become efficient and increasingly cost effective and competitive. 147 2.3 Challenges of Complexity 148 Page 7 of 25 Page 7 of 25 29-Aug-14 29-Aug-14 Complex services are challenging for managers as they may make local changes in good faith 159 expecting coherent system level changes to occur and yet experience the opposite effect. 160 Management of complex services requires organisational structures which are able to provide 161 rigour to operational processes in order to maintain control, yet also remain flexible enough to 162 enable managers to respond to and address unexpected issues (Schuh et al. 2008). 2 Servitization 84 In the extant literature, the emergent 184 deviations to a proposed business model are largely ignored as the business moves from 185 formulation to implementation – (Demil and Lecocq, 2010). It is proposed that the business 186 model is the sum of three interacting elements: the value proposition, value realization and 187 worth capture. 188 Value has been ascribed many meanings and this work will follow Bowman and Ambrosini 189 (2000) who provide a definition which spans many interpretations and proposes that value is 190 the perception of how ‘good’ something is within a situated context. Value is not a naturally 191 occurring property, but is determined by how it is perceived (Ng et al., 2010). The process of 192 value creation operates across and between the individual, organization and society (Lepak et 193 al., 2007). It is proposed that there are three parts to the value creation process which are; 194 creating a value proposition, value realisation, worth capture (O’Cass and Ngo, 2011; 195 Osterwalder and Pigneur, 2010). The authors have arranged the value elements into a 196 framework , figure 1, which presents the three facets of the business model interacting to 197 form the value creation process. 198 further problems arise - Edelman and Yli-Renko (2010). In the extant literature, the emergent 184 deviations to a proposed business model are largely ignored as the business moves from 185 formulation to implementation – (Demil and Lecocq, 2010). It is proposed that the business 186 model is the sum of three interacting elements: the value proposition, value realization and 187 worth capture. 188 Value has been ascribed many meanings and this work will follow Bowman and Ambrosini 189 (2000) who provide a definition which spans many interpretations and proposes that value is 190 the perception of how ‘good’ something is within a situated context. Value is not a naturally 191 occurring property, but is determined by how it is perceived (Ng et al., 2010). The process of 192 value creation operates across and between the individual, organization and society (Lepak et 193 al., 2007). It is proposed that there are three parts to the value creation process which are; 194 creating a value proposition, value realisation, worth capture (O’Cass and Ngo, 2011; 195 Osterwalder and Pigneur, 2010). 2 Servitization 84 Managers 163 must understand the system when it is under control (Taylor and Tofts, 2009) and develop the 164 ability to respond to emergence, coping with both environmental, task and customer 165 requirement changes. 166 Complex services are challenging for managers as they may make local changes in good faith 159 expecting coherent system level changes to occur and yet experience the opposite effect. 160 The focus of study for this paper is that of manufactures moving to offer service to support an 168 asset and deliver a desired outcome. The contracts put in place are generally either for an 169 assured level of asset availability in service, or are designed to deliver an outcome for the 170 customer. It is proposed that the creation of value through service is different to that of 171 manufacture, due to the level of “co-opted” resource across the extended enterprise, and so a 172 different business model is required. 173 Business models narrate the business operation and describe the structure and strategy 174 employed by a firm to differentiate themselves and compete (Magretta, 2002). Many authors 175 make the link between business models and value creation. Zott et al (2011) propose that 176 business models are the descriptors of value creation. Business models are described by 177 Baden-Fuller and Morgan (2010) as the process of customer engagement with a product or 178 service, specifically focussing on how value is created and worth value is captured sufficient 179 that the firm can achieve greater returns. Business model innovation is considered as the 180 reconfiguring the firm’s capabilities to increase value capture (Sabatier et al 2010). 181 Baden Fuller and Morgan (2010) state that over 66% of firms have not given thought to their 182 Baden-Fuller and Morgan (2010) state that over 66% of firms have not given thought to their 182 business model and cannot articulate it. In addition, if the focus is incorrect or changes, then 183 Baden-Fuller and Morgan (2010) state that over 66% of firms have not given thought to their 182 business model and cannot articulate it. In addition, if the focus is incorrect or changes, then 183 29-Aug-14 Page 8 of 25 29-Aug-14 Page 8 of 25 further problems arise - Edelman and Yli-Renko (2010). 2 Servitization 84 212 The value Realisation occurs when the proposition is enacted for the benefit of a customer. 213 The proposition may be a product or services, but the proposition does not create value until 214 the customer uses it, integrating the proposition into their enterprise to realise value. Value is 215 determined by the cost and timing of deployment of resource and is realized through the 216 outcomes achieved through the process of the application of the resource base for a stated 217 benefit (Zott, 2003). Value realisation occurs in the specific context of resource use by and for 218 the benefit of the customer firm. 219 Worth Capture is the ability of both providers and customers to capture worth following the 220 realisation of the value of a proposition. Worth is usually the monetary exchange; the focus of 221 good dominant logic (Vargo and Lusch, 2004). Sustaining value creation depends upon the 222 producer capturing value sufficient to exceed costs and the amount is determined by the user 223 as a function of their perception of their increased benefit compared to alternates (Lepak et 224 al., 2007). Without these antecedents, the user will not engage in future value realisation and 225 The value Realisation occurs when the proposition is enacted for the benefit of a customer. 213 The proposition may be a product or services, but the proposition does not create value until 214 the customer uses it, integrating the proposition into their enterprise to realise value. Value is 215 determined by the cost and timing of deployment of resource and is realized through the 216 outcomes achieved through the process of the application of the resource base for a stated 217 benefit (Zott, 2003). Value realisation occurs in the specific context of resource use by and for 218 the benefit of the customer firm. 219 Worth Capture is the ability of both providers and customers to capture worth following the 220 realisation of the value of a proposition. Worth is usually the monetary exchange; the focus of 221 good dominant logic (Vargo and Lusch, 2004). Sustaining value creation depends upon the 222 producer capturing value sufficient to exceed costs and the amount is determined by the user 223 as a function of their perception of their increased benefit compared to alternates (Lepak et 224 al., 2007). 2 Servitization 84 The authors have arranged the value elements into a 196 framework , figure 1, which presents the three facets of the business model interacting to 197 form the value creation process. 198 Value has been ascribed many meanings and this work will follow Bowman and Ambrosini 189 (2000) who provide a definition which spans many interpretations and proposes that value is 190 the perception of how ‘good’ something is within a situated context. Value is not a naturally 191 occurring property, but is determined by how it is perceived (Ng et al., 2010). The process of 192 value creation operates across and between the individual, organization and society (Lepak et 193 al., 2007). It is proposed that there are three parts to the value creation process which are; 194 creating a value proposition, value realisation, worth capture (O’Cass and Ngo, 2011; 195 Osterwalder and Pigneur, 2010). The authors have arranged the value elements into a 196 framework , figure 1, which presents the three facets of the business model interacting to 197 form the value creation process. 198 199 Figure 1. The three facets of value creation in business models 200 199 Figure 1. The three facets of value creation in business models 200 Page 9 of 25 Page 9 of 25 29-Aug-14 29-Aug-14 The value Proposition is the system of valued resource necessary to deliver the purpose of the 201 enterprise and includes materials and equipment, people, information and knowledge (Ireland, 202 Hitt, and Sirmon, 2003; Ng et al., 2011). From a resource based perspective the firm creates its 203 value offering based upon the resources which it is able to coordinate. A portfolio of 204 potentially valuable resources does not mean that a firm can create value (Barney & Arikan, 205 2001; Priem & Butler, 2001). The resources under a firms control are defined as the resource 206 portfolio and the maximum value creating potential of the firm is defined by its portfolio 207 (Maddock, 2003).The value proposition cannot be offered and delivered in all potential 208 contexts. The firm is limited in the number of resources which it may employ and so it is 209 limited as to the value it may offer. Vargo and Lusch (2004, 2008) propose that all propositions 210 (or offerings) are service offerings, where the word service reflects the process of using 211 resource for the benefit of another entity. 2 Servitization 84 Without these antecedents, the user will not engage in future value realisation and 225 Worth Capture is the ability of both providers and customers to capture worth following the 220 realisation of the value of a proposition. Worth is usually the monetary exchange; the focus of 221 good dominant logic (Vargo and Lusch, 2004). Sustaining value creation depends upon the 222 producer capturing value sufficient to exceed costs and the amount is determined by the user 223 as a function of their perception of their increased benefit compared to alternates (Lepak et 224 al., 2007). Without these antecedents, the user will not engage in future value realisation and 225 29-Aug-14 Page 10 of 25 29-Aug-14 exchanges, making the business unsustainable. Lepak et al. (2007) use the term value slippage 226 to describe the situation when the value creator is unable to capture worth. Those who create 227 value may find that other individuals, organisations or society benefits more from their efforts 228 than they do. Slippage acts to disincentivize long term value creation. 229 exchanges, making the business unsustainable. Lepak et al. (2007) use the term value slippage 226 to describe the situation when the value creator is unable to capture worth. Those who create 227 value may find that other individuals, organisations or society benefits more from their efforts 228 than they do. Slippage acts to disincentivize long term value creation. 229 3 Research Methodology 230 The research uses case studies to capture the business models from three complex deployed 231 services offered by engineering firms. Two of the cases pertain to the military domain, aero 232 engines and surface ships and the third to civilian commercial aero engines. The cases were 233 produced by the senior managers from the firms involved in providing the services through a 234 method of co-operative enquiry (Heron, 1966). A workshop was held where the theory of the 235 business model and the value framework was explained and materials giving details of the 236 theories from literature provided. Guided by the theory the managers then created case 237 materials, providing background on the context of the service and detailed operational 238 information on the three service value elements: production, realisation and worth capture. 239 The reports all contained KPIs and an Enterprise Image (Mills et al., 2012), a method for 240 creating a visual depiction of a service enterprise. The image helped to show the organisational 241 resources and business units employed in creating the service and acknowledge both client 242 and service provider roles in enabling behaviours that promote value co-creation (Vargo & 243 Lusch, 2008). Due to commercial sensitivity it is not possible to show images in this paper. 244 Once complete the cases were presented back to the group and scrutinized in a workshop. The 245 authors then codified the case studies and documented them here. 246 Page 11 of 25 29-Aug-14 4 Complex Deployed Responsive Service Case Studies 247 The traditional view of the business model of all the engineering firms was one of manufacture 248 of a unit, undertaken within the firm’s facility with contribution from suppliers. With regards 249 power units, once the unit was complete the equipment was transferred to the business 250 contracted to manufacture the platform and installed. Ownership was transferred to the 251 customer and value for the unit realised at the point of exchange. Financial reward was given 252 upon delivery and installation of the power unit. Following a process of servitization the case 253 study firms now offer a number of different services in support of their assets. Three of these 254 complex services are now described. 255 4 Complex Deployed Responsive Service Case Studies 247 4 Complex Deployed Responsive Service Case Studies 247 The firm is a provider of civil aviation engines to the airline industry. They have a traditional 257 business model of asset sales and aftermarket support services with spares sales but have 258 been one of the first major engineering firms to engage in servitization. The EHM service is 259 offered as part of a service package to large civil airlines to enable them to gain most benefit 260 from the assets under control. 261 The Value Proposition in EHM is achieved by turning aircraft data into information and then 262 communicating that information to the correct person in the customer organisation in a timely 263 manner. The EHM service exploits data and seeks to offer value through analysis and 264 monitoring of the resource in operation, effectively allowing the airline access to the 265 knowledge base of the engine OEM. The service is complex as data from assets is complicated 266 and requires processing, the assets are globally dispersed, and responses to the data in terms 267 of advice must be provided quickly to the person capable of acting and with limited false alerts 268 and no missed events. The service value proposition is both proactive and reactive. 269 The reactive service provides a non-intrusive direct warning of impending problems to the 270 Page 12 of 25 29-Aug-14 29-Aug-14 service they offer. When a data trend emerges from the data that is deemed ‘of interest’ and 272 an expert makes a recommendation to the airline to investigate. The action may require the 273 airline, service provider and/or a third party to provide service such as support, logistics, 274 spares etc. 275 service they offer. When a data trend emerges from the data that is deemed ‘of interest’ and 272 an expert makes a recommendation to the airline to investigate. The action may require the 273 airline, service provider and/or a third party to provide service such as support, logistics, 274 spares etc. 275 The proactive service provides suitable information for the operator to understand the 276 operation of their fleet and the general health of the assets under control. This includes 277 provision of data and analytics of their operations, such as any mechanical issues, speed and 278 temperature usage of the asset. 279 Close interaction with the customer base ensures that analysis provided is fit for purpose. 4 Complex Deployed Responsive Service Case Studies 247 Due 280 to the interdependence of the business process success of the service operation requires a 281 strong customer relationship and close relationships with the supply side partners. The 282 enterprise necessarily draws upon business units in both provider and customer organisations 283 as well as third parties for spares, maintenance provision and logistics. Due to the inherent 284 complexity of the value proposition to facilitate management a single service model is offered 285 to the market with minimal bespoke elements. These limits make it difficult to offer the value 286 proposition to all operators in all markets and to maximise worth capture for specific service 287 applications. 288 Value is realized through both proactive and reactive offers. The reactive service facilitates the 289 management of any operational issues ‘in-service’ and in a controlled manner, preventing any 290 unplanned maintenance events. This represents co-created value as the proactive service 291 helps the airline to more efficiently run their operation and hence improve margin. The OEM is 292 able to understand the ‘normal’ operation of the resources at the fleet level, operator level 293 Close interaction with the customer base ensures that analysis provided is fit for purpose. Due 280 to the interdependence of the business process success of the service operation requires a 281 strong customer relationship and close relationships with the supply side partners. The 282 enterprise necessarily draws upon business units in both provider and customer organisations 283 as well as third parties for spares, maintenance provision and logistics. Due to the inherent 284 complexity of the value proposition to facilitate management a single service model is offered 285 to the market with minimal bespoke elements. These limits make it difficult to offer the value 286 proposition to all operators in all markets and to maximise worth capture for specific service 287 applications. 288 Value is realized through both proactive and reactive offers. The reactive service facilitates the 289 management of any operational issues ‘in-service’ and in a controlled manner, preventing any 290 unplanned maintenance events. This represents co-created value as the proactive service 291 helps the airline to more efficiently run their operation and hence improve margin. The OEM is 292 able to understand the ‘normal’ operation of the resources at the fleet level, operator level 293 and individual asset level. 4 Complex Deployed Responsive Service Case Studies 247 Decisions are able to be made 321 is in the operator’s interest to keep the assets flying and earning revenue for the airline. 297 Operators do not react in a consistent manner to the information presented potentially 298 resulting in unplanned disruption. Education is required to ensure appropriate response is 299 made to all levels of information provided. 300 Worth is captured at multiple levels. Primarily financial worth is captured through payment for 301 the service. The service has mutual dependency and both parties benefit from more efficient 302 operations. Disruption costs money to both operator and provider. Engine failures financially 303 cost the operator in terms of aircraft on the ground and the provider in terms of repair costs. 304 Failures also have a potential reputational cost to both companies. The data collected as part 305 of EHM services allows the OEM to build on its knowledge base, increasing their operational 306 awareness and helping them enhance their service offer in the future, potentially capturing 307 worth from additional customers. 308 Worth is captured at multiple levels. Primarily financial worth is captured through payment for 301 the service. The service has mutual dependency and both parties benefit from more efficient 302 operations. Disruption costs money to both operator and provider. Engine failures financially 303 cost the operator in terms of aircraft on the ground and the provider in terms of repair costs. 304 Failures also have a potential reputational cost to both companies. The data collected as part 305 of EHM services allows the OEM to build on its knowledge base, increasing their operational 306 awareness and helping them enhance their service offer in the future, potentially capturing 307 worth from additional customers. 308 4.2 Military Engine Service 309 The firm’s value proposition is a service contract guaranteeing engine availability to air force 310 operators. The operation of the service requires co-operative working in the front office space 311 and also draws upon numerous resources and business units in both provider and customer 312 organisations back office in addition to third party suppliers. There is a service delivery centre 313 manned by both provider and customer personnel, supported by the provider operations 314 centre and their engine overhaul facility. The on-site technical support includes trouble 315 shooting, EHM and technical policy experts. The contracted goal is to keep engines on the 316 aircraft as long as possible. 4 Complex Deployed Responsive Service Case Studies 247 This is not without its challenges, not least that not all events evolve 294 through a ‘standard pattern’. However, over time accumulated knowledge accelerates the 295 identification of issues which is mutually beneficial. Under the terms of the service contract it 296 29-Aug-14 Page 13 of 25 29-Aug-14 Page 13 of 25 29-Aug-14 Page 14 of 25 is in the operator’s interest to keep the assets flying and earning revenue for the airline. 297 Operators do not react in a consistent manner to the information presented potentially 298 resulting in unplanned disruption. Education is required to ensure appropriate response is 299 made to all levels of information provided. 300 Worth is captured at multiple levels. Primarily financial worth is captured through payment for 301 the service. The service has mutual dependency and both parties benefit from more efficient 302 operations. Disruption costs money to both operator and provider. Engine failures financially 303 cost the operator in terms of aircraft on the ground and the provider in terms of repair costs. 304 Failures also have a potential reputational cost to both companies. The data collected as part 305 of EHM services allows the OEM to build on its knowledge base, increasing their operational 306 awareness and helping them enhance their service offer in the future, potentially capturing 307 worth from additional customers. 308 4.2 Military Engine Service 309 The firm’s value proposition is a service contract guaranteeing engine availability to air force 310 operators. The operation of the service requires co-operative working in the front office space 311 and also draws upon numerous resources and business units in both provider and customer 312 organisations back office in addition to third party suppliers. There is a service delivery centre 313 manned by both provider and customer personnel, supported by the provider operations 314 centre and their engine overhaul facility. The on-site technical support includes trouble 315 shooting, EHM and technical policy experts. The contracted goal is to keep engines on the 316 aircraft as long as possible. On-site operations are supported offsite by the firm’s operations 317 centre at their manufacturing and service facilities. The offer proposes more predictable 318 operations, shorter turnaround time and greater asset availability for the customer. 319 Value is realised through the use of serviceable engines. The service is delivered through the 320 service delivery centre situated at the assets operational base. 4 Complex Deployed Responsive Service Case Studies 247 On-site operations are supported offsite by the firm’s operations 317 centre at their manufacturing and service facilities. The offer proposes more predictable 318 operations, shorter turnaround time and greater asset availability for the customer. 319 Value is realised through the use of serviceable engines. The service is delivered through the 320 service delivery centre situated at the assets operational base. Decisions are able to be made 321 4.2 Military Engine Service 309 The firm’s value proposition is a service contract guaranteeing engine availability to air force 310 operators. The operation of the service requires co-operative working in the front office space 311 and also draws upon numerous resources and business units in both provider and customer 312 organisations back office in addition to third party suppliers. There is a service delivery centre 313 manned by both provider and customer personnel, supported by the provider operations 314 centre and their engine overhaul facility. The on-site technical support includes trouble 315 shooting, EHM and technical policy experts. The contracted goal is to keep engines on the 316 aircraft as long as possible. On-site operations are supported offsite by the firm’s operations 317 centre at their manufacturing and service facilities. The offer proposes more predictable 318 operations, shorter turnaround time and greater asset availability for the customer. 319 Value is realised through the use of serviceable engines. The service is delivered through the 320 29-Aug-14 Page 14 of 25 29-Aug-14 Page 14 of 25 rapidly and action may be taken on site upon receipt of technical support from either onsite or 322 back office experts. 323 Worth is captured directly from the money paid to the firm for providing the service. The 324 longitudinal nature of support contracts guarantees long term revenue streams to the 325 provider. However, the contract incentivises the provider to keep the engine on the aircraft. 326 This leads to an increased maintenance burden, which can mean higher costs for the provider 327 and potentially decreases aircraft availability. Efforts are made to deliver zero in-service 328 disruption through review of every in-service event and constant risk management to identify 329 emerging reliability threats and reduce their impact. The aim is to balance engine reliability 330 with maintenance burden to ensure optimum service. Worth is also captured for both provider 331 the air force operator through improved return on capital employed through personnel 332 reduction and redeployment. 4 Complex Deployed Responsive Service Case Studies 247 333 4.3 Warship Propulsion Support 334 The support service seeks to minimize the total cost of ownership across a fleet of warships by 335 providing high levels of operational availability and capability, whilst minimizing the cost of 336 operating the vessels. The naval customer has partnered with an industry consortium to 337 achieve these aims as part of a future service provision. 338 The value proposition is the support of the propulsion system by the multi organisational 339 enterprise from a technical perspective, targeting capability and empowering the system 340 maintainers while providing a cost effective solution. The service will achieve a high level of 341 availability across committed platforms with a reduced level of availability across non 342 committed platforms. It provides for technical support via a helpdesk with both remote and 343 local assistance. Condition Monitoring via analysis of available data informs programme risk, 344 maintenance need and inventory decision making. Knowledge is further transferred via work 345 with training providers. The enterprise that provides the support service is multi- 346 29-Aug-14 Page 15 of 25 29-Aug-14 organisational. The service is provided by a partnered organisation comprising the naval 347 operator and a consortium of manufacturing firms but this necessarily draws upon naval 348 personnel and military support services together with a large number of materials, provision 349 and logistics organisations both commercial and governmental. 350 The value will be realised in use as the improvement in the customers operational 351 performance. This service has yet to be deployed but indicators of value are recognised 352 through KPIs: Availability %, Capability %, timely management of significant issues, and 353 customer satisfaction, though the last element is not quantified. 354 Worth is captured by the organisations through the payments made for the contracted service. 355 Worth capture for the customer is delivered through cost savings in spares supply, overhaul 356 costs, personnel costs and level of operational disruption compared with other programmes/ 357 competitors and is identified and quantified through comparison with calculations of 358 alternative approaches. Savings made as a result of costs lower than a baseline prediction from 359 cost models will be jointly shared with the service provider consortium to incentivise further 360 savings. 361 5 Discussion 362 organisational. 4 Complex Deployed Responsive Service Case Studies 247 The service is provided by a partnered organisation comprising the naval 347 operator and a consortium of manufacturing firms but this necessarily draws upon naval 348 personnel and military support services together with a large number of materials, provision 349 and logistics organisations both commercial and governmental. 350 organisational. The service is provided by a partnered organisation comprising the naval 347 operator and a consortium of manufacturing firms but this necessarily draws upon naval 348 personnel and military support services together with a large number of materials, provision 349 and logistics organisations both commercial and governmental. 350 The value will be realised in use as the improvement in the customers operational 351 performance. This service has yet to be deployed but indicators of value are recognised 352 through KPIs: Availability %, Capability %, timely management of significant issues, and 353 customer satisfaction, though the last element is not quantified. 354 Worth is captured by the organisations through the payments made for the contracted service. 355 Worth capture for the customer is delivered through cost savings in spares supply, overhaul 356 costs, personnel costs and level of operational disruption compared with other programmes/ 357 competitors and is identified and quantified through comparison with calculations of 358 alternative approaches. Savings made as a result of costs lower than a baseline prediction from 359 cost models will be jointly shared with the service provider consortium to incentivise further 360 savings. 361 5 Discussion 362 381 These particular services have been further identified as complex deployed responsive service, 382 previous classified by Parry et al., (2011). These are particularly challenging offerings as they 383 are not undertaken in the providers environment but are rather services which are created 384 primarily ‘out’ in the customers operating environment. From the three cases we can see that 385 the three value elements of the business model have distinct focus and these shall be 386 discussed using the business model value framework; value proposition, value realisation and 387 worth capture. 388 The value propositions of the three case study services are to offer a capability/availability 389 service. Compared to the traditional model of manufacture focussed upon delivery of a 390 manufactured unit, here the unit/asset is still present but the servitized offer is for an 391 operational unit/asset and support for the customer should a problem arise in the use of that 392 asset. Creating the resource base necessary for the service a multi-organisational enterprise is 393 required (Purchase et al., 2011a). 394 proposition. 371 Creation and delivery of the service proposition is further ‘complicated’ by being offered 372 within the context laden operating environment of the customer, which in these cases are 373 global and hence the contracted services are all global in reach. The offerings all rely heavily 374 upon information technology to relay communications of both the data from the engine giving 375 information of the state of equipment’s and the required actions. Data must be transformed 376 into knowledge and then further into advice which is relayed to the customer and supporting 377 facilities to ensure that action is taken, responding rapidly to changing customer context. All 378 three services require a knowledgeable customer and supplier partners to act as partner in 379 supporting and ensuring optimal operation of the asset to deliver desired and contracted 380 levels of capability. This requires transformation of people in terms of training. 381 supporting and ensuring optimal operation of the asset to deliver desired and contracted 380 levels of capability. This requires transformation of people in terms of training. 381 These particular services have been further identified as complex deployed responsive service, 382 previous classified by Parry et al., (2011). 5 Discussion 362 The three case studies describe the current service offer by large manufacturing engineering 363 firms to provide service capability. The servitization of the firms is illustrated by the 364 transformation described by Ng et al. (2012) from a manufacturing organisation transforming 365 materials and equipment to a service provider co-ordinating the simultaneous transformation 366 of materials and equipment, information and people and therefore meets the criteria of 367 complex engineering service systems (Ng et al., 2012). The manufactured asset is still evident 368 for all the services in terms of a power unit, representing the transformation of materials and 369 The three case studies describe the current service offer by large manufacturing engineering 363 firms to provide service capability. The servitization of the firms is illustrated by the 364 transformation described by Ng et al. (2012) from a manufacturing organisation transforming 365 materials and equipment to a service provider co-ordinating the simultaneous transformation 366 of materials and equipment, information and people and therefore meets the criteria of 367 complex engineering service systems (Ng et al., 2012). The manufactured asset is still evident 368 for all the services in terms of a power unit, representing the transformation of materials and 369 Page 16 of 25 Page 16 of 25 29-Aug-14 29-Aug-14 Page 17 of 25 equipment into a functional engine. Provision of that engine is only part of the value 370 proposition. 371 Creation and delivery of the service proposition is further ‘complicated’ by being offered 372 within the context laden operating environment of the customer, which in these cases are 373 global and hence the contracted services are all global in reach. The offerings all rely heavily 374 upon information technology to relay communications of both the data from the engine giving 375 information of the state of equipment’s and the required actions. Data must be transformed 376 into knowledge and then further into advice which is relayed to the customer and supporting 377 facilities to ensure that action is taken, responding rapidly to changing customer context. All 378 three services require a knowledgeable customer and supplier partners to act as partner in 379 supporting and ensuring optimal operation of the asset to deliver desired and contracted 380 levels of capability. This requires transformation of people in terms of training. 5 Discussion 362 The three propositions 398 all require much closer working relationships between the provider firms enterprise, to the 399 extent that their offer is only made to those customers with whom the provider has 400 sufficiently close relationships and trust already exists. 401 The value of the service propositions is realised in their use. In the manufacturing model, due 402 to the simultaneous nature of the delivery of the unit and financial reward, value realisation 403 and worth capture were considered to be simultaneous. The simultaneity of value exchange 404 and worth capture may have led the firms to believe that value was realised within the 405 exchange, which led to a focus on exchange as the source of worth and the construct that the 406 asset or unit of production was inherently valuable. Resources are not inherently valuable and 407 value can only be realised in use and in context (Ng, 2013). In complex deployed service the 408 customer uses the service as part of their dynamic operational context. The services allow the 409 customer firm to achieve the desired outcome through the use of their assets. This is 410 consistent with Lapierre et al. (2008) who describe a hierarchical construct of value where 411 customers realise the value of providers’ propositions in order to achieve higher-level ‘end- 412 states’. Such service propositions are challenging to realise as they operate in the dynamic 413 situated context of the customer’s operational environment. However, the contracted service 414 refocuses the service provider and their partners away from the exchange relationship and 415 onto the value realised in the use of the service. 416 Worth capture was traditionally at the point of exchange, when a customer bought an asset 417 The value proposition is not an extension of the manufacturers offer; rather it is a 395 reconceptualization of the business model. The knowledge required is not an extension of the 396 knowledge base of manufacture (Ng et al., 2012) but rather requires a paradigm shift in the 397 business model and service enterprise required (Barnett et al., 2013). The three propositions 398 all require much closer working relationships between the provider firms enterprise, to the 399 extent that their offer is only made to those customers with whom the provider has 400 sufficiently close relationships and trust already exists. 401 The value of the service propositions is realised in their use. 5 Discussion 362 These are particularly challenging offerings as they 383 are not undertaken in the providers environment but are rather services which are created 384 primarily ‘out’ in the customers operating environment. From the three cases we can see that 385 the three value elements of the business model have distinct focus and these shall be 386 discussed using the business model value framework; value proposition, value realisation and 387 worth capture. 388 The value propositions of the three case study services are to offer a capability/availability 389 service. Compared to the traditional model of manufacture focussed upon delivery of a 390 manufactured unit, here the unit/asset is still present but the servitized offer is for an 391 operational unit/asset and support for the customer should a problem arise in the use of that 392 asset. Creating the resource base necessary for the service a multi-organisational enterprise is 393 required (Purchase et al., 2011a). 394 These particular services have been further identified as complex deployed responsive service, 382 previous classified by Parry et al., (2011). These are particularly challenging offerings as they 383 are not undertaken in the providers environment but are rather services which are created 384 primarily ‘out’ in the customers operating environment. From the three cases we can see that 385 the three value elements of the business model have distinct focus and these shall be 386 discussed using the business model value framework; value proposition, value realisation and 387 worth capture. 388 The value propositions of the three case study services are to offer a capability/availability 389 service. Compared to the traditional model of manufacture focussed upon delivery of a 390 manufactured unit, here the unit/asset is still present but the servitized offer is for an 391 operational unit/asset and support for the customer should a problem arise in the use of that 392 asset. Creating the resource base necessary for the service a multi-organisational enterprise is 393 required (Purchase et al., 2011a). 394 Page 17 of 25 29-Aug-14 29-Aug-14 The value proposition is not an extension of the manufacturers offer; rather it is a 395 reconceptualization of the business model. The knowledge required is not an extension of the 396 knowledge base of manufacture (Ng et al., 2012) but rather requires a paradigm shift in the 397 business model and service enterprise required (Barnett et al., 2013). 5 Discussion 362 In the manufacturing model, due 402 to the simultaneous nature of the delivery of the unit and financial reward, value realisation 403 and worth capture were considered to be simultaneous. The simultaneity of value exchange 404 and worth capture may have led the firms to believe that value was realised within the 405 exchange, which led to a focus on exchange as the source of worth and the construct that the 406 asset or unit of production was inherently valuable. Resources are not inherently valuable and 407 value can only be realised in use and in context (Ng, 2013). In complex deployed service the 408 customer uses the service as part of their dynamic operational context. The services allow the 409 customer firm to achieve the desired outcome through the use of their assets. This is 410 consistent with Lapierre et al. (2008) who describe a hierarchical construct of value where 411 customers realise the value of providers’ propositions in order to achieve higher-level ‘end- 412 states’. Such service propositions are challenging to realise as they operate in the dynamic 413 situated context of the customer’s operational environment. However, the contracted service 414 refocuses the service provider and their partners away from the exchange relationship and 415 onto the value realised in the use of the service. 416 Worth capture was traditionally at the point of exchange, when a customer bought an asset 417 from a firm. The change in worth capture reflects a change in the perception of value of the 418 customer. In the pre-servitization asset purchase the asset was valued. Asset value was 419 Worth capture was traditionally at the point of exchange, when a customer bought an asset 417 from a firm. The change in worth capture reflects a change in the perception of value of the 418 customer. In the pre-servitization asset purchase the asset was valued. Asset value was 419 Worth capture was traditionally at the point of exchange, when a customer bought an asset 417 from a firm. The change in worth capture reflects a change in the perception of value of the 418 customer. In the pre-servitization asset purchase the asset was valued. Asset value was 419 Worth capture was traditionally at the point of exchange, when a customer bought an asset 417 from a firm. 5 Discussion 362 The change in worth capture reflects a change in the perception of value of the 418 customer. In the pre-servitization asset purchase the asset was valued. Asset value was 419 Page 18 of 25 Page 18 of 25 29-Aug-14 29-Aug-14 assessed as an input to the customer process and a decision to purchase or not taken by the 420 customer firm. At the point of purchase exchange value was realised by the seller. The value of 421 the asset in terms of value realisation was not recorded or part of the seller’s asset worth 422 capture, but rather the use of the asset would generate revenue for the provider through sales 423 of spares and servicing only if it failed – a perverse incentive (Bowman and Ambrosini, 2000). 424 In the case studies described the customers and providers have sought to address this 425 anomaly by jointly benefiting from the successful use of the providers assets in the outcome of 426 the customers operation. The KPIs ensure that worth capture is contractually linked to these 427 outcomes. In this way effort to ensure reliability is repaid to the parties who have invested 428 effort, preventing value slippage (Lepak et al., 2007). To ensure that worth is captured the 429 provider has assumed part of the role traditionally held by the customer (Baines et al., 2011). 430 The provider must both integrate their operations into the dynamic context of the customer’s 431 environment and act on their behalf. The provider has had to both align with, and in many 432 instances taken control over, the customers’ performance management activity. This changes 433 the power dynamic in the relationship, from one of buyer/supplier competing for power by 434 seeking to leverage value from each other, to one where both partners empower each other 435 as both have a vested interest in working to achieve a common goal (Cox 1999). 436 6 Summary and future work 437 This paper builds upon previous literature for business models based upon three elements; 438 value proposition, value co-creation and worth capture (O’Cass and Ngo, 2011; Osterwalder 439 and Pigneur, 2010) and develops a framework for value in business models. Through repeated 440 application by industry the value framework has become known as business CPR (Capture, 441 Proposition, Realisation) and helps managers consider the different interacting aspects of their 442 business model. The work presented here was undertaken through a process of co-operative 443 This paper builds upon previous literature for business models based upon three elements; 438 value proposition, value co-creation and worth capture (O’Cass and Ngo, 2011; Osterwalder 439 and Pigneur, 2010) and develops a framework for value in business models. Through repeated 440 application by industry the value framework has become known as business CPR (Capture, 441 This paper builds upon previous literature for business models based upon three elements; 438 value proposition, value co-creation and worth capture (O’Cass and Ngo, 2011; Osterwalder 439 and Pigneur, 2010) and develops a framework for value in business models. Through repeated 440 application by industry the value framework has become known as business CPR (Capture, 441 Proposition, Realisation) and helps managers consider the different interacting aspects of their 442 business model. The work presented here was undertaken through a process of co-operative 443 Proposition, Realisation) and helps managers consider the different interacting aspects of their 442 business model. The work presented here was undertaken through a process of co-operative 443 Page 19 of 25 Page 19 of 25 29-Aug-14 29-Aug-14 enquiry, working with senior managers in the creation of the case studies to help instil in them 444 greater understanding of their business and through the sharing of their knowledge develop 445 and test service theory. The business models studied were all business-to-business service 446 contracts where the proposition was to achieve an outcome in terms of a realised capability or 447 level of service availability set within the customers own dynamic context. 448 The value framework is used to describe the servitization transformation from traditional 449 manufacturing business model to the current endpoint of a complex deployed responsive 450 service (Parry et al., 2011b). The new service offers are understood through the lens of service 451 dominant logic (Vargo and Lusch 2004,2008) and centre on multiple firms working together to 452 co-create value in the use of resources. 6 Summary and future work 437 The services are interdependent and close 453 relationships are required between all parties in the enterprise (Purchase et al., 2011b) before 454 the services can be offered. 455 The case studies have demonstrated the utility of the proposed value framework (Figure 1) as 456 a business model which emphasises the differentiation between value realisation and worth 457 capture allowing servitized manufacturers to more effectively articulate opportunities and 458 competitive advantage. The framework highlights how, through servitization, the new 459 contracted forms have seen the provider taking over some of the traditional roles of the 460 customer (Baines et al., 2011). This has helped balance the power dynamic (Cox, 1999) as 461 efforts to provide efficient service are repaid to the parties who invest value slippage is 462 minimised (Lepak et al., 2007). 463 The case studies have demonstrated the utility of the proposed value framework (Figure 1) as 456 a business model which emphasises the differentiation between value realisation and worth 457 capture allowing servitized manufacturers to more effectively articulate opportunities and 458 competitive advantage. The framework highlights how, through servitization, the new 459 contracted forms have seen the provider taking over some of the traditional roles of the 460 customer (Baines et al., 2011). This has helped balance the power dynamic (Cox, 1999) as 461 efforts to provide efficient service are repaid to the parties who invest value slippage is 462 minimised (Lepak et al., 2007). 463 To summaries the challenges and requirement of CDRS:  Providers co-ordinate the simultaneous transformation of materials and equipment, 465 information (Ng et al., 2012) 466  Knowledge required is not an extension of manufacture (Ng et al., 2012)  Manufacturers require a paradigm shift in the business model to a service enterprise 468 (Barnett et al., 2013). 6 Summary and future work 437 469 Page 20 of 25 29-Aug-14 Page 20 of 25 29-Aug-14  Propositions are challenging to realise as they operate in the dynamic situated context 470 of the customer’s operational environment, as value is realised in use and in context 471 (Vargo and Lusch, 2004; Ng, 2013) 472  Close working relationship are required 473  Services require knowledgeable customer and supplier partners 474  Offerings rely heavily upon IT to transfer asset condition data and advice 475  Contracts must avoid perverse incentives which allow worth capture for activity which 476 doesn’t support value creation (Bowman and Ambrosini, 2000) 477  KPIs ensure that worth capture is contractually linked to desired outcomes 478 479 Further research is necessary to identify the extent to which the value framework for the 480 business model and characterisation of complex deployed responsive service can be 481 generalized to other public/private sector enterprises that are acknowledged to be highly 482 complex in their functioning and also to business-to-consumer case examples. Work should 483 examine the requirement and nature of trust in the relationships between the partners in such 484 complex enterprises, particularly how this evolves as the service propositions mature. This 485 work analyses how business model formulation and implementation impacts on value capture. 486 However, it does not analyze the changes in business models over time, a phenomena known 487 in the literature as business model experimentation (Chesbrough, 2010; McGrath, 2010), 488 analysis of which could provide valuable insight into the creation, adaptation and successful 489 operations management of CDRS. 490 References 491 Anderson, J.C. and Narus, J.A. 1995. Capturing the value of supplementary services”, Harvard 492 Business Review, 73 (1):75-83. 493 Angelis, J., Edson, P. 2011. Shifting from production to service experience based operations, in 494 Macintyre, M., G., Angelis, J. (Eds), Service Design and Delivery, New York: Springer, Chapter 6, 495 83-94. 496 Anderson, J.C. and Narus, J.A. 1995. Capturing the value of supplementary services”, Harvard 492 Business Review, 73 (1):75-83. 493 Angelis, J., Edson, P. 2011. Shifting from production to service experience based operations, in 494 Macintyre M G Angelis J (Eds) Service Design and Delivery New York: Springer Chapter 6 495 Anderson, J.C. and Narus, J.A. 1995. Capturing the value of supplementary services”, Harvard 492 Business Review, 73 (1):75-83. 493 Anderson, J.C. and Narus, J.A. 1995. Capturing the value of supplementary services”, Harvard 492 Business Review, 73 (1):75-83. 493 Angelis, J., Edson, P. 2011. Shifting from production to service experience based operations, in 494 Macintyre, M., G., Angelis, J. (Eds), Service Design and Delivery, New York: Springer, Chapter 6, 495 83-94. 496 Angelis, J., Edson, P. 2011. Shifting from production to service experience based operations, in 494 Macintyre, M., G., Angelis, J. (Eds), Service Design and Delivery, New York: Springer, Chapter 6, 495 83-94. 496 Page 21 of 25 Page 21 of 25 29-Aug-14 Baines, T.S., Lightfoot, H.W. and Kay, J.M. 2009. Servitized manufacture: practical challenges of 497 Baines, T.S., Lightfoot, H.W. and Kay, J.M. 2009. Servitized manufacture: practical challenges of 497 delivering integrated products and services, Proceedings of the Institution of Mechanical 498 i l f i i f 223(9) 20 2 99 Baines, T.S., Lightfoot, H.W. and Kay, J.M. 2009. Servitized manufacture: practical challenges of 497 delivering integrated products and services, Proceedings of the Institution of Mechanical 498 Engineers Part B-Journal of Engineering Manufacture, 223(9): 1207-1215. 499 Baines, T.S., Lightfoot, H.W. and Smart,P. 2011b. Servitization within manufacturing: Exploring the 500 provision of advanced services and their impact on vertical integration, Journal of Manufacturing 501 Technology Management, 22(7):947 - 954 502 Baines, T.S., Lightfoot, H.W. and Swink, M. 2011a. Servitization in action: findings from a study 503 of the extended Caterpillar enterprise, Proceedings of the EUROMA 2011 Conference, July 3- 504 11th Cambridge, paper 0167 505 engineering service availability: Is a paradigm shift in the business model and service 507 enterprise required?, Strategic Change: Briefings in Entrepreneurial Finance , 22 (3-4)145-156 508 Bowman, C., and Ambrosini, V. 2000. References 491 Value creation versus value capture: Towards a coherent 509 definition of value in strategy. British Journal of Management, 11(1): 1–15 510 enterprise required?, Strategic Change: Briefings in Entrepreneurial Finance , 22 (3-4)145-156 508 Bowman, C., and Ambrosini, V. 2000. Value creation versus value capture: Towards a coherent 509 definition of value in strategy. British Journal of Management, 11(1): 1–15 510 Chesbrough, H.W., 2010. Business model innovation: opportunities and barriers. Long Range 511 Planning, 43 (2), 354–363. 512 Cox, A., 1999. Power, value and supply chain management, Supply Chain Management: An 513 International Journal, 4(4)167-175 514 International Journal, 4(4)167-175 514 Daliwal, J.S., Macintyre, M., Parry, G. 2011. Understanding Services and the Customer 515 Response”, in Macintyre, M., G., Angelis, J. (Eds), Service Design and Delivery, New York: 516 Springer,, Chapter 1, 1-19 517 Heron, J. 1996. Cooperative Inquiry: Research into the human condition. London: Sage. 518 Lepak, D., Smith, K.G., Taylor, M.S. 2007. Value creation and value capture: A multilevel 519 perspective, Academy of Management Review, 32(1) 180–194 520 Page 22 of 25 29-Aug-14 Mattyssens, P., Vandenbempt, K. 1998. Creating competitive advantage in industrial services, 521 Journal of Industrial and Business Marketing, 13 (4/5):339-355 522 McGrath, R.G., 2010. Business models: a discovery driven approach. Long Range Planning 43 523 (2), 247–261. 524 Ng, ICL. 2013. Value and Worth: Creating New Markets in the Digital Economy, Cambridge UK : 525 Innovorsa Press 526 Ng, I, Parry, G., Smith, L., Maull, R., Briscoe, G. 2012. Transitioning from a goods-dominant to a 527 service-dominant logic: Visualising the value proposition of Rolls-Royce, Journal of Service 528 Management, 23(3):416 – 439 529 Ng, I., Parry, G., McFarlane, D., Tasker, P. 2011. Towards A Core Integrative Framework For 530 Complex Engineering Service Systems in Ng, I., Parry, G., Wilde, P., McFarlane, D., Tasker, P. 531 (Eds) Complex Engineering Service Systems: Concepts and Research, London: Springer 532 O'Cass, A., Ngo, L. (2011) Examining the Firm's Value Creation Process: A Managerial 533 Perspective of the Firm's Value Offering Strategy and Performance. British Journal of 534 Management, 22(4): 646-671 535 Osterwalder, A. & Pigneur, Y. (2010) Business Model Generation, New Jersey, John Wiley & 536 Sons Inc. 537 Osterwalder, A. & Pigneur, Y. (2010) Business Model Generation, New Jersey, John Wiley & 536 Sons Inc. 537 Parry, G., Newnes, L., Huang, X. 2011a. Goods, Products and Services: what do these terms 538 mean? in Macintyre, M., Parry, G., Angelis, J. References 491 (Eds) Service Design and Delivery, New York: 539 Springer 540 Parry, G., McLening, M., Caldwell, N., Thompson, R. 2011b. Complex Deployed Responsive 541 Service, in Macintyre, M., Parry, G., Angelis, J. (Eds) Service Design and Delivery, New York: 542 Springer 543 Prahalad, C.K. Ramaswamy, V. 2000. Co-opting customer competence, Harvard Business 544 Review, 78(1): 79–87. 545 Page 23 of 25 29-Aug-14 Prahalad, C.K. Ramaswamy, V. 2003. The new frontier of experience innovation, MIT Sloan 546 Management Review, 44(4):12–18. 547 Purchase, V., Parry, G.C., Valerdi, R. Nightingale, D. and Mills, J. 2011a. Enterprise 548 Transformation: what is it, what are the challenges and why are we interested?, Journal of 549 Enterprise Transformation, 1 (1):14-40 550 Purchase, V., Parry, G., Mills, J. 2011b. Service Enterprise Transformation, in Ng, I.C.L., Parry, 551 G.C., Wild, P., Mcfarlane, P., Tasker, P. Complex Engineering Service Systems: Concepts and 552 Research, London :Springer, Chapter 2, pp25-48 553 Ren, G. 2009. Service business development in manufacturing companies: Classification 554 characteristics and implications. PhD Dissertation, University of Cambridge 555 Slack, N., Brandon-Jones, A., Johnston, R. 2013 Operations Management, Pearson: Harlow 556 Vandermerwe, S., Rada, J. 1988. Servitization of Business: Adding Value by Adding Services, 557 European Management Journal, 6(4): 314-324 558 Vargo, S.L., Lusch, R.F. 2008. Service-dominant logic: continuing the evolution, Journal of the 559 Academy of Marketing Science, 36(1): 1-10 560 Zeithaml, V.A., Parasuraman, A. and Berry, L.L. 1985. Problems and strategies in service 561 marketing, Journal of Marketing, 49(2):33-46 562 563 Prahalad, C.K. Ramaswamy, V. 2003. The new frontier of experience innovation, MIT Sloan 546 Management Review, 44(4):12–18. 547 Prahalad, C.K. Ramaswamy, V. 2003. The new frontier of experience innovation, MIT Sloan 546 Management Review, 44(4):12–18. 547 Purchase, V., Parry, G.C., Valerdi, R. Nightingale, D. and Mills, J. 2011a. Enterprise 548 Transformation: what is it, what are the challenges and why are we interested?, Journal of 549 Enterprise Transformation, 1 (1):14-40 550 Purchase, V., Parry, G., Mills, J. 2011b. Service Enterprise Transformation, in Ng, I.C.L., Parry, 551 G.C., Wild, P., Mcfarlane, P., Tasker, P. Complex Engineering Service Systems: Concepts and 552 G.C., Wild, P., Mcfarlane, P., Tasker, P. Complex Engineering Service Systems: Concepts and 552 Research, London :Springer, Chapter 2, pp25-48 553 Vargo, S.L., Lusch, R.F. 2008. Service-dominant logic: continuing the evolution, Journal of the 559 Academy of Marketing Science, 36(1): 1-10 560 Zeithaml, V.A., Parasuraman, A. and Berry, L.L. 1985. References 491 Problems and strategies in service 561 marketing, Journal of Marketing, 49(2):33-46 562 Page 24 of 25 29-Aug-14 Glenn Charles Parry is Associate Professor of Strategy and Operations Management and works 571 with firms to help them achieve higher performance by redesigning their processes and 572 through making strategic decisions based upon data driven analysis. His work spans music, 573 aerospace, automotive, digital media and construction industries. He has written for 574 international journals and edited the books, “Build to Order: The Road to the 5-day Car”, 575 “Complex Engineering Service Systems” and "Service Design and Delivery" - ranked in the IIJ top 576 20 upcoming design books for innovators. 577 Glenn Charles Parry is Associate Professor of Strategy and Operations Management and works 571 with firms to help them achieve higher performance by redesigning their processes and 572 through making strategic decisions based upon data driven analysis. His work spans music, 573 aerospace, automotive, digital media and construction industries. He has written for 574 international journals and edited the books, “Build to Order: The Road to the 5-day Car”, 575 “Complex Engineering Service Systems” and "Service Design and Delivery" - ranked in the IIJ top 576 20 upcoming design books for innovators. 577 578 Paul Tasker is Visiting Professor in Integrated System Design with the University of Kent and 579 Cranfield University where he is also Director, TES Services within the EPSRC National Centre for 580 Through-life Engineering Services. He has an interest in product and service innovation and 581 asset management supporting the development of industrial capability in complex service 582 systems. He is also Chair of the Industrial Advisory Board for WMG’s HAT Project. 583 584 585 Paul Tasker is Visiting Professor in Integrated System Design with the University of Kent and 579 Cranfield University where he is also Director, TES Services within the EPSRC National Centre for 580 Through-life Engineering Services. He has an interest in product and service innovation and 581 asset management supporting the development of industrial capability in complex service 582 systems. He is also Chair of the Industrial Advisory Board for WMG’s HAT Project. 583 584 Page 25 of 25 Page 25 of 25 29-Aug-14
https://openalex.org/W2756609766	https://dro.deakin.edu.au/articles/journal_contribution/Cross-sectional_and_prospective_mediating_effects_of_dietary_intake_on_the_relationship_between_sedentary_behaviour_and_body_mass_index_in_adolescents/20830792/1/files/37087408.pdf	English	null	Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents	BMC public health	2,017	cc-by	8,338	Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents AUTHOR(S) Eloise Fletcher, Karen Lamb, Sarah McNaughton, S P Garnett, D W Dunstan, L A Baur, Jo Salmon PUBLICATION DATE 29-09-2017 HANDLE 10536/DRO/DU:30103171 Downloaded from Deakin University’s Figshare repository Deakin University CRICOS Provider Code: 00113B Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents AUTHOR(S) Eloise Fletcher, Karen Lamb, Sarah McNaughton, S P Garnett, D W Dunstan, L A Baur, Jo Salmon 10536/DRO/DU:30103171 Downloaded from Deakin University’s Figshare repository Deakin University CRICOS Provider Code: 00113B DOI: http://www.dx.doi.org/10.1186/s12889-017-4771-0 ©2017, The Authors Reproduced by Deakin University under the terms of the Creative Commons Attribution Licence Reproduced by Deakin University under the terms of the Creative Commons Attribution Licence Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents Citation: Fletcher, Elly A, Lamb, Karen E, McNaughton, Sarah A, Garnett, Sarah P, Dunstan, David W, Baur, Louise A and Salmon, Jo 2017, Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents, BMC public health, vol. 17, Article number: 751, pp. 1-10. Fletcher, Elly A, Lamb, Karen E, McNaughton, Sarah A, Garnett, Sarah P, Dunstan, David W, Baur, Louise A and Salmon, Jo 2017, Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents, BMC public health, vol. 17, Article number: 751, pp. 1-10. DOI: http://www.dx.doi.org/10.1186/s12889-017-4771-0 Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents Citation: Fletcher, Elly A, Lamb, Karen E, McNaughton, Sarah A, Garnett, Sarah P, Dunstan, David W, Baur, Louise A and Salmon, Jo 2017, Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents, BMC public health, vol. 17, Article number: 751, pp. 1-10. DOI: http://www.dx.doi.org/10.1186/s12889-017-4771-0 Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents DOI: http://www.dx.doi.org/10.1186/s12889-017-4771-0 * Correspondence: elly.fletcher@deakin.edu.au 1Deakin University, Geelong, Australia, Institute for Physical Activity and Nutrition (IPAN), School of Exercise and Nutrition Sciences, Geelong, Australia Full list of author information is available at the end of the article © The Author(s). 2017 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. Cross-sectional and prospective mediating effects of dietary intake on the relationship between sedentary behaviour and body mass index in adolescents Elly A. Fletcher1* , Karen E. Lamb1, Sarah A. McNaughton1, Sarah P. Garnett2, David W. Dunstan1,3,4,5,6,7,8, Louise A. Baur2 and Jo Salmon1 Elly A. Fletcher1* , Karen E. Lamb1, Sarah A. McNaughton1, Sarah P. Garnett2, David W. Dunstan1,3,4,5,6,7,8, Louise A. Baur2 and Jo Salmon1 Downloaded from DRO: http://hdl.handle.net/10536/DRO/DU:30103171 Downloaded from DRO: DRO Deakin Research Online, Deakin University’s Research Repository Deakin University CRICOS Provider Code: 00113B DRO Deakin Research Online, Deakin University’s Research Repository Deakin University CRICOS Provider Code: 00113B Fletcher et al. BMC Public Health (2017) 17:751 DOI 10.1186/s12889-017-4771-0 Background [15, 16]. However, a major limitation of both of these studies was their cross-sectional design which limits causal inference. Whereas, a longitudinal design would allow both the temporal order of associations to be ex- amined and many other aspects of a mediation model to be explored. g Adolescent obesity is a major public health concern. The combined rates of overweight and obesity among adoles- cents have increased over the last two decades world- wide [1]. In the United States, the proportion of obese adolescents has risen from 10.5% in 1988–1994 to 20.6% in 2013–2014 [2]. Australia has experienced similar in- creases with almost one in three adolescents currently overweight or obese [3]. Given that obesity tracks from adolescence to adulthood [4], it is imperative to under- stand the lifestyle risk factors associated with adolescent obesity, particularly prospectively, in order to inform effective interventions. Against this background, the primary aim of the study was to explore both the cross-sectional and prospective mediating effects of the consumption of discretionary foods, SSB and takeaway foods on the association be- tween TV viewing and BMI z-score (zBMI) in a cohort of Australian adolescents, and to examine whether these findings differ when total sedentary time and sedentary bout duration are examined. The secondary aims of the study were to explore the individual associations between the sedentary behaviour variables (TV viewing, sedentary time and sedentary bout duration) and dietary intake vari- ables (discretionary foods, SSB and takeaway foods), as well as their associations with zBMI. Based on the existing evidence, we hypothesised that consumption of discretion- ary foods, SSB and takeaway foods would partially mediate the cross-sectional and prospective association between TV viewing and zBMI, but would not mediate the cross- sectional or prospective association between total seden- tary time and sedentary bouts with zBMI. Sedentary behaviour – defined as any waking behav- iours characterised by low energy expenditure (< 1.5 METS) while in a sitting or reclining posture – has emerged as a new research focus for obesity prevention [5]. High amounts of television (TV) viewing, a common leisure-time sedentary behaviour, during adolescence have both immediate and long-term health conse- quences, including a higher risk of obesity [6]. However, there are inconsistent associations between total time spent in sedentary time [7, 8], or time spent in periods, or ‘bouts’, of sedentary time [9], and indicators of adipos- ity (e.g. BMI, waist circumference) in adolescents. Abstract Background: Cross-sectional evidence suggests TV viewing, but not objectively-measured sedentary time or bouts of sedentary time, is consistently associated with body mass index (BMI) in adolescents. However, it is unclear whether dietary intake is a potential mediator of these relationships. The aim of this study was to explore the cross-sectional and prospective mediating effects of dietary intake on the association of sedentary behaviour with BMI z-score (zBMI) in a cohort of Australian adolescents. Methods: Cross-sectional and prospective analyses were conducted in adolescents aged 12–15 years participating in the 2002/03 (baseline) and 2004/05 (follow-up) Nepean Growing Up Study. The independent variables were television (TV) viewing, an objective measure of total sedentary time and average sedentary bout duration, and the outcome variable zBMI. Using the Sobel-Goodman method with bootstrapping, mediation analyses were conducted examining three dietary components (discretionary foods, sugar- sweetened beverages [SSB] and takeaway foods) as mediators of associations between TV viewing and zBMI (n = 259) and between total sedentary time and average sedentary bout duration with zBMI (n = 140). Results: No significant cross-sectional or prospective total or direct associations were observed for TV viewing, total sedentary time and average sedentary bout duration with zBMI. However, TV viewing was positively associated with consumption of takeaway foods cross-sectionally (β = 0.06; 95% CI 0.01 to 0.12), prospectively at baseline (β = 0.07; 95% CI 0.01 to 0.12) and prospectively at follow-up (β = 0.10; 95% CI 0.04, 0.16), and average sedentary bout duration was inversely associated with SSB consumption both cross-sectionally (β = −0.36; 95% CI -0.69 to −0.02) and prospectively at baseline (β = −0.36; 95% CI -0.70 to −0.02). No mediation effects were identified. Conclusions: TV viewing, total sedentary time and bouts of sedentary time were not associated cross- sectionally or prospectively with adolescents’ zBMI, and three elements of dietary intake (e.g. intake of discretionary foods, SSB and takeaway foods) did not mediate this relationship. The role of dietary intake and sedentary behaviour in relation to adolescent health requires further clarification. Keywords: Television viewing, Sedentary behaviour, Dietary intake, BMI, Adolescents Fletcher et al. BMC Public Health (2017) 17:751 Page 2 of 10 Page 2 of 10 Page 2 of 10 Study design y g In 2002/03 (baseline), 348 adolescents aged 12–13 years participated in the Nepean Kids Growing-Up Study. The adolescents were originally part of a birth cohort study (the “Nepean Study”) which involved 2314 infants born between 1989 and 1990 at the Nepean Hospital (western Sydney, Australia). Written consent was obtained from the parent or guardian and assent from the adolescent. The Ethics Committees of The Children’s Hospital at Westmead and Wentworth Area Health Service gave ethical approval. Full details about the original study and eligibility criteria have been previously published [17]. Briefly, the study involved adolescents attending the clinic at Nepean Hospital, where they had their height and weight measured and completed a questionnaire on their demographics and physical activity levels, and a semi-quantitative food frequency questionnaire (FFQ). Afterwards, adolescents wore an accelerometer for 7 days during all waking hours. In 2004/05, the adoles- cents were recontacted and invited to participate in the follow-up study. In total, 63 adolescents were unable to be contacted or withdrew from the study, leaving 285 adolescents participating at both time points. A systematic review examining whether associations between sedentary behaviour and health outcomes in adolescents were independent of dietary intake found TV viewing, screen time and overall sedentary time were positively related to BMI, independent of dietary intake [14]. The systematic review also identified very few studies had specifically examined the mediating role of dietary intake in the TV viewing and BMI rela- tion; with only two studies (out of the 21 studies identi- fied) exploring this and reporting no mediation effects Background y g One potential behavioural mechanism that could explain why TV viewing has more consistent associa- tions with body mass index (BMI) when compared to total or bouts of sedentary time in adolescents, is an in- crease in energy-dense, nutrient poor foods and sugar- sweetened beverages (SSB). For example, TV viewing has consistently been reported to be associated with a higher energy intake, and an increased consumption of dis- cretionary foods, SSB, and fast food/takeaway foods in adolescents [10], whereas few studies have reported as- sociations between objectively-measures of sedentary time with dietary intake [11, 12]. In addition, no study to date has explored whether prolonged bouts of sedentary time are related to dietary intake among adolescents. The latter is important as studies with adults have shown that, independent of how much total sedentary time is accumulated, those with fewer interruptions in sedentary time (i.e., prolonged bouts) have poorer car- diometabolic health profiles [13]. Independent variables (sedentary behaviour) Adolescents completed a self-report questionnaire on their time spent watching TV (hours/day) on a usual school day (Monday to Friday) and a usual weekend day (Saturday and Sunday). The questionnaire has previously been shown to have good to excellent reliability (per- centage agreement = 70%–99%) [20]. To calculate ave- rage daily hours spent watching TV over a usual week, daily weekday TV hours was multiplied by five and daily weekend TV hours was multiplied by 2, then summed together and divided by seven. Sedentary time was measured objectively by an ActiGraph AM-7164 accelerometer (ActiGraph Inc., Florida). At both time points, adolescents were asked to wear the monitor on their right hip during all waking hours for 7 days, except when bathing, swimming and sleeping. Data were downloaded in 1-min epochs and non-wear time was defined as at least 20 min of zero counts. Sedentary time was defined as all wear-time mi- nutes with an average activity count of ≤100 counts per minute (cpm), and was standardised for wear time using the residual method [21]. Average sedentary bout duration was calculated by summing all uninterrupted minutes ≤100 cpm, and then taking the midpoint of all sedentary bouts that lie on the accumulation curve for each individual [22]. Analyses were limited to partici- pants who had ≥8 h of wear time on ≥3 week days and ≥7 h of wear time on ≥1 weekend day [23]. Covariates The covariates considered for the analyses included age at baseline, sex, maternal education (an indicator for family socioeconomic status), pubertal status and accelerometry-measured moderate-to-vigorous physical activity (MVPA) collected at baseline. Maternal educa- tion was collapsed into three categories: “low” (some secondary education or less); “medium” (completing sec- ondary school, an apprenticeship or technical certifi- cate); and “high” (university or tertiary qualification). Pubertal status was self-assessed using the ‘Tanner Stages’ of breast development and commencement of menses (girls) and pubic hair and genitalia (boys) [25]. For analyses, participants were categorised as early pu- berty, mid-pubertal, late-pubertal and post-pubertal. Accelerometry-measured MVPA at baseline was calcu- lated based on the Freedson accelerometer age-cut points [26] and considered as a covariate for the analyses involving sedentary time and sedentary bouts with zBMI. Mediating variables (dietary intake) Usual dietary intake was measured using a 56-item semi-quantitative FFQ, which was developed based on data from the 1995 Australian National Nutrition Survey [24]. Adolescents were asked to report how often they ate certain foods and beverages (either in times per week or per day) over the previous 7 days. The 8-item frequency response scale was converted to times per week as follows: 1) “not consumed last week” = 0; 2) “consumed once last week” = 0.143; 3) “consumed 2–3 times last week” (average number used) = 0.357; 4) “con- sumed 4–6 times last week” (average number used) = 0.714; 5) “consumed once a day” = 1; 6) “con- sumed 2 times a day” = 2; 7) “consumed 3 times a Outcome variable (zBMI) Height and weight were measured at both time points by a trained research assistant and study dietitian at the Fletcher et al. BMC Public Health (2017) 17:751 Page 3 of 10 Page 3 of 10 day” = 3; 8) “consumed 4–6 times a day” (average number used) = 5. For the analyses, a combination of food and beverage items were summed together to create three dietary mediators at both time points: 1) frequency of consuming discretionary foods, which in- cluded any savoury or plain biscuits, sweet pastries, cakes, doughnuts, chocolate, confectionary, and potato chips; 2) frequency of consuming SSB, which included non-diet soft drink, non-diet cordial and fruit juice and not sweetened milk drinks or energy drinks; and 3) fre- quency of consuming takeaway foods, which included savoury pastries (e.g. meat pies and sausage rolls), ham- burgers, pizza, hot chips and spring rolls/dim sims. The FFQ was tested for reproducibility and overall showed fair to excellent reliability for sweet snacks (ICC = 0.61), savoury snacks (ICC = 0.63), SSBs (ICC = 0.77), and fast food (ICC = 0.44). Frequency of consuming discretionary foods and frequency of consuming SSB were multiplied by 7 to convert to times per day. Participants with miss- ing data for any of the dietary items listed were excluded (n = 2). In all analyses, the dietary mediators were treated as continuous variables. clinic. Height was measured to the nearest 0.1 cm and weight was measured without shoes and in light clothing to the nearest 0.1 kg with electronic scales (Wedderburn, Summer Hill, NSW, Australia). Height and weight were used to calculate each participants’ BMI and zBMI was determined using the age- and sex-specific CDC 2000 reference data [18]. Overweight and obesity was deter- mined using the International Obesity Task Force (IOTF) criteria [19]. In all analyses, zBMI was treated as a continuous variable. Statistical analyses As shown in Fig. 1, to be included in the analyses involv- ing TV viewing and zBMI, participants were required to have complete data for age, sex, maternal education, pu- bertal status, dietary intake and TV viewing at baseline, and complete data for zBMI at baseline and follow-up (n = 259). To examine the association of total and bouts of sedentary time, dietary intake and zBMI, separate ana- lyses were undertaken from a subsample of participants who met the previous inclusion criteria as well as Page 4 of 10 Fletcher et al. BMC Public Health (2017) 17:751 Fig. 1 Flow diagram of participants for analyses Fig. 1 Flow diagram of participants for analyses meeting the accelerometry inclusion criteria at baseline (n = 140). Prior to the main analyses, all variables were checked for normality. Discretionary foods, SSB and takeaway food intake at baseline were not normally dis- tributed and were log-transformed. Fig. 2 Theoretical diagram of the cross-sectional (Model 1) and prospective (Model 2) mediation pathway g Analyses were conducted using Stata/SE v14.0 (StataCorp LP, College Station, Texas, 2015). Figure 2 illustrates two theoretical models of the cross-sectional and prospective mediation pathways examined [27]. For all mediation analyses, the Sobel-Goodman mediation method with bootstrapping with 5000 replications was used to estimate standard errors and 95% confidence intervals [28]. For the cross-sectional analyses, only baseline variables were used to test the following associ- ations: 1) association between the independent variable and the mediator (a-coefficient pathway); 2) association between the mediator and the outcome variable, adjust- ing for the independent variable (b-coefficient pathway); 3) total association between the independent variable and outcome variable (c-coefficient pathway); 4) direct association between the independent variable and out- come variable, accounting for each mediator (c′-coeffi- cient pathway); and 5) indirect association (e.g. mediating effect) of the mediator on the independent variable and outcome variable. For the prospective ana- lyses, similar pathways were tested. However, dietary in- take at baseline and dietary intake at follow-up were examined separately as potential mediators in the associ- ations between the independent variable at baseline and the outcome variable at follow-up. All analyses were adjusted for age, sex, pubertal status and maternal edu- cation, with objectively-measured MVPA adjusted for in the analyses involving total and bouts of sedentary time Fig. 2 Theoretical diagram of the cross-sectional (Model 1) and prospective (Model 2) mediation pathway Fig. Associations between TV viewing, dietary intake and zBMI Cross-sectionally Overall, 259 and 140 adolescents were included in the TV viewing and zBMI, and the sedentary time and zBMI analytic samples, respectively (Table 1). Those excluded in the TV viewing and zBMI analyses had mothers with a lower maternal education and those excluded in the y There was no evidence of total or direct cross-sectional as- sociations of TV viewing with zBMI (Table 2). A small, positive association was observed between hours spent watching TV per day and frequency of consuming take- away foods each week (a-coefficient pathway); with each additional hour of TV viewing, adolescents consumed an additional 0.06 serves (95% CI 0.01 to 0.12; p < 0.05) of takeaway foods each week. An inverse association was also observed for consuming discretionary foods each day and zBMI (b-coefficient pathway); with each additional serving of discretionary foods consumed each day, zBMI was lower by −0.39 units (95% CI -0.65 to −0.13; p < 0.01). None of the dietary variables were significant mediators of the cross-sectional association of TV viewing with zBMI. Prospectively h There was no evidence of total or direct prospective associations between TV viewing at baseline and zBMI at follow-up. Similar to the cross-sectional associations, a positive prospective association was observed between hours spent watching TV per day and frequency of con- suming takeaway foods each week at baseline and at follow-up. For example, for each additional hour of TV viewing, adolescents consumed an additional 0.07 (95% CI 0.01 to 0.12; p < 0.05) serves of take-away foods each week at baseline, and an additional 0.10 (95% CI 0.04 to 0.16; p < 0.05) serves of take-away foods each week at follow-up. However, no significant association remained for any of the dietary variables consumed at baseline and at follow-up with zBMI at follow-up (b-coefficient pathway). None of the dietary variables were significant mediators of the pro- spective association between TV viewing and zBMI. Associations between TV viewing, dietary intake and zBMI Cross-sectionally Table 1 Baseline characteristics of participants included in analyses Variables TV and zBMI (n = 259) Sedentary time and zBMI (n = 140) Age, years 12.9 (12.9, 13.0) 12.9 (12.9, 13.0) Sex, % Male 47.5 (41.4, 53.6) 50.0 (41.7, 58.3) Female 52.5 (46.4, 58.4) 50.0 (41.7, 58.3) Maternal education, % Low 6.9 (4.4, 10.8) 5.0 (2.3, 10.2) Medium 65.3 (59.2, 70.8) 62.8 (54.4, 70.5) High 27.8 (22.7, 33.6) 32.1 (24.8, 40.4) Pubertal status, % Early puberty 19.3 (8.0, 4.6) 14.3 (35.1, 56.2) Mid-pubertal 35.5 (30.0, 41.6) 41.4 (33.5, 50.0) Late-pubertal 35.5 (30.0, 41.6) 35.0 (27.5, 43.4) Post-pubertal 27.0 (21.9, 32.8) 22.1 (16.0, 30.0) BMI kg/m2 20.7 (20.2, 21.3) 20.3 (19.7, 21.0) z-score 0.4 (0.3, 0.5) 0.3 (0.1, 0.5) Overweight, % 18.9 (14.6, 24.2) 13.6 (8.8, 20.4) Obese, % 9.2 (6.3, 13.5) 9.2 (5.4, 15.4) Sedentary behaviour TV viewing (hours/day) 3.2 (3.0, 3.4) Total sedentary time (hours/day) 6.1 (6.0, 6.2) Average sedentary bout (minutes/day) 6.6 (6.5, 6.6) MVPA (minutes/day) 48.2 (35.0, 61.4) Dietary intake Discretionary foods (freq/day) 2.4 (2.2, 2.6) 2.4 (2.1, 2.7) Sugar-sweetened beverages (freq/day) 2.1 (1.8, 2.3) 1.9 (1.6, 2.2) Takeaway food (freq/week) 3.4 (3.1, 3.6) 3.2 (2.8, 3.6) BMI body mass index, zBMI BMI z-score, TV television, MVPA moderate-to- vigorous physical activity; freq: frequency Footnote: Maternal education: low = some secondary education or less; medium = completing secondary school, an apprenticeship or technical certificate; high = university or tertiary qualification. Pubertal status self-assessed using the ‘Tanner Stages’ [25]. BMI z-score determined using the age- and sex-specific CDC 2000 reference data [18]. Overweight and obesity determined using the International Obesity Task Force (IOTF) criteria [19] Table 1 Baseline characteristics of participants included in Statistical analyses 2 Theoretical diagram of the cross-sectional (Model 1) and prospective (Model 2) mediation pathway Page 5 of 10 Fletcher et al. BMC Public Health (2017) 17:751 with zBMI. The prospective analyses were additionally adjusted for zBMI at baseline. The significance level was set at p < 0.05 for all statistical tests. sedentary time and zBMI analyses were older, had a lower maternal education and had a higher proportion of overweight participants (Additional file 1: Tables S1 and S2 in the online supplement file). Cross-sectionally There was no evidence of total or direct cross-sectional associations for total sedentary time (hours/day) and zBMI, or between total sedentary time and any of the diet- ary mediators (a-coefficient pathway) (Table 3). An inverse association was observed between frequency of consum- ing discretionary foods per day and zBMI; with each additional discretionary food consumed each week, zBMI was lower by −0.42 units (95% CI -0.77 to −0.07; p < 0.05) (b-coefficient). There were no significant mediating effects for any of the dietary variables in the cross-sectional asso- ciation of total sedentary time with zBMI. Fletcher et al. BMC Public Health (2017) 17:751 Page 6 of 10 Table 2 Cross-sectionala and prospectiveb associations of dietary intake (mediator), TV viewing (independent) and zBMI (outcome) (n = 259) Outcome: zBMI c`-coefficient (direct) β (95% CI) a-coefficient β (95% CI) b-coefficient β (95% CI) ab/indirect (mediated/indirect)c Independent: TV viewing Uncorrected β (95% CI) Bias-corrected β (95% CI) Cross-sectional mediators Discretionary food (freq/day) 0.05 (−0.02, 0.13) 0.01 (−0.03, 0.05) −0.39 (−0.65, −0.13)** −0.04 (−0.18, 0.10) −0.04 (−0.23, 0.12) SSB (freq/day) 0.05 (−0.03, 0.13) 0.02 (−0.03, 0.07) 0.03 (−0.15, 0.22) 0.01 (−0.04, 0.05) 0.01 (−0.07, 0.09) Takeaway food (freq/week) 0.05 (−0.02, 0.13) 0.06 (0.01, 0.12)* −0.07 (−0.24, 0.09) −0.05 (−0.16, 0.07) −0.05 (−0.22, 0.06) Prospective mediators At baseline Discretionary foods (freq/day) −0.01 (−0.04, 0.02) 0.01 (−0.02, 0.05) 0.02 (−0.09, 0.13) 0.01 (−0.01, 0.02) 0.01 (−0.02, 0.04) SSB (freq/day) −0.01 (−0.04, 0.02) 0.02 (−0.03, 0.07) 0.03 (−0.04, 0.11) 0.01 (−0.02, 0.03) 0.01 (−0.02, 0.04) Takeaway food (freq/week) −0.01 (−0.04, 0.02) 0.07 (0.01, 0.12)* −0.01 (−0.07, 0.06) −0.01 (−0.04, 0.05) −0.01 (−0.06, 0.05) At follow-up Discretionary foods (freq/day) −0.01 (−0.04, 0.02) 0.02 (−0.02, 0.07) −0.04 (−0.13, 0.06) −0.01 (−0.04, 0.02) −0.01 (−0.06, 0.02) SSB (freq/day) −0.01 (−0.04, 0.02) −0.01 (−0.05, 0.04) 0.03 (−0.06, 0.11) −0.001 (−0.01, 0.01) −0.001 (−0.03, 0.02) Takeaway food (freq/week) −0.01 (−0.04, 0.02) 0.10 (0.04, 0.16)* 0.02 (−0.05, 0.08) 0.02 (−0.05, 0.08) 0.02 (−0.04, 0.09) aCross-sectional total effects (c-pathway) of TV viewing and zBMI β = 0.05 (95% CI -0.03 to 0.13), adjusting for age, sex, mother’s education and pubertal status. bProspective total effects (c-pathway) of TV viewing and zBMI β = −0.01 (95% CI -0.04 to 0.02), adjusting for age, sex, mother’s education, pubertal status and zBMI at baseline. cDue to the small units of measure, the indirect effects have been multiplied by 10 Significant *p < 0.01, p < 0.05. Cross-sectionally TV: television; SSB: sugar-sweetened beverages; zBMI: body mass index z-score; freq: frequency Prospectively Total sedentary time at baseline was not significantly associated with zBMI at follow-up, accounting for mediation by dietary intake at baseline and at follow-up. No significant associations were also observed for any of the total, direct and indirect pathways, nor with the a- coefficient and b-coefficient pathways. No significant total or direct cross-sectional associations were observed for average sedentary bout duration (minutes/day) with zBMI (Table 4). Average sedentary bout duration was inversely related to frequency of Table 3 Cross-sectionala and prospectiveb associations of dietary intake (mediator), sedentary time (independent) and zBMI (outcome) (n = 140) Outcome: zBMI c`-coefficient (direct) β (95% CI) a-coefficient β (95% CI) b-coefficient β (95% CI) ab/indirect (mediated/indirect)c Independent: total sedentary time Uncorrected β (95% CI) Bias-corrected β (SE) Cross-sectional mediators Discretionary food (freq/day) 0.27 (−0.03, 0.58) −0.01 (−0.15, 0.14) −0.42 (−0.77, −0.07)* 0.01 (−0.61, 0.62) 0.01 (−0.62, 0.69) SSB (freq/day) 0.27 (−0.05, 0.58) −0.16 (−0.38, 0.06) −0.04 (−0.29, 0.20) 0.07 (−0.33, 0.46) 0.07 (−0.31, 0.61) Takeaway food (freq/week) 0.27 (−0.05, 0.58) −0.06 (−0.30, 0.18) −0.04 (−0.26, 0.18) 0.02 (−0.13, 0.18) 0.02 (−0.36, 0.35) Prospective mediators At baseline Discretionary food (freq/day) 0.05 (−0.07, 0.17) 0.03 (−0.12, 0.17) 0.03 (−0.10, 0.17) 0.01 (−0.05, 0.07) 0.01 (−0.08, 0.15) SSB (freq/day) 0.05 (−0.06, 0.17) −0.15 (−0.37, 0.07) 0.02 (−0.07, 0.11) −0.03 (−0.17, 0.11) −0.03 (−0.22, 0.13) Takeaway food (freq/week) 0.05 (−0.07, 0.17) −0.05 (−0.30, 0.19) −0.03 (−0.12, 0.05) 0.02 (−0.08, 0.11) 0.02 (−0.12, 0.16) At follow-up Discretionary food (freq/day) 0.05 (−0.07, 0.17) 0.08 (−0.09, 0.24) 0.01 (−0.11, 0.13) 0.01 (−0.05, 0.10) 0.01 (−0.16, 0.17) SSB (freq/day) 0.05 (−0.06, 0.17) −0.04 (−0.23, 0.15) 0.01 (−0.10, 0.12) −0.01 (−0.05, 0.04) −0.01 (−0.11, 0.09) Takeaway food (freq/week) 0.05 (−0.06, 0.17) 0.21 (−0.57, 0.47) −0.01 (−0.08, 0.07) −0.01 (−0.16, 0.15) −0.01 (−0.19, 0.17) aCross-sectional total effects (c-pathway) of total sedentary time and zBMI β = 0.27 (95% CI -0.04 to 0.58), adjusting for age, sex, mother’s education and pubertal status. bProspective total effects (c-pathway) of total sedentary time and zBMI β = 0.05 (95% CI -0.06 to 0.17), adjusting for age, sex, mother’s education, pubertal status and zBMI at baseline. cDue to the small units of measure, the indirect effects have been multiplied by 10. Significant p < 0.05. aCross-sectional total effects (c-pathway) of total sedentary time and zBMI β = 0.27 (95% CI -0.04 to 0.58), adjusting for age, sex, mother’s education and pubertal status. bProspective total effects (c-pathway) of total sedentary time and zBMI β = 0.05 (95% CI -0.06 to 0.17), adjusting for age, sex, mother’s education, pubertal status and zBMI at baseline. cDue to the small units of measure, the indirect effects have been multiplied by 10. Significant p < 0.05. SSB: sugar-sweetened bever- ages; zBMI: body mass index z-score; freq: frequency Prospectively SSB: sugar-sweetened bever- ages; zBMI: body mass index z-score; freq: frequency nala and prospectiveb associations of dietary intake (mediator), sedentary time (independent) and zBMI Table 3 Cross-sectionala and prospectiveb associations of dietary intake (mediator), sedentary time (indepen (outcome) (n = 140) aCross-sectional total effects (c-pathway) of total sedentary time and zBMI β = 0.27 (95% CI -0.04 to 0.58), adjusting for age, sex, mother’s education and pubertal status. bProspective total effects (c-pathway) of total sedentary time and zBMI β = 0.05 (95% CI -0.06 to 0.17), adjusting for age, sex, mother’s education, pubertal status and zBMI at baseline. cDue to the small units of measure, the indirect effects have been multiplied by 10. Significant p < 0.05. SSB: sugar-sweetened bever- ages; zBMI: body mass index z-score; freq: frequency Fletcher et al. BMC Public Health (2017) 17:751 Page 7 of 10 Table 4 Cross-sectionala and prospectiveb associations of dietary intake (mediator), sedentary bouts (independent) and zBMI (outcome) (n = 140) Outcome: zBMI c`-coefficient (direct) β (95% CI) a-coefficient β (95% CI) b-coefficient β (95% CI) ab/indirect (mediated/indirect)c Independent: sedentary bouts Uncorrected β (95% CI) Bias-corrected β (SE) Cross-sectional mediators Discretionary food (freq/day) 0.12 (−0.36, 0.61) −0.08 (−0.31, 0.15) −0.42 (−0.77, −0.06) 0.32 (−0.67, 1.32) 0.32 (−0.55, 1.37) SSB (freq/day) 0.14 (−0.36, 0.64) −0.36 (−0.69, −0.02)* −0.06 (−0.30, 0.19) 0.20 (−0.71, 1.11) 0.20 (−0.68, 0.12) Takeaway food (freq/week) 0.16 (−0.33, 0.65) 0.03 (−0.34, 0.41) −0.05 (−0.28, 0.18) −0.02 (−0.21, 0.18) −0.02 (−0.62, 0.34) Prospective mediators At baseline Discretionary food (freq/day) 0.08 (−0.10, 0.26) −0.06 (−0.29, 0.17) 0.04 (−0.10, 0.17) −0.02 (−0.14, 0.10) −0.02 (−0.24, 0.14) SSB (freq/day) 0.09 (−0.09, 0.27) −0.36 (−0.70, −0.02)* 0.02 (−0.07, 0.11) −0.09 (−0.42, 0.24) −0.09 (−0.52, 0.24) Takeaway food (freq/week) 0.08 (−0.10, 0.26) 0.04 (−0.34, 0.41) −0.04 (−0.12, 0.04) −0.01 (−0.15, 0.13) −0.01 (−0.27, 0.16) At follow-up Discretionary food (freq/day) 0.08 (−0.10, 0.26) −0.08 (−0.34, 0.17) 0.02 (−0.10, 0.14) −0.02 (−0.12, 0.09) −0.02 (−0.28, 0.16) SSB (freq/day) 0.08 (−0.10, 0.26) 0.11 (−0.18, 0.40) 0.01 (−0.010, 0.11) 0.01 (−0.11, 0.12) 0.01 (−0.19, 0.20) Takeaway food (freq/week) 0.08 (−0.10, 0.26) 0.06 (−0.35, 0.48) −0.01 (−0.08, 0.07) −0.01 (−0.05, 0.04) −0.01 (0.18, 0.17) aCross-sectional total effects (c-pathway) of average sedentary bout and zBMI β = 0.16 (95% CI -0.33 to 0.65), adjusting for age, sex, mother’s education and pubertal status. Prospectively b Prospective total effects (c-pathway) of average sedentary bout and zBMI β = 0.08 (95% CI -0.10 to 0.26), adjusting for age, sex, mother’s education, pubertal status and zBMI at baseline. cDue to the small units of measure, the indirect effects have been multiplied by 10. Significant p < 0.05. SSB: sugar-sweetened beverages; zBMI: body mass index z-score; freq: frequency aCross-sectional total effects (c-pathway) of average sedentary bout and zBMI β = 0.16 (95% CI -0.33 to 0.65), adjusting for age, sex, mother’s education and pubertal status. b Prospective total effects (c-pathway) of average sedentary bout and zBMI β = 0.08 (95% CI -0.10 to 0.26), adjusting for age, sex, mother’s education, pubertal status and zBMI at baseline. cDue to the small units of measure, the indirect effects have been multiplied by 10. Significant p < 0.05. SSB: sugar-sweetened beverages; zBMI: body mass index z-score; freq: frequency with TV viewing, average sedentary bout duration and zBMI, none of the dietary variables significantly medi- ated the relationships between the sedentary variables and zBMI cross-sectionally or prospectively. consuming SSB each day (a-coefficient pathway); with each additional minute spent in a sedentary bout, the frequency of consuming SSB was lower by nearly half a serve each day (β = −0.36, 95% CI -0.69 to −0.02; p < 0.05). An inverse association was also observed for frequency of consuming discretionary foods each week and zBMI (b-coefficient pathway); with each additional discretionary food consumed each week, zBMI was lower by −0.42 units (95% CI -0.77 to −0.06; p < 0.05). None of the dietary variables significantly mediated cross-sectional associations of sedentary bout duration with zBMI. consuming SSB each day (a-coefficient pathway); with each additional minute spent in a sedentary bout, the frequency of consuming SSB was lower by nearly half a serve each day (β = −0.36, 95% CI -0.69 to −0.02; p < 0.05). An inverse association was also observed for frequency of consuming discretionary foods each week and zBMI (b-coefficient pathway); with each additional discretionary food consumed each week, zBMI was lower by −0.42 units (95% CI -0.77 to −0.06; p < 0.05). None of the dietary variables significantly mediated cross-sectional associations of sedentary bout duration with zBMI. Prospectively The null finding for the association of TV viewing with zBMI in the current study is in contrast to previous studies in youth that have consistently shown significant and positive associations both cross-sectionally [6] and prospectively [29, 30]. The differences in findings could be attributed to the homogeneity of the current sample being examined, with higher than average number of hours spent watching TV [31] and a lower zBMI com- pared to the population average [32]. The null associ- ation for total sedentary time and average sedentary bout duration with zBMI is supported by some previous studies [8, 33–35], but not others [7]. Research examin- ing accelerometer-measured sedentary time with health indictors among children and youth appears to be mixed; it is unclear whether an association exists in only some populations or if there are inconsistencies in meas- uring sedentary time and/or the analytical approaches undertaken. Prospectively When examining the prospective associations between average sedentary bout duration at baseline and zBMI at follow-up, no significant total or direct associations were observed. However, a significant inverse association remained for average sedentary bout duration with f- requency of consuming SSB at baseline (β = −0.36; 95% CI -0.70 to −0.02; p < 0.05), but not at follow-up for the a-coefficient pathways. No significant associations were observed for the b-coefficient pathways or mediating effects. The positive association observed for TV viewing with the consumption of takeaway foods, both cross-sectionally and prospectively is consistent with previous research [36, 37]. This link could be partially explained by the large extent of TV advertising of foods high in fat and energy during peak times when children and adolescents are likely to be watching TV [38]. In contrast, no evidence of an association was observed for total sedentary time with any of the dietary Discussion This study found no evidence of direct or indirect associations for TV viewing, total sedentary time and average sedentary bout duration with adolescents’ zBMI, either cross-sectionally or prospectively. Although some of the dietary variables were independently associated Fletcher et al. BMC Public Health (2017) 17:751 Page 8 of 10 Page 8 of 10 (both subjective and objectively measured) and health indicators (e.g. BMI, metabolic syndrome). variables. The null finding is consistent with previous lit- erature in youth [11, 12] where, compared to TV viewing, fewer significant associations are observed for total seden- tary time with elements of a less healthy diet. The null finding could be due to the measure used to capture total sedentary time. For example, accelerometers measure all time spent being sedentary, and thus may capture times where adolescents may not be eating/drinking (e.g. sitting in school, sitting in the car). In addition, due to accelerom- eters being unable to determine posture (e.g. standing still versus sitting), time spent standing may have been included in total sedentary time and thus may diminish the oppor- tunity to engage in an eating occasion. Unexpectedly, the study found higher sedentary bout duration was inversely associated with a lower consumption of SSB, both cross- sectionally and prospectively. One possible reason for this could be due to the adolescents not breaking up their sed- entary time in order to retrieve a SSB from another room (e.g. the kitchen, school canteen). However, given the current study is one of the first to examine individual asso- ciations between sedentary bout duration and dietary in- take, further research is urgently needed in this area. To our knowledge, the current study is the first to examine both the cross-sectional and prospective medi- ating effects of dietary intake on the association between sedentary behaviour and zBMI in an adolescent popula- tion, and the first to examine this using objective measures of sedentary time. Other strengths include ad- justment for a variety of confounders, including mater- nal education and adolescent pubertal status, and examining the dietary mediators (discretionary foods, SSB, takeaway foods) separately in all models. Limita- tions of the study include participants self-reporting their dietary intake using an FFQ and hours spent watching TV, and the low sample size for mediation ana- lyses. In addition, the semi-quantitative FFQ used in the current study was limited with the number of healthy food items included (e.g. Conclusion In conclusion, despite identifying some significant associations between TV viewing and average sedentary bout duration with frequency of consuming takeaway foods and SSB, and between frequency of consuming discretionary foods and zBMI, no significant associations were observed for any of the sedentary behaviour variables with zBMI, either cross-sectionally or prospect- ively. In addition, none of the dietary variables were found to be significant mediators of the associations between sedentary behaviour and zBMI. Given the unacceptably high levels of adolescent overweight and obesity, further studies are warranted to elucidate the complex relationships between TV viewing, sedentary time, dietary intake and health indicators. gy g g Only one other study has explored the prospective as- sociations between TV viewing, dietary intake and BMI; however, that study examined a younger population of pre-school children aged 0–5 years over a 2-year period [42]. In contrast to the current study, Fuller-Tyszkiewicz et al. reported a significant positive association between TV viewing and BMI that was bi-directional, with those children characterised with high amounts of TV viewing having higher BMI, and children with higher BMI watching a greater amount of TV. In addition, that study reported the prospective associations between TV viewing and BMI among 4 year olds were mediated by discretionary foods and soft drink consumption [42]. The differences in findings between the current study and previous study could be contributed to the different study populations and dietary mediators being examined, and that the previous study only examined TV view- ing and BMI. Thus, further prospective studies are needed to explore whether dietary intake mediates the relationship between various sedentary behaviours Discussion fruit and vegetables) and did not have information on portion sizes. There were also differences in those that were included and excluded in the analysis of this study and thus may limit the repre- sentative of the findings. Further, the data presented in the current study was collected more than a decade ago and thus the behaviours reported in the study may not reflect the contemporary sedentary and dietary behav- iours adolescents are engaging in today. In contrast, the cross-sectional, inverse association found for discretionary foods with zBMI, and the null finding for SSB and takeaway foods is in contrast with previous studies that have found positive associations for unhealthy dietary intake with BMI [39, 40]. Our findings could be a consequence of under-reporting, where overweight or obese children and adolescents have been found to under-report their energy intake by 20–40% [41]. Alternatively, it is possible that some participants in the current study with a higher zBMI may have changed their behaviour by decreasing their discretionary food intake over time as a strategy to manage their weight. Additional file Additional file 1: Supplement file. Comparison of baseline characteristics of participants included in the TV viewing and zBMI analyses compared to those excluded in the analyses. (PDF 29 kb) Additional file 1: Supplement file. Comparison of baseline characteristics of participants included in the TV viewing and zBMI analyses compared to those excluded in the analyses. (PDF 29 kb) Competing interests Competing interests The authors declare that they have no competing interests. Abbreviations BMI: Body mass index; Cpm: Counts per minute; FFQ: Food frequency questionnaire; MVPA: Moderate-to-vigorous physical activity; SSB: Sugar- sweetened beverages; TV: Television; zBMI: Body mass index z-score Page 9 of 10 Page 9 of 10 Fletcher et al. BMC Public Health (2017) 17:751 Funding h 6. Tremblay MS, LeBlanc AG, Kho ME, Saunders TJ, Larouche R, Colley RC, Goldfield G, Connor GS. Systematic review of sedentary behaviour and health indicators in school-aged children and youth. Int J Behav Nutr Phys Act. 2011;8(98):98. The Nepean longitudinal study was funded by The Children’s Hospital at Westmead Grant Research Scheme, a National Health and Medical Research Council Project Grant #206501 and Meat and Livestock Australia. SAM is funded by an NHMRC Career Development Fellowship Level 2, ID1104636 and was previously funded by an ARC Future Fellowship (2011-2015, FT100100581). DWD is funded by NHMRC Senior Research Fellowship (APP1078360). JS is funded by a NHMRC Principal Research Fellowship (APP1026216). This work was also partially funded by an OIS grant from the Victorian State Government and a National Health & Medical Research Council, Centre of Research Excellence grant (APP1057608). 7. Martínez-Gómez D, Eisenmann JC, Gómez-Martínez S, Veses A, Marcos A, Veiga OL. Sedentary behavior, adiposity, and cardiovascular risk factors in adolescents: the AFINOS study. Rev Esp Cardiol. 2010;63(3):277–85. 8. Chaput JP, Lambert M, Mathieu ME, Tremblay MS, O’ Loughlin J, Tremblay A. Physical activity vs. sedentary time: independent associations with adiposity in children. Pediatr Obes. 2012;7(3):251–8. 9. Colley R, Garriguet D, Janssen I, Wong S, Saunders T, Carson V, Tremblay M. The association between accelerometer-measured patterns of sedentary time and health risk in children and youth: results from the Canadian Health Measures Survey. BMC Public Health. 2013;13(1):1–9. Publisher’s Note 17. Garnett SP, Cowell CT, Baur LA, Fay RA, Lee J, Coakley J, Peat JK, Boulton TJ. Abdominal fat and birth size in healthy prepubertal children. Int J Obes Relat Metab Disord. 2001;25(11):1667–73. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. 18. Kuczmarski RJ, Ogden CL, Guo SS, et al. 2000 CDC growth charts for the United States: Methods and development. National Center for Health Statistics Vital Health Stat 11. 2002;11(246):1–190. Authors’ contributions JS, LB, and SG were involved in the conception and design of the study. EF carried out the data analysis with statistical guidance from KL. EF led the writing of the article, with contributions from authors KL, SM, JS, LB, SG, and DD. All authors read and approved the final manuscript. 12. Borghese MM, Tremblay MS, Leduc G, Boyer C, Bélanger P, LeBlanc AG, Francis C, Chaput J-P. Independent and combined associations of total sedentary time and television viewing time with food intake patterns of 9- 11 year-old Canadian children. Appl Physiol Nutr Metab. 2014;39(8):937–43. 13. Healy GN, Dunstan DW, Salmon J, Cerin E, Shaw JE, Zimmet PZ, Owen N. Breaks in sedentary time: beneficial associations with metabolic risk. Diabetes Care. 2008;31(4):661–6. Availability of data and materials The dataset supporting the conclusions of this article has not been approved to be made publicly available by the ethics committees. To require information about the content and analyses of the dataset, contact Sarah Garnett: sarah.garnett@health.nsw.gov.au 10. Pearson N, Biddle S. Sedentary behavior and dietary intake in children, adolescents, and adults: a systematic review. Am J Prev Med. 2011;41(2):178–88. 11. Vissers PAJ, Jones AP, van Sluijs EMF, Jennings A, Welch A, Cassidy A, Griffin SJ. Association between diet and physical activity and sedentary behaviours in 9–10-year-old British White children. Public Health. 2013;127(3):231–40. Received: 29 January 2017 Accepted: 18 September 2017 24. Australian Bureau of Statistics: National Nutrition Survey 1997: Nutrient intakes and physical measurements (Report No. 4805.0). Canberra, Australia: Australian Bureau of Statistics; 1998. Acknowledgements h h ld l k 4. Singh AS, Mulder C, Twisk JW, van Mechelen W, Chinapaw MJ. Tracking of childhood overweight into adulthood: a systematic review of the literature. Obes Rev. 2008;9(5):474–88. The authors would like to thank Dr. Vanessa Shrewsbury for collecting the data and her work on the project. We would like to thank all the families that generously donated their time to participate in this study. 5. Sedentary Behaviour Research Network. Letter to the Editor: standardized use of the terms “sedentary” and “sedentary behaviours”. Appl Physiol Nutr Metab. 2012;37(3):540–2. Author details 1 1Deakin University, Geelong, Australia, Institute for Physical Activity and Nutrition (IPAN), School of Exercise and Nutrition Sciences, Geelong, Australia. 2The Children’s Hospital at Westmead Clinical School, University of Sydney, Westmead, Australia. 3Baker IDI Heart and Diabetes Institute, Melbourne, VIC, Australia. 4Department of Medicine, Monash University, Melbourne, Australia. 5The University of Queensland, School of Public Health, Brisbane, Australia. 6Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia. 7School of Sport Science, Exercise and Health, The University of Western Australia, Perth, Australia. 8Mary MacKillop Institute for Health Research, Australian Catholic University, Melbourne, Australia. 19. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ. 2000;320(7244):1240. 20. Hardy L, Baur L, Garnett S, Crawford D, Campbell K, Shrewsbury V, Cowell C, Salmon J. Family and home correlates of television viewing in 12–13 year old adolescents: The Nepean Study. Int J Behav Nutr Phys Act. 2006;3(1):1–9. 1. Willett WC, Howe GR, Kushi LH. Adjustment for total energy intak 21. Willett WC, Howe GR, Kushi LH. Adjustment for total energy intake in epidemiologic studies. Am J Clin Nutr. 1997;65(4 Suppl):1220S–8S. 22. Chastin SF, Granat MH. Methods for objective measure, quantification and analysis of sedentary behaviour and inactivity. Gait & Posture. 2010;31(1):82–6. 23. Mattocks C, Ness A, Leary S, Tilling K, Blair SN, Shield J, Deere K, Saunders J, Kirkby J, Smith GD, et al. Use of accelerometers in a large field-based study of children: protocols, design issues, and effects on precision. J Phys Act Health. 2008;5(s1):S98–S111. Received: 29 January 2017 Accepted: 18 September 2017 Ethics approval and consent to participate Ethical approval was provided by the Ethics Committees of The Children’s Hospital at Westmead (Project No: 2001/095) and Wentworth Area Health Service (Registration No 2001/065). Written consent was obtained from the adolescent’s parent or guardian and each adolescent signed a study agreement form. 14. Fletcher E, Leech R, McNaughton SA, Dunstan DW, Lacy KE, Salmon J. Is the relationship between sedentary behaviour and cardiometabolic health in adolescents independent of dietary intake? A systematic review. Obes Rev. 2015;16(9):795–805. 15. Carson V, Janssen I. The mediating effects of dietary habits on the relationship between television viewing and body mass index among youth. Pediatr Obes. 2012;7(5):391–8. Consent for publication Not applicable. Not applicable. 16. Fletcher EA, McNaughton SA, Lacy KE, Dunstan DW, Carson V, Salmon J. Mediating effects of dietary intake on associations of TV viewing, body mass index and metabolic syndrome in adolescents. Obes Sci & Prac. 2016;2(3):232–40. Competing interests Fletcher et al. BMC Public Health (2017) 17:751 Fletcher et al. BMC Public Health (2017) 17:751 27. MacKinnon DP. Introduction to statistical mediation analysis. New York: Taylot & Francis Group; 2008. 28. Preacher KJ, Hayes AF. SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behav Res Methods Instrum Comput. 2004;36(4):717–31. 29. Krahnstoever Davison K, Marshall SJ, Birch LL. Cross-sectional and longitudinal associations between TV viewing and girls’ body mass index, overweight status, and percentage of body fat. J Pediatr. 2006;149(1):32–7. 30. Danner FW. A national longitudinal study of the association between hours of TV viewing and the trajectory of BMI growth among US children. J Pediatr Psychol. 2008;33(10):1100–7. 31. Australian Health Survey: Physical activity, 2011–12 cat no. 4364.0.55.004. Canberra: Commonwealth of Australia; 2013. http://www.abs.gov.au 32. Australian Health Survey: updated results, 2011-2012. Overweight and obesity. http://www.abs.gov.au/ausstats/abs@.nsf/Lookup/ 33C64022ABB5ECD5CA257B8200179437?opendocument 33. Carson V, Janssen I. Volume, patterns, and types of sedentary behavior and cardio-metabolic health in children and adolescents: a cross-sectional study. BMC Public Health. 2011;11:274. 34. Ekelund U, Luan J, Sherar LB, Esliger DW, Griew P, Cooper A. Moderate to vigorous physical activity and sedentary time and cardiometabolic risk factors in children and adolescents. JAMA. 2012;307(7):704–12. 35. Collings PJ, Wijndaele K, Corder K, Westgate K, Ridgway CL, Sharp SJ, Atkin AJ, Bamber D, Goodyer I, Brage S, et al. Prospective associations between sedentary time, sleep duration and adiposity in adolescents. Sleep Med. 2015;16(6):717–22. 36. Barr-Anderson DJ, Larson NI, Nelson MC, Neumark-Sztainer D, Story M. Does television viewing predict dietary intake five years later in high school students and young adults? Int J Behav Nutr Phys Act. 2009;6:7. 37. Larson NI, Neumark-Sztainer DR, Story MT, Wall MM, Harnack LJ, Eisenberg ME. Fast food intake: longitudinal trends during the transition to young adulthood and correlates of intake. J Adolesc Health. 2008;43(1):79–86. 38. Kelly B, Halford JCG, Boyland EJ, Chapman K, Bautista-Castaño I, Berg C, Caroli M, Cook B, Coutinho JG, Effertz T, et al. Television food advertising to children: a global perspective. Am J Pub Health. 2010;100(9):1730–6. 39. Rosenheck R. Fast food consumption and increased caloric intake: a systematic review of a trajectory towards weight gain and obesity risk. Obes Rev. 2008;9(6):535–47. 40. Malik VS, Schulze MB, Hu FB. Intake of sugar-sweetened beverages and weight gain: a systematic review. Am J Clin Nutr. 2006;84(2):274–88. 41. Collins CE, Watson J, Burrows T. Measuring dietary intake in children and adolescents in the context of overweight and obesity. Int J Obes. 2010;34(7):1103–15. Fletcher et al. BMC Public Health (2017) 17:751 References 1. Wang Y, Lobstein T. Worldwide trends in childhood overweight and obesity. Int J Pediatr Obes. 2006;1(1):11–25. 1. Wang Y, Lobstein T. Worldwide trends in childhood overweight and obesity. Int J Pediatr Obes. 2006;1(1):11–25. 25. Duke PM, Litt IF, Gross RT. Adolescents' self-assessment of sexual maturation. Pediatrics. 1988;66(6):918–20. 2. Ogden CL, Carroll MD, Lawman HG, et al. Trends in obesity prevalence among children and adolescents in the united states, 1988-1994 through 2013-2014. JAMA. 2016;315(21):2292–9. 26. Trost SG, Pate RR, Sallis JF, Freedson PS, Taylor WC, Dowda M, Sirard J. Age and gender differences in objectively measured physical activity in youth. Med Sci Sports Exerc. 2002;34(2):350–5. 3. Australian Bureau of Statistics. Profiles of Health, Australia, 2011–13. 2013. http://www.abs.gov.au/. Page 10 of 10 Fletcher et al. BMC Public Health (2017) 17:751 42. Fuller-Tyszkiewicz M, Skouteris H, Hardy LL, Halse C. The associations between TV viewing, food intake, and BMI. A prospective analysis of data from the Longitudinal Study of Australian Children. Appetite. 2012;59(3):945–8. • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries
https://openalex.org/W2112857583	https://jbiol.biomedcentral.com/track/pdf/10.1186/1475-4924-2-16	English	null	Controlling how many cells make a fly.	Journal of biology	2,003	cc-by	3,538	Journal of Biology Journal of Biology Journal of Biology BioMed Central BioMed Central Research news Controlling how many cells make a fly Pete Moore Published: 21 August 2003 Journal of Biology 2003, 2:16 The electronic version of this article is the complete one and can be found online at http://jbiol.com/content/2/3/16 Published: 21 August 2003 Journal of Biology 2003, 2:16 Published: 21 August 2003 Journal of Biology 2003, 2:16 The electronic version of this article is the complete one and can be found online at http://jbiol.com/content/2/3/16 The electronic version of this article is the complete one and can be found online at http://jbiol.com/content/2/3/16 © 2003 BioMed Central Ltd © 2003 BioMed Central Ltd Studies in Drosophila have revealed the Forkhead-family transcription factor FOXO to be a crucial mediator of the branch of the insulin-signaling pathway that controls cell number. The bottom line • The homologous transcription factors FOXO and DAF-16 are known to lie on the insulin-signaling pathway, but it was unclear precisely how this pathway regulates cell size, cell number, and development in dif- ferent organisms. • In Drosophila, FOXO mutants have no growth phenotype, but are more sensitive than wild-type flies to oxidative stress. • Mutations in chico, an upstream component of the insulin-signaling pathway, reduce both cell size and cell number; an additional FOXO mutation rescues the reduction in cell number, indicating that wild- type FOXO negatively regulates this aspect of growth. But a FOXO- chico double mutant still has its cell size reduced. • Cell size is regulated by the S6 kinase branch of the insulin-signaling pathway, while FOXO regulates cell number, in part by up-regulating a protein involved in the regulation of translation. • The insulin-signaling pathway is highly conserved in mammals, C. elegans and Drosophila, and may have evolved in the ancestor of metazoans to allow regulation of growth and development in response to stress and nutrient availability. Studies in Drosophila have revealed the Forkhead-family transcription factor FOXO to be a crucial mediator of the branch of the insulin-signaling pathway that controls cell number. Studies in Drosophila have revealed the Forkhead-family transcription factor FOXO to be a crucial mediator of the branch of the insulin-signaling pathway that controls cell number. In our earliest biology lessons we learnt that all living organisms grow, and that growth requires an increase in both cell number and cell size. But how is this controlled? Insulin and mechanism of action at the molecular level will have important conse- quences not only for our knowledge of biology, but for pathology as well. insulin-like growth factors (IGFs; see the ‘Background’ box) play a critical role, and they are also implicated in medical conditions such as cancer and diabetes. So understanding their Working at the University of Zürich, Switzerland, Ernst Hafen heads a team that is looking at the control of growth. “You can think of our work in terms of a triangle,” he explains. “At the three corners are Homo sapiens, Caenorhabditis elegans and Drosophila melanogaster, and at the center of the triangle is the insulin-signaling pathway.” Hafen’s team has learnt important lessons about the pathway from each species, and their new findings, published in this issue of Journal of Biology [1], add significant evidence in support of the idea that the key functions of the pathway have been powerfully con- served through evolution. The new results also serve to tie together con- trols of cell size and cell number with how organisms respond to oxidative stress and nutrient availability (see ‘The bottom line’ box for a summary of their work). The bottom line Drosophila and growth p g Into this arena of confusion comes Drosophila. The clearest indication of the way that insulin signaling affects this species comes from the so-called chico mutant. Wild-type Chico protein func- tions in the insulin-signaling pathway, and flies lacking it are small with delayed development. In many ways this is similar to the situation in mammals, where mutations in the insulin/IGF pathway affect growth and body size. The flies have fewer cells, and the cells they do have are smaller in size. “This [growth] reduction is something that was never seen in C. elegans,” says Hafen. “So, before our recent work, the best concept was that the initial pathway was the same in all species, but the readout was different,” leading to growth in mammals but pre- venting dauer formation in C. elegans. • Protein kinase B (PKB, also known as AKT) phosphorylates FOXO and turns off its transcriptional activity. PKB also regulates growth through a pathway independent of FOXO but including the S6 kinase. • Protein kinase B (PKB, also known as AKT) phosphorylates FOXO and turns off its transcriptional activity. PKB also regulates growth through a pathway independent of FOXO but including the S6 kinase. signaling activity within the pathway, which in turn drives the worms into the developmentally arrested ‘dauer stage’ (DAF denotes ‘dauer forma- tion’). Dauer larvae alter their metabo- lism, stockpile fat and can survive in this state for at least four to eight times longer than the normal two-week life- span of C. elegans. student in Hafen’s lab. Decades of experiments have shown that insulin regulates energy metabolism, and more recent results show that it plays a key role in embryonic [2] and post- embryonic [3] growth, as well as the determination of lifespan [4]. Studies in mammalian cells have also shown that insulin negatively reg- ulates FOXO (Forkhead box, subclass O) transcription factors, which in turn arrest the cell cycle and, in some types of cell, induce cell death. FOXO transcription factors therefore have a negative influence on growth, and their function is turned off by the insulin effector protein kinase B (PKB, which is also known as AKT [5]). The evidence that dauer formation is dependent on the transcription factor DAF-16 comes from genetic experiments showing that if the insulin-signaling pathway is mutated, C. elegans enters the dauer stage. Drosophila and growth But in a double mutant in which DAF-16 is also disabled, the worms develop as normal. The clear implication is that in normal animals the insulin pathway has its effects on dauer formation via negative regulation of DAF-16. “But the Sorting out size and number The insulin-signaling pathway is nor- mally triggered by insulin binding to the insulin receptor, which then phospho- rylates Chico, an intracellular adapter protein (see Figure 1). Chico then recruits the phosphatidylinositol (PI) 3-kinase, which in turn phosphorylates The worm and its dauer stage The link between insulin and FOXO proteins initially came from experi- ments in C. elegans, where insulin signals to the FOXO equivalent, DAF- 16 (see Table 1 for the names of corre- sponding proteins in the different species discussed in this article). In worms, the effect of modulating the insulin-signaling pathway is quite unique: rather than affecting size, it induces a change in the nematode’s developmental program. Adverse con- ditions, such as starvation, decrease Background • Both insulin, first identified for its role in energy metabolism, and insulin-like growth factors (IGFs) signal through the insulin receptor, a transmembrane protein kinase that initiates a signaling cascade that includes transcriptional regulation by FOXO, a member of the Forkhead family of transcription factors. • The insulin-signaling pathway has roles in growth and development in many animal species, and is implicated in the control of lifespan, ini- tially from studies of the genes controlling the formation of the devel- opmentally arrested, stress-resistant dauer form in C. elegans. Insulin and IGF in mammals Insulin and IGF in mammals “We know most of the biochemistry of the system from mammalian cell- culture experiments and knockout mice,” explains Martin Jünger, a PhD Journal of Biology 2003, 2:16 16.2 Journal of Biology 2003, Volume 2, Issue 3, Article 16 Moore http://jbiol.com/content/2/3/16 link to growth [in worms] is not clear,” says Hafen. “Because this strange worm is built by a precisely fixed number of cells, there is no relation between body size and insulin signaling.” This appar- ent difference in action threw into question the idea that the insulin pathway has a conserved role in worms and mammals. Table 1 Terms for equivalent proteins in different species Human C. elegans Drosophila Forkhead transcription factors Three different DAF-16 dFOXO hFOXO proteins Insulin effector kinases, PDK1 and PDK1, Akt -1 dPDK1 and containing pleckstrin PKB/AKT 1-3 and Akt-2 dPKB/dAktfs homology (PH) domains Terms for equivalent proteins in different species Journal of Biology 2003, 2:16 Journal of Biology 2002, Volume 2, Issue 3, Article 16 Moore 16.3 http://jbiol.com/content/2/3/16 Figure 1 The key molecules of the insulin-signaling pathway, as discussed in the text. Insulin receptor PIP3 PIP2 S6K FOXO Cell number Cell size Insulin or IGF Cytoplasm Membrane Chico PI 3-kinase PDK1 PKB cells are small. The answer to the cell number question came from the paper by Jünger et al. [1], which ini- tially set out to characterize the fly DAF-16 homolog and to assess both whether and how it fitted into the fly insulin-signaling pathway and also its growth-modulating capabilities. When the Zürich team produced dFOXO mutants they were initially sur- prised. The flies were viable and normal-sized; there was no apparent phenotype, other than that the flies were more susceptible to oxidative stress than were their wild-type cousins. Jünger and colleagues had anticipated that removing the pre- sumed negative influence would cause the flies to grow bigger. At first, they questioned whether they really had mutated dFOXO, but the genetic and molecular evidence was compelling. As a next step, Jünger started to test the mutants in a genetic background in which other aspects of the insulin pathway were compromised. In this case, a normal fly would produce fewer, smaller cells. But take dFOXO away and the flies have small cells, but almost the normal number. “The reduced cell number [in insulin- pathway mutants] is rescued by the absence of the transcription factor, because [wild-type] dFOXO has a neg- ative influence,” he explains. Figure 1 The key molecules of the insulin-signaling pathway, as discussed in the text. cytoplasm where they cannot stimulate the initiation of transcription of target genes. “In C. elegans, we know that this [part of the pathway] influences devel- opment, not size, so the question for us was if size was mediated through DAF-16 in flies.” the membrane-bound phospholipid phosphatidylinositol (4,5)-bisphos- phate (PIP2) to phosphatidylinositol (3,4,5)-trisphosphate (PIP3). Hafen explains that the next key event is that PIP3 causes kinases like PDK1 and PKB, which contain plekstrin-homol- ogy (PH) domains, to be translocated from the cytoplasm to the membrane. Journal of Biology 2003, 2:16 What prompted the work? On top of this, Tjian’s group had another striking result. “We found that FOXO also regulates expression of the insulin receptor,” says Tjian. “This means that in the absence of insulin, FOXO is produced. This not only limits growth, but it also up-regulates sensitivity for insulin. The system is now primed to look for lower concen- trations of insulin.” Team members in the lab had a long-running interest in growth regulation and had performed extensive genetic screens for growth-affecting muta- tions. They had found many components of the insulin-signaling cascade, but did not find FOXO. As FOXO is such an established target in mammals and worms, it was an obvious issue to address. My involvement started with my PhD thesis. I got my degree in bio- chemistry in Berlin and became interested in signal transduction during my diploma work. I moved to Ernst’s lab for the beginning of my thesis to combine signal transduction and genetics. Behind the scenes Journal of Biology asked Martin Jünger about how and why he set out to study dFOXO and its role in regulating growth. How long did it take to do the experiments, and what was the team’s reaction to the results? A third study, by Jamie Kramer and colleagues at the Memorial University of Newfoundland, Canada, presents a slightly different picture. Kramer et al. [10] agree with Jünger et al. and Tjian’s group that dFOXO is the fly homolog of DAF-16 and hFOXO (see Table 1). But, in a key difference, Kramer et al. found that overexpres- sion of dFOXO leads to reductions in both cell size and cell number. “We have seen this effect in both the eye and the wing of Drosophila,” says Kramer. He believes that this differ- ence between his results and those of the other groups is most likely to arrive from his use of overexpression analysis whereas Jünger used loss-of- function techniques. “In total it took about three years, although when I started in December 2000, several months work had already been invested by Michael Green- berg’s team at Harvard. [When we saw the results] we were surprised and excited, mainly because of FOXO’s double role, the absence of a growth phenotype and the effect within the mutant context - it was a very interesting project. What are the next steps? We will certainly follow up on some of the results, for example the oxida- tive stress issue and the control of cell proliferation. More extensive expression-profiling studies should help to clarify the molecular mecha- nisms underlying these effects. The rather small microarray experiment in our dFOXO paper was something of a sidetrack. Personally, I will invest much of my time in studying the insulin pathway in cultured cells in more detail at the transcriptome and pro- teome level. We have a couple of exciting collaborations going on. “A general problem,” agrees Jünger, “is that overexpression studies are prone to artefacts, because over- expressed proteins often start doing things which under normal, physiolog- ical protein concentrations they do not.” Tjian agrees; “If I got results from overexpression experiments that differ from loss-of-function work I would be inclined to trust the loss-of-function study,” he says. At the same time, Tjian points out that his team’s findings also came from overexpression studies. He is now keen to study the exact differ- ences in method between his own and Kramer’s work to see if this sheds light on the differences. FOXO element taking charge of cell number (see Figure 1). It’s more like a ‘proof-of-principle’ experiment, showing that we can find physiologically relevant targets in our rather artificial cell culture system, where we stimulate Drosophila cultured cells with bovine insulin! But recent microarray studies (by Puig et al. [8] and Ramaswamy et al. [9]) suggest that FOXO proteins work by modulating the transcription of large sets of target genes.” Table 1 Jünger went on to show that dFOXO operates in part by up-regulat- ing the gene for a binding protein called d4E-BP. With larger quantities of this binding protein produced, the translation-initiation factor eIF4E is effectively removed from the transla- tion machinery, in turn inhibiting the initiation of protein synthesis. This shows that insulin operates not only by regulating pre-existing 4E-BP protein via phosphorylation [7], but also by influencing the intracellular abundance of 4E-BP at the gene expression level. One part of the answer to this question - dealing with the size of each cell - came from a paper previ- ously published in Science [6]. This showed that cell size is controlled in Drosophila by the S6 kinase (dS6K), an enzyme that apparently acts down- stream of dPDK1 and dPKB and is named for its effects on ribosomal protein S6. Mutating dS6K produces small flies that have the same number of cells as in the wild type but whose Now, Jünger, Hafen and colleagues have looked at what happens in Drosophila downstream of PKB [1] (see the ‘Behind the scenes’ box for more discussion of the background to the work). From work in mammalian cells, they knew that PKB phosphorylates transcription factors of the FOXO family, causing them to leave the nucleus and become trapped in the “We have shown that d4E-BP is a relevant target [of the pathway],” says Jünger, “but we absolutely don’t postulate that it is the only one. Journal of Biology 2003, 2:16 16.4 Journal of Biology 2003, Volume 2, Issue 3, Article 16 Moore http://jbiol.com/content/2/3/16 His group reports that “targeted expres- sion of dFOXO in fly tissues regulates organ size by specifying cell number with no effect on cell size”. Moreover, they also found and validated d4E-BP as a target gene. This nicely comple- ments the findings of Jünger et al. [1]. References 1. Jünger MA, Rintelen F, Stocker H, Wasserman JD, Végh M, Radimerski T, Greenberg ME, Hafen E: The Drosophila Forkhead transcription factor FOXO mediates the reduction in cell number associated with reduced insulin signaling. J Biol 2003, 2:20. 2. g g J , 2. Takahashi Y, Kadowaki H, Momomura K, Fukushima Y, Orban T, Okai T, Taketani Y, Akanuma Y, Yazaki Y, Kadowaki T: A homozygous kinase-defective muta- tion in the insulin receptor gene in a patient with leprechaunism. Dia- betologia 1997, 40:412-420. g 3. Baker J, Liu JP, Robertson EJ, Efstratiadis A: Role of insulin-like growth factors in embryonic and postnatal growth. Cell 1993, 75:73-82. 4. Holzenberger M, Dupont J, Ducos B, Leneuve P, Geloen A, Even PC, Cervera P, Le Bouc Y: IGF-1 receptor regu- lates lifespan and resistance to oxidative stress in mice. Nature 2003;421:182-187. 5. He postulates that his group didn’t see the full effect of the dFOXO mutants because the flies were growing in unnatural conditions: because the flies are fed the whole time, the insulin pathway is constantly activated. A con- stantly starving wild fly with a dFOXO mutation might have an impaired ability to limit its rate of growth to suit the nutrient availability. ; 5. Brunet A, Bonni A, Zigmond MJ, Lin MZ, Juo P, Hu LS, Anderson MJ, Arden KC, Blenis J, Greenberg ME: Akt promotes cell survival by phosphorylating and inhibiting a Forkhead transcription factor. Cell 1999, 96:857-868. 6. Montagne J, Stewart MJ, Stocker H, Hafer E, Kozma SC, Thomas G: Drosophila S6 kinase: a regulator of cell size. Science 1999, 285:2126-2129. 7. Miron M, Verdu J, Lachance PE, Birnbaum MJ, Lasko PF, Sonenberg N: The trans- lational inhibitor 4E-BP is an effec- tor of PI(3)K/Akt signalling and cell growth in Drosophila. Nat Cell Biol 2001, 3:596-601. Hafen likens the situation to driving a car when you know that the tank is running out of fuel. “You don’t go at hundred and forty kilometers an hour, you reduce speed to reduce fuel con- sumption,” he comments. “This is what animals had to learn to do during evo- lution - and they do it at least in part via the insulin-IGF pathway. The main goal of this pathway is to adjust growth rates, or the developmental program in the case of C. Completing the triangle role that FOXO plays in morphogenesis has far-reaching implications in both the laboratory and medical practice. p g g For Hafen, the new data complete the triangle. “In the worm, fly and human, FOXO is [a] negative [regulator of growth],” he says. “Now the pictures do not look different at all. What we see is a great underlying evolutionary conserva- tion of this pathway.” In Hafen’s view, this pathway governs one of the most fundamental controls that the ancestors of multicellular organisms had to evolve. “Wild flies are not like our labo- ratory flies, fed on delicious food day in, day out. In nature animals often have too little food, so they have to evolve mechanisms to deal with the issue. They can’t just run their metabolism at maximal speed, irrespective of whether there is food around or not; they have to find ways to adjust their metabolic rate and their speed of development accord- ing to the availability of nutrients.” Related studies At the same time as the Jünger et al. paper [1] was published, two other groups were publishing findings that support the same idea. Robert Tjian and colleagues at the University of Cal- ifornia, Berkeley, presented biochemi- cal evidence that when insulin is applied, dFOXO is phosphorylated by dPKB, leading to it being retained in the cytoplasm and therefore not being capable of initiating transcription [8]. The picture that emerges for Drosophila is that the insulin signaling pathway forks at PKB, with an S6K element controlling cell size, and a Journal of Biology 2003, 2:16 Journal of Biology 2002, Volume 2, Issue 3, Article 16 Moore 16.5 http://jbiol.com/content/2/3/16 Pete Moore is a science writer based in Surrey, UK. E-mail: moorep@mja-uk.org References elegans, with respect to availability of food, and the mechanism is conserved right down to the level of the DAF-16 transcription factor.” 8. , 8. Puig O, Marr MT, Ruhf ML, Tjian R. Control of cell number by Drosophila FOXO: Downstream and feedback regulation of the insulin receptor pathway. Genes Dev 2003, 17:2006-2020. 9. Ramaswamy S, Nakamura N, Sansal I, Bergeron L, Sellers WR: A novel mech- anism of gene regulation and tumor suppression by the transcription factor FKHR. Cancer Cell 2002, 2:81-91. 10. Kramer JM, Davidge JT, Lockyer JM, Staveley BE: Expression of Drosophila FOXO regulates growth and can phenocopy starvation. BMC Dev Biol 2003, 3:5. Pete Moore is a science writer based in Surrey, UK. E-mail: moorep@mja-uk.org Tjian is also excited by the findings. “We are starting to get a better idea of how transcription factors affect organ size and how they are used to decide when to stop putting new cells into organs,” he says. And understanding the E-mail: moorep@mja-uk.org Journal of Biology 2003, 2:16
https://openalex.org/W2161618199	https://europepmc.org/articles/pmc4118783?pdf=render	English	null	Performance assessment of aquatic macrophytes for treatment of municipal wastewater	Journal of environmental health science & engineering	2,014	cc-by	5,702	© 2014 Shah 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 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. RESEARCH ARTICLE Open Access Performance assessment of aquatic macrophytes for treatment of municipal wastewater Mumtaz Shah1, Hashim Nisar Hashmi2, Arshad Ali3 and Abdul Razzaq Ghumman2 Abstract The objective of the study was to evaluate the performance of three different aquatic macrophytes for treatment of municipal wastewater collected from Taxila (Pakistan). A physical model of treatment plant was constructed and was operated for six experimental runs with each species of macrophyte. Every experimental run consist of thirty days period. Regular monitoring of influent and effluent concentrations were made during each experimental run. For the treatment locally available macrophyte species i.e. water hyacinth, duckweed & water lettuce were selected to use. To evaluate the treatment performance of each macrophyte, BOD5, COD, and Nutrients (Nitrogen and Phosphorus) were monitored in effluent from model at different detention time of every experimental run after ensuring steady state conditions. The average reduction of effluent value of each parameter using water hyacinth were 50.61% for BOD5, 46.38% for COD, 40.34% for Nitrogen and 18.76% for Phosphorus. For duckweed the average removal efficiency for selected parameters were 33.43% for BOD5, 26.37% for COD, 17.59% for Nitrogen and 15.25% for Phosphorus and for Water Lettuce the average removal efficiency were 33.43% for BOD5, 26.37% for COD, 17.59% for Nitrogen and 15.25% for Phosphorus. The mechanisms of pollutant removal in this system include both aerobic and anaerobic microbiological conversions, sorption, sedimentation, volatilization and chemical transformations. The rapid growth of the biomass was measured within first ten days detention time. It was also observed that performance of macrophytes is influenced by variation of pH and Temperature. A pH of 6-9 and Temperature of 15-38°C is most favorable for treatment of wastewater by macrophytes. The option of macrophytes for treatment of Municipal sewage under local environmental conditions can be explored by further verifying the removal efficiency under variation of different environmental conditions. Also this is need of time that macrophyte system should be used for treatment of wastewater because their performance is comparable to conventional wastewater treatment plants and also the system has very low O&M costs. Keywords: Aquatic macrophytes, BOD5, Duckweed, Phytoplankton, NEQS, Plant growth, Waste stabilization ponds solids and nutrients such as Nitrogen (N) and Phosphor- ous (P). The actual proportion of each constituent within any given wastewater varies depending on the spatial and temporal differences [1]. Correspondence: mumtazshah@gmail.com 1Department of Civil and Environmental Engineering, UET, Taxila, Pakistan Full list of author information is available at the end of the article JOURNAL OF ENVIRONMENTAL HEALTH SCIENCE & ENGINEERING JOURNAL OF ENVIRONMENTAL HEALTH SCIENCE & ENGINEERING Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 JOURNAL OF ENVIRONMENTAL HEALTH SCIENCE & ENGINEERING Introduction Wastewater is any liquid that has been adversely affected in quality by anthropogenic influence. It comprises liquid waste discharged by domestic residences, commercial prop- erties, industry, or agriculture and can encompass a wide range of potential contaminants and concentrations. In the most common usage, it refers to the municipal wastewater that contains a broad spectrum of contaminants resulting from the mixing of wastewater from different sources. Urban wastewater contains 99% water, and other materials make up the portion. The potential pollutants include pathogens, oil and grease, metals, organic matter (OM), In recent years, the amount of wastewater produced from several activities has increased as a result of the rapid improvement of living standards [2]. Although some com- munities treat their wastewater in a suitable way, others lack convenient treatment systems, thus discharging untreated wastewater into the natural environment. Pollutants (e.g. heavy metals) enter aquatic systems via numerous pathways, including effluent discharge, urban and agricultural run-off. Contaminants present in sewage * Correspondence: mumtazshah@gmail.com 1Department of Civil and Environmental Engineering, UET, Taxila, Pakistan Full list of author information is available at the end of the article Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 2 of 12 Figure 1 Experimental setup for the study. Figure 1 Experimental setup for the study. specifically constructed for the purpose of pollution con- trol and waste management, at a location other than exist- ing natural wetlands” is known as constructed wetland. Wetlands have many unique benefits as a wastewater treatment process, including the ability to operate on am- bient solar energy, self-organize and increase treatment capacity over time, rich in biodiversity, produce oxygen and consume carbon dioxide, and achieve high levels of treatment with minimum maintenance [5]. Macrophytes have been used effectively to treat different types of commonly include a wide range of metallic and organic compounds [3]. Wastewater treatment technology needs to be appro- priate and sustainable. It also needs to be less costly, easy to operate and maintain, and very efficient in removing both organic matter and heavy metals. In developing countries natural treatment systems, are more suitable. Natural treatment systems are considered one of the best treatment options, particularly in warm climates [4]. Wetlands with macrophytes are one of the many types of natural systems that can be used for treatment of municipal wastewater. Introduction According to Trepanier, a wetland Figure 2 Reduction of BOD5 with water hyacinth. Figure 2 Reduction of BOD5 with water hyacinth. municipal wastewater. According to Trepanier, a wetland Table 1 Characteristics of raw wastwater Sr. # Parameter Unit Value 1. BOD5 mg/l 132 2. COD 236 3. Nitrogen 2.65 4. Phosphorus 2.1 Figure 2 Reduction of BOD5 with water hyacinth. Table 1 Characteristics of raw wastwater Sr. # Parameter Unit Value 1. BOD5 mg/l 132 2. COD 236 3. Nitrogen 2.65 4. Phosphorus 2.1 Table 1 Characteristics of raw wastwater Figure 2 Reduction of BOD5 with water hyacinth. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 3 of 12 Figure 3 Reduction of BOD5 with duckweed. Figure 3 Reduction of BOD5 with duckweed. ponds. The investigation was conducted to evaluate the role of solar intensity, pH, dissolved oxygen, temperature, sedimentation, and attachment of faecal coliforms on Eichhornia crassipes on disappearance of bacteria in water hyacinths ponds. The results showed that environmental factors such as solar intensity and pH were the key factors when water hyacinths ponds have a large exposed surface area [9]. wastewaters. This is mainly due to their nutrient absorb- ing capacity, simplicity, low construction/operation and maintenance cost, low energy demand, process stability, potential benefits of the harvested materials [6]. The macrophytes have several properties in relation to the treatment processes. The most important ef- fects of the macrophytes in relation to the wastewater treatment processes are the physical effects of the plant tissues give rise to filtration effect and provide of surface area for attached microorganisms. The pol- lutants removal of macrophytes by plant uptake and oxygen release affects the wastewater treatment processes in different extends. The macrophytes provide habitat for wildlife [7]. Water pollution is becoming a serious issue of the en- tire world due to the rapid population growth, unsuit- able treatment technology and inadequate management. In Pakistan untreated municipal wastewater is indiscrim- inately discharged into water bodies. Rapid urbanization and industrialization have resulted in increased pollution load in the rivers and streams. In large cities municipal wastewaters from almost whole city along with commer- cial/industrial effluents is being discharged into water bodies in the immediate vicinity of the city (i.e. rivers, sur- face drains & canals). Introduction As a result pollution level in the water bodies is ever increasing due to the increase in the population and commercial/industrial development. There are several sophisticated treatment systems available, such as activated sludge process, rotating biological contactor, Zhang et al. [8] conducted investigation regarding the efficiency of macrophytes based treatment system in China for Municipal Wastewater Treatment. According to his findings Large-scale centralized wastewater treat- ment systems often prevail in industrial countries and have been regarded as a successful approach during the last century [8]. According to Mayo et al. [9] the removal of faecal coliforms was investigated in pilot-scale water hyacinths Figure 4 Reduction of BOD5 with water lettuce. Figure 4 Reduction of BOD5 with water lettuce. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 4 of 12 Figure 5 BOD removal efficiency versus OLR. igure 5 BOD removal efficiency versus OLR. and aerated lagoon, but they require high capital, oper- ational, and maintenance costs. In Pakistan treatment plant based of trickling filter was installed in Karachi while other based on activated sludge process was installed in Islamabad but both the treatment plant are non functional now a days due to high maintenance and no skilled staff availability. Waste stabilization ponds are currently func- tioning in Peshawar. due to the experience of wastewater treatment plants, biological treatment system with low operational and capital costs is preferred especially for developing countries like Pakistan, with warm climate all year round. Pakistan has sufficient land for natural waste- water treatment technology in the outskirts of cities. Thus, the maximum advantages of climate and land availability should be taken for wastewater treatment purpose. The attractive method is combination of waste stabi- lization pond along with macrophytes. The mechanism of treatment in this system is same as in constructed wetland. This method has been used as the effective low-cost technologies that require minimum energy to operate that are suitable for urban as well as for rural areas in Pakistan. Though, the treatment of wastewater by macrophyte plants has been started long before. The question how low aquatic plants can decrease the wastewater quality indicators still remain unanswered. Mainly macrophytes treat wastewater by organic matter uptake from the wastewater. at the roots of macrophytes small zones exist which arrest organic matter from wastewater. Introduction It is of interest to determine the lower bounds of pollutant contents that can reached because of their removal by aquatic plants and under what condition removal occur. This reflects on the range of application of aquatic plants (macrophytes) for wastewater treatment. There- fore, a physical Macrophyte based treatment plant model was constructed to treat municipal wastewater from University of Engineering& Technology (UET), Taxila. Wastewater was discharged into physical model con- taining Macrophytes. The objectives of the study aimed to evaluate the removal performance of pollutants as Figure 6 Reduction of COD with water hyacinth. Figure 6 Reduction of COD with water hyacinth. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 5 of 12 Figure 7 Reduction of COD with duckweed. Figure 7 Reduction of COD with duckweed. all the factors three macrophytes i.e. Water Hyacinth, Duckweed & Water Lettuce were selected. all the factors three macrophytes i.e. Water Hyacinth, Duckweed & Water Lettuce were selected. COD, BOD5 and nutrients (N&P) by different macro- phytes species. Municipal wastewater The municipal wastewater for the study was collected from the municipal sewer of UET Taxila containing wastewater of university and university colony. Analytical procedure The study was carried for 18 months in physical model. The BOD5, COD, and ammonia, Phosphorus were ana- lyzed according to Standard Methods for the Examination of Water and Wastewater. A circular storage tank (having volume of 2280 L) was used for collection of municipal wastewater from univer- sity sewer. Form that storage tank raw wastewater was distributed to individual model compartments (1.52 m (L)×1.83 m (W)×0.91 m (D)) containing different macro- phyte species. Figure 1 shows the layout of the experi- mental setup. Results and discussions The removal of pollutants by macrophytes may occur through a number of processes, including sedimentation, filtration, plant uptake/removal efficiency, adsorption, formation of solid compounds, and microbial-mediated reaction. BOD5 removal (i.e. 50.61% average reduction) of BOD5 as compared to Duckweed and Water Lettuce. Figures 2, 3 & 4 shows removal of BOD by water hyacinth, Duckweed and Water Lettuce respectively. It was ob- served that removal of BOD is mainly in first 10 days of each Experimental Run, after that the removal is at a slower rate. It is mainly due to plant update is much high in first 10 days similarly significant plant growth was observed during this period as well. In the study it was also found that the desirable concen- tration of BOD5 (i.e. <80 mg/l) as prescribed by National Environmental Quality Standards (NEQS, Pakistan) was achieved in water hyacinth system at 10th day in each run. Biochemical oxygen demand (BOD5) is a measure of the oxygen consumption of microorganisms in the oxidation of organic matter. Settleable organics are rap- idly removed in experimental system by quiescent condi- tions, deposition, and filtration. Attached and suspended microbial growth is responsible for removal of soluble BOD5. The influent concentrations were ranged in 95-160 mg/l showing the medium strength sewage. The BOD5 effluent concentrations at 30th day of each experiment run for Water Hyacinth system were 70 mg/l (Run-1), 76 mg/l (Run-2), 45 mg/l (Run-3), 59 mg/l (Run-4), 63 (Run-5) and 79 mg/l (Run-6). Where for Duckweed based system showed 81 mg/l (Run-1), 105 mg/l (Run-2), 60 mg/l (Run-3), 84 mg/l (Run-4), 84 (Run-5) and 110 mg/l (Run-6) at 30th day of each experimental run. Similarly Water Lettuce showed, 91 mg/l (Run-1), 110 mg/l (Run-2), 65 mg/l (Run-3), 89 mg/l (Run-4), 93 (Run-5) and 119 mg/l (Run-6) at 30th day of each experimental run. The re- sults showed that water hyacinth showed the maximum Biochemical oxygen demand (BOD5) is a measure of the oxygen consumption of microorganisms in the oxidation of organic matter. Settleable organics are rap- idly removed in experimental system by quiescent condi- tions, deposition, and filtration. Attached and suspended microbial growth is responsible for removal of soluble BOD5. The influent concentrations were ranged in 95-160 mg/l showing the medium strength sewage. The BOD5 effluent concentrations at 30th day of each experiment run for Water Hyacinth system were 70 mg/l (Run-1), 76 mg/l (Run-2), 45 mg/l (Run-3), 59 mg/l (Run-4), 63 (Run-5) and 79 mg/l (Run-6). Raw wastewater characteristics The selection of macrophytes for the study was done on the availability of macrophytes locally as well as the environmental conditions of the area. Keeping in view of The wastewater for study has been collected from main foul sewer line of UET Taxila, Pakistan. The characteris- tics of the untreated wastewater is described Table 1; Figure 8 Reduction of COD with water lettuce. Figure 8 Reduction of COD with water lettuce. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 6 of 12 Figure 9 Reduction of ammonia nitrogen with water hyacinth. Figure 9 Reduction of ammonia nitrogen with water hyacinth. BOD5 Similarly in Duckweed, it was observed that reduction of COD from initial value of 130 mg/l to final 87 mg/l (33.43% average reduction) and in Water Lettuce, the COD was reduced from initial 131 to final 94 mg/l (28.59% average reduc- tion) respectively. Around 30-40% of decrease in the parameters occurred within the first 10 days of the experi- ment. Result confirmed that the growth of macrophytes and showed high performance to remove COD mainly COD COD is the amount of oxygen necessary to oxidize the Organic Compound (OC) completely to CO2, H2O and NH3. COD is measured via oxidation with potassium di- chromate (K2Cr2O7) in the presence of sulfuric acid and silver and is expressed in mg/l. Thus, COD is a measure of the O2 equivalent of the organic matter as well as micro-organisms in the wastewater. If the COD value is much higher than the BOD value, the sample contains large amounts of organic compounds that are not easily biodegraded. Water Hyacinth was capable to decrease COD from its initial value to final value below National Environmental Quality Standards (NEQS), Pakistan i.e. <150 mg/L (50.61% average reduction). Similarly in Duckweed, it was observed that reduction of COD from initial value of 130 mg/l to final 87 mg/l (33.43% average reduction) and in Water Lettuce, the COD was reduced from initial 131 to final 94 mg/l (28.59% average reduc- tion) respectively. Around 30-40% of decrease in the parameters occurred within the first 10 days of the experi- ment. Result confirmed that the growth of macrophytes and showed high performance to remove COD mainly BOD5 Where for Duckweed based system showed 81 mg/l (Run-1), 105 mg/l (Run-2), 60 mg/l (Run-3), 84 mg/l (Run-4), 84 (Run-5) and 110 mg/l (Run-6) at 30th day of each experimental run. Similarly Water Lettuce showed, 91 mg/l (Run-1), 110 mg/l (Run-2), 65 mg/l (Run-3), 89 mg/l (Run-4), 93 (Run-5) and 119 mg/l (Run-6) at 30th day of each experimental run. The re- sults showed that water hyacinth showed the maximum BOD removal efficiencies was also observed against Organic Loading rate (OLR). The removal efficiencies resulting in different OLR are presented in Figure 5. It shows increase in removal efficiency with increase in OLR and also the optimum OLR of 112-113 kg BOD/ ha-d and by further increase in OLR results in reduction of system removal efficiency. Figure 10 Reduction of ammonia nitrogen with duckweed. Figure 10 Reduction of ammonia nitrogen with duckweed. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 7 of 12 Figure 11 Reduction of ammonia nitrogen with water lettuce. Figure 11 Reduction of ammonia nitrogen with water lettuce. Figure 11 Reduction of ammonia nitrogen with water lettuce. because of well-developed root system. Similarly it was observed that a major part of the degradation of COD in the wastewater is attributed to micro-organisms which may have established a symbiotic relationship with the plants. Figures 6, 7 & 8 shows removal of COD by Water Hyacinth, Duckweed and Water Lettuce respectively. COD COD is the amount of oxygen necessary to oxidize the Organic Compound (OC) completely to CO2, H2O and NH3. COD is measured via oxidation with potassium di- chromate (K2Cr2O7) in the presence of sulfuric acid and silver and is expressed in mg/l. Thus, COD is a measure of the O2 equivalent of the organic matter as well as micro-organisms in the wastewater. If the COD value is much higher than the BOD value, the sample contains large amounts of organic compounds that are not easily biodegraded. Water Hyacinth was capable to decrease COD from its initial value to final value below National Environmental Quality Standards (NEQS), Pakistan i.e. <150 mg/L (50.61% average reduction). Ammonia nitrogen (NH3-N) Urban wastewater contains high concentration of nutri- ents in addition to other pollutants. The major nutrients found in wastewater are N and P, which if not treated would cause number of problems to the environment es- pecially receiving water bodies. Excess of nutrients in water body cause overproduction of phytoplankton and resulted in O2 depletion. N is an essential nutrient and it enters a wetland in particulate and dissolved inorganic and organic forms . Particulate forms of N are removed through a series of process including ammonification, nitrification, denitrification and ammonia volatilization. In fresh sewerage, about 25% of the “N” is in the organic form and 75% in the ammonium form. The organic nitro- gen fraction is converted almost entirely to NH3-N, and Figure 12 Reduction of phosphorus with water hyacinth. Figure 12 Reduction of phosphorus with water hyacinth. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 8 of 12 Figure 13 Reduction of phosphorus with duckweed. Figure 13 Reduction of phosphorus with duckweed. further converted to nitrate- N (NO3-N) via microbial oxi- dation. Like COD, we have measured the N removal by different set of experiments. In Water Hyacinth system, the reduction of N ranging from 2.42 to 1.45 mg/l (40.34% average reduction) whereas the reduction in Duckweed ranges from 2.37 to 1.95 mg/l (17.59% average reduction) while in Water Lettuce reduced N from 2.42 to 2.09 mg/l (14.45% average reduction). The Water Hyacinth showed the highest level of N removal as compared to Duckweed and Water Lettuce. Result of present study showed that plants are significantly more efficient to remove N. In addition to plant uptake, N removal can occur by NH3 volatilization favoured by high pH, nitrification under aer- obic conditions and denitrification under anaerobic condi- tions and formation of organic films. In the present study, N removal was occurred by volatilization because pH was higher than 6.5. It has been observed that that a greater ratio of plant biomass to model volume can enhance the contact between plant roots and wastewater resulting in a greater nutrient removal. presence of plants significantly reduced the ammonia-N from their initially levels. General decline of ammonia- ni- trogen was found in all the experimental set up to 10 days thereafter the reduction is much lesser. Ammonia nitrogen (NH3-N) Reduction by Water hyacinth is much more than Duckweed and Water Lettuce. Phosphorus (PO4 −3) Phosphorus (P) is an essential nutrient for all life forms, and is the eleventh-most abundant mineral in the earth’s crust. P is needed for plant growth and is required for many metabolic reactions in plants and animals. Organic phosphorus (OP) is a part of living plants and animals, their by-products, and their remains. Generally, P is the limiting nutrient in freshwater aquatic systems. That is, if all P is used, plant growth will cease, no matter how much N is available. P typically functions as the “growth-limiting” factor because it is usually present in very low concentrations. The natural scarcity of phos- phorus can be explained by its attraction to organic matter and soil particles. Any unattached or “free” P is quickly removed from the aquatic system by aquatic plants. Excessive concentrations of P can quickly cause extensive Figures 9, 10 and 11 shows the removal of Nitrogen achieved in the study. At the end of experiment the Figure 14 Reduction of phosphorus with water lettuce. Figure 14 Reduction of phosphorus with water lettuce. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 Page 9 of 12 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 9 of 12 Figure 15 Effect of temperature variation on performance of macrophytes. Figure 15 Effect of temperature variation on performance of macrophytes. settling. Although initial level of P was low but the plant like Water Hyacinth with a well-developed root system further purify the wastewater. growth of aquatic plants. A normal adult excretes 1.3 - 1.5 g of P per day. Primary treatment removes only 10% of the P in the waste stream; secondary treatment removes only 30%. Tertiary treatment is required to remove add- itional P from the water. The amount of additional P that can be removed varies with the success of the treatment technologies used. Available technologies include biolo- gical removal and chemical precipitation. Factors affecting the performance Temperature In order to check the performance of macrophytes under variation of temperature condition additional experiments were conducted with pH 7.5. Temperature has been maintained in laboratory under controlled con- dition at different levels to check the performance variation in treatment of macrophytes. Figure 15 shows effect of temperature variation on performance of Water Hyacinth, Duckweed and Water Lettuce regarding BOD5 removal. During the experiments it was observed that macrophytes are sensitive to temperature and shows no growth and pollutant (BOD5) removal at a temperature below 10°C. Almost all three species cease to survive at this temperature. As the growth of species is negligible at temperature below 10°C therefore, there was no up- take of nutrients (N&P) by the plants. The temperature g Figures 12, 13 and 14 showed the removal of P in Water Hyacinth, Duckweed and Water Lettuce based systems respectively. Water Hyacinth showed maximum removal of P (18.76% average reduction) whereas Duckweed showed 15.25% average reduction and Water Lettuce showed 10.69% average reduction within the 30 days experimental period. The highest removal was observed inWater Hyacinth. This was due to synergistic effect of aquatic plant. Plants and micro-organisms all utilize P as an essential nutrient, and contain P in their tissues though the portion of tissue P is very small compared with C and N. Reduction of T-P may be due to uptake of soluble P, Filtration of particulate matter through the roots, and Figure 16 Effect of pH variation on performance of macrophytes. Figure 16 Effect of pH variation on performance of macrophytes. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 10 of 12 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Table 2 Plants (Macrophyte) height measurement Time (days) Plant height (ft) Run1 Run 2 Run 3 Run 4 Run 5 Run 6 WH WL WH WL WH WL WH WL WH WL WH WL 0 0.65 0.51 0.74 0.53 0.69 0.54 0.70 0.55 0.62 0.61 0.62 0.67 5 0.70 0.55 0.79 0.57 0.74 0.60 0.75 0.59 0.67 0.65 0.68 0.72 10 0.78 0.60 0.87 0.61 0.81 0.66 0.82 0.65 0.75 0.72 0.76 0.79 20 0.81 0.64 0.91 0.63 0.85 0.68 0.84 0.68 0.80 0.75 0.81 0.81 30 0.83 0.69 0.94 0.65 0.87 0.70 0.86 0.70 0.84 0.77 0.83 0.84 (WH = Water Hyacinth; WL = Water Lettuce). Plants biomass productivity d d p y Primary productivity and biomass are the important parameters. In general, the productivity of macro- phytes is higher than that of terrestrial communities and agricultural crops because they do not suffer from shortage of water. Macrophytes have high tolerance for the fluctuations in environment conditions and show high photosynthetic efficiencies. Uptake of nutrients by macro- phytes is an essential for their growth and reproduction. The high productivity of macrophytes enables substantial amounts of nutrients to be stored in plant biomass. Mea- surements of biomass were made after 5th, 10th, 20th and 30th day of each experimental run for Water Hyacinth, Water Lettuce and Duckweed. The plants biomass growth in the model was plotted and shown in Figures 17, 18 & 19. It is quite clear from the results that there is major rise in plants biomass in first ten however, in remaining twenty days it much less as compared to initial growth. pH In order to monitor the effect of the variation of pH on performance of macrophytes experiments in laboratory were conducted at a Temperature of 25°C and at differ- ent pH. At a pH below 5 macrophytes performance (BOD5 removal) in almost zero. This is mainly due to highly acidic nature of the wastewater. On the other hand when pH gradually increase performance improves upto 7.5 and by further increase in pH again start retarding macrophytes performance (BOD5 removal) and at a pH of 10 (at high Alkalinity) the performance of macrophytes was again decrease to zero. Therefore, a pH of 6-9 is most suitable for macrophytes performance. Figure 16 shows the effect of pH variation on the perfor- mance of Water Hyacinth, Duckweed and Water Lettuce regarding BOD5 removal. Factors affecting the performance Temperature Table 2 Plants (Macrophyte) height measurement between 15-38°C is suitable for treatment of municipal wastewater by macrophytes as high growth was observed at this temperature. filtered some filthy. They reduced inflow and they in- duce particles accumulated or precipitated in the sys- tems. Plant heights were observed to investigate plant growth during each trial. The measured plant heights has been given below in Table 2. Plants growth The plant stems filtered and reduced some particles in wastewater. When they died, they acted as net that Figure 17 Growth of water hyacinth in study model. Figure 17 Growth of water hyacinth in study model. Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 Page 11 of 12 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 Page 11 of 12 Figure 18 Growth of water lettuce in study model. Figure 18 Growth of water lettuce in study model. The results showed that the optimum period for har- vesting can be found to be 8-10 days. At the optimum point, the growth rate of the plant is lowest. was found that macrophytes gave optimum perform- ance at pH 6-9. Temperature is another factor that severely affect macrophytes performance and it was found a Temperature below 10°C macrophytes were unable to perform treatment and the favorable tempe- rature for treatment is 15-38°C. the growth of macro- phytes is more in first ten days of an experimental run because in first ten days maximum treatment to wastewater is provided by macrophytes by up-taking the organic matter. Pre-treatment of wastewater be- fore the plant acclimatization could be potentially ef- fective. The option of macrophytes for treatment of Municipal sewage under local environmental condi- tions can be explored by further verifying the removal efficiency under variation of different environmental conditions. Also this is need of time that macrophyte system should be used for treatment of wastewater be- cause their performance is comparable to conventional wastewater treatment plants and also the system has very low O&M costs. Author details 1 1Department of Civil and Environmental Engineering, UET, Taxila, Pakistan. 2Department of Civil Engineering, UET, Taxila, Pakistan. 3Civil Engineering Department, MCE, Rasalpur, Pakistan. Received: 21 July 2013 Accepted: 9 July 2014 Published: 16 July 2014 Authors’ contributions This research is a part of the thesis by MS who prepared the literature survey and performed the experiments. AA and ARG participated in the design of the study, data analysis, and manuscript preparation. HNH was the advisor. All the authors have read and approved the final manuscript. Acknowledgement The authors acknowledge the financial and scientific support provided by Department of Civil and Environmental Engineering, University of Engineering and Technology, Taxila, Pakistan. Competing interests Competing interests The authors declare that they have no competing interests. References 1. IWMI: A Framework for Global Assessment of the Extent of Wastewater Irrigation. The Need for a Common Wastewater Typology. Netherlands: International Water Management Institute; 2004:958. g 2. Trepanier C, Parent S, Comeau Y, Bouvrette J: Phosphorous budget as a 2. Trepanier C, Parent S, Comeau Y, Bouvrette J: Phosphorous budge t lit t t l f l d ti 2. Trepanier C, Parent S, Comeau Y, Bouvrette J: Phosphorous budget as a water quality management tool for closed aquatic mesocosms. J Water Res 2002, 36:1007–1017. water quality management tool for closed aquatic mesocosms. J Water Res 2002, 36:1007–1017. 3. Montaigne F, Essick P: Water pressure. Natl Geog 2002, 202:2 4. Duenas JF, Alonso JR, Rey AF, Ferrer AS: Characterization of phosphorous forms in wastewater treatment plants. J Hazard Mater 2003, 97:1–3. 5. Bitton G: Wastewater Microbiology. 3rd edition. Hoboken, NJ: John Wiley & Sons, Inc; 2005:7–746. 6. Metacalf and Eddy: Wastewater Engineering. New York: McGraw- Hill Inc; 1991. 7. Reed SC, Middlebrooks EJ, Crites RW: Natural Systems for Waste Management and Treatment. 2nd edition. New York: McGraw- Hill; 1988:3–307. 8. Zhang XB, Liu P, Yang YS, Chen WR: Phytoremediation of urban wastewater by model wetlands with ornamental hydrophytes. J Environ Sci 2007, 19:902–909. 8. Zhang XB, Liu P, Yang YS, Chen WR: Phytoremediation of urban wastewater by model wetlands with ornamental hydrophytes. J Environ Sci 2007, 19:902–909. 9. Mayo AW, Kalibbala M: Modelling faecal coliform mortality in water hyacinths ponds. Phys Chem Earth 2007, 32:1212–1220. 9. Mayo AW, Kalibbala M: Modelling faecal coliform mortality in water hyacinths ponds. Phys Chem Earth 2007, 32:1212–1220. doi:10.1186/2052-336X-12-106 Cite this article as: Shah et al.: Performance assessment of aquatic macrophytes for treatment of municipal wastewater. Journal of Environmental Health Science & Engineering 2014 12:106. Conclusion As far removal efficiencies are concerned Water Hyacinth is found most effective while considerable removals of pollutants were also found with Duckweed and Water Lettuce. Performance of Water Hyacinth based system was found to be 50.61% for BOD5, 46.38% for COD, 40.34% for Nitrogen and 18.76% for Phosphorus. For Duckweed based system the efficiencies were 33.43% for BOD5, 26.37% for COD, 17.59% for Nitrogen and 15.25% for Phosphorus. Similarly for Water Lettuce 33.43% for BOD5, 26.37% for COD, 17.59% for Nitrogen and 15.25% for Phosphorus. The mechanisms of pollutant removal in the system include both aerobic and anaerobic microbiological conversions, sorption, sedimentation, volatilization and chemical transformations. pH for the wastewater affect the performance of macrophytes and it Figure 19 Growth of duckweed in study model. Figure 19 Growth of duckweed in study model. Page 12 of 12 Page 12 of 12 Shah et al. Journal of Environmental Health Science & Engineering 2014, 12:106 http://www.ijehse.com/content/12/1/106 doi:10.1186/2052-336X-12-106 Cite this article as: Shah et al.: Performance assessment of aquatic macrophytes for treatment of municipal wastewater. Journal of Environmental Health Science & Engineering 2014 12:106. References 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/W4289519250	https://www.frontiersin.org/articles/10.3389/fnut.2022.938356/pdf	English	null	A putative causality of vitamin D in common diseases: A mendelian randomization study	Frontiers in nutrition	2,022	cc-by	6,189	TYPE Original Research PUBLISHED 02 August 2022 DOI 10.3389/fnut.2022.938356 TYPE Original Research PUBLISHED 02 August 2022 DOI 10.3389/fnut.2022.938356 OPEN ACCESS OPEN ACCESS EDITED BY Qi Feng, University of Oxford, United Kingdom REVIEWED BY Andrew Grant, University of Cambridge, United Kingdom Armin Zittermann, Heart and Diabetes Center North Rhine-Westphalia, Germany Ningjian Wang, Shanghai Jiao Tong University, China CORRESPONDENCE Xia Jiang xia.jiang@ki.se Xiujun Cai srrsh_cxj@zju.edu.cn SPECIALTY SECTION This article was submitted to Nutritional Epidemiology, a section of the journal Frontiers in Nutrition RECEIVED 07 May 2022 ACCEPTED 14 July 2022 PUBLISHED 02 August 2022 CITATION Hui Liu1, Xudan Shen2, Tunan Yu3, Yifan Wang3, Sheng Cai2, Xia Jiang4 and Xiujun Cai1,3* Hui Liu1, Xudan Shen2, Tunan Yu3, Yifan Wang3, Sheng Cai2, Xia Jiang4* and Xiujun Cai1,3* 1Zhejiang Provincial Key Laboratory of Laparoscopic Technology, Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University, Hangzhou, China, 2Zhejiang Province Key Laboratory of Anti-Cancer Drug Research, Institute of Drug Metabolism and Pharmaceutical Analysis, Zhejiang University, Hangzhou, China, 3Department of General Surgery, Sir Run Run Shaw Hospital, School of Medicine, Zhejiang University, Hangzhou, China, 4Department of Clinical Neuroscience, Center for Molecular Medicine, Karolinska Institute, Stockholm, Sweden Backgrounds: Vitamin D is considered as a nutrient protecting individuals against an array of diseases based on observational studies. Such a protective efect, however, has not been demonstrated by randomized controlled trials. This study aims to explore a putative causal role of vitamin D in common diseases through a two-sample Mendelian randomization (MR) framework. Methods: Circulating vitamin D was predicted by 41 genetic variants discovered in European populations. Common diseases were veriﬁed through two ways, using information from Japanese patients of Biobank Japan and using information from European patients of FinnGen project. We additionally validated the results by replacing vitamin D-associated instrumental variables (IVs) of European population with that of an independent Japanese population and of an independent Indian population. Inverse-variance weighted method was used as the primary analytical approach while a series of MR methods including MR-Egger regression, weighted median, maximum likelihood, MR-PRESSO and multivariate MR were adopted to guarantee MR model assumptions and to detect horizontal pleiotropy. CITATION Liu H, Shen X, Yu T, Wang Y, Cai S, Jiang X and Cai X (2022) A putative causality of vitamin D in common diseases: A mendelian randomization study. Front. Nutr. 9:938356. doi: 10.3389/fnut.2022.938356 COPYRIGHT © 2022 Liu, Shen, Yu, Wang, Cai, Jiang and Cai. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). vitamin D, Mendelian randomization, Biobank Japan, UK Biobank, FinnGen study Data for IV-exposure Vitamin D, mainly existing in the body as 25- hydroxyvitamin D [25(OH)D], has been reported to be negatively associated with skeletal diseases such as fractures and falls, as well as various non-skeletal chronic diseases including cancer, cardiovascular diseases, and metabolic disorders (1, 2). However, ﬁndings of clinical trials did not support the protective role of vitamin D supplementation in decreasing the risk of falls and fractures (3–5). In addition, a few studies found that higher vitamin D level was related to a higher risk of colon cancer (6) and prostate cancer (7). Gathering above evidences, the role of vitamin D in diverse diseases may be diﬀerent and complex. Besides, it is uncertain whether there is a putative causal link between vitamin D and diseases, since observational studies may be susceptible to confounding and reverse causation. Therefore, Mendelian randomization (MR), a method using instrumental variable (IV) to unveil the casual relationship, has been comprehensively applied to explain the eﬀect of modiﬁable exposures (8, 9). We retrieved the summary statistics for the eﬀect size of SNPs signiﬁcantly associated with circulating 25(OH)D concentration from a meta-GWAS of vitamin D involving a GWAS in 401,460 white British subjects from UK Biobank (UKBB) cohort and another independent GWAS in 42,274 European participants (13). Since 25(OH)D level was ﬁrst log- transformed and then standardized in meta-GWAS analysis, all eﬀect estimates in this MR-PheWAS study corresponded to log-transformed 25(OH)D concentration per standard deviation increase. Totally, 138 conditionally independent SNPs passing genome-wide signiﬁcance (P < 5 × 10−8) were identiﬁed by GCTA-COJO v.1.91.1 using the meta-GWAS, and used as our exposure of interest (15). The total variance explained by these vitamin D associated SNPs was 4.9%. In addition, we performed quality control on genetic variants of exposure of interest. The selection criterion was biallelic SNPs with P value reaching genome-wide signiﬁcance (P < 5.0 × 10−8), minor allele frequency (MAF) >0.01, call rate >95%, linkage disequilibrium r2 >0.9, and P value for Hardy- Weinberg equilibrium >1.0 × 10−6. Besides, the palindromic SNPs that could not verify whether alleles were correctly orientated, and the pleiotropic SNPs which were associated with other traits other than the exposure and outcomes of interest according to GWAS Catalog (accessed on June 2022) were excluded. Detailed information on vitamin D instruments was shown in Supplementary Table 1. Several MR studies have interrogated the causal role of vitamin D in various diseases. Ong et al. Data for IV-exposure have assessed the association between vitamin D and the susceptibility of 10 cancers and revealed no relationship between vitamin D and cancers, with the exception of ovarian cancer (OR = 0.89, 95%CI:0.82–0.96) (10). Additionally, an earlier phenome- wide MR (MR-PheWAS) analysis integrating the phenome- wide association study (PheWAS) and MR method found there was no evidence of a moderate or large (OR>1.2) causal eﬀect of vitamin D on plenty of health outcomes (11). Yet this MR-PheWAS study used six SNPs as IVs, which only explained 2.84% variance of vitamin D, and did not have enough power to detect small causal eﬀects of vitamin D on diseases. Since an updated genome-wide association study (GWAS) of vitamin D was published, a recent MR study investigated the causal links between genetically predicted vitamin D by 143 variants and 106 diseases/traits, and discovered the protective roles of vitamin D in height, ovarian cancer, multiple sclerosis, and fracture (12). Despite that, all three MR studies identiﬁed common diseases in European populations, limiting the generalization to other populations. In order to ensure a valid MR result, strong IV is essential element, which was reﬂected by F-statistics. It is calculated using formula F = R2(n−1−k) (1−R2)k , where R2 is the proportion of variance explained by the IVs, k refers to the number of IVs, and n indicates the sample size (16). The F-statistics for 25(OH)D level was 203.9, indicating strong IVs (F-statistics > 10) for our exposure of interest. OPEN ACCESS 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. Results: Genetically predicted vitamin D was signiﬁcantly associated with an increased risk of Graves’ disease (OR = 1.71, 95%CI: 1.25–2.33, P = 0.001) and cataract (OR = 1.14, 95%CI: 1.03–1.28, P = 0.016); while with a decreased risk of esophageal cancer (OR = 0.66, 95%CI: 0.46–0.93, P = 0.019). This signiﬁcant causal link between vitamin D and cataract was validated replacing IVs identiﬁed in the European population with those from Japanese population. No notable associations of vitamin D with other diseases were observed. Conclusions: Our ﬁndings indicate a potential causal role of vitamin D in common diseases, which needs further validation. KEYWORDS 01 Frontiers in Nutrition frontiersin.org Liu et al. Liu et al. 10.3389/fnut.2022.938356 Frontiers in Nutrition Statistical analysis eﬀect estimates and reference alleles between vitamin D GWAS and outcome GWAS. Sample sizes for the 42 diseases ranged from 90,336 to 212,453 (Supplementary Table 2). MR analysis, which uses SNPs as IVs, can infer the causal eﬀects of an exposure [25(OH)D concentration] on outcomes (multiple diseases), since SNPs are assumed to be randomly allotted and occurs before the onset of disease. It is crucial to assure valid IVs so that to deduce the right causal link, where three essential assumptions should be satisﬁed (19): (1) IVs should be associated with the exposure; (2) IVs should not be associated with any confounders of exposure-outcome relationship; (3) only through the exposure of interest can IVs aﬀect the outcome. Brieﬂy, 42 diseases covered seven categories including thirteen types of cancer (lung cancer, breast cancer, gastric cancer, colorectal cancer, prostate cancer, gallbladder/cholangiocarcinoma, cervical cancer, uterine cancer, esophageal cancer, hematopoietic tumor, liver cancer, ovarian cancer and pancreas cancer), four respiratory tract diseases (bronchial asthma, chronic obstructive pulmonary disease, interstitial lung disease/pulmonary ﬁbrosis and pulmonary tuberculosis), three cardiovascular diseases (arrhythmia, coronary artery disease and heart failure), three chronic liver diseases (chronic hepatitis B, chronic hepatitis C and liver cirrhosis), two cerebrovascular disorders (cerebral aneurysm and cerebral infarction), two eye diseases (cataract and glaucoma), ﬁfteen others disease (atopic dermatitis, drug eruption, endometriosis, epilepsy, Graves’ disease, keloid, nephrotic syndrome, osteoporosis, peripheral artery disease, periodontitis, hay fever, rheumatoid arthritis, diabetes mellitus, uterine ﬁbroids and urolithiasis). To establish the causation of vitamin D with the risk of common diseases, two-sample MR analyses were performed, in which the populations of exposure of interest and outcomes of interest came from two non-overlapping populations. Inverse- variance weighted (IVW) approach was applied as our main MR method, which calculated the ratio of SNP-outcome association to SNP-exposure association and then combined the multiple IVs weighted by the inverse of their variances (20). Cochran’s Q test was used to assess the potential heterogeneity and P value <0.05 was considered as the existence of heterogeneity. As complements of IVW method, weighted median method and maximum likelihood method were applied to test the robustness of IVW approach. Weighted median method used the inverse of the variance of the ratio estimates as weights and calculated the median as the estimated casual eﬀect, under the condition that at least 50% genetic variants were valid IVs (21). Data for IV-outcome Diseases were identiﬁed from BBJ Project, which established a large-scale database containing genomic information (covering 8,712,794 autosomes genetic variants) and diseases status in non-European population (14). BBJ database recruited 260,000 Japanese patients representing 440,000 cases of 51 types of common diseases, where each individual was followed up with an average of 10 years. The variant annotation (SNP rsID, chromosome position, eﬀect and reference allele) and summary-level data (beta coeﬃcients, standard error, and P value) for each SNP with 42 diseases were extracted (https://humandbs.biosciencedbc.jp/en/hum0014- v21#42diseases; https://biobankjp.org/), and harmonized the As an extension of these evidences, a MR-PheWAS analysis is performed to discover the causal links between vitamin D and common diseases in Biobank Japan cohort. One hundred thirty-eight independent SNPs identiﬁed by the most recently published vitamin D GWAS study were used as IVs, which was conducted in a total of 443,734 European individuals (13). In addition, Biobank Japan (BBJ) project is a largest Biobank project with a total of 260,000 patients of non-European population covering 51 common diseases (14). This study aims to provide a comprehensive understanding of the causal role of vitamin D in multiple diseases. Frontiers in Nutrition frontiersin.org 02 Liu et al. 10.3389/fnut.2022.938356 10.3389/fnut.2022.938356 Statistical analysis Maximum likelihood method assumed a liner regression between the risk factor and outcome, as well as a bivariate normal distribution for the genetic estimates (22). Validation in other populations Given that the population of exposure was European, whereas the population of outcome was Japanese, we replicated analyses in populations of similar genetic background. With regard to IV-exposure, we extracted summary level data of the associations of ﬁve SNPs with serum 25(OH)D concentration from Japanese patients with rheumatoid arthritis (N = 1957) (17). Due to the limitation of sample size and the number of valid IVs, we then extracted summary statistics of SNPs associated with 25(OH)D concentration from a GWAS study of vitamin D in Asian Indian diabetic heart study/Sikh diabetes study (AIDHS/SDS) (18). AIDHS/SDS comprised of 3,538 subjects from South Asian population and found 24 putative SNPs signiﬁcantly associated with 25(OH)D level using a two stage GWAS analysis. The main issue of MR analyses was horizontal pleiotropy which the variant had an eﬀect on outcome outside of its eﬀect on the exposure, herein, two MR methods were adopted. Firstly, MR-Egger regression, a regression approach taking the existence of intercept terms into account, detected the potential pleiotropy by the non-zero intercept under the instrument strength independent of direct eﬀect (InSIDE) assumption (23). Secondly, MR-pleiotropy residual sum and outlier (MR- PRESSO) method, which identiﬁed a speciﬁc SNP by the diﬀerence between calculated and expected values of the residual sum of squares using a leave-one-out approach, detecting horizontal pleiotropy by a global test (P<1.0×10−6) and correcting the estimates excluding SNP outliers (24). Regarding the IV-outcome, summary statistics of genetic variants associated with the risk of outcome from FinnGen database were used. The FinnGen study is a large personalized medicine project covering 500,000 Finnish biobank participants, and aims to provide the evidence of genomic eﬀect on human health. We extracted relative data from data freeze 7 of the FinnGen study, which consisted of 321,302 individuals with almost 17 million genetic variants detected (https://console.cloud.google.com/ storage/browser/ﬁnngen-public-data-r7/summary_stats/). In the FinnGen study, summary statistics of ﬁve diseases including Graves’ disease, senile cataract, other cataract, benign esophageal cancer and malignant esophageal cancer were extracted. Moreover, we conducted multivariable MR (MVMR) analysis, which provided estimations of independent direct eﬀects of multiple and potentially related exposures [i.e., 25(OH)D level and body mass index (BMI)] on an outcome (25). Considering that 25(OH)D levels are closely related with BMI, summary statistics for BMI estimated from UKBB participants of GIANT Consortium were used in the MVMR model (26). frontiersin.org Validation in other populations Statistical signiﬁcance was regarded as P value <0.05 according to IVW method, while directional consistency Frontiers in Nutrition frontiersin.org 03 10.3389/fnut.2022.938356 Liu et al. Liu et al. and signiﬁcance were veriﬁed by other MR methods. Bonferroni correction method was used to account for multiple comparisons, where we set a more conservative P value threshold using 0.05 divided by the number of outcomes (0.05/42 ≈1.0 × 10−3) to avoid false positive results taken by an insuﬃciently rigorous threshold. heterogeneity according to IVW method (P > 0.05), neither obvious pleiotropy based on MR-Egger regression (P > 0.05) and MR-PRESSO method (P > 1.0 × 10−6). Findings of the casual links between vitamin D and multiple diseases were shown in Supplementary Table 3. Despite signiﬁcant association between vitamin D and atopic dermatitis was noted using MR-Egger regression (OR = 0.40, 95%CI: 0.24– 0.66) and weighted median method (OR = 0.56, 95%CI: 0.35– 0.89), non-signiﬁcant relationship was observed using IVW. According to our criteria of statistical signiﬁcance, we regarded the above result as null association. According to the results of MR-PRESSO method (Supplementary Table 4), two outcomes including coronary artery disease and diabetes mellitus indicated the presence of pleiotropy (P < 1.0 × 10−6), however, the results did not change after removing outlier SNPs. Since the number of SNPs used as IVs dramatically decreased in the MVMR model, the causal associations of vitamin D with the risk of Graves’ disease, cataract and esophageal cancer were attenuated toward null (Supplementary Table 5). Results The procedure of SNPs selection is displayed in Figure 1. Among the 138 genetic variants signiﬁcantly associated with circulating 25(OH)D concentrations, 41 SNPs that were successfully matched to common diseases in BBJ project were included in our primary MR analysis. According to IVW approach, we found vitamin D was associated with a higher risk of Graves’ disease (OR = 1.71, 95%CI: 1.25–2.33) and cataract (OR = 1.14, 95%CI: 1.03–1.28); while with a lower risk of esophageal cancer (OR = 0.66, 95%CI: 0.46–0.93). After multiple comparisons correction, the causal relationship between vitamin D and Graves’ disease remained signiﬁcant (P = 0.001). All results remained directional consistent and signiﬁcant in other MR methods including weighted median method, maximum likelihood method, MR-Egger regression and MR-PRESSO (Table 1). Moreover, we found no obvious Considering the population heterogeneity between exposure and outcomes of interest, we ﬁrstly sought summary statistics of genetic variants predicting 25(OH)D concentration in Japanese population. As presented in Table 2, the strength of the causal estimate between vitamin D and cataract was stronger, the odds ratio calculated using the IVW method FIGURE 1 Flowchart on the selection of instrumental variables. FIGURE 1 Flowchart on the selection of instrumental variables. FIGURE 1 Flowchart on the selection of instrumental variables. 04 04 Frontiers in Nutrition frontiersin.org Liu et al. Liu et al. 10.3389/fnut.2022.938356 TABLE 1 Causal link between genetically predicted vitamin D in European population and diseases identiﬁed in individuals from Biobank Japan. n genetically predicted vitamin D in European population and diseases identiﬁed in individuals from Biobank Japan. TABLE 1 Causal link between genetically predicted vitamin D in European population and diseases identiﬁed in individuals from Biobank Japan. Results Disease Method N of IVs OR (95%CI) P P* Graves’ disease Inverse-variance weighted 41 1.71 (1.25–2.33) 0.001 0.657 Weighted median 41 1.83 (1.18–2.83) 0.007 Maximum likelihood 41 1.71 (1.24–2.38) 0.001 MR-Egger 41 2.10 (1.29–3.42) 0.005 0.264 MR-PRESSO 41 1.71 (1.25–2.33) 0.002 0.696 Cataract Inverse-variance weighted 41 1.14 (1.03–1.28) 0.016 0.425 Weighted median 41 1.14 (0.99–1.32) 0.076 Maximum likelihood 41 1.14 (1.03–1.27) 0.014 MR-Egger 41 1.16 (0.98–1.36) 0.089 0.851 MR-PRESSO 41 1.14 (1.03–1.28) 0.020 0.454 Esophageal cancer Inverse-variance weighted 41 0.66 (0.46–0.93) 0.019 0.921 Weighted median 41 0.66 (0.38–1.12) 0.123 Maximum likelihood 41 0.66 (0.43–1.00) 0.050 MR-Egger 41 0.61 (0.33–1.13) 0.124 0.733 MR-PRESSO 41 0.66 (0.46–0.93) 0.024 0.943 P indicates P value of heterogeneous from IVW approach, or P value of intercept from MR-Egger regression, or P value from Mendelian randomization pleiotropy residual sum and outlier (MR-PRESSO) global test. ausal link between genetically predicted vitamin D in Asian population and diseases identiﬁed in individuals from Biobank Japan. Disease Data source of vitamin D N of IVs OR (95%CI) P P Graves’ disease Japanese population 2 1.34 (0.040–50.66) 0.875 0.040 Indian population 14 1.05 (0.74–1.49) 0.773 0.374 Cataract Japanese population 2 1.94 (1.90–1.98) <0.001 0.973 Indian population 14 1.08 (0.98–1.19) 0.106 0.735 Esophageal cancer Japanese population 2 1.00 (0.08–13.13) 0.998 0.253 Indian population 14 1.67 (1.08–2.57) 0.020 0.438 P indicates P value of heterogeneous from IVW approach. P indicates P value of heterogeneous from IVW approach. outcomes identiﬁed in European population were in general null. outcomes identiﬁed in European population were in general null. was 1.94 (95%CI: 1.90–1.98). Yet, the causal associations of vitamin D with the risk of Graves’ disease and esophageal cancer were attenuated toward null. Since there were only 2 SNPs regarded as IVs in the Japanese population, the results were further replicated in another Asian population (Indian) that provided more valid IVs to genetically predict vitamin D. Unfortunately, we did not observe any statistically signiﬁcant causal association of vitamin D with the risk of either Graves’ disease or cataract in the Indian population. Nevertheless, an increased risk of esophageal cancer was found in this population (OR = 1.67, 95%CI:1.08–2.57). Results of the causal estimates between vitamin D and common diseases identiﬁed in Asian populations were shown in Supplementary Tables 6, 7. Frontiers in Nutrition Discussion This phenome-wide MR study explored the causal role of vitamin D in a large quantity of common diseases, including cancers, cardiovascular disease, respiratory tract disease, eye disease, and chronic liver disease, which were identiﬁed among more than 200,000 Japanese participants. We found the potential causal associations of vitamin D with the risk of Graves’ disease, cataract and esophageal cancer in individuals from Biobank Japan. Moreover, the causal link between vitamin D and Graves’ disease remained signiﬁcant at a more conservative P value threshold (P = 0.001). In addition, the signiﬁcant causal relationship of vitamin D with the risk of cataract was veriﬁed by replacing vitamin D instruments of European population with Japanese population. Regarding the outcome of interest, summary statistics of Japanese population from Biobank Japan were replaced with European population from FinnGen study for additional validation analysis. As noted in Table 3, the causal links between vitamin D and diﬀerent Frontiers in Nutrition frontiersin.org 05 Liu et al. Liu et al. 10.3389/fnut.2022.938356 TABLE 3 Causal link between genetically predicted vitamin D in European population and diseases identiﬁed in European individuals from FinnGen study. Disease Method N of IVs OR (95%CI) P P* Graves’ disease Inverse-variance weighted 43 1.34 (0.91–1.98) 0.136 0.132 Weighted median 43 1.25 (0.77–2.05) 0.370 Maximum likelihood 43 1.34 (0.95–1.90) 0.096 MR-Egger 43 1.35 (0.82–2.20) 0.241 0.980 Senile cataract Inverse-variance weighted 43 0.99 (0.90–1.10) 0.911 0.030 Weighted median 43 0.94 (0.83–1.08) 0.389 Maximum likelihood 43 0.99 (0.91–1.08) 0.893 MR-Egger 43 1.01 (0.89–1.15) 0.823 0.601 Other cataract Inverse-variance weighted 43 0.84 (0.72–0.97) 0.021 0.060 Weighted median 43 0.73 (0.59–0.89) 0.002 Maximum likelihood 43 0.84 (0.73–0.95) 0.007 MR-Egger 43 0.79 (0.65–0.95) 0.019 0.322 Benign esophageal cancer Inverse-variance weighted 43 0.90 (0.42–1.95) 0.793 0.933 Weighted median 43 1.52 (0.44–5.25) 0.507 Maximum likelihood 43 0.90 (0.36–2.27) 0.827 MR-Egger 43 1.43 (0.45–4.52) 0.546 0.197 Malignant esophageal cancer Inverse-variance weighted 43 1.00 (0.48–2.08) 0.998 0.185 Weighted median 43 0.77 (0.32–1.86) 0.563 Maximum likelihood 43 1.00 (0.51–1.96) 0.997 MR-Egger 43 0.68 (0.27–1.67) 0.401 0.163 P indicates P value of heterogeneous from IVW approach, or P value of intercept from MR-Egger regression. P indicates P value of heterogeneous from IVW approach, or P value of intercept from MR-Egger regression. Age-Related Eye Disease Study (OR = 0.97, 95%CI: 0.65–1.45) (32). Discussion As reported by the IVW method, we found vitamin D to be associated with an increased 14% risk of cataract (OR = 1.14, 95%CI: 1.03–1.28), which was in accordance with a previous MR study (12). It indicated an increased risk of cataract using 60 SNPs genetically predicting vitamin D in a population from UK Biobank, where the odds ratio was 1.01(95%CI: 1.00–1.02) with nominal statistical signiﬁcance. Our ﬁndings refuted the protective role of vitamin D suggested by previous observational studies. Vitamin D deﬁciency has been considered to play a role in the occurrence and development of Graves’ disease, which was indicative of lower circulating 25(OH)D level for Graves’ disease patients (55.0 ± 23.2 vs. 87.2 ± 27.6 nmol/L) (27, 28). Another case- control study including 51 cases with Graves’ disease and 51 healthy controls found there was no signiﬁcant diﬀerence of 25(OH)D levels (81.77 ± 5.60 vs. 83.49 ± 6.24 nmol/L) (29). Although these results contradicted with ours which indicated vitamin D was associated with an 71% increased risk of Graves’ disease (OR = 1.71, 95%CI: 1.25–2.33), further large-scale studies need to be performed. Previous investigations demonstrated that vitamin D had no association with the risk of esophageal cancer. A MR study using six SNPs associated with vitamin D level as IVs found the odds ratio for esophageal cancer was 0.68 (95%CI:0.39–1.19) among 4,112 patients with esophageal cancer and 17,159 controls (33). Using a larger set of variants associated with vitamin D level (76 SNPs) identiﬁed by UKBB, Ong et al. did not ﬁnd vitamin D was to be a causal risk factor for esophageal cancer (OR = 0.97, 95%CI: 0.78–1.20) (10). Yet our study indicated that genetically predicted vitamin D was associated with a 34% decreased risk of esophageal cancer. Previous investigations considered vitamin D had a protective role in preventing diseases, such as cataract and esophageal cancer. A case-control study discovered mean 25(OH)D level in cataract patients were signiﬁcantly lower than that in control group (7.6 ± 5.5 vs. 18.5 ± 9.6 ng/mL) (30). Another study based on Korean National Health and Nutrition Examination Survey found the inverse relationship between serum 25(OH)D level and the risk of nuclear cataract where the odd ratio was 0.86(95%CI:0.75–0.99) (31). Frontiers in Nutrition Funding This study was supported by National Natural Science Foundation of China (No. 82103807). Acknowledgments We thank participants and fellows involving in UK Biobank Project, Biobank Japan Project, and FinnGen Project. Data availability statement example, a protective eﬀect of vitamin D predicted by 143 independent loci on ovarian cancer in participants from UKBB was found by Ye et al. with the odds ratio of 0.96 (95%CI: 0.94–0.98) (12), which was not conﬁrmed in our study (OR = 0.77, 95%CI:0.54–1.10). The discrepancies may be explained by diﬀerent IVs used to genetically predict vitamin D level, population heterogeneity from both exposure and outcome of interest, or diﬀerent threshold of statistical signiﬁcance. 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. In conclusion, this MR study indicated a potential causal relationship between vitamin D and multiple diseases, and a more obvious causal link between vitamin D and Graves’ disease. It may have implications for disease prevention; however, this causal link needs to validate or refute in a large prospective study. Ethics statement According to traditional theory, the protective role of vitamin D was mainly played by regulation of the immune system, such as activation of cytolytic T cells, and up-regulation of the anti-microbial peptide CAMP or the plasma membrane- anchored glycoprotein CD14 (34). Nevertheless, one possible reason for the raised risk of cataract with high vitamin D level was that a higher vitamin D concentration yields to a stronger absorption of dietary calcium, a risk factor for cataract (35). Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements. A major strength of our investigation was performing a phenome-wide MR study with a large number of genetic variants and common diseases in a large-scale Japanese population. Yet several limitations cannot be ruled out. Firstly, we performed a large amount of health outcomes in this MR study, which increased the risk of false positive ﬁndings. Thus, we deﬁned the threshold of statistical signiﬁcance as 1.0 × 10−6 using Bonferroni correction method. Secondly, 41 independent loci associated with circulating 25(OH)D concentration was used as IVs, which might violate the exclusion restriction assumption of a valid MR analysis, because some IVs may have an eﬀect on diseases not mediated by vitamin D. In order to test the horizontal pleiotropy, two MR methods including MR- Egger regression and MR-PRESSO method were conducted, where the results by these methods were consistent with that by the IVW approach. In addition, we assumed that the relationships between vitamin D and common diseases were linear, however, the shape of association between vitamin D and disease such as prostate cancer (36), cardiovascular disease and all-cause mortality (37) was non-linear. Last but not least, two sample populations of exposure and outcomes of interest came from diﬀerent ancestry, where exposure was from white British but outcome was from Japanese. Although this met the non-overlap sample requirement of two sample MR method, it may introduce the population heterogeneity and limit the generalizability of results. However, the signiﬁcant causal link between vitamin D and cataract remained by replicating the analyses in a population which exposure and outcome were both from Japanese. Herein, a large-scale phenome-wide MR study composed with mixed populations was warranted. Author contributions HL, XJ, and XC designed the study. HL, TY, and YW did literature research and collected data. HL, XS, and SC undertook analyses and interpreted the results. HL and XS wrote the ﬁrst draft of the manuscript. XJ and XC had primary responsibility for ﬁnal content. All authors reviewed the manuscript. All authors contributed to the article and approved the submitted version. Discussion On the contrary, there was no notable association of vitamin D and nuclear cataract among women of all ages in the Carotenoids in In general, the null ﬁndings for multiple health outcomes including most types of cancers, cardiovascular and cerebrovascular disease, autoimmune disease, and metabolic disease of our MR analysis were in accordance with previous MR results. On the contrary, there are some disparities. For Frontiers in Nutrition frontiersin.org 06 Liu et al. 10.3389/fnut.2022.938356 Frontiers in Nutrition References Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent Estimation in Mendelian Randomization with Some Invalid Instruments Using a Weighted Median Estimator. Genet Epidemiol. (2016) 40:304–14. doi: 10.1002/gepi.21965 5. Kahwati LC, Weber RP, Pan H, Gourlay M, LeBlanc E, Coker-Schwimmer M, et al. Vitamin D, calcium, or combined supplementation for the primary prevention of fractures in community-dwelling adults: evidence report and systematic review for the US preventive services task force. JAMA. (2018) 319:1600–12. doi: 10.1001/jama.2017.21640 22. S. Burgess A, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol. (2013) 37:658–65. doi: 10.1002/gepi.21758 6. Weinstein SJ, Yu K, Horst RL, Ashby J, Virtamo J, Albanes D. Serum 25- hydroxyvitamin D and risks of colon and rectal cancer in Finnish men. Am J Epidemiol. (2011) 173:499–508. doi: 10.1093/aje/kwq398 23. Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: eﬀect estimation and bias detection through Egger regression. Int J Epidemiol. (2015) 44:512–25. doi: 10.1093/ije/dyv080 7. Xu Y, Shao X, Yao Y, Xu L, Chang L, Jiang Z, et al. Lin. Positive association between circulating 25-hydroxyvitamin D levels and prostate cancer risk: new ﬁndings from an updated meta-analysis. J Cancer Res Clin Oncol. (2014) 140:1465– 77. doi: 10.1007/s00432-014-1706-3 24. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. (2018) 50:693–8. doi: 10.1038/s41588-018-0164-2 8. Burgess S, Butterworth A, Malarstig A, Thompson SG. Use of Mendelian randomisation to assess potential beneﬁt of clinical intervention. BMJ. (2012) 345:e7325. doi: 10.1136/bmj.e7325 25. Sanderson E. Multivariable mendelian randomization and mediation. Cold Spring Harb Perspect Med. (2021) 11:a038984. doi: 10.1101/cshperspect.a038984 9. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. (2014) 23:R89– 98. doi: 10.1093/hmg/ddu328 26. Yengo L, Sidorenko J, Kemper KE, Zheng Z, Wood AR, Weedon MN, et al. Meta-analysis of genome-wide association studies for height and body mass index in approximately 700000 individuals of European ancestry. Hum Mol Genet. (2018) 27:3641–9. doi: 10.1093/hmg/ddy271 10. Ong JS, Dixon-Suen SC, Han X, An J, Esophageal Cancer C, Me Research T, et al. A comprehensive re-assessment of the association between vitamin D and cancer susceptibility using Mendelian randomization. Nat Commun. (2021) 12:246. doi: 10.1038/s41467-020-20368-w 27. Planck T, Shahida B, Malm J, Manjer J. Publisher’s note All claims expressed in this article are solely those of the authors and do not necessarily represent those 07 frontiersin.org Liu et al. Liu et al. 10.3389/fnut.2022.938356 Supplementary material 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. The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/ fnut.2022.938356/full#supplementary-material References 1. Theodoratou E, Tzoulaki I, ZgagaL, Ioannidis JP. Vitamin D and multiple health outcomes: umbrella review of systematic reviews and meta- analyses of observational studies and randomised trials. BMJ. (2014) 348:g2035. doi: 10.1136/bmj.g2035 1. Theodoratou E, Tzoulaki I, ZgagaL, Ioannidis JP. Vitamin D and multiple health outcomes: umbrella review of systematic reviews and meta- analyses of observational studies and randomised trials. BMJ. (2014) 348:g2035. doi: 10.1136/bmj.g2035 17. Yoshida S, Ikari K, Furuya T, Toyama Y, Taniguchi A, Yamanaka H, et al. A GC polymorphism associated with serum 25-hydroxyvitamin D level is a risk factor for hip fracture in Japanese patients with rheumatoid arthritis: 10-year follow-up of the Institute of Rheumatology, Rheumatoid Arthritis cohort study. Arthritis Res Ther. (2014) 16:R75. doi: 10.1186/ar4516 2. Bouillon R, Marcocci C, Carmeliet G, Bikle D, White JH, Dawson-Hughes B, et al. Skeletal and extraskeletal actions of vitamin d: current evidence and outstanding questions. Endocr Rev. (2019) 40:1109–51. doi: 10.1210/er.2018-00126 18. Sapkota BR, Hopkins R, Bjonnes A, Ralhan S, Wander GS, Mehra NK, et al. Genome-wide association study of 25(OH) Vitamin D concentrations in Punjabi Sikhs: results of the Asian Indian diabetic heart study. J Steroid Biochem Mol Biol. (2016) 158:149–56. doi: 10.1016/j.jsbmb.2015.12.014 3. Molmen KS, Hammarstrom D, Pedersen K, Lian Lie v, Steile RB, Nygaard H, et al. Vitamin D3 supplementation does not enhance the eﬀects of resistance training in older adults. J Cachexia Sarcopenia Muscle. (2021) 12:599– 628. doi: 10.1002/jcsm.12688 3. Molmen KS, Hammarstrom D, Pedersen K, Lian Lie v, Steile RB, Nygaard H, et al. Vitamin D3 supplementation does not enhance the eﬀects of resistance training in older adults. J Cachexia Sarcopenia Muscle. (2021) 12:599– 628. doi: 10.1002/jcsm.12688 19. Pierce BL, Burgess S. Eﬃcient design for Mendelian randomization studies: subsample and 2-sample instrumental variable estimators. Am J Epidemiol. (2013) 178:1177–84. doi: 10.1093/aje/kwt084 19. Pierce BL, Burgess S. Eﬃcient design for Mendelian randomization studies: subsample and 2-sample instrumental variable estimators. Am J Epidemiol. (2013) 178:1177–84. doi: 10.1093/aje/kwt084 4. LeBoﬀMS, Murata EM, Cook NR, Cawthon P, Chou SH, Kotler G, et al. VITamin D and OmegA-3 TriaL (VITAL): eﬀects of vitamin D supplements on risk of falls in the US population. J Clin Endocrinol Metab. (2020) 105:2929– 38. doi: 10.1210/clinem/dgaa311 20. Greco MF, Minelli C, Sheehan NA, Thompson JR. Detecting pleiotropy in Mendelian randomisation studies with summary data and a continuous outcome. Stat Med. (2015) 34:2926–40. doi: 10.1002/sim.6522 21. 32. Rao P, Millen AE, Meyers KJ, Liu Z, Voland R, Sondel S, et al. The Relationship Between Serum 25-Hydroxyvitamin D Levels and Nuclear Cataract in the Carotenoid Age-Related Eye Study (CAREDS), an Ancillary Study of the Women’s Health Initiative. Invest Ophthalmol Vis Sci. (2015) 56:4221– 30. doi: 10.1167/iovs.15-16835 33. Dong J, Gharahkhani P, Chow WH, Gammon MD, Liu G, Caldas C, et al. No Association between Vitamin D status and risk of barrett’s esophagus or esophageal adenocarcinoma: a mendelian randomization study. Clin Gastroenterol Hepatol. (2019) 17:2227–35 e1. doi: 10.1016/j.cgh. 2019.01.041 34. Carlberg C, Velleuer E. Vitamin D and the risk for cancer: a molecular analysis. Biochem Pharmacol. (2021) 196:114735. doi: 10.1016/j.bcp. 2021.114735 37. E.-C.V.D.V.D.S.C. Emerging Risk Factors Collaboration. Estimating dose- response relationships for vitamin D with coronary heart disease, stroke, and all-cause mortality: observational and Mendelian randomisation analyses. Lancet Diabetes Endocrinol. (2021) 9:837–46. doi: 10.1016/S2213-8587(21)00263-1 References Vitamin D in graves disease: levels, correlation with laboratory and clinical parameters, and genetics. Eur Thyroid J. (2018) 7:27–33. doi: 10.1159/000484521 11. Meng X, Li X, Timofeeva MN, He Y, Spiliopoulou A, Wei WQ, et al. Phenome-wide Mendelian-randomization study of genetically determined vitamin D on multiple health outcomes using the UK Biobank study. Int J Epidemiol. (2019) 48:1425–34. doi: 10.1093/ije/dyz182 28. Antonelli A, Ferrari SM, Ragusa F, Elia G, Paparo SR, Ruﬃlli I, et al. Graves’ disease: Epidemiology, genetic and environmental risk factors and viruses. Best Pract Res Clin Endocrinol Metab. (2020) 34:101387. doi: 10.1016/j.beem.2020.101387 12. Ye Y, Yang H, Wang Y, Zhao H. A comprehensive genetic and epidemiological association analysis of vitamin D with common diseases/traits in the UK Biobank. Genet Epidemiol. (2021) 45:24–35. doi: 10.1002/gepi.22357 29. Ke W, Sun T, Zhang Y, He L, Wu Q, Liu J, et al. 25-Hydroxyvitamin D serum level in Hashimoto’s thyroiditis, but not Graves’ disease is relatively deﬁcient. Endocr J. (2017) 64:581–7. doi: 10.1507/endocrj.EJ16-0547 13. Manousaki D, Mitchell R, Dudding T, Haworth S, Harroud A, Forgetta V, et al. Genome-wide Association Study for Vitamin D Levels Reveals 69 Independent Loci. Am J Hum Genet. (2020) 106:327–37. doi: 10.1016/j.ajhg.2020.01.017 30. Abdellah MM, Mohamed Mostafa E, Salama EH, Roshdy Mohamed E. Association of Serum 25-Hydroxyl Vitamin D Deﬁciency and Age-Related Cataract: a case-control study. J Ophthalmol. (2019) 2019:9312929. doi: 10.1155/2019/9312929 14. Nagai A, Hirata M, Kamatani Y, Muto K, Matsuda K, Kiyohara Y, et al. Overview of the BioBank Japan Project: Study design and proﬁle. J Epidemiol. (2017) 27:S2–8. doi: 10.1016/j.je.2016.12.005 31. Park S, Choi NK. Serum 25-hydroxyvitamin D and age-related cataract. Ophthalmic Epidemiol. (2017) 24:281–6. doi: 10.1080/09286586.2017. 1281427 15. Yang J, Lee SH, Goddard ME, Visscher PM. GCTA: a tool for genome-wide complex trait analysis. Am J Hum Genet. (2011) 88:76-82. doi: 10.1016/j.ajhg.2010.11.011 32. Rao P, Millen AE, Meyers KJ, Liu Z, Voland R, Sondel S, et al. The Relationship Between Serum 25-Hydroxyvitamin D Levels and Nuclear Cataract in the Carotenoid Age-Related Eye Study (CAREDS), an Ancillary Study of the Women’s Health Initiative. Invest Ophthalmol Vis Sci. (2015) 56:4221– 30. doi: 10.1167/iovs.15-16835 16. Pierce BL, Ahsan H, Vanderweele TJ. Power and instrument strength requirements for Mendelian randomization studies using multiple genetic variants. Int J Epidemiol. (2011) 40:740–52. doi: 10.1093/ije/dyq151 Frontiers in Nutrition 08 frontiersin.org Liu et al. Liu et al. 10.3389/fnut.2022.938356 lens crystallins: new insights into the pathogenesis of cataract disorders. 35. Ghahramani M, YouseﬁR, Khoshaman K, Alavianmehr MM. The impact of calcium ion on structure and aggregation propensity of peroxynitrite-modiﬁed 36. Tuohimaa P, Tenkanen L, Ahonen M, Lumme S, Jellum E, Hallmans G, et al. Both high and low levels of blood vitamin D are associated with a higher prostate cancer risk: a longitudinal, nested case-control study in the Nordic countries. Int J Cancer. (2004) 108:104–8. doi: 10.1002/ijc.11375 References Colloids Surf B Biointerfaces. (2015) 125:170–80. doi: 10.1016/j.colsurfb.2014.11.002 36. Tuohimaa P, Tenkanen L, Ahonen M, Lumme S, Jellum E, Hallmans G, et al. Both high and low levels of blood vitamin D are associated with a higher prostate cancer risk: a longitudinal, nested case-control study in the Nordic countries. Int J Cancer. (2004) 108:104–8. doi: 10.1002/ijc.11375 Frontiers in Nutrition 09 Frontiers in Nutrition frontiersin.org
https://openalex.org/W2245497221	https://inferensi.iainsalatiga.ac.id/index.php/inferensi/article/download/209/168	Indonesian	null	MODEL PENGUKURAN KINERJA INTELLECTUAL CAPITAL DENGAN IB-VAIC DI PERBANKAN SYARIAH	Inferensi/Inferensi	2,013	cc-by-sa	5,620	MODEL PENGUKURAN KINERJA INTELLECTUAL CAPITAL DENGAN IB-VAIC DI PERBANKAN SYARIAH Ihyaul Ulum Program Doktor Ilmu Ekonomi Universitas Diponegoro Semarang E-mail: mas_ulum@yahoo.com Abstract The aim of this study is to develop a measurement of intellectual capital performance of Indonesian syariah banking by modifying the Pulic’s model that popular with the name of VAIC (value added intellectual coefficient). This study was done by documentation and focus group discussion with some financial accounting experts and public accountants. The result show that the main formula to measure IC performance of syariah banking is not too different with Pulic model, it was: iB-VAIC™ = iB-VACA + IB- VAHU + iB-STVA. The different one is the accounts to develop VA. VA in Pulic model was constructed by total revenue, while in iB-VAIC, VA was constructed from syariah activities. Keywords: iB-VAIC, Syariah Banking, Intellectual Capital Kata kunci: iB-VAIC, Perbankan Syariah, Intellectual Capital. Abstrak Tujuan penelitian ini adalah untuk mengembangkan suatu ukuran kinerja intellectual capital (IC) untuk perbankan syariah di Indonesia dengan memodifikasi model Pulic yang populer dengan sebutan VAIC (value added intellectual coefficient). Penelitian ini dilakukan melalui dokumentasi dan FGD (focus group discussion) bersama para pakar di bidang akuntansi keuangan dan akuntan publik. Hasil penelitian menunjukkan bahwa rumus utama untuk mengukur kinerja IC perbankan syariah tidak jauh berbeda dengan model Pulic, yaitu: iB- VAIC™ = iB-VACA + IB-VAHU + iB-STVA. Perbedaannya terletak pada akun-akun yang digunakan untuk mengembangkan rumus VA. VA dalam model Pulic dikonstruksi dari total pendapatan, sementara dalam iB-VAIC, VA dikonstruksi dari aktivitas-aktivitas syariah. Kata kunci: iB-VAIC, Perbankan Syariah, Intellectual Capital. Vol. 7, No. 1, Juni 2013 185 Ihyaul Ulum Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Pertama, industri perbankan merupakan salah satu dari 4 industri yang merupakan IC intencive industry sector (Firer & Williams, 2003;353). Selain itu, dari aspek intelektual, secara ke seluruhan karyawan di sektor perbankan lebih homogen dibanding kan dengan sektor ekonomi lainnya (Kubo dan Saka, 2002: 265). Kedua,hasil penelitian di berbagai negara termasuk di Indonesia, menunjukkan bahwa IC memiliki peran dalam meng gerakkan nilai perusahaan (firm’s value). IC berpengaruh positif terhadap kinerja keuangan perusahaan – yang merupakan ukuran jangka pendek dan yang paling mudah dilihat, baik pada masa kini maupun di masa yang akan datang. Artinya, IC dapat pula digunakan dalam memprediksi kinerja keuangan perusahaan (lihat misalnya: Chen et al., 2005, Tan et al., 2007 Ulum 2008a, Wang 2011, Kamal et al., 2011; Ahmad & Mushraf, 2011; Salman et al., 2012; Latif et al., 2012). Berdasarkan uraian tersebut di atas, maka penelitian ini dirancang untuk menjawab masalah tentang bagaimana model atau metode untuk menilai kinerja IC perbankan syariah di Indonesia. Teori tentang Intellectual Capital Ketertarikan akan IC bermula ketika Tom Stewart, pada Juni 1991, menulis sebuah artikel (”Brain Power - How Intellectual Capital Is Becoming America’s Most Valuable Asset”), yang mengantar IC kepada agenda manajemen. Tabel 1 meringkas kronologi beberapa kontribusi signifikan terhadap pengidentifikasian, pengukuran dan pelaporan IC. Pendahuluan Intellectual capital (IC) adalah kajian penelitian baru yang mendapatkan perhatian cukup besar dari para ahli di berbagai disiplin seiring dengan pertumbuhan ekonomi yang berbasis pada pengetahuan (knowledge-based economy) (Ståhle et al. 2011;541). Dari sisi akuntansi, sejumlah penelitian telah dilakukan di berbagai negara untuk mengkaji bagaimana metode untuk mengidentifikasi, mengukur, melaporkan dan menyajikannya dalam laporan perusahaan. Berbagai metode juga telah coba ditawarkan, salah satunya adalah VAIC™ (value added intellectual coefficient). VAIC™ dikonstruksi oleh Pulic (2000) untuk menilai kinerja IC pada perusahaan konvensional (private sector, profit motive, non syariah). Akun-akun yang digunakan dalam menghitung kinerja IC dengan VAIC™ adalah akun-akun yang lazim pada perusahaan konvensional. Sejauh ini, belum ada instrumen (sejenis VAIC™) yang dapat digunakan untuk menilai kinerja IC perbankan syariah. Sementara di Indonesia, perkembangan perbankan syariah cukup signifikan. Sepanjang tahun 2012 (per Oktober) perbankan syariah tumbuh dengan 37% sehingga total asetnya menjadi Rp 174,09 triliun. Pembiayaan telah mencapai 135,58 triliun (40,06%, yoy) dan penghimpunan dana menjadi Rp 134,45 triliun (32,06%). Dari sisi kelembagaan, jumlah bank yang melakukan kegiatan usaha syariah meningkat seiring dengan munculnya pemain-pemain baru. Sampai akhir 2012, terdapat 11 Bank Umum Syariah (BUS) dan 24 unit usaha syariah (UUS) dengan 508 Kantor Cabang. Jumlah kantor pelayanan (Kantor Kas, Kantor Kas Pembantu) sampai akhir 2012 mencapai 2.188. Model penilaian kinerja IC untuk perbankan syariah ini (yang akan diberi nama iB-VAIC – dibaca Islamic banking VAIC) penting untuk dihasilkan sebagai modifikasi dari model yang telah ada, yaitu Value Added Intellectual Coefficient – VAIC™. VAIC™ didesain untuk mengukur kinerja IC perusahaan-perusahaan dengan jenis transaksi yang umum. Sementara perbankan syariah memiliki jenis transaksinya sendiri yang relatif berbeda dari perbankan umum/konvensional. Model pengukuran kinerja IC untuk perbankan syariah (iB- VAIC) ini menjadi penting untuk dihasilkan setidaknya karena dua alasan: INFERENSI, Jurnal Penelitian Sosial Keagamaan 186 Pertengahan 1990-an Tabel: 1 Kronologi Kontribusi Signifikan terhadap Pengidentifikasian, Pengukuran dan Pelaporan IC Periode Perkembangan Awal 1980-an Muncul pemahaman umum tentang Intangible value (biasanya disebut “goodwill”) Pertengahan 1980-an Era informasi (information age) memegang peranan, dan selisih (gap) antara nilai buku dan nilai pasar semakin tampak jelas di beberapa perusahaan. Kronologi Kontribusi Signifikan terhadap Pengidentifikasian, Pengukuran dan Pelaporan IC Periode Perkembangan Awal 1980-an Muncul pemahaman umum tentang Intangible value (biasanya disebut “goodwill”) Pertengahan 1980-an Era informasi (information age) memegang peranan, dan selisih (gap) antara nilai buku dan nilai pasar semakin tampak jelas di beberapa perusahaan. Vol. 7, No. 1, Juni 2013: 185-206 187 Ihyaul Ulum Periode Perkembangan Akhir 1980-an Awal usaha para konsultan (praktisi) untuk membangun laporan/akun yang mengukur intellectual capital (Sveiby, 1988). Perkembangan g 0-an Awal usaha para konsultan (praktisi) untuk membangun laporan/akun yang mengukur intellectual capital (Sveiby, 1988). Kaplan dan Norton memperkenalkan konsep tentang balanced scorecard (1992). Nonaka dan Takeuchi (1995) mempresentasikan karya yang sangat berpengaruh terhadap “penciptaan pengetahuan perusahaan”. Meskipun buku ini berkonsentrasi pada ‘knowledge’, pembedaan antara pengetahuan dan intellectual capital dalam buku ini cukup menunjukkan bahwa mereka fokus pada intellectual capital. Pada tahun 1994, suplemen laporan tahunan Skandia dihasilkan. Suplemen ini fokus pada penyajian dan penilaian Persediaan perusahaan atas intellectual capital. Visualisasi IC menarik minat perusahaan lain untuk mengikuti petunjuk Skandia. Sensasi lainnya terjadi pada tahun 1995 ketika Celemi menggunakan knowledge audit untuk menawarkan suatu taksiran detail atas pernyataan intellectual capitalnya. Para pioner intellectual capital mempublikasikan buku- buku laris dengan topik IC (Kaplan dan Norton, 1996; Edvinsson and Malone, 1997; Sveiby, 1997). Karya Edvinsson and Malone lebih banyak mengupas tentang proses dan ‘bagaimana’ pengukuran IC. Akhir 1990-an Intellectual capital menjadi topik populer dengan konferensi para peneliti dan akademisi, working paper, dan publikasi lainnya menemukan audien. Akhir 1990-an Intellectual capital menjadi topik populer dengan konferensi para peneliti dan akademisi, working paper, dan publikasi lainnya menemukan audien. n Intellectual capital menjadi topik populer dengan konferensi para peneliti dan akademisi, working paper, dan publikasi lainnya menemukan audien. INFERENSI, Jurnal Penelitian Sosial Keagamaan 188 Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Periode Perkembangan Peningkatan jumlah proyek-proyek besar (misalnya the MERITUM project; Danish; Stockholm) yang diselenggarakan dengan tujuan, antara lain, untuk memperkenalkan beberapa penelitian tentang intellectual capital. Pada tahun 1999, OECD menyelenggarakan simposium internasional tentang intellectual capital di Amsterdam. Tabel: 1 Sumber: Petty and Guthrie (2000;166); Ulum (2009b;154) Perkembangan Peningkatan jumlah proyek-proyek besar (misalnya the MERITUM project; Danish; Stockholm) yang diselenggarakan dengan tujuan, antara lain, untuk memperkenalkan beberapa penelitian tentang intellectual capital. Pada tahun 1999, OECD menyelenggarakan simposium internasional tentang intellectual capital di Amsterdam. Sumber: Petty and Guthrie (2000;166); Ulum (2009b;154) Stewart mendefinisikan dalam artikelnya IC adalah “The sum of everything everybody in your company knows that gives you a competitive edge in the market place. It is intellectual material - knowledge, information, intellectual property, experience - that can be put to use to create wealth”. Brooking (1996: 34) misalnya mendefinisikan IC sebagai “the term given to the combined intangible assets of market, intellectual property, human-centred and infrastructure – which enable the company to function”. Sedangkan Roos et al. (1997;76) menyatakan bahwa: “IC includes all the processes and the assets which are not normally shown on the balance-sheet and all the intangible assets (trademarks, patent and brands) which modern accounting methods consider…”. Adapaun menurut Bontis (1998;69) mendefinisikan bahwa: “IC is elusive, but once it is discovered and exploited, it may provide an organisation with a new resource-base from which to compete and win” Salah satu definisi IC yang banyak digunakan adalah yang ditawarkan oleh Organisation for Economic Co-operation and Development (OECD, 1999) yang menjelaskan IC sebagai nilai ekonomi dari dua kategori aset tak berwujud: (1) organisational (structural) capital; dan (2) human capital. Vol. 7, No. 1, Juni 2013: 185-206 189 Ihyaul Ulum Tabel: 2 Perbandingan Konsep IC Menurut Beberapa Peneliti Tabel: 2 Perbandingan Konsep IC Menurut Beberapa Peneliti Brooking (UK) Roos (UK) Stewart (USA) Bontis (Kanada) Human-centered Skills, abilities and expertise, problem solving abilities and leadership styles Human capital Competence, attitude, and intellectual agility Human capital Employees are an organization’s most important asset Human capital The individual level knowledge that each employee possesses Infrastructure assets All the technologies, process and methodologies that enable company to function Orgnistional capital All organizational, innovation, processes, intellectual property, and cultural assets Structural capital Knowledge embedded in information technology Structural capital Non-human assets or organizational capabilities used to meet market requirements Intellectual property Know-how, trademarks and patents Renewal and development capital New patents and training efforts Structural capital All patents, plans and trademarks Intellectual property Unlike, IC, IP is a protected asset and has a legal definition Market assets Brands, customers, customer loyalty and distribution channels Relational capital Relationship which include internal and external stakeholders Customer capital Market information used to capture and retain customers Relational capital Customer capital is only one feature of the knowledge embedded in organizational relationships Sumber : Bontis et al. (2000;92) Orgnistional capital All organizational, innovation, processes, intellectual property, and cultural assets Knowledge embedded in information technology Sumber : Bontis et al. (2000;92) Sumber : Bontis et al. (2000;92) The International Federation of Accountan – IFAC (1998) mengklasifikasikan IC menjadi tiga kategori, yaitu Organizational Capital, Relational Capital, dan Human Capital. Organizational Capital meliputi intellectual property dan infrastructure assets. Table 3 menyajikan secara lebih detil tentang klasifikasi tersebut. 190 INFERENSI, Jurnal Penelitian Sosial Keagamaan Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Table: 3 Klasifikasi Intellectual Capital menurut IFAC Organizational Capital Relational Capital Human Capital Intellectual Property: • Patents • Copyrights • Design rights • Trade secret • Trademarks • Service marks Infrastructure Assets: • Management philosophy • Corporate culture • Management processes • Information systems • Networking systems • Financial relations • Brands • Customers • Customer loyalty • Backlog orders • Company names • Distribution channels • Business collaborations • Licensing agreements • Favourable contracts • Franchising agreements • Know-how • Education • Vocational qualification • Work-related knowledge • Work-related competencies • Entrepreneurial spirit, innovativeness, proactive and reactive abilities, changeability • Psychometric valuation Sumber: IFAC, 1998 Table: 3 Klasifikasi Intellectual Capital menurut IFAC Sumber: IFAC, 1998 Bontis et al. (2000: 87) menyatakan bahwa secara umum, para peneliti mengidentifikasi tiga konstruk utama dari IC, yaitu: human capital (HC), structural capital (SC), dan customer capital (CC). Menurut Bontis et al. (2000: 90), secara sederhana HC merepresentasikan individual knowledge stock suatu organisasi yang direpresentasikan oleh karyawannya. HC merupakan kombinasi dari genetic inheritance; education; experience, and attitude tentang kehidupan dan bisnis. Lebih lanjut Bontis et al. (2000: 90) menyebutkan bahwa SC meliputi seluruh non-human storehouses of knowledge dalam organisasi. Termasuk dalam hal ini adalah database, organisational charts, process manuals, strategies, routines dan segala hal yang membuat nilai perusahaan lebih besar daripada nilai materialnya. Sedangkan tema utama dari CC adalah pengetahuan yang melekat dalam marketing channels dan customer relationship dimana suatu organisasi mengembangkannya melalui jalannya bisnis (Bontis et al., 2000: 91). Vol. 7, No. 1, Juni 2013: 185-206 191 Ihyaul Ulum Ihyaul Ulum Value Added Intellectual Coefficient (VAIC™) Hal terpenting dalam manajemen di abad ke-20 adalah peningkatan hingga 50 kali lipat produktivitas pekerja manual dalam memproduksi. Kontribusi penting manajemen yang harus dibuat di abad ke-21 adalah dengan cara yang sama meningkatkan produktivitas pekerjaan pengetahuan (knowledge work) dan pekerja berpengetahuan (knowledge workers). Aset yang paling berharga bagi perusahaan di abad ke-20 adalah peralatan produksinya. Menurut Peter Drucker, aset yang paling berharga institusi di abad ke-21 adalah pekerja berpengetahuan (knowledge workers) dan produkvitasnya (Pulic, 2000). Metode value added intellectual coefficient (VAIC™) dikembangkan oleh Pulic pada tahun 1997 yang didesain untuk menyajikan informasi tentang value creation efficiency dari aset berwujud (tangible asset) dan aset tidak berwujud (intangible assets) yang dimiliki perusahaan. VAIC™ merupakan instrumen untuk mengukur kinerja intellectual capital perusahaan. Pendekatan ini relatif mudah dan sangat mungkin untuk dilakukan, karena dikonstruksi dari akun-akun dalam laporan keuangan perusahaan (neraca, laba rugi) (Ulum 2009b: 111). Metode ini dimulai dengan kemampuan perusahaan untuk menciptakan value added (VA). Value added adalah indikator paling objektif untuk menilai keberhasilan bisnis dan menunjukkan kemampuan perusahaan dalam penciptaan nilai (value creation). VA dihitung sebagai selisih antara output dan input. Output (OUT) merepresentasikan revenue dan mencakup seluruh produk dan jasa yang dijual di pasar, sedangkan input (IN) mencakup seluruh beban yang digunakan dalam memperoleh revenue. Hal penting dalam metode ini adalah bahwa beban karyawan (labour expenses) tidak termasuk dalam IN. Karena peran aktifnya dalam proses value creation, intellectual potential (yang direpresentasikan dengan labour expenses) tidak dihitung sebagai biaya (cost) dan tidak masuk dalam komponen IN. Karena itu, aspek kunci dalam metode Pulic adalah memperlakukan tenaga kerja sebagai entitas penciptaan nilai (value creating entity). INFERENSI, Jurnal Penelitian Sosial Keagamaan 192 Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC VA dipengaruhi oleh efisiensi dari Human Capital (HC) dan Structural Capital (SC). Hubungan lainnya dari VA adalah capital employed (CE), yang dalam hal ini dilabeli dengan VACA. VACA adalah indikator untuk VA yang diciptakan oleh satu unit dari physical capital. Pulic (2000) mengasumsikan bahwa jika 1 unit dari CE meng hasilkan return yang lebih besar daripada perusahaan yang lain, maka berarti perusahaan tersebut lebih baik dalam memanfaatkan CE-nya. Dengan demikian, pemanfaatan CE yang lebih baik meru pakan bagian dari IC perusahaan. Hubungan selanjutnya adalah VA dan HC. ‘Value Added Human Capital’ (VAHU) menunjukkan berapa banyak VA dapat dihasilkan dengan dana yang dikeluarkan untuk tenaga kerja. Hubungan antara VA dan HC mengindikasikan kemampuan dari HC untuk menciptakan nilai di dalam perusahaan. Value Added Intellectual Coefficient (VAIC™) Konsisten dengan pandangan para penulis IC lainnya, Pulic berargumen bahwa total salary and wage costs adalah indikator dari HC perusahaan. Hubungan ketiga adalah “structural capital coefficient” (STVA), yang menunjukkan kontribusi structural capital (SC) dalam penciptaan nilai. STVA mengukur jumlah SC yang dibutuhkan untuk menghasilkan 1 rupiah dari VA dan merupakan indikasi bagaimana keberhasilan SC dalam penciptaan nilai. SC bukanlah ukuran yang independen sebagaimana HC, ia dependen terhadap value creation. Artinya, semakin besar kontribusi HC dalam value creation, maka akan semakin kecil kontribusi SC dalam hal tersebut. Lebih lanjut Pulic menyatakan bahwa SC adalah VA dikurangi HC, yang hal ini telah diverifikasi melalui penelitian empiris pada sektor industri tradisional (Pulic, 2000). Rasio terakhir adalah menghitung kemampuan intelektual perusahaan dengan menjumlahkan koefisien-koefisien yang telah dihitung sebelumnya. Hasil penjumlahan tersebut diformulasikan dalam indikator baru yang unik, yaitu VAIC™ (Tan et al., 2007: 83). Keunggulan metode VAIC™ adalah karena data yang dibutuhkan relatif mudah diperoleh dari berbagai sumber dan jenis perusahaan. Data yang dibutuhkan untuk menghitung berbagai rasio tersebut adalah angka-angka keuangan yang standar yang umumnya Vol. 7, No. 1, Juni 2013: 185-206 193 Ihyaul Ulum tersedia dari laporan keuangan perusahaan. Alternatif pengukuran IC lainnya terbatas hanya menghasilkan indikator keuangan dan non-keuangan yang unik yang hanya untuk melengkapi profil suatu perusahaan secara individu. Indikator-indikator tersebut, khususnya indikator non-keuangan, tidak tersedia atau tidak tercatat oleh perusahaan yang lain (Tan et al., 2007;85). Konsekuensinya, kemampuan untuk menerapkan pengukuran IC alternatif tersebut secara konsisten terhadap sample yang besar dan terdiversifikasi menjadi terbatas (Firer dan Williams, 2003;359). Sejumlah penelitian di berbagai negara telah menggunakan VAIC sebagai proksi atas IC. Mavridis (2004) dan Kamath (2007) misalnya, menggunakan VAIC untuk merangking kinerja IC perbangkan di Jepang dan India. Hal yang sama kemudian dilakukan oleh Ulum (2009a) untuk konteks perbankan yang terdaftar di Bursa Efek Indonesia. Hasilnya menemukan bahwa secara umum, kinerja IC perusahaan perbankan di Indonesia tahun 2004 masuk dalam kategori “top performers” berdasarkan klasifikasi yang dibuat oleh Mavridis (2004) dan Kamath (2007). Lebih lanjut, VAIC juga banyak digunakan untuk meneliti pengaruh IC terhadap kinerj perusahaan. Firer dan Williams (2003) menguji hubungan VAIC™ dengan kinerja perusahaan di Afrika Selatan, sementara Chen et al. (2005) menggunakan VAIC™ untuk menguji hubungan antara IC dengan nilai pasar dan kinerja keuangan perusahaan dengan menggunakan sampel perusahaan publik di Taiwan. Tan et al. (2007) menggunakan 150 perusahaan yang terdaftar di bursa efek Singapore sebagai sampel penelitian untuk melihat pengaruh IC terhadap kinerja keuangan perusahaan. INFERENSI, Jurnal Penelitian Sosial Keagamaan INFERENSI, Jurnal Penelitian Sosial Keagamaan Value Added Intellectual Coefficient (VAIC™) Di Indonesia, kajian yang sama juga mulai banyak dilakukan. Ulum (2007, 2008a) misalnya menganalisis hubungan antara IC (diproksikan dengan VAIC™) dan kinerja perusahaan perbankan di Indonesia, baik terdaftar maupun tidak terdaftar di BEI. Berdasarkan hasil pengujian dengan PLS diketahui bahwa secara statistik (baik nilai t-statistics seluruh path antara VAIC dan PERF maupun nilai R-square) terbukti terdapat pengaruh VAIC terhadap kinerja keuangan perusahaan, baik kinerja masa kini maupun masa yang akan datang. Artinya, IC dapat digunakan untuk memprediksi kinerja perusahaan di masa yang akan datang. INFERENSI, Jurnal Penelitian Sosial Keagamaan 194 Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Ulum (2008b) dalam penelitian berjudul Intellectual Capital dan Return Finansial Perusahaan Publik Sektor Perbankan di Indo nesia berusaha menguji kembali daya pengaruh VAIC™ terhadap kinerja keuangan dengan mengkhususkan pada perusahaan perbankan yang go publik (terdaftar di Bursa Efek Indonesia). Hasilnya konsisten dengan kajian sebelumnya bahwa VAIC™ berpengaruh terhadap kinerja keuangan. Dalam perkembangan selanjutnya, Ulum (2012) berusaha menginvestigasi hubungan antara kinerja modal intelektual (VAIC) dan praktik pengungkapannya (IC disclosure) dalam laporan tahu nan perusahaan. Hasilnya menemukan indikasi bahwa jika suatu perusahaan memiliki kinerja IC yang baik (VAIC-nya tinggi), ad a kecenderungan untuk meminimalisir jumlah (kuantitas) informasi tentang IC yang diungkapkan dalam laporan tahunan perusahaan. Manajemen mungkin berasumsi bahwa tingginya kinerja IC dapat menjadi sinyal bagi kompetitor tentang kekuatan perusahaan dalam memenangi kompetisi di pasar. Hasil penelitian ini konsisten dengan temuan Williams (2001) yang menggunakan 30 perusahaan publik di Inggris yang masuk dalam kelompok FTSE 100 dalam kurun waktu 1996-2000. Metode Penelitian Jenis penelitian yang digunakan dalam penelitian ini adalah pene litian eksploratif yang ingin menemukan ”teori baru” atau mengem bangan teori melalui formulasi, pengujian, dan pengembangan ulang preposisi selama penyusunan teori yang bersifat grounded. Penelitian ini berusaha untuk mengkonstruksi suatu metode ukuran kinerja IC untuk perbankan syariah dengan melakukan kajian men dalam terhadap metode Penilaian yang telah ada (VAIC™) dan di sesuaikan dengan akun-akun dalam laporan keuangan perbankan syariah. Unit analisis dalam penelitian ini adalah (1) metode Penilaian kinerja IC yang dikonstruksi oleh Ante Pulic (VAIC™). Metode ini akan dianalisis untuk dilakukan modifikasi; (2) standar akuntansi keuangan syariah yang akan digunakan untuk memodifikasi metode VAIC™; dan (3) literatur-literatur akuntansi keuangan yang relevan. Vol. 7, No. 1, Juni 2013: 185-206 195 Ihyaul Ulum Ihyaul Ulum Data yang digunakan adalah data sekunder yang terdiri dari: Data yang digunakan adalah data sekunder yang terdiri dari 1. Data jenis-jenis akun yang digunakan untuk mengkonstruksi metode VAIC™. Data ini diperoleh dari kajian literatur dan internet. 2. Data jenis-jenis akun dan format pelaporan keuangan perbankan syariah. Data ini dapat diperoleh dari standar akuntansi keuangan (SAK) syariah yang diterbitkan DSAK Syariah Ikatan Akuntan Indonesia (IAI) dan Bank Indonesia. Tahapan analisis data yang dilakukan adalah sebagai berikut: 1. mengidentifikasi akun-akun dalam perhitungan VAIC™-nya Pulic yang sejauh ini digunakan untuk perusahaan konvensional. Hal ini dilakukan oleh tim peneliti melalui sebuah diskusi dengan tim dari Kantor Akuntan Publik yang memiliki keahlian tentang seluk-beluk akun; 2. mengidentifikasi akun-akun yang disajikan dalam laporan keuangan perbankan syariah melalui kajian terhadap SAK Syariah dan peraturan terkait, serta laporan keuangan bank/ unit usaha syariah. Hal ini dilakukan oleh tim peneliti melalui sebuah diskusi dengan tim dari Kantor Akuntan Publik yang memiliki keahlian dalam hal ‘mengutak-atik’ akun; 3. Melakukan modifikasi dan penyesuaian akun-akun dalam perhitungan VAIC™ dengan akun-akun dalam laporan keuangan perbankan syariah. Hal ini dilakukan melalui diskusi mendalam dengan para ahli di bidangnya (akademisi dan praktisi perbankan syariah) dalam bentuk Focus Group Discussion (FGD); dan 4. Menyusun metode Penilaian kinerja IC untuk perbankan syariah dengan pendekatan iB-value added intellectual coefficient (iB- VAIC). Gambaran Umum Perbankan Syariah Perkembangan perbankan syariah selama satu tahun terakhir, sampai dengan bulan Oktober 2012 (yoy) cukup menggembirakan. Perbankan syariah mampu tumbuh ± 37% sehingga total asetnya menjadi Rp174,09 triliun. Pembiayaan telah mencapai Rp135,58 INFERENSI, Jurnal Penelitian Sosial Keagamaan 196 Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC triliun (40,06%, yoy) dan penghimpunan dana menjadi Rp134,45 triliun (32,06%). Pertumbuhan penghimpunan dana cukup baik diimbangi dengan pertumbuhan penyaluran dana kepada sektor riil baik berupa pembiayaan (Mudharabah dan Musyarakah), piutang (Murabahah, Istisna, dan Qardh), dan dalam bentuk pembiayaan Ijarah. Dengan demikian fungsi intermediasi perbankan dapat relatif terjaga yang tercermin dari FDR agregat perbankan syariah tercatat cukup tinggi yaitu sebesar 100,84% meningkat lebih tinggi dari tahun sebelumnya yang tercatat sebesar 95,08%. Selain fungsi intermediasi, untuk memberikan pelayanan dengan jangkauan yang lebih luas bagi masyarakat, akses jaringan perkantoran meningkat menjadi 2.188 (29,31%) dari 1.692 kantor pada tahun sebelumnya. Perluasan jaringan kantor tersebut telah mampu meningkatkan pengguna bank syariah yang tercermin dari peningkatan jumlah total rekening (pembiayaan + DPK) yaitu sebesar 3,4 juta rekening dari 9 juta rekening menjadi 12,4 juta rekening (Oktober 2012, yoy). Selama periode tahun 2012, jumlah Bank Umum Syariah (BUS) dan Unit Usaha Syariah (UUS) sampai dengan Oktober 2012 tidak mengalami perubahan, namun demikian jumlah jaringan kantor meningkat. Meskipun dengan jumlah BUS (11 buah) maupun UUS (24 buah) yang sama, namun pelayanan kebutuhan masyarakat akan perbankan syariah menjadi semakin meluas yang tercermin dari bertambahnya Kantor Cabang dari sebelumnya sebanyak 452 menjadi 508 Kantor, sementara Kantor Cabang Pembantu (KCP) dan Kantor Kas (KK) telah bertambah sebanyak 440 kantor pada periode yang sama (Oktober 2012, yoy). Secara keseluruhan jumlah kantor perbankan syariah yang beroperasi sampai dengan bulan Oktober 2012 dibandingkan tahun sebelumnya meningkat dari 1.692 kantor menjadi 2.188 kantor. BPRS sebagai bagian dari lembaga perbankan syariah juga mengalami perkembangan yang cukup menggembirakan. Aset BPRS selama kurun waktu satu tahun terakhir meningkat sebesar 33,09% menjadi sebesar Rp4,46 triliun (yoy), dengan share pembiayaan merupakan 77,68% dari total aktiva. Penghimpunan dana BPRS juga meningkat tinggi yaitu sebesar 41,47% menjadi Rp2,77 triliun. Vol. 7, No. 1, Juni 2013: 185-206 197 Ihyaul Ulum BPRS telah menjalankan fungsi intermediasi perbankan dengan baik, tercermin dari rasio FDR agregat BPRS yang mencapai 124,80%. Pertumbuhan penyaluran dana tersebut cukup terkendali dengan kualitas pembiayaan yang baik dengan penurunan rasio NPF (net) dari 5,90% menjadi 5,60%. Rasio permodalan BPRS cukup memadai yang tercermin dari agregat rasio CAR yang tinggi mencapai 25%. Gambaran Umum Perbankan Syariah Keunggulan karakteristik BPRS yang beroperasi di daerah- daerah terpencil bahkan sampai pada daerah remote area sehingga dapat memberikan pelayanan dengan jangkauan yang lebih luas kepada masyarakat. Luasnya demografi BPRS ternyata berperan cukup signifikan dalam perolehan laba untuk menjaga tingkat rentabilitas. Rasio ROE meningkat dari 16,10% menjadi 22,30%, ROA meningkat dari 2,40% menjadi 2,80%, meskipun rasio BOPO lebih tinggi dari rata-rata BUS dan UUS, namun dapat dijaga dalam kisaran 86,20%. Menghitung iB-Value Added (VA) Menggunakan data laporan keuangan, standar pelaporan, dan regulasi terkait tentang perbankan syariah, kami mengidentikasi akun-akun dalam laporan keuangan bank syariah untuk menyusun model iB-VAIC. Berdasarkan hasil focus grup discussion (FGD) yang telah dilakukan, maka rumus yang digunakan untuk menghitung iB-VAIC adalah sebagai berikut: Tahap pertama dengan menghitung iB-Value Added (iB-VA). IB-VA dihitung dengan menggunakan cara yaitu sebagai berikut : iB-VA = OUT - IN OUT (Output) : Total pendapatan, diperoleh dari: 1. Pendapatan bersih kegiatan syariah = pendapatan operasi utama kegiatan syariah + pendapatan operasi lainnya - hak pihak ketiga atas bagi hasil dan syirkah temporer. Pendapatan operasi utama kegiatan syariah terdiri: a. Pendapatan penyaluran dana a. Pendapatan penyaluran dana 1) Dari pihak ketiga bukan bank a) Pendapatan dari jual beli (pendapatan marjin mura bahah) INFERENSI, Jurnal Penelitian Sosial Keagamaan 198 Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC b) Pendapatan bersih salam parallel c) Pendapatan bersih istishna parallel d) Pendapatan sewa ijarah e) Pendapatan pendapatan bagi hasil musyarakah f) Pendapatan bagi hasil mudharabah g) Pendapatan dari penyertaan 2) Dari Bank Indonesia a. Bonus SBIS 3) Dari bank-bank lain di Indonesia a. Bonus dari bank syariah lain b. Pendapatan bagi hasil mudharabah c. Tabungan mudharabah d. Deposito mudharabah e. Sertifikat investasi mudharabah antar bank b. Pendapatan operasi lainnya b. Pendapatan operasi lainnya b. Pendapatan operasi lainnya 1) Jasa investasi terikat (mudharabah muqayyadah) 1) Jasa investasi terikat (mudharabah muqayyadah) 2) Jasa layanan 3) Pendapatan dari transaksi valuta asing 4) Koreksi PPAP 5) Koreksi penyisihan penghapusan transaksi rek. Adminis trasi 6) Lainnya c. Hak pihak ketiga atas bagi hasil syirkah temporer 1) Pihak ketiga bukan bank a. Tabungan mudharabah b. Deposito mudharabah c. Lainnya 2) Bank Indonesia Vol. 7, No. 1, Juni 2013: 185-206 199 Ihyaul Ulum a) FPJP syariah a) FPJP syariah b) Lainnya 3) Bank-bank lain di Indonesia dan di luar Indonesia a) Tabungan mudharabah b) Deposito mudharabah c) Sertifikat investasi mudharabah antar bank d) Lainnya b) Lainnya 3) Bank-bank lain di Indonesia dan di luar Indonesia b) Deposito mudharabah 2. Pendapatan non operasional INFERENSI, Jurnal Penelitian Sosial Keagamaan 2. Pendapatan non operasional 2. Pendapatan non operasional IN (input): Beban usaha/operasional dan beban non operasional kecuali beban kepegawaian/karyawan Beban usaha/operasional kecuali beban kepegawaian: a. Beban penyisihan kerugian asset produktif-bersih b. Beban estimasi kerugian komitmen dan kontijensi c. Beban operasi lainnya d. Beban bonus titipan wadiah e. Beban administrasi dan umum f. Beban penurunan nilai surat nerharga g. Beban transaksi valuta asing h. Beban promosi i. Beban lainnya IN (input): Beban usaha/operasional dan beban non operasiona kecuali beban kepegawaian/karyawan Beban usaha/operasional kecuali beban kepegawaian: a. Beban penyisihan kerugian asset produktif-bersih b. Beban estimasi kerugian komitmen dan kontijensi e. Beban administrasi dan umum f. Beban penurunan nilai surat nerharga g. Beban transaksi valuta asing h. Beban promosi iB-VAHU = iB-VAHU = Keterangan : iB-VAHU : Value added Human Capital : rasio dari iB-VA terhadap HC iB-VA : Value added HC : Human capital : beban karyawan i. Beban lainnya i. Beban lainnya Value added (iB-VA) juga dapat dihitung dari akun-akun perusahaan sebagai berikut: iB-VA= OP + EC + D + A iB-VA= OP + EC + D + A Keterangan : OP : operating profit (laba operasi/laba usaha) EC : employee costs (beban karyawan) D : depreciation (depresiasi) A : amortization (amortisasi) Keterangan : OP : operating profit (laba operasi/laba usaha) EC : employee costs (beban karyawan) D : depreciation (depresiasi) A : amortization (amortisasi) OP : operating profit (laba operasi/laba usaha) EC : employee costs (beban karyawan) D : depreciation (depresiasi) INFERENSI, Jurnal Penelitian Sosial Keagamaan 200 Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Menghitung iB-Value Added Capital Employed (iB-VACA) Tahap kedua dengan menghitung Value Added Capital Employed (iB-VACA). iB-VACA adalah indikator untuk iB-VA yang diciptakan oleh satu unit dari human capital. Rasio ini menunjukkan kontibusi yang dibuat oleh setiap unit dari CE terhadap value added perusahaan. iB-VACA = iB-VACA = Keterangan: iB-VACA : Value Added Capital Employed : rasio dari iB-VA ter hadap CE iB-VA : value added CE : Capital Employed : dana yang tersedia (total ekuitas) iB-VAHU menunjukkan berapa banyak iB-VA dapat dihasil kan dengan dana yang dikeluarkan untuk tenaga kerja. Rasio ini menunjukkan kontribusi yang dibuat oleh setiap rupiah yang di investasikan dalam HC terhadap value added organisasi. iB-VAHU menunjukkan berapa banyak iB-VA dapat dihasil kan dengan dana yang dikeluarkan untuk tenaga kerja. Rasio ini menunjukkan kontribusi yang dibuat oleh setiap rupiah yang di investasikan dalam HC terhadap value added organisasi. Menghitung Structural Capital Value Added (iB-STVA) Rasio ini mengukur jumlah SC yang dibutuhkan untuk menghasilkan satu rupiah dari iB-VA dan merupakan indikasi bagaimana keberhasilan SC dalam penciptaan nilai. Vol. 7, No. 1, Juni 2013: 185-206 201 Ihyaul Ulum iB-STVA = Keterangan : STVA : Structural Capital Value Added : rasio dari SC terhadap IB-VA SC : Structural capital : IB-VA – HC IB-VA : Value Added Menghitung Value Added Intellectual Coefficient (iB-VAIC™) IB-VAIC™ mengindikasikan kemampuan intelektual orga nisasi yang dapat juga dianggap sebagai BPI (Business Performance Indikator). iB-VAIC™ merupakan penjumlahan dari tiga komponen sebelumnya, yaitu iB-VACA, iB-VAHU, dan iB-STVA. iB-VAIC™ = iB-VACA + IB-VAHU + iB-STVA INFERENSI, Jurnal Penelitian Sosial Keagamaan iB-VAIC™ = iB-VACA + IB-VAHU + iB-STVA Dengan menggunakan formula ini, kinerja IC perbankan syariah dapat diukur. Hasil pengukuran tersebut dapat menjadi indikasi bagi pengambil keputusan tentang bagaimana perusahaan mengelola IC yang dimiliki untuk memaksimalkan value bagi perusahaan. iB-VAIC™ = iB-VACA + IB-VAHU + iB-STVA iB-VAIC yang dirumuskan dalam penelitian ini dapat diguna kan untuk mengukur kinerja IC perbankan syariah di Indonesia. Perhitungan yang berbasis pada akun-akun dalam laporan keungan tradisional ini akan dengan mudah dapat dilakukan dan dapat memberikan gambaran tentang kinerja IC yang dimiliki oleh perbankan syariah. Untuk dapat dilakukan pemeringkatan terhadap sejumlah perbankan, hasil perhitungan iB-VAIC (untuk selanjutnya dapat disebut BPI) dapat diranking berdasarkan skor yang dimiliki. Sejauh ini, belum ada standar tentang skor kinerja IC tersebut, namun penelitian Ulum (2008) telah merumuskan untuk memberikan kategori dari hasil perhitungan VAIC, yaitu: 1. Top performers – skor VAICTM di atas 3,00 2. Good performers – skor VAICTM antara 2,0 sampai 2,99 3. Common performers – skor VAICTM antara 1,5 sampai 1,99 4. Bad performers – skor VAICTM di bawah 1,5 INFERENSI, Jurnal Penelitian Sosial Keagamaan 202 Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Kesimpulan Formula perhitungan iB-VAIC pada dasarnya tidak banyak berbeda dengan formula VAICTM yang dirumuskan oleh Pulic (1998). Perbedaan mendasar diantara keduanya terletak pada akun-akun untuk menghitung VA. Dalam iB-VAIC, VA dikonstruksi dari akun-akun pendapatan yang semuanya adalah berbasis syariah, yaitu pendapatan bersih kegiatan syariah dan pendapatan non-operasional yang syar’iy. Formula akhir dalam menghitung iB-VAIC adalah sebagai berikut: Daftar Pustaka Ahmad, S. & A.M. Mushraf. 2011. “The Relationship between Intellectual Capital and Business Performance: An Empirical Study in Iraqi Industry”. Conference Proceeding 2011 International Conference on Management and Artificial Intelligence. Bali, Indonesia. Bontis, N. 1998. “Intellectual Capital: an Exploratory Study that Develops Measures and Models”. Management Decision. Vol. 36 No. 2. p. 63. Bontis, N., W.C.C. Keow, S. Richardson. 2000. “Intellectual Capital and Business Performance in Malaysian Industries”. Journal of Intellectual Capital. Vol. 1 No. 1. pp. 85-100. Brooking, A. 1996. Intellectual Capital: Core Asset for Third Millennium Enterprise. International Thomson Business Press, London. Chen, M.C., S.J. Cheng, Y. Hwang. 2005. “An Empirical Investigation of the Relationship between Intellectual Capital and Firms’ Market Value and Financial performance”. Journal of Vol. 7, No. 1, Juni 2013: 185-206 203 Ihyaul Ulum Intellectual Capital. Vol. 6 N0. 2. pp. 159-176. Intellectual Capital. Vol. 6 N0. 2. pp. 159-176. Firer, S., and S.M. Williams. 2003. “Intellectual Capital and Traditional Measures of Corporate Performance”. Journal of Intellectual Capital. Vol. 4 No. 3. pp. 348-360. IFAC. 1998. “The Measurement and Management of Intellectual Capital”. Available from www.ifac.org (cited November 2012) Kamal, M.H.M., R.C. Mat, N.A. Rahim, N. Husin & I. Ismail. 2011. “Intellectual Capital and Firm Performance of Commercial Banks in Malaysia”. Asian Economic and Finance Review. Vol 2/4 pp. 577-590 Kamath, G.B. 2007. “The Intellectual Capital Performance of Indian Banking Sector”. Journal of Intellectual Capital. Vol. 8 No. 1. pp. 96-123. Kubo, I., and A. Saka. 2002. “An inquairy into the motivations of knowledge workers in the Japanese financial industry”. Journal of Knowledge Management. Vol. 6 No. 3. pp. 262-271. Latif, M., M.S. Malik & S. Aslam. 2012. “Intellectual Capital Efficiency And Corporate Performance In Developing Countries: A Comparison Between Islamic And Conventional Banks Of Pakistan” Interdisciplinary Journal of Contemporary Research in Business. Vol 4/11 pp. 405-420. Mavridis, D.G. 2004. “The intellectual capital performance of the Japanese banking sector”. Journal of Intellectual Capital. Vol. 5 No. 3. pp. 92-115. Petty, R. and J. Guthrie. .2000. “Intellectual capital literature review: measurement, reporting and management”. Journal of Intellectual Capital, Vol. 1 No. 2/3, pp. 155-76. Pulic, A. 2000. “Basic Information on VAIC™”. available online at: www.vaic-on.net. (accessed November 2006). Roos, J., G. Roos, N.C. Dragonetti, and L. Edvinsson. 1997. Intellectual Capital: Navigating in the New Business Landscape. Macmillan Business, Houndsmills. Salman, R.T., M. Mansor & A.D. Babatunde. 2012. INFERENSI, Jurnal Penelitian Sosial Keagamaan INFERENSI, Jurnal Penelitian Sosial Keagamaan Daftar Pustaka “Impact of Intellectual Capital on Return on Asset in Nigerian INFERENSI, Jurnal Penelitian Sosial Keagamaan 204 Model Pengukuran Kinerja Intellectual Capital dengan iB-VAIC Manufacturing Companies”. Interdisciplinary Journal of Research in Business. Vol 2/4 pp 21-30 Manufacturing Companies”. Interdisciplinary Journal of Research in Business. Vol 2/4 pp 21-30 Ståhle, P., S. Ståhle & S. Aho. 2011. “Value Added Intellectual Coefficient (VAIC): a Critical Analysis”. Journal of Intellectual Capital. Vol 12/4 pp. 531-551. Stewart, T.A. (1997) Intellectual Capital: The New Wealth of Organizations, Doubleday/Currency, New York: United States of America. Tan, H.P., D. Plowman, P. Hancock. 2007. “Intellectual Capital and Financial Returns of Companies. Journal of Intellectual Capital. Vol. 8 No. 1. pp. 76-95. Ulum, I. 2007. “Pengaruh Intellectual Capital terhadap Kinerja Keuangan Perusahaan Publik Sektor Perbankan”. Thesis magister akuntansi Undip, unpublished. _______. 2008a. “Intellectual Capital dan Kinerja Keuangan Perusahaan; Sebuah Analisis dengan Pendekatan Partial Least Squares”. Call for paper Simposium Nasional Akuntansi XI. Ikatan Akuntan Indonesia. Pontianak. ___. 2008b. Intellectual Capital and Financial Return of Listed Indonesian Banking Sector. Proceeding international research seminar and exhibition. Lemlit UMM. Malang. ___. 2009a. “Intellectual Capital Performance Sektor Perbankan di Indonesia”. Jurnal Akuntansi dan Keuangan (terakreditasi dikti). Vol 10/2. Februari 2009. ISSN: 1411-0288. ___. 2009b. Intellectual Capital; Konsep dan Kajian Empiris. PT. Graha Ilmu, Yogyakarta. ______. 2009c. Investigasi Hubungan antara Intellectual Capital dan Nilai Pasar Perusahaan serta Kinerja Keuangan. Program Penelitian Unggulan FE UMM, Malang. ____. 2012. “Investigasi Hubungan antara Modal Intelektual dan Praktik Pengungkapannya (IC Disclosure) dalam Laporan Tahunan Perusahaan”. Jurnal Ekonomi Bisnis. Tahun 17 No. 1, pp 36-45. ______. 2012. “Investigasi Hubungan antara Modal Intelektual dan Praktik Pengungkapannya (IC Disclosure) dalam Laporan Tahunan Perusahaan”. Jurnal Ekonomi Bisnis. Tahun 17 No. 1, pp 36-45. Vol. 7, No. 1, Juni 2013: 185-206 205 Ihyaul Ulum Wang, M.S. 2011. “Intellectual Capital and Firm Performance”. Proceeding Annual Conference on Innovations in Business & Management. London, The Center for Innovations in Business and Management Practice. INFERENSI, Jurnal Penelitian Sosial Keagamaan 206
https://openalex.org/W4205662306	https://www.intechopen.com/citation-pdf-url/79728	English	null	Imaging of Vascular Aphasia	IntechOpen eBooks	2,022	cc-by	4,070	1. Introduction Aphasia is an acquired language disorder caused by damage to language regions of the brain that can affect the ability of a person to understand and/or produce language. It is often accompanied by impairment in reading (alexia) and writing (agraphia). Aphasia is one of the most common and debilitating consequences of stroke and is associated with a higher risk of mortality, a poor functional prognosis, and an augmented risk of vascular dementia. Fortunately, some degree of recovery of language function occurs for about 70% of patients with post-stroke aphasia thanks to neural plasticity and speech and language therapy [1–3]. Brain imaging is essential for the initial diagnosis of stroke and the assessment of stroke severity and prognosis. Stroke location and extension are associated with different patterns of aphasia with diverse functional outcomes [2]. Functional imaging, such as positron emission tomography (PET), functional magnetic resonance imaging (fMRI), and magnetoencephalography (MEG), has also been extensively evaluated for its ability to predict recovery from aphasia [4]. The purpose of this chapter is to review the state of the art of morphological and functional imaging of vascular aphasia and to illustrate the MRI profiles of different aphasic syndromes. Abstract Brain imaging is essential for the diagnosis of acute stroke and vascular aphasia. Magnetic resonance imaging (MRI) is the modality of choice for the etiological diagnosis of aphasia, the assessment of its severity, and the prediction of recovery. Diffusion weighted imaging is used to detect, localize, and quantify the extension of the irreversibly injured brain tissue called ischemic core. Perfusion weighted imaging (from MRI or CT) is useful to assess the extension of hypoperfused but salvageable tissue called penumbra. Functional imaging (positron emission tomog- raphy (PET), functional MRI (fMRI)) may help predicting recovery and is useful for the understanding of language networks and individual variability. This chapter is meant to review the state of the art of morphological and functional imaging of vascular aphasia and to illustrate the MRI profiles of different aphasic syndromes. Keywords: aphasia, stroke, imaging, MRI, recovery Keywords: aphasia, stroke, imaging, MRI, recovery Imaging of Vascular Aphasia Loïc Duron, Augustin Lecler, Dragos Catalin Jianu, Raphaël Sadik and Julien Savatovsky 3. Imaging of aphasia Brain imaging is critical to the initial management of ischemic stroke. Magnetic Resonance (MR) diffusion and perfusion imaging as well as computed tomography (CT) perfusion imaging are commonly used in clinical routine for estimation of ischemic core, penumbra, outcome prediction, and treatment decision-making in the acute stroke setting. Functional MRI is used after the initial phase to assess the severity of aphasia and predict recovery. 2. Brain language areas Aphasia is caused by a brain damage localized in one or several language areas of the left hemisphere for 95% of right-handed people and 75% of left-handed 1 Aphasia Compendium people. Functional neuroimaging techniques have highlighted Brodman Areas (BA, Figure 1) associated with language functions. They are clustered into two main language networks: (1) a language reception/understanding system, includ- ing a “core Wernicke’s area” involved in word recognition (BA21, BA22, BA41, and BA42), and a peripheral area (“extended Wernicke’s area:” BA20, BA37, BA38, BA39, and BA40) involved in language associations; (2) a language production system (“Broca’s complex:” BA44, BA45, and also BA46, BA47, partially BA6—mainly its mesial supplementary motor area—and extending toward the basal ganglia and the thalamus. The insula (BA13) may also play a coordinating role in interconnecting these two brain language networks [5–7]. A given language impairment can result from damage or dysfunction of several Figure 1. Lateral view of Brodmann areas. Figure 1. g Lateral view of Brodmann areas. g Lateral view of Brodmann areas. people. Functional neuroimaging techniques have highlighted Brodman Areas (BA, Figure 1) associated with language functions. They are clustered into two main language networks: (1) a language reception/understanding system, includ- ing a “core Wernicke’s area” involved in word recognition (BA21, BA22, BA41, and BA42), and a peripheral area (“extended Wernicke’s area:” BA20, BA37, BA38, BA39, and BA40) involved in language associations; (2) a language production system (“Broca’s complex:” BA44, BA45, and also BA46, BA47, partially BA6—mainly its mesial supplementary motor area—and extending toward the basal ganglia and the thalamus. The insula (BA13) may also play a coordinating role in interconnecting these two brain language networks [5–7]. A given language impairment can result from damage or dysfunction of several different brain areas due to the impact of the lesion not only on the function of the affected region but also on the many regions connected to it within the language networks. 3.1 Ischemic core The ischemic core corresponds to brain tissue that has been irreversibly injured and has already turned or will inevitably turn into infarction regardless of treatment [8]. It can be assessed on non-contrast CT as hypo-attenuating areas and on Diffusion Weighted Imaging (DWI) of MRI exams as hyperintense areas with decreased Apparent Diffusion Coefficient (ADC) values on ADC map. The ischemic core can also be assessed using CT or MR perfusion imaging and is currently defined as areas with 2 Imaging of Vascular Aphasia Imaging of Vascular Aphasia DOI: http://dx.doi.org/10.5772/intechopen.101581 relative Cerebral Blood Flow (CBF) < 30% compared with the baseline level. However, hyperintense lesions on DWI, hypodense areas on CT, and areas with CBF < 30% on perfusion CT or MRI can be partially reversible, particularly if reperfusion is per- formed within 30 min, so that there is no gold standard for ischemic core imaging. 3.2 Penumbra The ischemic penumbra is defined as an area of nonfunctioning but viable brain tissue that may recover its function if blood flow is restored. Therefore, it is the main target of reperfusion treatments. The ischemic penumbra has been widely investigated because of its potential to personalize therapeutic opportunities. It can be assessed using CT or MR perfusion and is defined as the area outside the ischemic core with a time-to-maximum (Tmax) > 6 s (Figure 2). A large area of penumbra associated with a small ischemic core represents a good candidate for reperfusion therapeutics such as intravenous thrombolysis and mechanical thrombectomy. This reperfusion is highly correlated with improvements on specific language tasks [1]. Figure 2. Left sylvian acute ischemic stroke in a 44-year-old man with global aphasia. DWI (A) and FLAIR (B) are normal in the left sylvian area and show an isolated ischemic spot in the right sylvian area. Perfusion maps show a decrease in cerebral blood flow (C, dark blue) in the left sylvian area, with a consistent increase in Tmax > 6 s (D, red). This corresponds to ischemic penumbra, i.e., salvageable tissue that may benefit from reperfusion treatments. Figure 2. Left sylvian acute ischemic stroke in a 44-year-old man with global aphasia. DWI (A) and FLAIR (B) are normal in the left sylvian area and show an isolated ischemic spot in the right sylvian area. Perfusion maps show a decrease in cerebral blood flow (C, dark blue) in the left sylvian area, with a consistent increase in Tmax > 6 s (D, red). This corresponds to ischemic penumbra, i.e., salvageable tissue that may benefit from reperfusion treatments. 3 Aphasia Compendium 4.1 Broca’s aphasia Broca’s aphasia is generally produced by infarcts or severe hypoperfusion of the superior division of the left middle cerebral artery [2, 14, 15]. Brain areas involved in Broca’s aphasia are classically: 1. Broca’s area: the posterior part of the third frontal gyrus—BA44 and BA45. Lesions in this area determine transitory apraxia of speech. Larger lesions, extending to the subjacent white matter, produce transitory mutism, which is replaced by a rapidly improving syndrome with prominent arthric deforma- tions and deficits in action naming that are more severe than deficits in object naming (Figure 3). 2. Rolandic operculum: lower part of motor area: Fa (Figure 4). 3. Lesions can extend or separately affect insular cortex (Figure 5) and subjacent white matter, centrum semi-ovale, capsule-striatum (caudate nucleus head and putamen), and periventricular areas. Infarctions involving together these struc- tures and Broca’s area may also produce the complete syndrome of Broca’s aphasia. 3.3 Recovery prediction Depending on the location and extent of the left hemisphere lesion, different mechanisms may concur to recovery from aphasia: (1) right hemisphere reorga- nization, (2) implication of residual left hemisphere language areas, (3) recruit- ment of left hemisphere regions not previously involved in language function, and (4) reorientation of domain-general networks not specifically dedicated to language [1, 9–11]. Intensive and targeted language therapy may interact with brain plasticity to favor recovery from aphasia. Several neuroimaging findings were associated with aphasia severity and poor recovery, such as a large volume of ischemia, a cortical involvement, a non-fluent profile of aphasia, and a high National Institutes of Health Stroke Scale (NIHSS) score at 2 weeks [12]. Functional imaging has also been extensively investigated for its potential to predict recovery from aphasia. However, the generalizability, variability, and interpretability of group-based approaches that most imaging studies use have been criticized because of the variability in mapping function onto macro-anatomy across neurologically healthy individuals, which hinders the interpretation of results at the individual level. Therefore, the methods used in studies must be care- fully validated to safely generalize the findings [13]. 4. Illustration of aphasic syndromes The location and extension of the stroke lesions are the main determinants of the aphasic profile. Eight patterns of aphasia will be illustrated below: Broca’s aphasia, Wernicke’s aphasia, conduction, transcortical, global, anomic, crossed, and subcortical aphasia. Figure 4. g Left inferior frontal gyrus acute ischemic stroke in a 82-year-old man presenting with Broca’s aphasia, hyperintense on DWI (A) and FLAIR (B). A22) part of the temporal lobe and inferior parietal lobule [2, 14, 15]. Brain areas involved in Wernicke’s aphasia are classically: 1. Wernicke’s area: posterior part of the first two temporal gyri-T1/T2 (BA22) (Figure 6). 2. Inferior parietal lobes: angular gyrus (BA39) and supramarginal gyrus (BA40). 3. Lesions can extend to the insular-external capsule region and anterior part of temporal gyri (BA22). Besides the cortical destructions from these areas, subjacent white matter can be also affected. 4.2 Wernicke’s aphasia Wernicke aphasia is generally produced by infarcts or severe hypoperfusion of the inferior division of the left middle cerebral artery, which supplies the posterior 4 Imaging of Vascular Aphasia Imaging of Vascular Aphasia DOI: http://dx.doi.org/10.5772/intechopen.101581 Figure 3. Broca and left Rolandic operculum acute ischemic stroke in a 65-year-old woman presenting with Broca’s aphasia, hyperintense on DWI (A) and FLAIR (B). Figure 3. Figure 3. Figure 3. Broca and left Rolandic operculum acute ischemic stroke in a 65-year-old woman presenting with Broca’s aphasia, hyperintense on DWI (A) and FLAIR (B). Figure 4. Left inferior frontal gyrus acute ischemic stroke in a 82-year-old man presenting with Broca’s aphasia, hyperintense on DWI (A) and FLAIR (B). Figure 7. Figure 7. Acute ischemic stroke of the left external capsule in a 88-year-old man presenting with conduction aphasia, hyperintense on DWI (A) and FLAIR (B). memory)—so, the associated lesions are situated in areas critical for working memory: inferior parietal lobule (supramarginal and angular gyri), inferior frontal cortex, posterior temporal lobe, and/or their white matter connections (the exter- nal capsule (Figure 7). Conduction aphasia is the result of an embolic infarct of the inferior divi- sion (posterior temporal or parietal) of the left middle cerebral artery. It is rarely observed at the acute stage of stroke and more frequently affects younger patients [16]. 4.4 Transcortical aphasias Cortical lesions isolating the spared peri-sylvian language areas (watershed territory between the left anterior cerebral artery and middle cerebral artery in addition to the watershed territory between the left middle cerebral artery and posterior cerebral artery). p y Subcortical lesions: large thalamic hemorrhage interrupting the temporal isthmus; infarcts in the left thalamus, putamen, and periventricular white matter [17]. 4.3 Conduction aphasia The lesions affect the inferior parietal lobes, especially the supramarginal gyrus and/or the external capsule; they classically disrupt the arcuate fascicu- lus, although its role remains debated for the repetition impairments: probably 5 Aphasia Compendium Figure 5. Acute ischemic stroke of the left insula in a 62-year-old man with hyperacute Broca’s aphasia, hyperintense on DWI (A) but still normal on FLAIR (B). Figure 5. Acute ischemic stroke of the left insula in a 62-year-old man with hyperacute Broca’s aphasia, hyperintense on DWI (A) but still normal on FLAIR (B). disconnection between the superior temporal cortex and the inferior frontal gyri, respectively [2, 14, 15]. Other explanations for the repetition impairments have been noted, such as short-term memory syndrome (the repetition impairment due to limited working Figure 6. Acute ischemic stroke of the left temporal gyri in a 63-year-old man presenting with acute Wernicke’s aphasia, hyperintense on DWI (A) and FLAIR (B). The left middle cerebral artery is occluded (C, arrow). Dilated cortical veins are visible in the larger hypoperfused area (D), known as “cortical brush sign.” Figure 6. Acute ischemic stroke of the left temporal gyri in a 63-year-old man presenting with acute Wernicke’s aphasia, hyperintense on DWI (A) and FLAIR (B). The left middle cerebral artery is occluded (C, arrow). Dilated cortical veins are visible in the larger hypoperfused area (D), known as “cortical brush sign.” disconnection between the superior temporal cortex and the inferior frontal gyri, respectively [2, 14, 15]. disconnection between the superior temporal cortex and the inferior frontal gyri, respectively [2, 14, 15]. disconnection between the superior temporal cortex and the inferior frontal gyri, respectively [2, 14, 15]. p y [ , , ] Other explanations for the repetition impairments have been noted, such as short-term memory syndrome (the repetition impairment due to limited working 6 Imaging of Vascular Aphasia Imaging of Vascular Aphasia DOI: http://dx.doi.org/10.5772/intechopen.101581 Figure 7. Acute ischemic stroke of the left external capsule in a 88-year-old man presenting with conduction aphasia, hyperintense on DWI (A) and FLAIR (B). 4.5 Global aphasia Extended lesions (including left peri-sylvian anterior and posterior language areas), which are the result of a left middle cerebral artery or carotid artery occlu- sion (with a total left middle cerebral artery infarct), produce global aphasia with hemiplegia, hemisensory deficits, and hemianopia [2, 14, 15]. Broca’s and Wernicke’s areas may be simultaneously hypoperfused in the acute period. Thus, global aphasia can be the initial aphasic syndrome (Figure 8). g y g Early involution into Broca’s aphasia (with early recovery of comprehension) may result from reperfusion of Wernicke’s area. In this case, the patient presents only left frontal lobe, left basal ganglia, and left insula ischemic lesions (diffusion- weighted image shows infarct in superior division of left middle cerebral artery territory, which includes Broca’s area), sparing in the same time the left temporo- parietal region (global aphasia with hemiplegia and early improvement of compre- hension). Later recovery of comprehension may appear from the reorganization of the language network: 7 Aphasia Compendium • Frontal and temporoparietal lesions (two lesions) produce global aphasia without hemiplegia. When sensory-motor deficit is missing, we should search for mixed transcortical aphasia. • Subcortical infarct extended into basal ganglia. 4.6 Anomic aphasia Acute anomic aphasia may be noted after stroke in many locations. It also represents a stage of all aphasic syndromes when they improve. 4.7 Crossed aphasias This is a very rare condition (1% of all acute ischemic stroke aphasias), defined by an aphasic syndrome in a right-handed patient (free from developmental disorders and previous brain lesions, fully lateralized, which is demonstrated using a questionnaire such as Edinburgh Inventory), caused by a right hemisphere lesion (non-dominant hemisphere) [2, 14, 15]. p The anatomical determinants are similar to those observed in left hemisphere lesion, although a higher proportion of deviant cases are observed, particularly with mild aphasia contrasting with the large lesion. This fact is usually reported as evidence Figure 8. Acute ischemic stroke of the left sylvian territory in a 76-year-old man with ischemic core involving the Broca’s area on DWI (A) and FLAIR (B) and penumbra involving the Wernicke’s area (C and D). Global aphasia improved after recanalization, resulting in a chronic non-fluent Broca’s aphasia. Figure 8. Acute ischemic stroke of the left sylvian territory in a 76-year-old man with ischemic core involving the Broca’s area on DWI (A) and FLAIR (B) and penumbra involving the Wernicke’s area (C and D). Global aphasia improved after recanalization, resulting in a chronic non-fluent Broca’s aphasia. Figure 8. Acute ischemic stroke of the left sylvian territory in a 76-year-old man with ischemic core involving the Broca’s area on DWI (A) and FLAIR (B) and penumbra involving the Wernicke’s area (C and D). Global aphasia improved after recanalization, resulting in a chronic non-fluent Broca’s aphasia. 8 Imaging of Vascular Aphasia DOI: http://dx.doi.org/10.5772/intechopen.101581 Imaging of Vascular Aphasia Figure 9. Acute ischemic stroke of the right external capsule in a 76-year-old man presenting with global aphasia, hyperintense on DWI (A) and FLAIR (B). g 9 Acute ischemic stroke of the right external capsule in a 76-year-old man presenting with global aphasia, hyperintense on DWI (A) and FLAIR (B). for bilateral representation of the language. In the past, crossed aphasia was consid- ered to be non-fluent, although today is reported that all aphasic syndromes can be registered (some cases of crossed Wernicke’s aphasia in right-handed patients with lesions in the homologous area of the right cerebral hemisphere are noted) (Figure 9) 4.8 Subcortical aphasias Pure left striato-capsular infarcts (left deep middle cerebral artery infarcts) can produce different types of aphasias (mainly non-fluent, especially motor trans- cortical aphasia and Broca’s aphasia) (Figure 10). Frequently, hypophonia (poor speech volume) can be noted [2, 14, 15]. Fluent and non-fluent aphasias have been reported in thalamic lesions. Usually, a thalamic aphasia presents a significant impairment of spontaneous speech, with verbal paraphasias, but with oral comprehension and repetition relatively spared [1, 2, 5]. Patients with subcortical aphasias are older, because the main mechanism of ischemic stroke is small vascular disease. Figure 10. Acute ischemic stroke of the left striato-capsular area in a 66-year-old woman with subcortical aphasia, hyperintense on DWI (A) and FLAIR (B). Figure 10. g Acute ischemic stroke of the left striato-capsular area in a 66-year-old woman with subcortical aphasia, hyperintense on DWI (A) and FLAIR (B). 9 Aphasia Compendium Aphasia Compendium There are two distinct mechanisms concerning subcortical vascular aphasias: (a) a possible sustained cortical hypoperfusion and infarction not visible on structural imag- ing studies and (b) a possible thalamic disconnection, due to striato-capsular infarcts. Acknowledgements The authorship criteria are listed in our Authorship Policy: https://www.intecho- pen.com/page/authorship-policy. This section of your manuscript may also include funding information. 5. Conclusions Brain imaging, especially MRI, is the cornerstone of the etiological diagnosis and prognostic evaluation of vascular aphasia. Location and extent of the ischemic core are valuable information to assess the severity of aphasia and predict recovery. Despite overlap in MRI patterns between aphasic syndromes, two main networks are known to induce specific language deficit: the anterior network centered on the Broca’s area and the posterior network centered on the Wernicke’s area. Perfusion imaging is helpful to determine the mismatch between irreversibly injured tissue in the ischemic core and salvageable tissue in the ischemic penumbra that may benefit from reperfusion treatment and result in symptoms recovery. p y p y Future work may focus on the discovery of new imaging biomarkers to help predict aphasia recovery with better accuracy and orient specific treatments. Conflict of interest The authors declare no conflict of interest. Author details 10 Author details Loïc Duron1, Augustin Lecler1, Dragos Catalin Jianu2, Raphaël Sadik3 and Julien Savatovsky1 1 Department of Radiology, Foundation Rothschild Hospital, Paris, France 2 Department of Neurology, “Victor Babes” University of Medicine and Pharmacy, Timisoara, Romania 3 Foundation Rothschild Hospital, Paris, France Address all correspondence to: lduron@for.paris © 2021 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Loïc Duron1*, Augustin Lecler1, Dragos Catalin Jianu2, Raphaël Sadik and Julien Savatovsky1 10 Imaging of Vascular Aphasia DOI: http://dx.doi.org/10.5772/intechopen.101581 Imaging of Vascular Aphasia DOI: http://dx.doi.org/10.5772/intechopen.101581 Imaging of Vascular Aphasia DOI: http://dx.doi.org/10.5772/intechopen.101581 [8] Goyal M, Ospel JM, Menon B, Almekhlafi M, Jayaraman M, Fiehler J, et al. Challenging the ischemic core concept in acute ischemic stroke imaging. Stroke. 2020;51(10):3147-3155 [7] Jiménez de la Peña MM, Gómez Vicente L, García Cobos R, Martínez de Vega V. Neuroradiologic correlation with aphasias. Cortico-subcortical map of language. Radiología. 2018;60(3): 250-261 References of functional imaging studies of reorganization of language processing. Neurobiology of Language. 2021;2(1): 22-82 of functional imaging studies of reorganization of language processing. Neurobiology of Language. 2021;2(1): 22-82 [1] Thompson CK, den Ouden DB. Neuroimaging and recovery of language in aphasia. Current Neurology and Neuroscience Reports. 2008;8(6):475-483 [1] Thompson CK, den Ouden DB. Neuroimaging and recovery of language in aphasia. Current Neurology and Neuroscience Reports. 2008;8(6):475-483 [2] Abou Zeki D, Hillis A. Chapter 12. Acquired disorders of language and speech. In: Masud H, Schott JM, editors. Oxford Textbook of Cognitive Neurology and Dementia. UK: Oxford University Press; 2016. pp. 123-133 [2] Abou Zeki D, Hillis A. Chapter 12. Acquired disorders of language and speech. In: Masud H, Schott JM, editors. Oxford Textbook of Cognitive Neurology and Dementia. UK: Oxford University Press; 2016. pp. 123-133 [10] Lorca-Puls DL, Gajardo-Vidal A, PLORAS Team, Oberhuber M, Prejawa S, TMH H, et al. Brain regions that support accurate speech production after damage to Broca’s area. Brain Communications. 2021;3(4):fcab230 [3] Croquelois A, Godefroy O. Vascular aphasias. In: Godefroy O, editor. The Behavioral and Cognitive Neurology of Stroke. Cambridge: Cambridge University Press; 2013. pp. 65-75 [11] Chen X, Chen L, Zheng S, Wang H, Dai Y, Chen Z, et al. Disrupted brain connectivity networks in aphasia revealed by resting-state fMRI. Frontiers in Aging Neuroscience. 2021;13:666301 [4] Rosenberg MD, Song H. Predicting post-stroke aphasia from brain imaging. Nature Human Behaviour. 2020; 4:675-676 [12] Lee S, Na Y, Tae W-S, Pyun S-B. Clinical and neuroimaging factors associated with aphasia severity in stroke patients: diffusion tensor imaging study. Scientific Reports. 2020;10(1):12874 [4] Rosenberg MD, Song H. Predicting post-stroke aphasia from brain imaging. Nature Human Behaviour. 2020; 4:675-676 [4] Rosenberg MD, Song H. Predicting post-stroke aphasia from brain imaging. Nature Human Behaviour. 2020; 4:675-676 [5] Ardila A, Bernal B, Rosselli M. How localized are language brain areas? A review of brodmann areas involvement in oral language. Archives of Clinical Neuropsychology. 2016;31(1):112-122 [5] Ardila A, Bernal B, Rosselli M. How localized are language brain areas? A review of brodmann areas involvement in oral language. Archives of Clinical Neuropsychology. 2016;31(1):112-122 [13] Blank IA, Kiran S, Fedorenko E. Can neuroimaging help aphasia researchers? Addressing generalizability, variability, and interpretability. Cognitive Neuropsychology. 2017;34(6):377-393 [13] Blank IA, Kiran S, Fedorenko E. Can neuroimaging help aphasia researchers? Addressing generalizability, variability, and interpretability. Cognitive Neuropsychology. 2017;34(6):377-393 [13] Blank IA, Kiran S, Fedorenko E. Can neuroimaging help aphasia researchers? [9] Stephen MW, Schneck SM. Neuroplasticity in post-stroke aphasia: A systematic review and meta-analysis [3] Croquelois A, Godefroy O. Vascular aphasias. In: Godefroy O, editor. The Behavioral and Cognitive Neurology of Stroke. Cambridge: Cambridge University Press; 2013. pp. 65-75 [6] Fridriksson J, den Ouden D-B, Hillis AE, Hickok G, Rorden C, Basilakos A, et al. Anatomy of aphasia revisited. Brain: A Journal of Neurology. 2018;141(3):848-862 [8] Goyal M, Ospel JM, Menon B, Almekhlafi M, Jayaraman M, Fiehler J, et al. Challenging the ischemic core concept in acute ischemic stroke imaging. Stroke. 2020;51(10):3147-3155 [9] Stephen MW, Schneck SM. Neuroplasticity in post-stroke aphasia: A systematic review and meta-analysis References Addressing generalizability, variability, and interpretability. Cognitive Neuropsychology. 2017;34(6):377-393 [6] Fridriksson J, den Ouden D-B, Hillis AE, Hickok G, Rorden C, Basilakos A, et al. Anatomy of aphasia revisited. Brain: A Journal of Neurology. 2018;141(3):848-862 [14] Lapointe L, Sierwalt J. Aphasia and related neurogenic language disorders. New York: Thieme; 2018 [14] Lapointe L, Sierwalt J. Aphasia and related neurogenic language disorders. New York: Thieme; 2018 [7] Jiménez de la Peña MM, Gómez Vicente L, García Cobos R, Martínez de Vega V. Neuroradiologic correlation with aphasias. Cortico-subcortical map of language. Radiología. 2018;60(3): 250-261 [7] Jiménez de la Peña MM, Gómez Vicente L, García Cobos R, Martínez de Vega V. Neuroradiologic correlation with aphasias. Cortico-subcortical map of language. Radiología. 2018;60(3): 250-261 [15] Sebastian R, Breining B. Contributions of neuroimaging to understanding language deficits in acute stroke. Seminars in Speech and Language. 2018;39:066-078 [16] Benson DF, Sheremata WA, Bouchard R, Segarra JM, Price D, Geschwind N. Conduction aphasia. A clinicopathological study. Archives of Neurology. 1973;28(5):339-346 [17] Maeshima S, Osawa A. Thalamic lesions and aphasia or neglect. Current Neurology and Neuroscience Reports. 2018;18(7):39 [17] Maeshima S, Osawa A. Thalamic lesions and aphasia or neglect. Current Neurology and Neuroscience Reports. 2018;18(7):39 11
https://openalex.org/W3009306541	https://jamanetwork.com/journals/jamasurgery/articlepdf/2762479/jamasurgery_mctigue_2020_oi_200005.pdf	English	null	Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass	JAMA surgery	2,020	cc-by	11,556	Author Affiliations: Author affiliations are listed at the end of this article. Research Research Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass The National Patient-Centered Clinical Research Network (PCORNet) Bariatric Study Kathleen M. McTigue, MD; Robert Wellman, MS; Elizabeth Nauman, MPH, PhD; Jane Anau, BS; R. Yates Coley, PhD; Alberto Odor, MD; Julie Tice, MS; Karen J. Coleman, PhD; Anita Courcoulas, MD; Roy E. Pardee, JD; Sengwee Toh, ScD; Cheri D. Janning, MS; Neely Williams, MDiv; Andrea Cook, PhD; Jessica L. Sturtevant, MS; Casie Horgan, MPH; David Arterburn, MD; for the PCORnet Bariatric Study Collaborative IMPORTANCE Bariatric surgery can lead to substantial improvements in type 2 diabetes (T2DM), but outcomes vary across procedures and populations. It is unclear which bariatric procedure has the most benefits for patients with T2DM. OBJECTIVE To evaluate associations of bariatric surgery with T2DM outcomes. DESIGN, SETTING, AND PARTICIPANTS This cohort study was conducted in 34 US health system sites in the National Patient-Centered Clinical Research Network Bariatric Study. Adult patients with T2DM who had bariatric surgery between January 1, 2005, and September 30, 2015, were included. Data analysis was conducted from April 2017 to August 2019. INTERVENTIONS Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy (SG). MAIN OUTCOME AND MEASURES Type 2 diabetes remission, T2DM relapse, percentage of total weight lost, and change in glycosylated hemoglobin (hemoglobin A1c). RESULTS A total of 9710 patients were included (median [interquartile range] follow-up time, 2.7 [2.9] years; 7051 female patients [72.6%]; mean [SD] age, 49.8 [10.5] years; mean [SD] BMI, 49.0 [8.4]; 6040 white patients [72.2%]). Weight loss was significantly greater with RYGB than SG at 1 year (mean difference, 6.3 [95% CI, 5.8-6.7] percentage points) and 5 years (mean difference, 8.1 [95% CI, 6.6-9.6] percentage points). The T2DM remission rate was approximately 10% higher in patients who had RYGB (hazard ratio, 1.10 [95% CI, 1.04-1.16]) than those who had SG. Estimated adjusted cumulative T2DM remission rates for patients who had RYGB and SG were 59.2% (95% CI, 57.7%-60.7%) and 55.9% (95% CI, 53.9%-57.9%), respectively, at 1 year and 86.1% (95% CI, 84.7%-87.3%) and 83.5% (95% CI, 81.6%-85.1%) at 5 years postsurgery. Among 6141 patients who experienced T2DM remission, the subsequent T2DM relapse rate was lower for those who had RYGB than those who had SG (hazard ratio, 0.75 [95% CI, 0.67-0.84]). Estimated relapse rates for those who had RYGB and SG were 8.4% (95% CI, 7.4%-9.3%) and 11.0% (95% CI, 9.6%-12.4%) at 1 year and 33.1% (95% CI, 29.6%-36.5%) and 41.6% (95% CI, 36.8%-46.1%) at 5 years after surgery. JAMA Surgery \| Original Investigation Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass The National Patient-Centered Clinical Research Network (PCORNet) Bariatric Study JAMA Surg. 2020;155(5):e200087. doi:10.1001/jamasurg.2020.0087 Published online March 4, 2020. Invited Commentary Supplemental content Group Information: PCORnet Bariatric Study Collaborative members appear at the end of the article. Downloaded from jamanetwork.com by guest on 10/24/2024 Author Affiliations: Author affiliations are listed at the end of this article. Group Information: PCORnet Bariatric Study Collaborative members appear at the end of the article. Corresponding Author: Kathleen McTigue, MD, Department of Medicine, University of Pittsburgh, 230 McKee Pl, Ste 600, Pittsburgh, PA 15213 (kmm34@pitt.edu). Data Extraction The PCORnet Bariatric Study (PBS),25,26 one of the first scientific initiatives of PCORnet, the National Patient- Centered Clinical Research Network,27,28 was designed to examine the effectiveness of common bariatric procedures. This article compares T2DM outcomes in PCORnet up to 5 years following surgery for patients who had SG or RYGB. Secondary analyses assess the procedures’ outcomes on body weight and glycemic control independent of diabetes remission. The PCORnet sites store standardized electronic health rec- ord data and sometimes other data (eg, claims data), in PCORnet datamarts.28 Programming queries from the PCORnet Coordinating Center extracted relevant deidenti- fied data on eligible individuals from participating sites’ datamarts. Race/ethnicity, as recorded in electronic health records, was included, reflecting stakeholder input. Data were transmitted to the coordinating site for analysis. Codes from the ICD-9-CM and SNOMED identified diagnoses. Research Original Investigation Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass B a c m B ariatric surgery appears more effective than medical care alone for improving diabetes outcomes.1-3 Re- mission of type 2 diabetes (T2DM) is common after bariatric surgery4-7 and may reduce risk for subsequent mi- crovascular and macrovascular disease.8-11 However, T2DM remission rates after bariatric surgery vary substantially across procedures and populations4-7 and T2DM relapse has been re- ported in approximately a quarter to half of patients who have bariatric surgery and achieve remission.6,7,12 Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass The National Patient-Centered Clinical Research Network (PCORNet) Bariatric Study At 5 years, compared with baseline, hemoglobin A1c was reduced 0.45 (95% CI, 0.27-0.63) percentage points more for patients who had RYGB vs patients who had SG. RESULTS A total of 9710 patients were included (median [interquartile range] follow-up time, 2.7 [2.9] years; 7051 female patients [72.6%]; mean [SD] age, 49.8 [10.5] years; mean [SD] BMI, 49.0 [8.4]; 6040 white patients [72.2%]). Weight loss was significantly greater with RYGB than SG at 1 year (mean difference, 6.3 [95% CI, 5.8-6.7] percentage points) and 5 years (mean difference, 8.1 [95% CI, 6.6-9.6] percentage points). The T2DM remission rate was approximately 10% higher in patients who had RYGB (hazard ratio, 1.10 [95% CI, 1.04-1.16]) than those who had SG. Estimated adjusted cumulative T2DM remission rates for patients who had RYGB and SG were 59.2% (95% CI, 57.7%-60.7%) and 55.9% (95% CI, CONCLUSIONS AND RELEVANCE In this large multicenter study, patients who had RYGB had greater weight loss, a slightly higher T2DM remission rate, less T2DM relapse, and better long-term glycemic control compared with those who had SG. These findings can help inform patient-centered surgical decision-making. Group Information: PCORnet Bariatric Study Collaborative members appear at the end of the article. Corresponding Author: Kathleen McTigue, MD, Department of Medicine, University of Pittsburgh, 230 McKee Pl, Ste 600, Pittsburgh, PA 15213 (kmm34@pitt.edu). JAMA Surg. 2020;155(5):e200087. doi:10.1001/jamasurg.2020.0087 Published online March 4, 2020. JAMA Surg. 2020;155(5):e200087. doi:10.1001/jamasurg.2020.0087 Published online March 4, 2020. 1/12 Downloaded from jamanetwork.com by guest on 10/24/2024 Findings Findings In this cohort study of 9710 adults with T2DM who underwent bariatric surgery, most patients who had Roux-en-Y gastric bypass or sleeve gastrectomy experienced T2DM remission at some point over 5 years of follow-up. Patients who had Roux-en-Y gastric bypass showed slightly higher T2DM remission rates, better glycemic control, and fewer T2DM relapse events than patients who had sleeve gastrectomy. Studies focusing on the 2 most common bariatric proce- dures, sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB), show mixed evidence in terms of T2DM outcomes, es- pecially in the longer term.13-22 It is unclear how the choice be- tween them is likely to affect T2DM. The comparison is par- ticularly salient because SG has begun to supplant RYGB as the dominant bariatric procedure over the past decade, despite limited long-term comparative data.23-25 Meaning Understanding diabetes outcomes of different bariatric procedures will help surgeons and patients with diabetes make informed health care choices. Cohort Identification The PBS cohort was previously described.25 Patients in the T2DM analyses underwent a primary bariatric procedure at 34 PCORnet-affiliated health systems (eTable 1 in the Supplement) from January 1, 2005, through September 30, 2015. Procedures were identified from more than 59 million patient records using the International Classification of Dis- eases, Ninth Revision, Clinical Modification (ICD-9-CM), Cur- rent Procedure Terminology version 4, and Healthcare Com- mon Procedure Coding System codes. We defined patients with diabetes as having a hemoglobin A1c (HbA1c) level of 6.5% or more or a T2DM medication prescription in the year before surgery. Patients taking only metformin, thiazolidin- edione, or liraglutide needed an ICD-9-CM or Systematized Nomenclature of Medicine (SNOMED) code for T2DM or an HbA1c level of 6.5% or more in the year before surgery to be eligible for inclusion. We excluded patients 80 years or older, those without T2DM, and individuals without rel- evant outcomes data (eFigure 1 and eAppendix 1 in the Supplement). Statistical Analyses WecomparedtheassociationsofRYGBandSGwithtimetodia- betes remission. Pairwise analyses were restricted to sites with at least 1 patient of each procedure type at each point. Pos- sible confounding was addressed with direct adjustment for specific factors and deciles of an estimated propensity score. Analyses examining the adjustable gastric band procedure are provided in eAppendix 2 in the Supplement. The Kaiser Permanente Washington Health Research In- stitute obtained institutional review board approval for over- sight of data collection and analyses. Participating sites ob- tained approval or formal determination that these analyses was not human subjects research.25 A waiver of Health Insur- ancePortabilityandPrivacyActprivacyauthorization(andthus informed consent) was obtained for these analyses of deiden- tified data. Outcome Definitions Remission from T2DM was defined as the first postsurgical occurrence of an HbA1c level less than 6.5% (to convert to pro- portion of total hemoglobin, multiply by 0.04-0.07) follow- ing at least 6 months (presurgical and/or postsurgical time) withoutT2DMmedicationprescriptionorders.ThisHbA1clevel corresponds to a published, putative partial-remission threshold.29 It was identified by our clinical stakeholders as more clinically meaningful than the affiliated complete remis- sion threshold (a normal hemoglobin A1c level29 of <5.7%30), since an HbA1c level less than 6.5% corresponds to a T2DM diagnosis.30 The occurrence of levels of 6.5% or more and/or a prescription for T2DM medication after remission defined relapse. The absolute change in HbA1c level at 1 year, 3 years, and 5 years after surgery was calculated. The total weight loss percentage was estimated as (weight at surgery −weight at a postoperative point)/weight at surgery × 100). Key Points Question How do type 2 diabetes (T2DM) outcomes compare across the 2 most common bariatric procedures? JAMA Surgery May 2020 Volume 155, Number 5 jamasurgery.com Primary Analysis Cox proportional hazards models calculated the adjusted haz- ard ratio (HR) for remission and estimated the adjusted cumu- lative proportion of individuals remitting at 1 year, 3 years, and JAMA Surgery May 2020 Volume 155, Number 5 2/12 Downloaded from jamanetwork.com by guest on 10/24/2024 Original Investigation Research Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass the outcome and potential confounders (including baseline weight) and deciles of the propensity score as the indepen- dent variables. Adjusted total weight loss percentage was com- puted as the percentage change between the mean weight and the mean baseline weight. Time to T2DM relapse was as- sessed among patients who experienced diabetes remission, using the same methods as in the remission analyses. Ad- justed absolute changes in HbA1c level at 1 year, 3 years, and 5 years following surgery were estimated by procedure using a linear mixed-modeling framework with random effects for in- dividual (intercept) and follow-up time (slope). A b-spline ba- sis included a smooth function of follow-up time in the model, allowing nonlinearity in the trajectory of percentage change in HbA1c level following surgery. For HbA1c level, we consid- ered less than 7% as a goal range, consistent with American DiabetesAssociationgoalsforadultswhoarenotpregnant,and more than 8% (well above the goal for many adults, including those with advanced vascular complications) to indicate poor control.35 5 years following surgery. The proportional hazards assump- tion was tested by including an interaction between time and bariatric surgery group in the model, then inspecting Schoen- feld residuals over time. Models were adjusted for predeter- minedbaselinecovariates:age,sex,race,Hispanicethnicity,body massindexcategory(BMI;calculatedasweightinkilogramsdi- vided by height in meters squared), HbA1c category, Charlson/ Elixhausercomorbidityscore(range:−2to20;ahigherscoregen- erally indicates worse health),31 the health conditions listed in Table1,thenumberofdiabetesmedications,thenumberofdays hospitalized in the year before surgery, the year of surgery, and the site of surgery. 5 years following surgery. The proportional hazards assump- tion was tested by including an interaction between time and bariatric surgery group in the model, then inspecting Schoen- feld residuals over time. Models were adjusted for predeter- minedbaselinecovariates:age,sex,race,Hispanicethnicity,body massindexcategory(BMI;calculatedasweightinkilogramsdi- vided by height in meters squared), HbA1c category, Charlson/ Elixhausercomorbidityscore(range:−2to20;ahigherscoregen- erally indicates worse health),31 the health conditions listed in Table1,thenumberofdiabetesmedications,thenumberofdays hospitalized in the year before surgery, the year of surgery, and the site of surgery. Logistic regression models estimating treatment propen- sity scores included fixed main effects for the prespecified covariates plus baseline variables for automated selection. Subgroup Analyses Exploratoryhypothesis-generatinganalysesexaminedhetero- geneity of treatment outcomes. Following recommendations for use of risk-stratified analyses to detect differences in treat- ment outcome,33 subgroups defined by DiaRem score (Table 1) were assessed via interactions with procedure type. The Dia- Rem score is a widely validated approach to preoperative prog- nostication of T2DM remission after bariatric surgery; higher scoresdenotealowerprobabilityofT2DMremission.34Itiscal- culated based on age, HbA1c level, insulin use, and use of oral diabetes medications. The mean (SD) preoperative HbA1c was 7.2% (1.3%), and patients took a mean (SD) of 1.66 (1.1) diabetes medications (range, 0-7 medications). The mean (SD) preoperative sys- tolic and diastolic blood pressure were 130.5 (17.2) mm Hg and 73.7 (11.2) mm Hg, respectively. Weight-associated comorbidi- ties were common. Patients who had RYGB had higher preva- lence of some comorbidities, such as sleep apnea (RYGB: 3607 patients[57.9%];SG:1740patients[50.0%]),nonalcoholicfatty liver disease (RYGB: 1914 patients [30.7%]; SG: 730 patients [21.0%]), and gastroesophageal reflux disease (RYGB: 2609 patients [41.9%]; SG: 1264 patients [36.4%]). The mean (SD) Charlson/Elixhauser score was negative (−0.089 [0.99]), Sample In this unmatched surgical cohort, the analytic sample in- cluded 9710 adults, primarily female (7051 female patients [72.6%]) with a mean (SD) age of 49.8 (10.5) years (Table 1). A total of 6233 (64.2%) underwent RYGB, and 3477 (35.8%) had SGs. The mean (SD) preoperative BMI was 49.0 (8.4). Patients were primarily white (6040 [72.2%]). Most (7904 [81.4%]) sur- geries occurred between 2010 and 2014. Sensitivity Analyses l Sensitivity analyses considered 9-month and 12-month alter- native lags from the last observed T2DM medication order to define remission (HbA1c level <6.5%). To evaluate variability in medicationdatacaptureacrossdifferenthealthsystems,thepri- mary analyses were repeated using only data from 8 inte- grated health systems, where infrastructure may enable more complete access to medication orders. Additional sensitivity analyses assessed 2 alternate censoring scenarios for inpatient stays: (1) no removal of inpatient medications or HbA1c values and(2)censoringfollow-upatthedayofadmission.Similarsen- sitivity analyses were applied to the relapse analyses. Analy- ses were conducted using R version 3.4.3 (R Foundation for Statistical Computing). Follow-up for T2DM remission was calculated from the index procedure date to the last observable data point follow- ing surgery (ie, the last observed visit, weight, blood pres- sure, HbA1c laboratory value, or diabetes prescription). Remis- sion analyses’ censoring events included death, conversion to a second bariatric procedure (eg, SG to RYGB), pregnancy (at the delivery date minus 270 days), and an 18-month lapse in diabetes-specific health care at participating sites. The re- lapse analyses included an additional censoring event, lapse in provision of any care, because patients in remission from T2DM were not necessarily expected to receive HbA1c mea- sures or T2DM prescriptions but needed to receive care in the system to be observed for relapse. It was defined as more than 18 months without any recorded HbA1c levels, body weight measurement, blood pressure, diagnosis code, or procedure code. Since inpatient hospitalization can temporarily worsen glycemic control, we excluded HbA1c measurements from ad- mission date to 90 days after discharge and medication or- ders from admission dates to the day before discharge. Primary Analysis To allow for differing outcomes of confounding variables by pro- cedure site, propensity score models included subsets of all possible 2-way interactions between the listed variables and site. The subset of interactions and the additional covariates beyond the prespecified set were chosen using the least abso- lute shrinkage and selection operator method, with cross validation to select the most parsimonious model, with pre- diction error close to the minimum possible (within 1 SE).32 Secondary Analyses Estimates of trends in mean total weight loss percentage were obtained using linear mixed-effects modeling with weight as JAMA Surgery May 2020 Volume 155, Number 5 3/12 jamasurgery.com jamasurgery.com jamasurgery.com (continued) Downloaded from jamanetwork.com by guest on 10/24/2024 Downloaded from jamanetwork.com by guest on 10/24/2024 Research Original Investigation Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass Table 1. Sample Description of Adults Prior to Bariatric Surgery Characteristic No. (%) Standardized Difference Roux-en-Y Gastric Bypass Sleeve Gastrectomy Overall Patients 6233 (64.2) 3477 (35.8) 9710 (100.0) NA Follow-up time, y Mean (SD) 3.3 (2.1) 2.2 (1.4) 2.9 (1.9) NA Median (IQR) [range] 3.2 (1.55-4.64) [0.01-10.7] 2.0 (0.99-3.26) [0.01-7.2] 2.7 (1.26-4.19) [0.01-10.7] NA Female 4576 (73.4) 2475 (71.2) 7051 (72.6) 0.05 Age, mean (SD), y 49.9 (10.4) 49.7 (10.8) 49.8 (10.5) 0.01 Age category, y 20-44 1929 (31.0) 1117 (32.1) 3046 (31.4) 0.04 45-64 3819 (61.3) 2065 (59.4) 5884 (60.6) 65-80 485 (7.8) 295 (8.5) 780 (8.0) BMI, mean (SD) 49.0 (8.2) 49.0 (8.6) 49.0 (8.4) 0.01 BMI category 35-39 638 (10.2) 386 (11.1) 1024 (10.6) 0.06 40-49 3250 (52.1) 1781 (51.2) 5031 (51.8) 50-59 1739 (27.9) 917 (26.4) 2656 (27.4) ≥60 606 (9.7) 393 (11.3) 999 (10.3) Weight, mean (SD), kg 125.6 (25.6) 125.6 (27.1) 125.63 (26.1) 0.00 Weight, kg 45.4-90 253 (4.1) 165 (4.8) 418 (4.3) 0.06 90-135 4025 (64.6) 2238 (64.4) 6263 (64.6) 135-180 1743 (28.0) 927 (26.7) 2670 (27.5) 180-225 187 (3.0) 132 (3.8) 319 (3.3) 225-275 20 (0.3) 11 (0.3) 31 (0.3) Missing 5 (0.1) 4 (0.1) 9 (0.1) Year or year range of surgery 2005-2009 969 (15.6) 53 (1.5) 1022 (10.5) 0.75 2010 1049 (16.8) 216 (6.2) 1265 (13.0) 2011 1250 (20.1) 570 (16.4) 1820 (18.7) 2012 1037 (16.6) 657 (18.9) 1694 (17.5) 2013 798 (12.8) 743 (21.4) 1541 (15.9) 2014 744 (11.9) 840 (24.2) 1584 (16.3) 2015 386 (6.2) 398 (11.5) 784 (8.1) Hispanic ethnicity 1407 (22.9) 971 (28.3) 2378 (24.8) 0.12 Missing 91 (1.5) 42 (1.2) 133 (1.4) Race Asian 86 (1.6) 69 (2.4) 155 (1.9) 0.28 African American 900 (16.6) 800 (27.3) 1700 (20.3) Multiple 3 (0.1) 5 (0.2) 8 (0.1) White 4136 (76.2) 1904 (64.9) 6040 (72.2) Pacific Islander 32 (0.6) 19 (0.7) 51 (0.6) Native American 49 (0.9) 21 (0.7) 70 (0.8) Other 225 (4.1) 117 (4.0) 342 (4.1) Missing overall 802 (12.9) 542 (15.6) 1344 (13.8) Hemoglobin A1c, mean (SD) 7.3 (1.3) 7.1 (1.2) 7.2 (1.3) 0.17 (continued) Table 1. (continued) JAMA Surgery May 2020 Volume 155, Number 5 5/12 Downloaded from jamanetwork.com by guest on 10/24/2024 Downloaded from jamanetwork.com by guest on 10/24/2024 Sample Description of Adults Prior to Bariatric Surgery (continued) JAMA Surgery May 2020 Volume 155, Number 5 5/12 jamasurgery.com Research Original Investigation Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass Table 1. Sample Description of Adults Prior to Bariatric Surgery (continued) Characteristic No. (%) Standardized Difference Roux-en-Y Gastric Bypass Sleeve Gastrectomy Overall Nonalcoholic fatty liver disease 1914 (30.7) 730 (21.0) 2644 (27.2) 0.22 Osteoarthritis 148 (2.4) 93 (2.7) 241 (2.5) 0.02 Polycystic ovarian syndrome 257 (4.1) 147 (4.2) 404 (4.2) 0.01 Pulmonary embolism 87 (1.4) 39 (1.1) 126 (1.3) 0.03 Psychotic disorder 197 (3.2) 96 (2.8) 293 (3.0) 0.02 Sleep apnea 3607 (57.9) 1740 (50.0) 5347 (55.1) 0.16 Smoker 582 (9.3) 276 (7.9) 858 (8.8) 0.05 Substance use disorder 143 (2.3) 102 (2.9) 245 (2.5) 0.04 Inpatient hospital days in y before surgery, mean (SD) 0.67 (8.0) 0.83 (8.0) 0.73 (8.0) 0.02 Inpatient hospital days in categories 0 5758 (92.4) 3156 (90.8) 8914 (91.8) 0.06 1-7 373 (6.0) 253 (7.3) 626 (6.5) 8-14 45 (0.7) 36 (1.0) 89 (0.9) 15 or more 57 (0.9) 32 (0.9) 81 (0.8) DiaRem scorea 0-2 809 (13.0) 517 (14.9) 1326 (13.7) 0.11 3-7 2211 (35.5) 1251 (36.0) 3462 (35.7) 8-12 759 (12.2) 412 (11.9) 1171 (12.1) 13-17 2127 (34.1) 1185 (34.1) 3312 (34.1) 18-22 327 (5.3) 112 (3.2) 439 (4.5) Missing 0 0 0 0 0 0 Abbreviations: BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); GLP-1, glucagon-like peptide 1; IQR, interquartile range; NA, not applicable. a Score indicates preoperative prognostication of type 2 diabetes remission following Roux-en-Y gastric bypass surgery, where a higher score indicates lower probability of type 2 diabetes remission: 0 to 2 (88%-99%), 3 to 7 (64%-88%), 8 to 12 (23%-49%), 13 to 17 (11%-33%), and 18 to 22 (2%-16%). Table 1. Sample Description of Adults Prior to Bariatric Surgery (continued) consistent with the high hypertension prevalence in an oth- erwise relatively healthy sample. and 86.1% (95% CI, 84.7%-87.3%) vs 83.5% (95% CI, 81.6%- 85.1%) at 5 years (Table 3). Sensitivity analyses requiring 9-month and 12-month time frames without a diabetes medication prescription to define remission produced similar results to the primary analysis and its 6-month time frame, although differences between SG and RGB were not always statistically significant (eTable 2 in the Supplement). Downloaded from jamanetwork.com by guest on 10/24/2024 Analyses restricted to 8 integrated health sys- tems yielded qualitatively similar results to the primary analy- ses, despite slightly higher cumulative remission rates for SG and RYGB (eTable 3 in the Supplement). T2DM Relapse Atotalof6141patientswithdocumentedT2DMremissionwere eligiblefortherelapseanalyses.Preoperationdemographicand health features were similar to those of the larger T2DM co- hort (eTable 4 in the Supplement). Mean (SD) preoperation HbA1c levels were slightly lower (7.0% [1.1%]) vs 7.2% [1.3%]) as was the mean (SD) number of diabetes medications (1.5 (1.1) medications vs 1.7 [1.1] medications) and insulin use (2317 of 6141 [37.7%] vs 4692 of 9710 [48.3%]; eTable 4 in the Supple- ment). They were followed up for relapse for a median of 2.4 (0.003-10.35) years. Percentage of Total Weight Lost Patients who had each procedure showed considerable weight loss1yearaftersurgery(SG,−22.8%[95%CI,−23.1%to−22.5%]; RYGB, −29.1% [95% CI, −29.3% to −28.8%]); typically, weight regain then occurred. The groups maintained a mean body weight well below the baseline at 5 years (SG, −16.1% [95% CI, −17.3% to −14.8%]; RYGB, −24.1% [95% CI, −25.0% to −23.3%]). Typically, the RYGB group reflected 6.2% to 8.1% more total body weight loss than the SG group at each point (Figure 1; Table 2). This represents a 10.2-kg difference (95% CI, 8.3- 12.1kg;P < .001)inweightlossbetweenRYGBandSGat5years. Downloaded from jamanetwork.com by guest on 10/24/2024 Sample Description of Adults Prior to Bariatric Surgery 4/12 JAMA Surgery May 2020 Volume 155, Number 5 jamasurgery.com 4/12 Downloaded from jamanetwork.com by guest on 10/24/2024 Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass Original Investigation Research Original Investigation Research Table 1. Sample Description of Adults Prior to Bariatric Surgery (continued) Characteristic No. (%) Standardized Difference Roux-en-Y Gastric Bypass Sleeve Gastrectomy Overall Hemoglobin A1c category, % <6.5 1554 (24.9) 922 (26.5) 2476 (25.5) 0.19 6.5-6.9 1408 (22.6) 951 (27.4) 2359 (24.3) 7.0-7.9 1738 (27.9) 995 (28.6) 2733 (28.2) 8.0-8.9 834 (13.4) 354 (10.2) 1188 (12.2) ≥9.0 699 (11.2) 255 (7.3) 954 (9.8) Total diabetes medications, mean (SD), No. 1.70 (1.1) 1.60 (1.1) 1.66 (1.1) 0.09 Total diabetes medications, No. 0 1096 (17.6) 747 (21.5) 1843 (19.0) 0.11 1 1354 (21.7) 772 (22.2) 2126 (21.9) 2 2447 (39.3) 1266 (36.4) 3713 (38.2) 3 1048 (16.8) 546 (15.7) 1594 (16.4) 4-7 288 (4.6) 146 (4.2) 434 (4.5) Diabetes medications Biguanides 4109 (65.9) 2237 (64.3) 6346 (65.4) 0.03 GLP-1 receptor agonists 278 (4.5) 148 (4.3) 426 (4.4) 0.01 Insulins 3047 (48.9) 1645 (47.3) 4692 (48.3) 0.03 Sulfonylureas 2054 (33.0) 1058 (30.4) 3112 (32.1) 0.05 Thiazolidinediones 609 (9.8) 198 (5.7) 807 (8.3) 0.15 Other 477 (7.7) 260 (7.5) 737 (7.6) 0.01 Blood pressure, mean (SD) Systolic 130.1 (17.0) 131.3 (17.5) 130.5 (17.2) 0.07 Diastolic 73.8 (10.9) 73.5 (11.6) 73.7 (11.2) 0.02 Blood pressure category Normal 1473 (23.9) 779 (22.6) 2252 (23.4) 0.06 Prehypertensive 2991 (48.5) 1626 (47.1) 4617 (48.0) Stage 1 1320 (21.4) 812 (23.5) 2132 (22.2) ≥Stage 2 379 (6.2) 236 (6.8) 615 (6.4) Missing 70 (1.1) 24 (0.7) 94 (1.0) Charlson-Elixhauser category, mean (SD) −0.082 (0.97) −0.103 (1.02) −0.089 (0.99) 0.02 Health conditions Anxiety 1274 (20.4) 734 (21.1) 2008 (20.7) 0.02 Depression 2157 (34.6) 1053 (30.3) 3210 (33.1) 0.09 Diabetes 5952 (95.5) 3221 (92.6) 9173 (94.5) 0.12 Deep-vein thrombosis 38 (0.6) 28 (0.8) 66 (0.7) 0.02 Dyslipidemia 4775 (76.6) 2659 (76.5) 7434 (76.6) 0.00 Eating disorder 969 (15.6) 231 (6.6) 1200 (12.4) 0.29 Gastroesophageal reflux disease 2609 (41.9) 1264 (36.4) 3873 (39.9) 0.11 Hypertension 5113 (82.0) 2729 (78.5) 7842 (80.8) 0.09 Infertility 29 (0.5) 29 (0.8) 58 (0.6) 0.05 Kidney disease 1268 (20.3) 670 (19.3) 1938 (20.0) 0.03 (continued) Table 1. T2DM Remission Finding Total weight loss, % Sleeve gastrectomy 2404 −22.8 (−23.1 to −22.5) 2404 −19.2 (−20.0 to −18.5) 2404 −16.1 (−17.3 to −14.8) Roux-en-Y gastric bypass 4025 −29.1 (−29.3 to −28.8) 4025 −26.2 (−26.7 to −25.7) 4025 −24.1 (−25.0 to −23.3) Difference NA 6.2 (5.8-6.7) NA 7.0 (6.1-7.9) NA 8.1 (6.6-9.6) P Value NA <.001 NA <.001 NA <.001 Hemoglobin A1c mean difference (95% CI), %a Sleeve gastrectomy 2935 −0.89 (−0.93 to −0.86) 2935 −0.56 (−0.64 to −0.49) 2935 −0.35 (−0.51 to −0.19) Roux-en-Y gastric bypass 5428 −1.12 (−1.14 to −1.09) 5428 −1.01 (−1.06 to −0.97) 5428 −0.80 (−0.88 to −0.72) Difference NA −0.22 (−0.26 to −0.18) NA −0.45 (−0.54 to −0.36) NA −0.45 (−0.63 to −0.27) P Value NA <.001 NA <.001 NA <.001 Abbreviations: ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; NA, not applicable; SNOMED, Systematized Nomenclature of Medicine. a Difference is the baseline value minus the end point value; the model was adjusted for age, sex, race, Hispanic ethnicity, body mass index (calculated as weight in kilograms divided by height in meters squared), hemoglobin A1c value, blood pressure, number of inpatient hospital days in the year prior to surgery, number of diabetes medications excluding insulin, insulin use, Charlson/Elixhauser comorbidity score, year of procedure, days from hemoglobin A1c measurement to baseline, having an ICD-9-CM or SNOMED code for diabetes, smoking, having an ICD-9-CM or SNOMED code for other comorbidities (hypertension, dyslipidemia, sleep apnea, osteoarthritis, nonalcoholic fatty liver disease, gastroesophageal reflux disease, depression, anxiety, eating disorder, substance use, psychosis, kidney disease, infertility, polycystic ovarian syndrome, deep-vein thrombosis, and pulmonary embolism), having ICD-9-CM or SNOMED codes for specific diabetes medications (biguanides, glucagon-like peptide–1 agonists, sulfonylureas, thiazolidinediones, and others), site, and propensity-score deciles. Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass Original Investigation Research Figure 1. Adjusted Total Weight Loss and Change in Hemoglobin A1c Level by Procedure Over 5 Years of Follow-up 0 –10 –20 –30 –40 Total Weight Loss, % Time Since Bariatric Procedure, y Estimated total weight loss for RYGB and SG procedures A 5 4 3 2 1 0 0.5 0 –0.5 –1.0 –1.5 Change From Baseline in HbA1C Level, g/dL Time Since Procedure, y Estimated change in hemoglobin A1C level for all procedures B 5 4 3 2 1 0 SG RYGB SG RYGB Shaded areas represent 95% pointwise CIs for procedure-specific changes in hemoglobin A1c levels. T2DM Remission Finding Total weight loss, % Sleeve gastrectomy 2404 −22.8 (−23.1 to −22.5) 2404 −19.2 (−20.0 to −18.5) 2404 −16.1 (−17.3 to −14.8) Roux-en-Y gastric bypass 4025 −29.1 (−29.3 to −28.8) 4025 −26.2 (−26.7 to −25.7) 4025 −24.1 (−25.0 to −23.3) Difference NA 6.2 (5.8-6.7) NA 7.0 (6.1-7.9) NA 8.1 (6.6-9.6) P Value NA <.001 NA <.001 NA <.001 Hemoglobin A1c mean difference (95% CI), %a Sleeve gastrectomy 2935 −0.89 (−0.93 to −0.86) 2935 −0.56 (−0.64 to −0.49) 2935 −0.35 (−0.51 to −0.19) Roux-en-Y gastric bypass 5428 −1.12 (−1.14 to −1.09) 5428 −1.01 (−1.06 to −0.97) 5428 −0.80 (−0.88 to −0.72) Difference NA −0.22 (−0.26 to −0.18) NA −0.45 (−0.54 to −0.36) NA −0.45 (−0.63 to −0.27) P Value NA <.001 NA <.001 NA <.001 Abbreviations: ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; NA, not applicable; SNOMED, Systematized Nomenclature of Medicine. a Difference is the baseline value minus the end point value; the model was adjusted for age, sex, race, Hispanic ethnicity, body mass index (calculated as weight in kilograms divided by height in meters squared), hemoglobin A1c value, blood pressure, number of inpatient hospital days in the year prior to surgery, number of diabetes medications excluding insulin, insulin use, Charlson/Elixhauser comorbidity score, year of procedure, days from hemoglobin A1c measurement to baseline, having an ICD-9-CM or SNOMED code for diabetes, smoking, having an ICD-9-CM or SNOMED code for other comorbidities (hypertension, dyslipidemia, sleep apnea, osteoarthritis, nonalcoholic fatty liver disease, gastroesophageal reflux disease, depression, anxiety, eating disorder, substance use, psychosis, kidney disease, infertility, polycystic ovarian syndrome, deep-vein thrombosis, and pulmonary embolism), having ICD-9-CM or SNOMED codes for specific diabetes medications (biguanides, glucagon-like peptide–1 agonists, sulfonylureas, thiazolidinediones, and others), site, and propensity-score deciles. Table 2. Comparative Effectiveness of Gastric Bypass and Sleeve Gastrectomy for Percentage of Total Weight Loss and Absolute Difference in Hemoglobin A1c Level Among Adults With Diabetes With 1, 3, and 5 Years of Follow-upa Table 2. Comparative Effectiveness of Gastric Bypass and Sleeve Gastrectomy for Percenta in Hemoglobin A1c Level Among Adults With Diabetes With 1, 3, and 5 Years of Follow-upa Table 2. Comparative Effectiveness of Gastric Bypass and Sleeve Gastrectomy for Percentage of Total Weight Loss and Absolute n Hemoglobin A1c Level Among Adults With Diabetes With 1, 3, and 5 Years of Follow-upa Abbreviations: ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; NA, not applicable; SNOMED, Systematized Nomenclature of Medicine. T2DM Remission percentage point drop in HbA1c level (95% CI, 1.09-1.14 per- centage points) over 1 year. This change was 0.22 (95% CI, 0.18- 0.26) percentage points lower than seen for patients who had SG (Table 2). At 5 years, HbA1c levels remained 0.80 (95% CI, 0.72-0.88) percentage points below baseline among patients who had RYGB and 0.35 (95% CI, 0.19-0.51) percentage points belowbaselineamongpatientswhohadSG,adifferenceof0.45 (95% CI, 0.27-0.62) percentage points. The proportion with a poorly controlled HbA1c level (≥8.0%) declined from baseline through 1 year of follow-up for both groups (patients who had RYGB, 24.6% [95% CI, 23.5%-25.7%] to 6.7% [95% CI, 6.0%- 7.7%]; patients who had SG, 17.5% [95% CI, 16.24%-18.88%] relapse for the RYGB and SG groups, respectively, were 8.4% (95% CI, 7.4%-9.3%) and 11.0% (95% CI, 9.6%-12.4%) 1 year af- ter remission, 21.2% (95% CI, 19.1%-23.2%) and 27.2% (95% CI, 24.1%-30.1%) at 3 years, and 33.1% (95% CI, 29.6%-36.5%) and 41.6% (95% CI, 36.8%-46.1%) at 5 years (Table 3). Sensitivity analyses showed similar findings (eTable 5 and eTable 6 in the Supplement). T2DM Remission The cohort was followed up for a median of 2.7 (interquartile range,1.26-4.19)years.Type2diabetesremissionoccurredpri- marily in the first 2 years (Figure 2). Patients who underwent RYGB showed slightly (10%) higher T2DM remission rates than those who had SG (hazard ratio, 1.10 [95% CI, 1.04-1.16]; Table 3). We estimated that 59.2% (95% CI, 57.7%-60.7%) of patients who had RYGB vs 55.9% (95% CI, 53.9%-57.9%) of thosewhohadSGexperiencedremissionby1year,84.3%(95% CI, 82.9%-85.5%) vs 81.5% (95% CI, 79.6%-83.2%) at 3 years, The T2DM relapse rate was lower for RYGB than SG (haz- ard ratio, 0.75 [95% CI, 0.67-0.84]). Estimated proportions of JAMA Surgery May 2020 Volume 155, Number 5 6/12 jamasurgery.com jamasurgery.com Downloaded from jamanetwork.com by guest on 10/24/2024 relapse for the RYGB and SG groups, respectively, were 8.4% (95% CI, 7.4%-9.3%) and 11.0% (95% CI, 9.6%-12.4%) 1 year af- ter remission, 21.2% (95% CI, 19.1%-23.2%) and 27.2% (95% CI, 24.1%-30.1%) at 3 years, and 33.1% (95% CI, 29.6%-36.5%) and 41.6% (95% CI, 36.8%-46.1%) at 5 years (Table 3). Sensitivity percentage point drop in HbA1c level (95% CI, 1.09-1.14 per- centage points) over 1 year. This change was 0.22 (95% CI, 0.18- 0.26) percentage points lower than seen for patients who had SG (Table 2). At 5 years, HbA1c levels remained 0.80 (95% CI, 0.72-0.88) percentage points below baseline among patients Figure 1. Adjusted Total Weight Loss and Change in Hemoglobin A1c Level by Procedure Over 5 Years of Follow-up 0 –10 –20 –30 –40 Total Weight Loss, % Time Since Bariatric Procedure, y Estimated total weight loss for RYGB and SG procedures A 5 4 3 2 1 0 0.5 0 –0.5 –1.0 –1.5 Change From Baseline in HbA1C Level, g/dL Time Since Procedure, y Estimated change in hemoglobin A1C level for all procedures B 5 4 3 2 1 0 SG RYGB SG RYGB Shaded areas represent 95% pointwise CIs for procedure-specific changes in hemoglobin A1c levels. RYGB indicates Roux-en-Y gastric bypass; SG, sleeve gastrectomy. Table 2. Comparative Effectiveness of Gastric Bypass and Sleeve Gastrectomy for Percentage of Total Weight Loss and Absolute Difference in Hemoglobin A1c Level Among Adults With Diabetes With 1, 3, and 5 Years of Follow-upa Group Time Since Bariatric Procedure 1 y 3 y 5 y Patients, No. Finding Patients, No. Finding Patients, No. T2DM Remission RYGB indicates Roux-en-Y gastric bypass; SG, sleeve gastrectomy. Figure 1. Adjusted Total Weight Loss and Change in Hemoglobin A1c Level by Procedure Over 5 Years of Follow-up 0.5 0 –0.5 –1.0 –1.5 Change From Baseline in HbA1C Level, g/dL Time Since Procedure, y Estimated change in hemoglobin A1C level for all procedures B 5 4 3 2 1 0 SG RYGB 0 –10 –20 –30 –40 Total Weight Loss, % Time Since Bariatric Procedure, y Estimated total weight loss for RYGB and SG procedures A 5 4 3 2 1 0 SG RYGB epresent 95% pointwise CIs for procedure-specific changes in hemoglobin A1c levels. RYGB indicates Roux-en-Y gastric bypass; SG, sleeve Shaded areas represent 95% pointwise CIs for procedure-specific changes in hemoglobin A1c levels. RYGB indicates Roux-en-Y gastric bypass; SG, sleeve gastrectomy. relapse for the RYGB and SG groups, respectively, were 8.4% (95% CI, 7.4%-9.3%) and 11.0% (95% CI, 9.6%-12.4%) 1 year af- ter remission, 21.2% (95% CI, 19.1%-23.2%) and 27.2% (95% CI, 24.1%-30.1%) at 3 years, and 33.1% (95% CI, 29.6%-36.5%) and 41.6% (95% CI, 36.8%-46.1%) at 5 years (Table 3). Sensitivity analyses showed similar findings (eTable 5 and eTable 6 in the Supplement). Change in Glycosylated Hemoglobin percentage point drop in HbA1c level (95% CI, 1.09-1.14 per- centage points) over 1 year. This change was 0.22 (95% CI, 0.18- 0.26) percentage points lower than seen for patients who had SG (Table 2). At 5 years, HbA1c levels remained 0.80 (95% CI, 0.72-0.88) percentage points below baseline among patients who had RYGB and 0.35 (95% CI, 0.19-0.51) percentage points belowbaselineamongpatientswhohadSG,adifferenceof0.45 (95% CI, 0.27-0.62) percentage points. The proportion with a poorly controlled HbA level (≥8 0%) declined from baseline Table 2. Comparative Effectiveness of Gastric Bypass and Sleeve Gastrectomy for Percentage of Total Weight Loss and Absolute Difference in Hemoglobin A1c Level Among Adults With Diabetes With 1, 3, and 5 Years of Follow-upa Group Time Since Bariatric Procedure 1 y 3 y 5 y Patients, No. Finding Patients, No. Finding Patients, No. Change in Glycosylated Hemoglobin Patients who underwent RYGB experienced larger and more- sustained HbA1c reductions than those using SG (Figure 1). In adjusted comparisons, patients who had RYGB showed a 1.12 7/12 JAMA Surgery May 2020 Volume 155, Number 5 jamasurgery.com Downloaded from jamanetwork.com by guest on 10/24/2024 Figure 2. Cumulative Incidence Rates of Type 2 Diabetes Remission and Relapse Across 5 Years in the National Patient-Centered Clinical Research Network Bariatric Study Cohort 1.0 0.8 0.6 0.4 0.2 0 1.0 0.8 0.6 0.4 0.2 0 Adjusted Cumulative Remission, % Time Since Bariatric Procedure, y Adjusted cumulative remission A 5 4 3 2 1 0 Adjusted Cumulative Relapse, % Time Since Diabetes Remission, y Adjusted cumulative relapse B 5 4 3 2 1 0 RYGB SG RYGB SG Shaded areas represent 95% pointwise CIs for procedure-specific rates. RYGB indicates Roux-en-Y gastric bypass; SG, sleeve gastrectomy. Figure 2. Cumulative Incidence Rates of Type 2 Diabetes Remission and Relapse Across 5 Years in the National Patient-Cente Clinical Research Network Bariatric Study Cohort ntwise CIs for procedure-specific rates. RYGB indicates Roux-en-Y gastric bypass; SG, sleeve gastrectomy. Shaded areas represent 95% pointwise CIs for procedure-specific rates. RYGB indicates Roux-en-Y gastric bypass; SG, sleeve gastrectomy. Overall, these results indicate that RYGB is associated with better long-term T2DM and weight outcomes than SG in real- world clinical settings. This is at odds with recent random- ized clinical trials that compared T2DM outcomes of RYGB and SG and found no significant differences.19-21 Those trials had longer duration of follow-up but much smaller sample sizes, which may have limited their power to detect differences be- tween the procedures. Also, patients who are willing to un- dergo randomization between RYGB and SG and surgeons who have equal skill and equipoise for RYGB and SG are likely dif- ferent from those who choose either RYGB or SG in uncon- trolled settings. Thus, while the more rigorous, randomized clinical trial data indicate that RYGB and SG perform similarly in highly controlled environments, in everyday practice, the outcome differences may be larger. to 8.3% [95% CI, 7.05%-9.79%]); it then increased, with 16.2% of patients who had RYGB and 22.4% of patients who had SG having HbA1c levels greater than 8.0% 5 years after surgery. Following surgery, a well-controlled HbA1c level (<6.5%) was consistently more common among patients who had RYGB (eFigure 2 in the Supplement). T2DM Remission in Patient Subgroups Analyses for heterogeneity of treatment outcomes indicated that the likelihood of diabetes remission comparing RYGB vs SG varied significantly across DiaRem strata (eTable 7 in the Supplement). Patients with higher DiaRem scores showed greater likelihood of diabetes remission with RYGB compared with SG, with a statistically significant association for scores between 13 and 17. Among individuals with DiaRem scores in the 13-point to 17-point range, 83.4% (95% CI, 77.9%-87.6%) of patients who had RYGB had experienced T2DM remission by 5 years of follow-up vs 76.6% (95% CI, 70.0%-81.8%) of pa- tients who had SG (eTable 8 in the Supplement). Asexpected,1,6,7,22,36somepatientsubgroupsshowedlower rates of T2DM remission. Our findings indicate that patients with lower preoperative probability for T2DM remission (11%- 33%)maybemorelikelytoachieveT2DMremissionwithRYGB compared with SG. Estimating the likelihood of T2DM remis- sion could help inform patients’ and clinicians’ discussions of procedure choice. Preoperative insulin use, older age, higher HbA1c level, and more complex T2DM medication regimens predispose patients to lower probability of T2DM remission in the DiaRem scoring system.34 Informed decision-making for procedure choice should also consider other factors, such as the potential for adverse events. jamasurgery.com JAMA Surgery May 2020 Volume 155, Number 5 Discussion Covariates included age, sex, race, Hispanic ethnicity, body mass index, hemoglobin A1c level, blood pressure, days from body mass index measurement to baseline, a number of inpatient hospital days in the year prior to surgery, a number of diabetes medications excluding insulin, insulin use, Charlson/Elixhauser comorbidity score, the year of procedure, having an ICD-9-CM or SNOMED code for diabetes, smoking, having an ICD-9-CM or SNOMED code for other comorbidities (hypertension, dyslipidemia, sleep apnea, osteoarthritis, nonalcoholic fatty liver disease, gastroesophageal reflux disease, depression, anxiety, eating disorder, substance use, psychosis, kidney disease, infertility, polycystic ovarian syndrome, deep vein thrombosis, or pulmonary embolism), having ICD-9-CM or SNOMED codes for specific diabetes medications (biguanides, GLP-1 agonists, sulfonylureas, thiazolidinediones, and others), site, and propensity-score deciles. bNumber of people who had an event in the relevant time frame. c For Roux-en-Y gastric bypass vs sleeve gastrectomy; remission of diabetes was defined as hemoglobin A1c less than 6.5% after 6 months without any prescription order for a diabetes medication; covariates included age, sex, race, Hispanic ethnicity, body mass index (calculated as weight in kilograms divided by height in meters squared), hemoglobin A1c, blood pressure, days from body mass index measurement to baseline, number of inpatient hospital days in the year prior to surgery, number of diabetes medications excluding insulin, insulin use, Charlson/Elixhauser comorbidity score, year of procedure, having an ICD-9-CM or SNOMED code for diabetes, smoking, having an ICD-9-CM or SNOMED code for other comorbidities (hypertension, dyslipidemia, sleep apnea, osteoarthritis, nonalcoholic fatty liver disease, gastroesophageal reflux disease, depression, anxiety, eating disorder, substance use, psychosis, kidney disease, infertility, polycystic ovary syndrome, deep-vein thrombosis, or pulmonary embolism), having ICD-9-CM measured factors that impact the surgical effect on diabetes. Despite direct adjustment and the use of propensity scores, confoundingmaypersist.UsingICD-9-CMcodestoassessbase- line health may underestimate comorbidity prevalence. The PBS definitions for T2DM relapse and remission rely on medi- cation-prescribing data. To the extent that prescriptions were not filled, medication use may be overestimated. Some pa- tients may have had T2DM medications ordered outside of the health systems in the study. All dates were normalized to the date of surgery, so within a calendar year, we cannot differ- entiate patients with loss to follow-up from those for whom the study end date had been reached. Future work should ad- dress the potential role of weight loss in mediating diabetes remission and relapse. Discussion across a mix of procedure types and time frames; those ranges are consistent with PBS’s 5-year cumulative relapse rates. The large PBS sample and its comparison of remission and relapseratesacrossprocedures,extendedfollow-up,andevalu- ation of remission across patient subgroups contribute unique insight to the literature. Findings also contribute to ongoing dialogue about leveraging real-world evidence to understand health and improve care.42-44 Such data can reflect generaliz- able populations of patients and clinicians, as well as actual health care practices and settings.44 The data standardiza- tion and curation processes of PCORnet45 help mitigate data quality concerns that have been raised regarding analyses of electronic health record data,42,44 as do the consistency of our findings with prior literature. Our analyses suggest that, coupled with rigorous attention to study design and analytic methods, PCORnet data can be a valuable resource for health research. Similar to prior research,7 19% of the cohort was not pre- scribed diabetes medication preoperatively. Some people may have used lifestyle alone to treat diabetes.46 Undiagnosed dia- betes is common,47 and others may have been diagnosed dur- ing the preoperative evaluation—emphasizing the impor- tanceofcarecoordinationbetweenmedicalandsurgicalhealth professions among patients considering bariatric surgery. Discussion InthissampleofUSadultswithT2DMandbariatricsurgery,56% to 59% of those with RYGB or SG experienced T2DM remission in the year following surgery and 84% to 86% did so within 5 years of follow-up. However, T2DM relapse was common; 33% of patients who had RYGB and 42% of patients who had SG re- lapsed within 5 years of initial remission. The glycemic control of patients who had RYGB and SG showed sustained improve- mentsfromthesamples’baselinemeanHbA1clevelof7.2%,with an estimated mean HbA1c level 0.80 percentage points below baseline for the RYGB group 5 years after surgery vs 0.35 per- centage points below baseline for the SG group. While both groupsexperiencedconsiderableweightloss,patientswhohad RYGB lost more weight and maintained weight loss better than did patients who had SG. A range of T2DM remission rates are found in studies of bariatric surgery,6,7,12,37-41 reflecting varying follow-up time, remission definitions, and population characteristics (eg, in- sulin use, HbA1c level).38 The cumulative remission rates over 80% for SG or RYGB in PBS are consistent with or somewhat higher than estimates from systematic reviews or meta- analyses (54%-78%)4,37,40 and similar to findings (72%; all pro- cedures) from 3 US health systems.6 Literature on T2DM re- lapse is more limited. Published relapse estimates range from approximately 25% to 53%7,12,41 and are typically calculated JAMA Surgery May 2020 Volume 155, Number 5 8/12 jamasurgery.com Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass Original Investigation Research Table 3. Adjusted Hazard Ratios Comparing Time to Remission Since Surgery With Time to Relapse Since Remission for Roux-en-Y Gastric Bypass vs Sleeve Gastrectomy Outcome Total Patients, No. Time Since Bariatric Procedure Adjusted Hazard Ratio (95% CI) P Value 1 y 3 y 5 y No. at Riska Cumulative Eventsb Estimated Cumulative % (95% CI) No. at Risk Cumulative Events Estimated Cumulative % (95% CI) No. Discussion at Risk Cumulative Events Estimated Cumulative % (95% CI) Type 2 diabetes remission Roux-en-Y gastric bypass 5428 1800 2825 59.2 (57.7-60.7) 557 3593 84.3 (82.9-85.5) 215 3620 86.1 (84.7-87.3) 1.10 (1.04-1.16)c .007 Sleeve gastrectomy 2935 917 1519 55.9 (53.9-57.9) 211 1880 81.5 (79.6-83.2) 27 1889 83.5 (81.6-85.1) 1 [Reference] Type 2 diabetes relapsed Roux-en-Y gastric bypass 3352 2273 367 8.4 (7.4-9.3) 1053 665 21.2 (19.1-23.2) 264 772 33.1 (29.6-36.5) 0.75 (0.67-0.84)d <.001 Sleeve gastrectomy 1751 917 199 11.0 (9.6-12.4) 211 369 27.2 (24.1-30.1) 27 400 41.6 (36.8-46.1) 1 [Reference] Abbreviations: ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; NA, not applicable; SNOMED, Systematized Nomenclature of Medicine. a Number of people still being followed up at each point. bNumber of people who had an event in the relevant time frame. c For Roux-en-Y gastric bypass vs sleeve gastrectomy; remission of diabetes was defined as hemoglobin A1c less than 6.5% after 6 months without any prescription order for a diabetes medication; covariates included age, sex, race, Hispanic ethnicity, body mass index (calculated as weight in kilograms divided by height in meters squared), hemoglobin A1c, blood pressure, days from body mass index measurement to baseline, number of inpatient hospital days in the year prior to surgery, number of diabetes medications excluding insulin, insulin use, Charlson/Elixhauser comorbidity score, year of procedure, having an ICD-9-CM or SNOMED code for diabetes, smoking, having an ICD-9-CM or SNOMED code for other comorbidities (hypertension, dyslipidemia, sleep apnea, osteoarthritis, nonalcoholic fatty liver disease, gastroesophageal reflux disease, depression, anxiety, eating disorder, substance use, psychosis, kidney disease, infertility, polycystic ovary syndrome, deep-vein thrombosis, or pulmonary embolism), having ICD-9-CM or SNOMED codes for specific diabetes medications (biguanides, glucagon-like peptide–1 agonists, sulfonylureas, thiazolidinediones, and others), site, and propensity-score deciles. dRelapse of diabetes was defined as occurrence of any hemoglobin A1c level of 6.5% or more and/or prescription order for a diabetes medication. JAMA Surgery May 2020 Volume 155, Number 5 Discussion Covariates included age, sex, race, Hispanic ethnicity, body mass index, hemoglobin A1c level, blood pressure, days from body mass index measurement to baseline, a number of inpatient hospital days in the year prior to surgery, a number of diabetes medications excluding insulin, insulin use, Charlson/Elixhauser comorbidity score, the year of procedure, having an ICD-9-CM or SNOMED code for diabetes, smoking, having an ICD-9-CM or SNOMED code for other comorbidities (hypertension, dyslipidemia, sleep apnea, osteoarthritis, nonalcoholic fatty liver disease, gastroesophageal reflux disease, depression, anxiety, eating disorder, substance use, psychosis, kidney disease, infertility, polycystic ovarian syndrome, deep vein thrombosis, or pulmonary embolism), having ICD-9-CM or SNOMED codes for specific diabetes medications (biguanides, GLP-1 agonists, sulfonylureas, thiazolidinediones, and others), site, and propensity-score deciles. Table 3. Adjusted Hazard Ratios Comparing Time to Remission Since Surgery With Time to Relapse Since Remission for Roux-en-Y Gastric Bypass vs Sleeve Gastrectomy Outcome Total Patients, No. Time Since Bariatric Procedure Adjusted Hazard Ratio (95% CI) P Value 1 y 3 y 5 y No. at Riska Cumulative Eventsb Estimated Cumulative % (95% CI) No. at Risk Cumulative Events Estimated Cumulative % (95% CI) No. at Risk Cumulative Events Estimated Cumulative % (95% CI) Type 2 diabetes remission Roux-en-Y gastric bypass 5428 1800 2825 59.2 (57.7-60.7) 557 3593 84.3 (82.9-85.5) 215 3620 86.1 (84.7-87.3) 1.10 (1.04-1.16)c .007 Sleeve gastrectomy 2935 917 1519 55.9 (53.9-57.9) 211 1880 81.5 (79.6-83.2) 27 1889 83.5 (81.6-85.1) 1 [Reference] Type 2 diabetes relapsed Roux-en-Y gastric bypass 3352 2273 367 8.4 (7.4-9.3) 1053 665 21.2 (19.1-23.2) 264 772 33.1 (29.6-36.5) 0.75 (0.67-0.84)d <.001 Sleeve gastrectomy 1751 917 199 11.0 (9.6-12.4) 211 369 27.2 (24.1-30.1) 27 400 41.6 (36.8-46.1) 1 [Reference] or SNOMED codes for specific diabetes medications (biguanides, glucagon-like peptide–1 agonists, sulfonylureas, thiazolidinediones, and others), site, and propensity-score deciles. or SNOMED codes for specific diabetes medications (biguanides, glucagon-like peptide–1 agonists, sulfonylureas, thiazolidinediones, and others), site, and propensity-score deciles. or SNOMED codes for specific diabetes medications (biguanides, glucagon-like peptide–1 agonists, sulfonylureas, thiazolidinediones, and others), site, and propensity-score deciles. Abbreviations: ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; NA, not applicable; SNOMED, Systematized Nomenclature of Medicine. a Number of people still being followed up at each point. dRelapse of diabetes was defined as occurrence of any hemoglobin A1c level of 6.5% or more and/or prescription order for a diabetes medication. Limitations This study has limitations. Because of the observational study design, procedure choice may have been influenced by un- JAMA Surgery May 2020 Volume 155, Number 5 9/12 jamasurgery.com jamasurgery.com Published Online: March 4, 2020. doi:10.1001/jamasurg.2020.0087 Correction: This article was corrected on March 25, 2020, to fix an error in the name of a healthcare organization. The name was rendered as “the National Patient-Centered Clinical Research Network (PCORnet),” but it should have been “PCORnet, the National Patient-Centered Clinical Research Network.” This occurred once in the Introduction section and once in the Funding/ Support section of the Article Information section. Both have been fixed online. Conflict of Interest Disclosures: Dr Courcoulas reports grants from Covidien/Ethicon Johnson & Johnson, during the conduct of the study. Dr Tavakkoli reports personal fees from Medtronic and AMAG pharmaceuticals. Dr Jones reports personal fees from Allurion. Mr Nadglowski reports other support from the Obesity Action Coalition outside the submitted work. Funding/Support: The PCORnet Study reported in this article was conducted using PCORnet, the National Patient-Centered Clinical Research Network, an initiative funded by the Patient-Centered Outcomes Research Institute (grant OBS-1505-30683). Author Affiliations: Department of Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania (McTigue); Department of Epidemiology, University of Pittsburgh, Pittsburgh, Pennsylvania (McTigue); Kaiser Permanente Washington Health Research Institute, Seattle (Wellman, Anau, Coley, Pardee, Cook, Arterburn); Louisiana Public Health Institute, New Orleans (Nauman); Center for Health Technology, University of California, Davis, Davis (Odor); PaTH Clinical Data Research Network, Pennsylvania State University, Hershey (Tice); Department of Research and Evaluation, Kaiser Permanente Southern California, Pasadena (Coleman); Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania (Courcoulas); Department of Population Medicine, Harvard Medical School, Harvard Pilgrim Health Care Institute, Boston, Massachusetts (Toh, Sturtevant, Horgan); Duke Clinical & Translational Science Institute, Durham, North Carolina (Janning); Mid-South Clinical Data Research Network, Meharry-Vanderbilt Alliance Community Partner, Nashville, Tennessee (Williams); Now with Community Partners Network Inc, Nashville, Tennessee (Williams). Role of the Funder/Sponsor: The funder did not have a role in the study design; in the collection, management, analysis, and interpretation of data; in the preparation, review, or approval of the manuscript; and in the decision to submit the manuscript for publication. PCORnet Bariatric Study Collaborative: Corrigan L. McBride, MD, and James McClay, MD, University of Nebraska Medical Center, Omaha; Jeanne M. Clark, MD, Johns Hopkins University and Health Plan, Baltimore, Maryland; Thomas H. Inge, MD, Children’s Hospital Colorado and University of Colorado, Denver; Michelle R. Lent, PhD, Geisinger Health System, Danville, Pennsylvania; David G. Schlundt, PhD, Vanderbilt University, Nashville, Tennessee; Meredith Duke, MD, University of North Carolina–Chapel Hill; Steven R. Smith, MD, Florida Hospital–Translational Research Institute, Orlando; Andrew O. Downloaded from jamanetwork.com by guest on 10/24/2024 Research Original Investigation Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass lin use, more complex T2DM medications, and/or poor glyce- mic control) may expect larger improvements in T2DM with RYGB compared with SG. On the other hand, for patients with higher likelihood of T2DM remission, RYGB and SG are likely to yieldsimilar5-yearT2DMoutcomes.Forpatients,cliniciansand policy makers to make informed decisions about which proce- dure is best suited to patients’ personal situations, additional data are needed to understand the adverse event profile of the proceduresaswellaspatientvaluesregardingprocedurechoice andtheroleofsurgeryrelativetootheraspectsoflifelongweight management. Conclusions Sturtevant, MS, and Casie Horgan, MPH, Department of Population Medicine, Harvard Pilgrim Health Care Institute and Harvard Medical School, Boston, Massachusetts; Anita Courcoulas, MD, Department of Surgery, University of Pittsburgh, Pittsburgh, Pennsylvania, and Kathleen McTigue, MD, Departments of Medicine and Epidemiology, University of Pittsburgh, Pittsburgh, Pennsylvania; R. Yates Coley, PhD, David Arterburn, MD, Robert Wellman, MS, Jane Anau, BS, Roy E. Pardee, JD, and Andrea J. Cook, PhD, Kaiser Permanente Washington Health Research Institute, Seattle; Karen J. Coleman, PhD, Kaiser Permanente Southern California, Department of Research and Anau, Coley, Tice, Courcoulas, Pardee, Toh, Williams, Cook, Sturtevant, Horgan, Arterburn. Statistical analysis: Wellman, Coley, Toh, Cook. Obtained funding: McTigue, Anau, Arterburn. Administrative, technical, or material support: McTigue, Nauman, Anau, Coleman, Courcoulas, Pardee, Sturtevant, Horgan. Supervision: Coleman, Arterburn. Other—patient perspective: Janning. Conclusions Inconclusion,amongpatientswithT2DMwhounderwentRYGB or SG, most experienced T2DM remission at some point over 5 years of follow-up. While SG and RYGB resulted in similar rates of initial T2DM remission, RYGB was associated with larger and more persistent improvements in glycemic control and 25% lower rates of T2DM relapse compared with SG. Patients with more advanced T2DM at the time of surgery for whom remis- sion is more difficult to achieve (eg, those with older age, insu- Emiliano, MD, and Rhonda G. Kost, MD, The Rockefeller University, New York, New York; Caroline M. Apovian, MD, and Donald Hess, MD, Boston Medical Center, Boston, Massachusetts; Cynthia A. Blalock, APRN, Vanderbilt University Medical Center, Nashville, Tennessee; Elisha Malanga, BS, COPD Foundation, Miami, Florida; Jay R. Desai, MD, HealthPartners Institute, Bloomington, Minnesota; Joe Nadglowski, BS, Obesity Action Coalition, Tampa, Florida; John H. Holmes, PhD, University of Pennsylvania Perelman School of Medicine, Philadelphia; Joseph Vitello, MD, Jesse Brown VA Medical Center, Chicago, Illinois; Michael A. Horberg, MD, Kaiser Permanente Mid-Atlantic Permanente Medical Group, Rockville, Maryland; Robert T. Greenlee, PhD, Marshfield Clinic Research Institute, Marshfield, Wisconsin; Stephanie L. Fitzpatrick, PhD, Kaiser Permanente Center for Health Research, Portland, Oregon; Roni Zeiger, MD, Smart Patients, Inc, Mountain View, California; Molly B. Conroy, MD, University of Utah, Salt Lake City; Douglas S. Bell, MD, David Geffen School of Medicine at UCLA, Los Angeles, California; Jamy Ard, MD, Wake Forest School of Medicine, Salem, North Carolina; Jing Bian, PhD, University of Florida, Gainesville; Bipan Chan, MD, Loyola University Medical Center, Maywood, Illinois; Michael A. Edwards, MD, Temple University, Philadelphia, Pennsylvania; Christina Wee, MD, and Daniel B. Jones, Beth Israel Deaconess Medical Center, Boston, Massachusetts; Jennifer L. Kraschnewski, MD, Penn State University, College of Medicine, Hershey, Pennsylvania; Kirk Reichard, MD, Nemours AI DuPont Hospital for Children, Wilmington, Delaware; Howard S. Gordon, MD, and David O. Meltzer MD, University of Illinois, Chicago; Erin D. Roe, MD, Baylor Scott & White, Dallas, Texas; William Richardson, MD, Ochsner Clinic, New Orleans, Louisiana; Sameer Malhotra, MD, Weill Cornell Medicine, New York, New York; Lindsay G. Cowell, PhD, University of Texas Southwestern Medical Center, Dallas; Lydia A. Bazzano, MD, PhD, Tulane University, New Orleans, Louisiana; Jefferey S. Brown, Sengwee Toh, ScD, Jessica L. ARTICLE INFORMATION Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 McTigue KM et al. JAMA Surgery. Accepted for Publication: January 15, 2020. Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 McTigue KM et al. JAMA Surgery. ARTICLE INFORMATION Anau, Coley, Tice, Courcoulas, Pardee, Toh, Williams, Cook, Sturtevant, Horgan, Arterburn. Statistical analysis: Wellman, Coley, Toh, Cook. Obtained funding: McTigue, Anau, Arterburn. Administrative, technical, or material support: McTigue, Nauman, Anau, Coleman, Courcoulas, Pardee, Sturtevant, Horgan. Supervision: Coleman, Arterburn. Other—patient perspective: Janning. Anau, Coley, Tice, Courcoulas, Pardee, Toh, Williams, Cook, Sturtevant, Horgan, Arterburn. Statistical analysis: Wellman, Coley, Toh, Cook. Obtained funding: McTigue, Anau, Arterburn. Administrative, technical, or material support: McTigue, Nauman, Anau, Coleman, Courcoulas, Pardee, Sturtevant, Horgan. Supervision: Coleman, Arterburn. Other—patient perspective: Janning. Emiliano, MD, and Rhonda G. Kost, MD, The Rockefeller University, New York, New York; Caroline M. Apovian, MD, and Donald Hess, MD, Boston Medical Center, Boston, Massachusetts; Cynthia A. Blalock, APRN, Vanderbilt University Medical Center, Nashville, Tennessee; Elisha Malanga, BS, COPD Foundation, Miami, Florida; Jay R. Desai, MD, HealthPartners Institute, Bloomington, Minnesota; Joe Nadglowski, BS, Obesity Action Coalition, Tampa, Florida; John H. Holmes, PhD, University of Pennsylvania Perelman School of Medicine, Philadelphia; Joseph Vitello, MD, Jesse Brown VA Medical Center, Chicago, Illinois; Michael A. Horberg, MD, Kaiser Permanente Mid-Atlantic Permanente Medical Group, Rockville, Maryland; Robert T. Greenlee, PhD, Marshfield Clinic Research Institute, Marshfield, Wisconsin; Stephanie L. Fitzpatrick, PhD, Kaiser Permanente Center for Health Research, Portland, Oregon; Roni Zeiger, MD, Smart Patients, Inc, Mountain View, California; Molly B. Conroy, MD, University of Utah, Salt Lake City; Douglas S. Bell, MD, David Geffen School of Medicine at UCLA, Los Angeles, California; Jamy Ard, MD, Wake Forest School of Medicine, Salem, North Carolina; Jing Bian, PhD, University of Florida, Gainesville; Bipan Chan, MD, Loyola University Medical Center, Maywood, Illinois; Michael A. Edwards, MD, Temple University, Philadelphia, Pennsylvania; Christina Wee, MD, and Daniel B. Jones, Beth Israel Deaconess Medical Center, Boston, Massachusetts; Jennifer L. Kraschnewski, MD, Penn State University, College of Medicine, Hershey, Pennsylvania; Kirk Reichard, MD, Nemours AI DuPont Hospital for Children, Wilmington, Delaware; Howard S. Gordon, MD, and David O. Meltzer MD, University of Illinois, Chicago; Erin D. Roe, MD, Baylor Scott & White, Dallas, Texas; William Richardson, MD, Ochsner Clinic, New Orleans, Louisiana; Sameer Malhotra, MD, Weill Cornell Medicine, New York, New York; Lindsay G. Cowell, PhD, University of Texas Accepted for Publication: January 15, 2020. Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 McTigue KM et al. JAMA Surgery. Published Online: March 4, 2020. doi:10.1001/jamasurg.2020.0087 Accepted for Publication: January 15, 2020. Emiliano, MD, and Rhonda G. Kost, MD, The Rockefeller University, New York, New York; Caroline M. Apovian, MD, and Donald Hess, MD, Boston Medical Center, Boston, Massachusetts; Cynthia A. Blalock, APRN, Vanderbilt University Medical Center, Nashville, Tennessee; Elisha Malanga, BS, COPD Foundation, Miami, Florida; Jay R. Desai, MD, HealthPartners Institute, REFERENCES Johns Hopkins, Rachel Hess, Meghan Lynch, and Reid Holbrook, University of Utah, Jody McCullough, Matt Bolton, Wenke Hwang, Ann Rogers, Alison Bower, and Cynthia Chuang, Penn State, Cecilia Dobi, Mark Weiner, Anuradha Paranjape, Sharon J. Herring, and Patricia Bernard, Temple University, Janet Zahner, Parth Divekar, Keith Marsolo, and Lisa Boerger, Cincinnati Children’s Hospital, Kimberly J. Holmquist, Kaiser Permanente Southern California, Ray Pablo and Robynn Zender, University of California at Irvine, Lucila Ohno-Machado, Paulina Paul, and Michele Day, University of California at San Diego, Thomas Carton, Elizabeth Crull, and Iben McCormick-Ricket, Louisiana Public Health Institute, Ashley Vernon, Malcom Robinson, Scott Shikora, David Spector, Eric Sheu, Edward Mun, Matthew Hutter, Shawn Murphy, Jeffrey Klann, and Denise Gee, Partners Healthcare, Daniel Jones, Benjamin Schneider, Griffin Weber, and Robert Andrews, Beth Israel Deaconess Medical Center, Brian Carmine, Miguel Burch, and Galina Lozinski, Boston Medical Center, Ken Mandl, Jessica Lyons, 2. Khorgami Z, Shoar S, Saber AA, Howard CA, Danaei G, Sclabas GM. Outcomes of bariatric surgery versus medical management for type 2 diabetes mellitus: a meta-analysis of randomized controlled trials. Obes Surg. 2019;29(3):964-974. doi:10.1007/s11695-018-3552-x 3. Jakobsen GS, Småstuen MC, Sandbu R, et al. Association of bariatric surgery vs medical obesity treatment with long-term medical complications and obesity-related comorbidities. JAMA. 2018;319 (3):291-301. doi:10.1001/jama.2017.21055 16. Lee WJ, Chong K, Ser KH, et al. Gastric bypass vs sleeve gastrectomy for type 2 diabetes mellitus: a randomized controlled trial. Arch Surg. 2011;146 (2):143-148. doi:10.1001/archsurg.2010.326 17. Li P, Fu P, Chen J, Wang LH, Wang DR. Laparoscopic Roux-en-Y gastric bypass vs. laparoscopic sleeve gastrectomy for morbid obesity and diabetes mellitus: a meta-analysis of sixteen recent studies. Hepatogastroenterology. 2013;60 (121):132-137. 4. Buchwald H, Estok R, Fahrbach K, et al. Weight and type 2 diabetes after bariatric surgery: systematic review and meta-analysis. Am J Med. 2009;122(3):248-256.e5. doi:10.1016/j.amjmed. 2008.09.041 18. Celio AC, Wu Q, Kasten KR, Manwaring ML, Pories WJ, Spaniolas K. Comparative effectiveness of Roux-en-Y gastric bypass and sleeve gastrectomy in super obese patients. Surg Endosc. 2017;31(1): 317-323. doi:10.1007/s00464-016-4974-y 5. Sjöström L, Lindroos AK, Peltonen M, et al; Swedish Obese Subjects Study Scientific Group. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med. 2004; 351(26):2683-2693. doi:10.1056/NEJMoa035622 19. Peterli R, Wölnerhanssen BK, Peters T, et al. Effect of laparoscopic sleeve gastrectomy vs laparoscopic Roux-en-Y gastric bypass on weight loss in patients with morbid obesity: the SM-BOSS randomized clinical trial. JAMA. 2018;319(3):255-265. doi:10.1001/jama.2017.20897 6. REFERENCES 14. Abbatini F, Rizzello M, Casella G, et al. Long-term effects of laparoscopic sleeve gastrectomy, gastric bypass, and adjustable gastric banding on type 2 diabetes. Surg Endosc. 2010;24 (5):1005-1010. doi:10.1007/s00464-009-0715-9 Disclaimer: The views expressed in this article are solely those of the authors and do not reflect the views of PCORnet or PCORI. Dr McTigue attests that all listed authors meet authorship criteria and nobody meeting authorship criteria has been omitted. 1. Schauer PR, Bhatt DL, Kirwan JP, et al; STAMPEDE Investigators. Bariatric surgery versus intensive medical therapy for diabetes—5-year outcomes. N Engl J Med. 2017;376(7):641-651. doi: 10.1056/NEJMoa1600869 15. Peterli R, Wölnerhanssen B, Peters T, et al. Improvement in glucose metabolism after bariatric surgery: comparison of laparoscopic Roux-en-Y gastric bypass and laparoscopic sleeve gastrectomy: a prospective randomized trial. Ann Surg. 2009;250(2):234-241. doi:10.1097/SLA. 0b013e3181ae32e3 Additional Contributions: The study team also wishes to acknowledge the clinicians, analysts, and staff at the 34 health systems which contributed to the study: Stephen R. Perry, Kin Lam, David Hawkes, Thomas Dundon, and Kelli Kinsman, Kaiser Permanente Washington Health Research Institute, Shelly Sital, The Chicago Community Trust, Elizabeth Tarlov, University of Illinois at Chicago, Jasmin Phua, Medical Research Analytics and Informatics Alliance, Mia Gallagher, Lindsey Petro, Beth Syat, Harvard Pilgrim Health Care Institute and Harvard Medical School, Prakash Nadkarni, and Elizabeth Chrischilles, University of Iowa, Steffani Roush, and Laurel Verhagen, Marshfield Clinic Research Institute, Umberto Tachincardi, and Lawrence P. Hanrahan, University of Wisconsin, Phillip Reeder, Shiby Antony, Rania AlShahrouri, University of Texas–Southwestern Medical Center, Bret Gardner, James Campbell, Russell Buzalko, and Jay Pedersen, University of Nebraska Medical Center, Dan Connolly, and Russel Waitman, University of Kansas Medical Center, Russel Rothman, David Crenshaw, and Katie Worley, Vanderbilt University Medical Center, Emily Pfaff, Robert Bradford, Kellie Walters, Tim Carey, Timothy Farrell, and D. Wayne Overby, University of North Carolina, Maija Neville-Williams, The Rockefeller University, Elizabeth Shenkman, William Hogan, Kathryn McAuliffe, and Gigi Lipori, University of Florida, Rebecca Zuvich Essner, Florida Hospital, Howard Su, Michael George, Michael J. Becich, Barbara Postol, Giselle G. Hamad, Ramesh C. Ramanathan, Bestoun H. Ahmed, William F. Gourash, Bill Shirey, Chuck Borromeo, John Milnes, Nickie Cappella, and Desheng Li, University of Pittsburgh, Anthony T. Petrick, H. Lester Kirchner, Geisinger Health System, Daniel E. Ford, Michael A. Schweitzer, Karl Burke, Harold Lehmann, Megan E. Gauvey-Kern, and Diana Gumas. Published Online: March 4, 2020. doi:10.1001/jamasurg.2020.0087 Odegaard, PhD, University of California, Irvine; Nirav K. Desai, MD, Boston Children’s Hospital, Boston, Massachusetts; Ali Tavakkoli, MD, and Elizabeth Cirelli, MS, Brigham and Women’s Hospital, Boston, Massachusetts; Stavra A. Xanthakos, MD, Cincinnati Children's Medical Center, Cincinnati, Ohio; Laura J. Rasmussen-Torvik, PhD, Northwestern University Feinberg School of Medicine, Chicago, Illinois; Marc P. Michalsky, MD, Nationwide Children’s Hospital, Columbus, Ohio; Matthew F. Daley, MD, Institute for Health Research, Kaiser Permanente Colorado, Aurora; Gabrielle Purcell, MPH. University of California; San Francisco; Sameer Murali, MD, Southern California Permanente Medical Group, Fontana; Ana Author Contributions: Drs McTigue and Arterburn 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. Concept and design: McTigue, Wellman, Coley, Toh, Janning, Williams, Arterburn. Acquisition, analysis, or interpretation of data: McTigue, Wellman, Nauman, Anau, Coley, Tice, Coleman, Courcoulas, Pardee, Toh, Cook, Sturtevant, Horgan, Arterburn. Drafting of the manuscript: McTigue, Wellman, Anau, Coley, Coleman, Janning, Arterburn. Critical revision of the manuscript for important intellectual content: McTigue, Wellman, Nauman, 10/12 JAMA Surgery May 2020 Volume 155, Number 5 10/12 JAMA Surgery May 2020 Volume 155, Number 5 10/12 JAMA Surgery May 2020 Volume 155, Number 5 jamasurgery.com Downloaded from jamanetwork.com by guest on 10/24/2024 Original Investigation Research Original Investigation Research 13. Jiménez A, Casamitjana R, Flores L, et al. Long-term effects of sleeve gastrectomy and Roux-en-Y gastric bypass surgery on type 2 diabetes mellitus in morbidly obese subjects. Ann Surg. 2012;256(6):1023-1029. doi:10.1097/SLA. 0b013e318262ee6b 13. Jiménez A, Casamitjana R, Flores L, et al. Long-term effects of sleeve gastrectomy and Roux-en-Y gastric bypass surgery on type 2 diabetes mellitus in morbidly obese subjects. Ann Surg. 2012;256(6):1023-1029. doi:10.1097/SLA. 0b013e318262ee6b Evaluation, Pasadena; Cheri D. Janning, MS, Duke Clinical & Translational Science Institute, Durham, North Carolina; Neely Williams, MDiv, Mid-South Clinical Data Research Network and Meharry– Vanderbilt Alliance Community Partner, Nashville, Tennessee. Evaluation, Pasadena; Cheri D. Janning, MS, Duke Clinical & Translational Science Institute, Durham, North Carolina; Neely Williams, MDiv, Mid-South Clinical Data Research Network and Meharry– Vanderbilt Alliance Community Partner, Nashville, Tennessee. and Margaret Vella, Harvard Medical School, and Joseph Skelton and Kun Wei, Wake Forest Integrated Health System. Some of these individuals were compensated for their contributions. REFERENCES Arterburn D, Bogart A, Coleman KJ, et al. Comparative effectiveness of bariatric surgery vs. nonsurgical treatment of type 2 diabetes among severely obese adults. Obes Res Clin Pract. 2013;7 (4):e258-e268. doi:10.1016/j.orcp.2012.08.196 7. Arterburn DE, Bogart A, Sherwood NE, et al. A multisite study of long-term remission and relapse of type 2 diabetes mellitus following gastric bypass. Obes Surg. 2013;23(1):93-102. doi:10. 1007/s11695-012-0802-1 20. Salminen P, Helmiö M, Ovaska J, et al. Effect of laparoscopic sleeve gastrectomy vs laparoscopic Roux-en-Y gastric bypass on weight loss at 5 years among patients with morbid obesity: the SLEEVEPASS randomized clinical trial. JAMA. 2018; 319(3):241-254. doi:10.1001/jama.2017.20313 8. Coleman KJ, Haneuse S, Johnson E, et al. Long-term microvascular disease outcomes in patients with type 2 diabetes after bariatric surgery: evidence for the legacy effect of surgery. Diabetes Care. 2016;39(8):1400-1407. doi:10.2337/dc16-0194 21. Li J, Lai D, Wu D. Laparoscopic Roux-en-Y gastric bypass versus laparoscopic sleeve gastrectomy to treat morbid obesity-related comorbidities: a systematic review and meta-analysis. Obes Surg. 2016;26(2):429-442. doi:10.1007/s11695-015-1996-9 Geisinger Health System, Daniel E. Ford, Michael A. Schweitzer, Karl Burke, Harold Lehmann, Megan E. 9. Arterburn D, Gupta A. Comparing the outcomes of sleeve gastrectomy and roux-en-y gastric bypass for severe obesity. JAMA. 2018;319(3):235-237. doi:10.1001/jama.2017.20449 22. Aminian A, Brethauer SA, Andalib A, et al. Individualized metabolic surgery score: procedure selection based on diabetes severity. Ann Surg. 2017;266(4):650-657. doi:10.1097/SLA. 0000000000002407 10. Adams TD, Arterburn DE, Nathan DM, Eckel RH. Clinical Outcomes of metabolic surgery: microvascular and macrovascular complications. Diabetes Care. 2016;39(6):912-923. doi:10.2337/ dc16-0157 23. Reames BN, Finks JF, Bacal D, Carlin AM, Dimick JB. Changes in bariatric surgery procedure use in Michigan, 2006-2013. JAMA. 2014;312(9):959-961. doi:10.1001/jama.2014.7651 11. Billeter AT, Scheurlen KM, Probst P, et al. Meta-analysis of metabolic surgery versus medical treatment for microvascular complications in patients with type 2 diabetes mellitus. Br J Surg. 2018;105(3):168-181. doi:10.1002/bjs.10724 24. Arterburn DE, Courcoulas AP. Bariatric surgery for obesity and metabolic conditions in adults. BMJ. 2014;349:g3961. doi:10.1136/bmj.g3961 25. Toh S, Rasmussen-Torvik LJ, Harmata EE, et al; PCORnet Bariatric Surgery Collaborative. The National Patient-Centered Clinical Research Network (PCORnet) Bariatric Study cohort: rationale, methods, and baseline characteristics. JMIR Res Protoc. 2017;6(12):e222. doi:10.2196/ resprot.8323 12. Mingrone G, Panunzi S, De Gaetano A, et al. Bariatric-metabolic surgery versus conventional medical treatment in obese patients with type 2 diabetes: 5 year follow-up of an open-label, single-centre, randomised controlled trial. Lancet. 2015;386(9997):964-973. REFERENCES doi:10.1016/S0140-6736 (15)00075-6 JAMA Surgery May 2020 Volume 155, Number 5 jamasurgery.com jamasurgery.com Downloaded from jamanetwork.com by guest on 10/24/2024 Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass 26. Arterburn D, Wellman R, Emiliano A, et al; PCORnet Bariatric Study Collaborative. Comparative effectiveness and safety of bariatric procedures for weight loss: a PCORnet cohort study. Ann Intern Med. 2018;169(11):741-750. doi:10. 7326/M17-2786 34. Still CD, Wood GC, Benotti P, et al. Preoperative prediction of type 2 diabetes remission after Roux-en-Y gastric bypass surgery: a retrospective cohort study. Lancet Diabetes Endocrinol. 2014;2(1): 38-45. doi:10.1016/S2213-8587(13)70070-6 remission of diabetes after Roux-en-Y gastric bypass. Surg Obes Relat Dis. 2010;6(3):254-259. doi:10.1016/j.soard.2009.11.003 42. Corrigan-Curay J, Sacks L, Woodcock J. Real-world evidence and real-world data for evaluating drug safety and effectiveness. JAMA. 2018;320(9):867-868. doi:10.1001/jama.2018.10136 35. American Diabetes Association. 6. Glycemic targets: standards of medical care in diabetes—2019. Diabetes Care. 2019;42(suppl 1): S61-S70. doi:10.2337/dc19-S006 27. Collins FS, Hudson KL, Briggs JP, Lauer MS. PCORnet: turning a dream into reality. J Am Med Inform Assoc. 2014;21(4):576-577. doi:10.1136/ amiajnl-2014-002864 43. Jarow JP, LaVange L, Woodcock J. Multidimensional evidence generation and FDA regulatory decision making: defining and using “real-world” data. JAMA. 2017;318(8):703-704. doi: 10.1001/jama.2017.9991 36. Purnell JQ, Selzer F, Wahed AS, et al. Type 2 diabetes remission rates after laparoscopic gastric bypass and gastric banding: results of the Longitudinal Assessment of Bariatric Surgery study. Diabetes Care. 2016;39(7):1101-1107. doi:10.2337/ dc15-2138 28. Fleurence RL, Curtis LH, Califf RM, Platt R, Selby JV, Brown JS. Launching PCORnet, a national patient-centered clinical research network. J Am Med Inform Assoc. 2014;21(4):578-582. doi:10.1136/ amiajnl-2014-002747 44. Sherman RE, Anderson SA, Dal Pan GJ, et al. Real-world evidence—what is it and what can it tell us? N Engl J Med. 2016;375(23):2293-2297. doi: 10.1056/NEJMsb1609216 29. Buse JB, Caprio S, Cefalu WT, et al. How do we define cure of diabetes? Diabetes Care. 2009;32 (11):2133-2135. doi:10.2337/dc09-9036 37. Parikh M, Issa R, Vieira D, et al. Role of bariatric surgery as treatment for type 2 diabetes in patients who do not meet current NIH criteria: a systematic review and meta-analysis. J Am Coll Surg. 2013; 217(3):527-532. doi:10.1016/j.jamcollsurg.2013.04. 023 45. Qualls LG, Phillips TA, Hammill BG, et al. Evaluating foundational data quality in the National Patient-Centered Clinical Research Network (PCORnet®). EGEMS (Wash DC). 2018;6(1):3. doi:10. 5334/egems.199 30. American Diabetes Association. 2. Classification and diagnosis of diabetes: standards of medical care in diabetes—2019. Diabetes Care. 2019;42(suppl 1):S13-S28. doi:10.2337/dc19-S002 38. Dixon JB, le Roux CW, Rubino F, Zimmet P. Bariatric surgery for type 2 diabetes. Lancet. 2012; 379(9833):2300-2311. doi:10.1016/S0140-6736(12) 60401-2 46. Hippisley-Cox J, Pringle M. jamasurgery.com 12/12 JAMA Surgery May 2020 Volume 155, Number 5 Comparing the 5-Year Diabetes Outcomes of Sleeve Gastrectomy and Gastric Bypass Prevalence, care, and outcomes for patients with diet-controlled diabetes in general practice: cross sectional survey. Lancet. 2004;364(9432):423-428. doi:10.1016/ S0140-6736(04)16765-2 31. Gagne JJ, Glynn RJ, Avorn J, Levin R, Schneeweiss S. A combined comorbidity score predicted mortality in elderly patients better than existing scores. J Clin Epidemiol. 2011;64(7):749-759. doi:10.1016/j.jclinepi.2010.10.004 39. Courcoulas AP, Goodpaster BH, Eagleton JK, et al. Surgical vs medical treatments for type 2 diabetes mellitus: a randomized clinical trial. JAMA Surg. 2014;149(7):707-715. doi:10.1001/jamasurg. 2014.467 47. National Center for Health Statistics. Health United States, 2017: with special feature on mortality. https://www.cdc.gov/nchs/data/hus/ hus17.pdf. Published 2018. Accessed January 29, 2020. 32. Friedman J, Hastie T, Tibshirani R. Regularization paths for generalized linear models via coordinate descent. J Stat Softw. 2010;33(1):1-22. doi:10.18637/jss.v033.i01 40. Puzziferri N, Roshek TB III, Mayo HG, Gallagher R, Belle SH, Livingston EH. Long-term follow-up after bariatric surgery: a systematic review. JAMA. 2014;312(9):934-942. doi:10.1001/jama.2014.10706 33. Kent DM, Hayward RA. Limitations of applying summary results of clinical trials to individual patients: the need for risk stratification. JAMA. 2007;298(10):1209-1212. doi:10.1001/jama.298.10. 1209 41. Chikunguwo SM, Wolfe LG, Dodson P, et al. Analysis of factors associated with durable 12/12 JAMA Surgery May 2020 Volume 155, Number 5 12/12 JAMA Surgery May 2020 Volume 155, Number 5 jamasurgery.com Downloaded from jamanetwork.com by guest on 10/24/2024
https://openalex.org/W4220673375	https://www.researchsquare.com/article/rs-1370571/latest.pdf	English	null	Application of Machine Learning for Inter Turn Fault Detection in Pumping System	Research Square (Research Square)	2,022	cc-by	10,345	Application of Machine Learning for Inter Turn Fault Detection in Pumping System Nabanita Dutta Vellore Institute of Technology Palanisamy Kaliannan (  kpalanisamy@vit.ac.in ) Vellore Institute of Technology Nabanita Dutta Vellore Institute of Technology Palanisamy Kaliannan (  kpalanisamy@vit.ac.in ) Vellore Institute of Technology Paramasivam Shanmugam Esab India limited Introduction ` The induction motor is commonly used device which is indispensable for various industries and have received augmented attention for their robust construction, high performance, reliability and maintenance cost [1]. Any kind of fault in induction motor causes drastic consequences in the devices connected with the motor and entire system. If pump is connected with faulty induction motor the head value will change, flow rate will change and huge vibration will create severe damage [2]. Breakdown of entire system not only causes system damage but also creates huge energy loss and, unplanned sudden downtime causes and huge maintenance cost. It is reported that 30% to 40% failure is seen in induction motor for stator inter turn fault [3]. This is actually electrical fault and this electrical fault is very sensitive which causes severe damage. Only 10% to 20% inter turn fault causes huge rising of current in induction motor which causes insulation losses in the windings [4]. Electrical faults are categorized both as stator and rotor fault [5]. The rotor fault which is seen in induction motor is broken rotor bar. Stator faults are mainly three faults like phase to phase fault, inter turn fault and phase to ground fault. Among those inter turn fault is major and critical fault [6]. This inter turn fault not only hampers induction motor operation but also pumping operation. Besides mechanical and hydraulic fault, electrical fault also hampers the pump performance. The centrifugal pump is rotating machine used to transfer fluid through pipes. Sudden shutdown of pumping system causes huge loss for maintenance [7]. It has been analysed that 70% of the maintenance cost is seen for the pumping system. So it is required to improve the maintenance technology for reducing cost. Various researches have been done for inter turn fault detection in induction motor. Voltage between lines, neutral and star point of motor were used for fault detection. This was used as model of the motor and an imbalance was created due to inter turn short circuit fault. Before the total breakdown and major damage of the devices this imbalance should be identified. [8]. Negative sequence impedance was estimated and used as fault indicator in a research. The negative sequence impendence was seen due to imbalance in the motor. Oscillation used for Park transformation current was used for fault detection which were created for imbalance. ABSTRACT Pump fault diagnosis is essential for maintenance and safety of the device as it is an important appliance used in various major sectors. Fault diagnosis at proper time can reduce maintenance cost and save energy. In this article a Simulink model based on mathematical equations has been built for analyzing the effects of parameter estimation of three phase induction motor based centrifugal pump in inter turn fault condition. The inter turn fault causes huge increase of current which severely affects the parameters of both motor and pump and these have been analysed by simulation through Matlab Simulink model. Later, the results are verified by hardware in loop (HIL) based simulator. In this paper machine learning (ML) based artificial neural network (ANN) and ANFIS (ANN and Fuzzy) models have been applied for fault detection ANN and ANFIS based models provide a satisfactory level of accuracy. These models provide accurate training and testing results. Based on root mean square error (RMSE), R2, prediction accuracy and mean validation value these models are compared to find out which is more suitable for this experiment. Various supervised algorithms are compared with ANN, ANFIS and lastly found which is the most suitable for this experiment. ords: Induction motor, Pump, Inter turn fault, ANN, ANFIS, Machine learning Keywords: Induction motor, Pump, Inter turn fault, ANN, ANFIS, Machine learning Application of machine learning for inter turn fault detection in pumping system Nabanita Dutta1+, Palanisamy Kaliannan1+, Paramasivam Shanmugam2, + 1Department of Energy and Power Electronics, School of Electrical Engineering, Vellore Institute of Technology, Vellore 632014, India 2Esab India limited, Chennai, India kpalanisamy@vit.ac.in +these authors contributed equally to this work 1Department of Energy and Power Electronics, School of Electrical Engineering, Vell 632014, India 2Esab India limited, Chennai, India * kpalanisamy@vit.ac.in +these authors contributed equally to this work 1Department of Energy and Power Electronics, School of Electrical Engineering, Vellore Institute of Technology, Vellore 632014, India 2Esab India limited, Chennai, India * kpalanisamy@vit.ac.in +these authors contributed equally to this work Research Article Research Article Keywords: Induction motor, Pump, Inter turn fault, ANN, ANFIS, Machine learning Posted Date: March 8th, 2022 DOI: https://doi.org/10.21203/rs.3.rs-1370571/v1 License:   This work is licensed under a Creative Commons Attribution 4.0 International License. Read Full License Introduction To identify this problem space vector analysis is required [9]. Electrical fault can be detected by motor current signature analysis (MCSA) and vibration analysis. For the estimation of negative impedance in the motor, robustness with respect to unbalanced voltage supply was added as an approach [10]. Frequency spectrum and fast Fourier transformation (FFT) analysis also are helpful for fault detection in induction motor. Wavelet package transformations (WPT) and FFT were used along with some sort of classifier in some work [11] [12]. About faults in the stator for extracting knowledge higher order statistics (HOS) was used. For rotating machine, pump health monitoring is possible for analyzing supply voltage, current, and power, torque and speed value. Faults in the machine can be analysed and identified by harmonic analysis also. In this case stator current produces instantaneous harmonic current [13]. Parameter estimation technique also performs the pump health monitoring process. Harmonic frequencies can analyse machine health. In hazardous and extreme environmental conditions where access of the machine is difficult such as nuclear power plant, , paper mill, chemical plant and on shore and off shore facilities, this method is applied for health monitoring of the machines [14] [15]. MCSA is an electric machinery monitoring technique which has wide application mainly in heavy industry. It can be used for detecting motor faults. Still, the limitation of this method is it will perform better if it is used in conjunction with other technologies like fast Fourier transformation (FFT), Fuzzy logic, and Park's vector analysis. The latest development in artificial intelligence (AI) is 'transfer learning', which can detect failure patterns of different devices. This technique can be used instead of MCSA and able to find out localized anomalies and has been suggested in place of MCSA for diagnosing bearing faults, impeller breaking faults in an induction motor[16]. This proposed article presents stator inter turn fault of three phase induction motor based pumping system and specially inter turn fault has been described here as it is a major and vital problem in most of the industrial sectors. Reducing operational and maintenance cost is the main aim of most of the Industries, which leads to increase in importance of condition monitoring of the machine. Requirement of additional equipment and maintenance cost are the burden for industries due to sudden failure of machines. Introduction Health monitoring of the winding of the machine is very important to avoid the losses as inter turn fault causes damage of the windings. In this research article overall stator winding fault has been discussed in next sections and how the parameters of both motor and pump are affected by stator fault that is also described. As continuous monitoring and prediction of fault detection are possible through ML based algorithms in best suited way, ML algorithm based ANN and ANFIS models are applied for detecting inter turn stator fault in induction motor based pumping system. Both of these models are used in the proposed research and performances of both algorithms have been compared. Moreover various supervised algorithms are compared with ANN and ANFIS to find out the better result. Proposed Model The proposed research has been done based on some mathematical equations of pumping system which is coupled with induction motor. These equations are applied for building the matlab simulink model of induction motor based pumping system. As fault creation in real time system causes major damage in the system, so mathematical equation based simulink model has been developed for the experimental purpose for analyzing the healthy and faulty situation. At first the model has been analysed in healthy condition without changing any parameter value. Then to analyse the faulty situation phase A winding has been sorted such that the current in phase A suddenly increases and it helps to increase the phase B and phase C current also. Due to changes of current value, torque and speed value also change and the pump parameters like pressure, flow rate are forced to change as pump is coupled with induction motor and runs in same speedup to three levels of subheading are permitted. Subheadings should not be numbered. Mathematical model of induction motor based pumping system Subsection In this section various equations have been described which express voltage, current, and flux of stator and rotor of induction motor. S S S S V R I P   (1) 0 r r S R I P   (2) Where 1 2 [ ]T S as as bs cs V V V V V  (3) [ ( ) ]T S as as f bs cs I I I I I I   (4) [ ]T r ar br cr I I I I  (5)   1 2 T S as as bs cs       (6) S S S S V R I P   (1) 0 r r S R I P   (2) (5) (6) V represents the voltage, I shows current, flux is represented as , here, “s” and “r” represent stator and rotor respectively, a, b, c denote the three phase system. as1, as2 denote unfaulty and faulty part of stator respectively. Here P is Laplace operator, derivative operator d dt is replaced by P. V represents the voltage, I shows current, flux is represented as , here, “s” and “r” represent stator and rotor respectively, a, b, c denote the three phase system. as1, as2 denote unfaulty and faulty part of stator respectively. Here P is Laplace operator, derivative operator d dt is replaced by P. (7) (7) (8) (9) (8) (9) (8) (9) (9) (9) Shorted part of stator winding voltage is shown by these equations. β denotes shorted turn Shorted part of stator winding voltage is shown by these equations. β denotes shorted turn. (11) (11) (11) (11) (11) Here mutual inductance and self-inductance of stator winding have been represented by the equations [12-14]. (12) (12) (13) (14) (13) (14) (13) (14) β represents number of turn in phase a, represents rotor position, Ls shows self-inductance, Lr shows rotor self-inductance and Lsr represents stator to rotor mutual inductance. Water is pumped out from constant level water tank and the pumping system consists of water tank, asynchronous three phase induction motor and other parts. The tank receives liquid with input flow Output flow of control valve is represented by . With the help of fluid mechanics and fundamental laws of physics, analysis of the dynamics of plant has been done and a mathematical model is developed [17]. Mathematical model of induction motor based pumping system Subsection This mathematical model includes the mathematical models of centrifugal pump and tank. The counterpart of Newton’s law of force is that angular acceleration is proportional to the torque on the axis. So, the equations show the motion for the motor and pump set. J shows the moment of inertia. Here moment of inertia is constant of proportionality in specific case. Active torque of asynchronous motor is shown by , passive or resistive torque of pump is shown by and it is assumed that stator pole pair number is one. Here following equation shows the torque of the asynchronous motor. J shows the moment of inertia. Here moment of inertia is constant of proportionality in specific case. Active torque of asynchronous motor is shown by , passive or resistive torque of pump is shown by and it is assumed that stator pole pair number is one. Here following equation shows the torque of the asynchronous motor. Viscous torque and passive torque can be represented by Viscous torque and passive torque can be represented by Viscous torque and passive torque can be represented by Viscous torque and passive torque can be represented by M k    (17) M k    (17) 2 v p p gQ H M    (18) M k    (17) 2 v p p gQ H M    (18) (18) Equation 18 shows the basic parameters of the centrifugal pump and pump flow rate is shown by Q, H shows the pump head and the angular velocity is shown by ω. Peripheral cross section of the impeller channels and meridian component of velocity express the pump flow. Head value is proportional to angular velocity as flow rate is proportional to angular velocity [19]. In the last equation the pump efficiency coefficient is denoted by p which is constant and in different modes it changes to some extent and this reflects to the other parameters. The total operating system Total H can be defined as Here static head is represented by , dynamic head is shown by , pressure in the surface of the water in the receiving tank is shown by and the pressure in the surface of the water in the reservoir tank is represented by [20]. Mathematical model of induction motor based pumping system Subsection is associated with length of pipe ,friction and the diameter of the pipe. Here f shows the friction factor, L shows the pipe length and D is the pipe diameter. By modified version of Colebrook White equation the friction coefficient f can be found. Here roughness factor is k and Reynolds number is Re. The roughness factor k is a standard fixed value collected from standard tables and it depends on the material of the pipe and pipe condition. For any flow in pipe, the following formula is used for the calculation of Reynolds number [21]: is the kinematic viscosity. Operation of the pumping system is based on affinity law. First affinity law is shown in equation where flow Q is proportional to shaft speed N. As per, second affinity law, head is proportional to the square of the shaft speed. As per, second affinity law, head is proportional to the square of the shaft speed. The power of the pump can be calculated as The power of the pump can be calculated as Here P is the power requirement for pump, H is the head, is the acceleration gravity, is the density of water. Here P is the power requirement for pump, H is the head, is the acceleration gravity, is the density of water. Mathematical model of induction motor based pumping system Subsection Here static head is represented by , dynamic head is shown by , pressure in the surface of the water in the receiving tank is shown by and the pressure in the surface of the water in the reservoir tank is represented by [20]. Based on pump height pressure changes and it is considered negligible value.. But atmospheric pressure changes with the height. Equation shows the change in pressure and elevation difference between the reservoir and receiving tank. But this is not so significant and considered as negligible. (21) (21) (21) The difference between the point of discharge and the surface of the reservoir into the receiving tank is static head which is shown by . Between maximum and minimum head value, static head for the system will vary because water level of the reservoir also varies. Here top water level is TWL and bottom water level is BWL. Here top water level is TWL and bottom water level is BWL. Here top water level is TWL and bottom water level is BWL. Within the system as a result of friction dynamic head is generated. Basic Darcy Weisbach equation helps to calculate the dynamic head Here the loss coefficient is shown by K, velocity in the pipe is shown by and acceleration is . Here the loss coefficient is shown by K, velocity in the pipe is shown by and acceleration is . elocity is shown as Here flow rate is shown by Q through the pipe and cross sectional area is shown by A. Here flow rate is shown by Q through the pipe and cross sectional area is shown by A. Area A is shown as Here flow rate is shown by Q through the pipe and cross sectional area is shown by A. Area A is shown as The loss coefficient K is form of two elements: The loss coefficient K is form of two elements: The loss coefficient K is form of two elements: is shown as pumping the water from reservoir to receiver tank fittings used for the pipeworks of the system is shown as pumping the water from reservoir to receiver tank fittings used for the pipeworks of the system is shown as pumping the water from reservoir to receiver tank fittings used for the pipeworks of the system is associated with length of pipe ,friction and the diameter of the pipe. Experimental analysis p y The experiment has been done with 3 phase, 50Hz, 415 V,0.75 HP squirrel cage induction motor coupled with VFD based centrifugal pump with 2800 RPM speed and 23.5 meter head value. At healthy conditions three phase induction motor produces only positive sequence currents and it is symmetrical system. When symmetry is disturbed during fault situation it generates positive, negative, and zero sequence. The experiment has been done by creating inter turn fault in induction motor and analyzing the parameter changes both for motor and coupled pump. A Simulink model of three phase induction motor with turn fault in one phase winding has been built with the help of MATLAB software. As experimentally it is challenging to create fault due to shorting of high percentage value, the Simulink model has been developed. After completing the developed model, the model is verified both in healthy and faulty conditions. In different levels of shorting in one phase winding the model is simulated and the phase current values are stored in MATLAB workspace. Negative sequence current, positive sequence current, and zero sequence currents are calculated from these values. The next step is to verify how the inter turn fault affects various parameters of pump which is coupled with induction motor. After the simulation process is over the results are verified by OP5700 real- time simulator (hardware in loop) for validation. In another part of experiment, ML algorithms have been implemented on simulation data collected through MATLAB for identification and prediction of fault in induction motor based pumping system and for analyzing which algorithm is suitable for detection of fault. The Simulink model has been built based on the mathematical equations given in section 3. Figure 1 shows the block diagram of inter turn fault detection in induction motor based pumping system. Start Mathematical model of pumping system based on IM model for inter turn fault Simulation of the model and data collection Feature Extraction Application of Machine learning algorithm for training and testing the model. Find out best suited algorithm for inter turn fault detection in pumping system Verified the data and simulation results by HIL loop based device (OP5700) END Figure 1. Experimental analysis Flow chart of inter turn fault detection in induction motor based pumping system Mathematical model of pumping system based on IM model for inter turn fault Verified the data and simulation results by HIL loop based device (OP5700) Simulation of the model and data collection Feature Extraction Application of Machine learning algorithm for training and testing the model. Figure 2: Simulink model of induction motor based centrifugal pump ble 1: Magnitude of phase current and sequence component current in different % sh Table 1: Magnitude of phase current and sequence component current in different % de of phase current and sequence component current in different % shortings of Ph ase current and sequence component current in different % shortings of Phase A wi % Shorting in phase A windings Phase Current value (A) Sequence components of current values Ia Ib Ic Ipositive Inegetive Izero 0 1.54 1.52 1.53 1.601 0.0042 0.0042 1.5 1.62 1.59 1.60 1.628 0.0234 0.02312 2 1.628 1.59 1.60 1.635 0.0238 0.0235 5 1.64 1.59 1.61 1.646 0.0362 0.0369 10 1.641 1.59 1.61 1.679 0.0451 0.0457 15 1.66 1.60 1.615 1.681 0.0628 0.0630 20 1.67 1.60 1.62 1.684 0.0653 0.0654 25 1.68 1.605 1.621 1.691 0.0658 0.0661 30 1.74 1.65 1.63 1.695 0.0671 0.0669 35 1.82 1.68 1.69 1.699 0.0680 0.0670 40 1.85 1.70 1.71 1.7 0.0683 0.0675 Though the simulation model has been analysed from 0% to 40% short circuit fault, the HIL OP5700 results have been compared between healthy and 40% short circuit fault for checking the highest fluctuation during extreme fault condition. When inter turn fault occurs phase A current increases and it helps to increase the phase B and phase C current. During fault condition torque response of the motor suffers from oscillations. As torque is proportional to speed, if during fault condition torque increases, speed of the motor also increases. The motor is coupled with pump, so that speed is fed to pump also and once pump operates in fault condition, the flow rate value suddenly increases and pressure decreases. Now if the pressure goes below vapour pressure, cavitation problem will occur and sudden increase of flow rate creates vibrational problem in the overall system. Figure 1. Flow chart of inter turn fault detection in induction motor based pumping system 1. Flow chart of inter turn fault detection in induction motor based pumping system The details of the induction motor are : stator resistance Rs is 0.288 Ω, rotor resistance Rr is 0.158 Ω, stator inductance Ls and rotor inductance Lr are 0.0425 H and 0.0438 H respectively, mutual inductance Lm is 0.0412 H and inertia J is 0.4. Here number of poles are 2. Main input parameter shows per unit changes from positive sequence current and to negative sequence current for classification of severity of the fault level in phase windings. When the system is in healthy condition there will be no short circuit turn. But when the system is in fault condition the negative sequence current will increase once the turn fault percentage will increase. In the proposed research upto 40% inter turn fault has been measured. 40%, the value of Figure 2 shows the Simulink model of induction motor based centrifugal pump system which has three phase source, VFD drives and induction motor coupled with pump. Table 1 shows the magnitude of phase current and sequence component current. Figure 2: Simulink model of induction motor based centrifugal pump 2: Simulink model of induction motor based centrifugal pump Results and Discussion The figures show the performance curve of current, speed, torque with respect to time and pump curves for flowrate vs head value both in healthy and faulty condition. All the results are obtained from OP5700 HIL based device which verified the simulation results. Figure 3 (a) and 3 (b) show the healthy and 40% inter turn fault condition of stator current. Phase A, B and C current increases as fault occurs. Figure 4 (a) and figure 4 (b) show the healthy and 40% inter turn fault condition speed value. Similarly Figure 5(a) and figure 5(b) show the healthy and 40% inter turn fault condition torque value. During fault condition motor suffers from oscillations. The size of the oscillations changes when the percentage of turn increases at same load condition. As the oscillations lead to increase in the rated power of the machine, the oscillation in the torque also increases. Stator Current (A) Time (S) Phase A: 1.85 A Phase B: 1.70 A Phase C: 1.71 A Phase Current: 1.8 A/div Time: [0.06 s/div] (b) Stator Current (A) Time (S) Time: [0.04 s/div] Phase A: 1.60 A Phase B: 1.54 A Phase C: 1.535A Phase Current:1.6 A/div (a) Time: [0.06 s/div] (a) Stator current in healthy condition for induction motor, Figure 3 (b) Stator current after 40% inter turn induction motor. Stator Current (A) Ti (S) Time: [0.04 s/div] Phase A: 1.60 A Phase B: 1.54 A Phase C: 1.535A Phase Current:1.6 A/div (a) Time: [0.06 s/div] Stator Current (A) (a) Time (S) Phase A: 1.85 A Phase B: 1.70 A Phase C: 1.71 A Phase Current: 1.8 A/div Time: [0.06 s/div] (b) Stator Current (A) Time (S) (a) Stator Current (A) (b) Figure 3 (a) Stator current in healthy condition for induction motor, Figure 3 (b) Stator current after 40% inter turn fault in induction motor. Speed : 2800 RPM Time : [2s/div] Speed: 2000 RPM/div Speed (RPM) Time (s) (a) Speed : 2933 RPM Speed: 2000 RPM/div Time : [2s/div] Time (s) Speed (RPM) (b) Figure 4 (a): Speed for induction motor in healthy condition. Figure 4 (b): Speed for induction motor 40% stator inter turn fault occurs. Results and Discussion Pump Performance Curve Pump System Curve Head (m) Flow rate (m^3/s) Pump Head: 23.5 m Pump Flow rate: 2 m^3/s Pump head: 10 V/div Pump Flow rate: 10 V/div Time: [0.02 s/div] Pump Performance Curve Pump head: 10 V/div Pump System Curve Pump Flow rate: 10 V/div Pump Head: 18.2 m Pump Flow rate: 2.3 m^3/s Time: [0.02 s/div] Head (m) Flow rate (m^3/s) (a) (b) Figure 6(a) Pump performance curve and system curve in healthy condition Figure 6(b) Pump performance curve and system curve in 40% inter turn fault condition. RT lab software Probes Analog output pins OP5700 Real time HIL simulator Front view of OP5700 Results MSO Figure 7: Hardware setup of HIL device Pump Performance Curve Pump System Curve Head (m) Flow rate (m^3/s) Pump Head: 23.5 m Pump Flow rate: 2 m^3/s Pump head: 10 V/div Pump Flow rate: 10 V/div Time: [0.02 s/div] Pump Performance Curve Pump head: 10 V/div Pump System Curve Pump Flow rate: 10 V/div Pump Head: 18.2 m Pump Flow rate: 2.3 m^3/s Time: [0.02 s/div] Head (m) Flow rate (m^3/s) (a) (b) Pump Performance Curve Pump System Curve Head (m) Flow rate (m^3/s) Pump Head: 23.5 m Pump Flow rate: 2 m^3/s Pump head: 10 V/div Pump Flow rate: 10 V/div Time: [0.02 s/div] Pump Performance Curve Pump head: 10 V/div Pump System Curve Pump Flow rate: 10 V/div Pump Head: 18.2 m Pump Flow rate: 2.3 m^3/s Time: [0.02 s/div] Head (m) Flow rate (m^3/s) (a) (b) Figure 6(a) Pump performance curve and system curve in healthy condition Figure 6 Pump performance curve and system curve in 40% inter turn fault condition. Pump Performance Curve Pump System Curve Head (m) Flow rate (m^3/s) Pump Head: 23.5 m Pump Flow rate: 2 m^3/s Pump head: 10 V/div Pump Flow rate: 10 V/div Time: [0.02 s/div] (a) Head (m) Flow rate (m^3/s) Head (m) Flow rate (m^3/s) Flow rate (m^3/s) Flow rate (m 3/s) Figure 6(a) Pump performance curve and system curve in healthy condition Figure 6(b) Pump performance curve and system curve in 40% inter turn fault condition. RT lab software Probes Analog output pins OP5700 Real time HIL simulator Front view of OP5700 Results MSO Figure 7: Hardware setup of HIL device Figure 6(a) Pump performance curve and system curve in healthy condition Figure 6(b) Pump performance curve and system curve in 40% inter turn fault condition. Results and Discussion Speed : 2800 RPM Time : [2s/div] Speed: 2000 RPM/div Speed (RPM) Time (s) (a) Speed : 2933 RPM Speed: 2000 RPM/div Time : [2s/div] Time (s) Speed (RPM) (b) Speed : 2800 RPM Time : [2s/div] Speed: 2000 RPM/div Speed (RPM) Time (s) (a) Figure 4 (a): Speed for induction motor in healthy condition. Figure 4 (b): Speed for induction motor 40% stator inter turn fault occurs. Torque (Nm) Time (S) Time (S) Torque (Nm) (a) (b) Torque: 1.56 Nm Torque: 1.6 Nm/div Time : [0.06s/div] Torque: 1.61 Nm Torque: 1.6 Nm/div Time : [0.06 s/div] Figure 5 (a) Torque of the induction motor in healthy condition Figure 5 (b): Torque of the induction motor in faulty condition Time (S) (b) Figure 5 (a) Torque of the induction motor in healthy condition Figure 5 (b): Torque of the induction motor in faulty condition Figure 3 shows that current value increases for phase A. Once inter turn fault occurs and if the number of turn increases current value also increases. Increase in phase B and phase C current are accelerated also. Similarly speed and torque value also increase during fault condition as shown in figures 4 and 5. Once the motor speed increases pump speed also increases. . Increase in speed causes increase in flow rate and decrease in head value. Figure 6 (a) and 6 (b) show the pump performance curve and system curve in healthy and 40% inter turn fault condition. Figure 7 shows the hardware setup of HIL device. Figure 3 shows that current value increases for phase A. Once inter turn fault occurs and if the number of turn increases current value also increases. Increase in phase B and phase C current are accelerated also. Similarly speed and torque value also increase during fault condition as shown in figures 4 and 5. Once the motor speed increases pump speed also increases. . Increase in speed causes increase in flow rate and decrease in head value. Figure 6 (a) and 6 (b) show the pump performance curve and system curve in healthy and 40% inter turn fault condition. Figure 7 shows the hardware setup of HIL device. ormance curve and system curve in healthy and 40% inter turn fault condition. Figure 7 shows the ware setup of HIL device. Application of ML approach in the induction motor based pump fault detection Generally ML algorithms are of two types, supervised and unsupervised, and supervised algorithms have target variables which are formed from the predicted value of input variables. Figure 8 shows the generalized block diagram of proposed research after the data collection for finding out the best suited algorithm. Multistage VFD based parallal pump setup Feature extraction Data collection both in healthy and faulty situation in the form of vibrational signal, acceleration value with respect to time Feature selection Application of AI based ML classifier algorithm through Matlab Software. Training data set Testing data set Is the model trained? No Yes Classification process Finding out the fault occurs in the pump Classify healthy part Classify faulty part Find out the suitable ML algorithm for classification process Step 1: Data Collection Step 2: Data pre processing Step 3: Condition, classification and diagnosis of the pumping system Figure 8: Block diagram of the proposed research for choosing best suited algorithm Multistage VFD based parallal pump setup Data collection both in healthy and faulty situation in the form of vibrational signal, acceleration value with respect to time Step 1: Data Collection Application of AI based ML classifier algorithm through Matlab Software. Step 2: Data pre processing Classify faulty part Step 3: Condition, classification and diagnosis of the pumping system Find out the suitable ML algorithm for classification process Figure 8: Block diagram of the proposed research for choosing best suited algorithm The powerful technique ANN is used for the diagnosis of induction motor more accurately. Neural network (NN) is one of the pattern classifiers. Many problems can be solved by using pattern classification of NN k and these problems involve in variable recognition and for induction motor fault diagnosis it cannot be entirely described or predicted. Mathematical model based computational algorithm is ANN which behaves like human brain and thinking process. It has various features like similar parallel processing, self-organizing, self-learning, classification and non-linear mapping abilities. Combination of Fuzzy and ANN is ANFIS [22] [23], and it is combined to improve speed, fault tolerance, adaptiveness and to obtain the better modelling system. Based on RMSE, R2 value it can be compared which algorithm is suitable for inter turn fault detection in induction motor based pumping system. For the inter turn fault detection in induction motor ANN and ANFIS models are proposed. Results and Discussion RT lab software Probes Analog output pins OP5700 Real time HIL simulator Front view of OP5700 Results MSO Figure 7: Hardware setup of HIL device Figure 7: Hardware setup of HIL device Application of ML approach in the induction motor based pump fault detection Artificial immune system for ANN has self-adaptive control and performs better for continuous nonlinear function. The process can be done through online monitoring [24]. ANN is highly interconnected and the similarity is seen with human brain and it follows learning process like human being [25]. Units have interconnections between them and they have weights which are multiplied with the values which go through them. Unit has a fixed input which is known as bias and each unit form a weighted sum where bias is added. Transfer function analyzes this sum. Prediction of NN depends on training and testing data. The success of training is greatly affected by proper selection of inputs [26]. Learning process uses testing data and NN constructs input-output mapping. Iteration based on minimization or optimization of some error measured between the output produced and the desired output can be done adjusting the weights and bias. This process is repeated till an acceptable criterion for convergence is obtained. NN consists of input layer, hidden layer and output layer which is shown in figure 9. Output layer consists of six neurons like healthy condition, 5 turn short circuit, 10 turn short circuit, 20 turn short circuit, 30 turn short circuit and 40 turn short circuit. The algorithm can choose number of hidden layer by trial and error process. Levenberg Marquardt back propagation is chosen for the training purpose and training and testing data help to obtain average minimum square error (MSE) for ANN. For training purpose training rate is used for 70%, for cross validation 15% and for testing 15%. The average MSE values with respect to processing elements present in hidden layer are shown in table 2. With respect to processing elements percentage accuracy of hidden layers are shown for healthy and different turn fault condition in table2 and figure 10. Most of the cases the table and figure show that percentage of accuracy is 100 for healthy and faulty condition and average MSE is minimum 6.98e-8. W b W b   Input Layer Hidden layer neurons Output layer neurons Output layer Figure 9: Block diagram of neural network Output layer Figure 9: Block diagram of neural network Figure 9: Block diagram of neural network Table 2: Percentage accuracy of classification based on number of processing elements and MSE for ANN Table 2: Percentage accuracy of classification based on number of processing e ab e e ce tage accu acy o c ass cat o based o u be o p ocess g e e e ts a d S o No of Processing Elements MSE Percentage accuracy of classification Healthy condition 5 turn short circuit 10 turn short circuit 20 turn short circuit 30 turn short circuit 40 turn short circuit 1 0.691 100 43 39 46.2 45.8 46.7 2 0.540 85 54 45 67.1 56.9 54.3 3 0.439 76 78 72 49.8 69.34 67.5 4 0.231 100 96 61 78.1 78.9 79.2 5 0.195 93 100 74 99.8 80.9 89.3 6 0.0062 100 100 80 100 90.5 90.4 7 1.23e-2 86 100 100 100 95.6 100 8 3.67e-2 99 98 100 100 100 100 9 4.68e-6 100 100 100 99.5 100 100 10 5.78e-7 100 100 99 100 100 100 11 6.98e-8 100 100 100 100 100 100 Figure 10: Percentage of accuracy with respect to processing elements in healthy and faulty condition Figure 10: Percentage of accuracy with respect to processing elements in healthy and faulty condition An intelligent system Neuro –Fuzzy technique ANFIS is used for modelling and control of ill-defined and uncertain systems. Input/output data pairs of the system under consideration builds ANFIS. ANFIS is the combination o0f ANN and Fuzzy which is used for the learning ability of fuzzy system. ANFIS consists of five layers [27] [28]. Layer 1 is fuzzification layer which calculates membership function. Layer 2 represents rules layer whose output is the firing strength of each node. Layer 3 highlights the normalization layer which normalizes the calculated firing strength. Layer 4 shows the consequent layer whose output layer is the product of normalized firing strength and the fuzzy rules consequent polynomial. Layer 5 shows the overall output and defines defuzzification layer whose output is overall ANFIS output. The problems of continuous changes in mobile learning environments are solved by ANFIS [29]. The proposed ANFIS model can be used for modelling the learner context. Defining input and output values, Fuzzy sets for input values, Fuzzy rules, and creating and training the NN are the steps applying ANFIS to learner model [30]. Figure 11: Best fit output of ANN model Various previous works have been done for inter turn fault detection of induction motor and motor based pumping system. Till now researches on inter turn fault of induction motor analysed only changes of induction motor parameters after the fault occurs. But when the motor faces inter turn fault problem, the coupled pump is also affected and parameters of the pump also change. In the proposed research, including the change of parameters of inter turn fault affected motor, the change in pump parameters also have been analysed. Current coordinate transform algorithm for inter turn fault analysis in induction motor was developed. Mexbios development studio was built up to analyse the parameter changes in induction motor during fault condition. Though this is possible to implement in industrial applications, this process cannot predict the fault before the fault occurs and major damages are seen [31]. The proposed model is a simple easy process and helpful for less amount of data and predict the fault before massive failure. ANN algorithm for inter turn fault detection in induction motor with respect to various turns was used. Upto 10% inter turn fault was created in the experiment and phase A current changes were obtained. In the experiment per unit change of positive sequence current from negative sequence current was analysed. Experiment was done upto 54 epochs. Here experimental analysis was done for small level of values [32]. NN model in three different conditions which are no load, 50% load and full load condition for five different motors for inter turn fault analysis was developed.. Upto 15% inter turn fault was developed and the accuracy rate of NN model for various motors varied from 88% to 99% [33]. The novel wavelet analysis was developed in a research. Based on discrete wavelet transformation using Park’s vector transformation the model was built for analysis of inter turn fault. Performance analysis was done for healthy and various turn fault conditions. MSE was obtained for performance accuracy analysis of healthy and faulty situation [34]. Other researches are based on FFT analysis and Park transformation but these researches are not suitable for predictive control model and not useful for heavy industrial application [35]. The proposed method of ANN has been used in current research which developed the Matlab model for analysis of the parameter changes of induction motor and pump during inter turn fault. Figure 9: Block diagram of neural network But when the motor faces inter turn fault problem, the coupled pump is also affected and parameters of the pump also change. In the proposed research, including the change of parameters of inter turn fault affected motor, the change in pump parameters also have been analysed. Current coordinate transform algorithm for inter turn fault analysis in induction motor was developed. Mexbios development studio was built up to analyse the parameter changes in induction motor during fault condition. Though this is possible to implement in industrial applications, this process cannot predict the fault before the fault occurs and major damages are seen [31]. The proposed model is a simple easy process and helpful for less amount of data and predict the fault before massive failure. ANN algorithm for inter turn fault detection in induction motor with respect to various turns was used. Upto 10% inter turn fault was created in the experiment and phase A current changes were obtained. In the experiment per unit change of positive sequence current from negative sequence current was analysed. Experiment was done upto 54 epochs. Here experimental analysis was done for small level of values [32]. NN model in three different conditions which are no load, 50% load and full load condition for five different motors for inter turn fault analysis was developed.. Upto 15% inter turn fault was developed and the accuracy rate of NN model for various motors varied from 88% to 99% [33]. The novel wavelet analysis was developed in a research. Based on discrete wavelet transformation using Park’s vector transformation the model was built for analysis of inter turn fault. Performance analysis was done for healthy and various turn fault conditions. MSE was obtained for performance accuracy analysis of healthy and faulty situation [34]. Other researches are based on FFT analysis and Park transformation but these researches are not suitable for predictive control model and not useful for heavy industrial application [35]. The proposed method of ANN has been used in current research which developed the Matlab model for analysis of the parameter changes of induction motor and pump during inter turn fault. As it is Simulink model so high range that is upto 40% of inter turn fault is possible to be Figure 11: Best fit output of ANN model Figure 11: Best fit output of ANN model Figure 9: Block diagram of neural network In the proposed work ANN model and ANFIS model both are implemented and compared to find out the better performance. R2 and RMSE are used to find out the best suited model for fault detection of induction motor based pumping system. The performances of the obtained ANN and ANFIS models are also compared after building the model is done. RMSE and R2 have comparative statistical values for ANN and ANFIS models which are given in table 3. Validation data of the model is 0.05. Prediction accuracy is also measured by R2 and RMSE. Prediction accuracy for ANN (R2 is 100 and RMSE is 0.025) is better than ANFIS model (R2 is 98.61 and RMSE is 0.08). Table 3: Comparison of RMSE and R2 data for ANN and ANFIS Table 3: Comparison of RMSE and R2 data for ANN and ANFIS Parameters Training data Testing data ANN ANFIS ANN ANFIS RMSE 0.054 0.121 0.098 0.062 R2 0.998 0.934 0.969 0.897 ANN and ANFIS models both performed well and are compatible for fault detection and able to predict the fault but based on RMSE and R2 of training and testing data, ANN performed better than ANFIS in the proposed experiment. The ANN model has been applied upto 200 epochs and best validation has been received in 150 epochs. Figure 11 shows the best fit value of the proposed model of ANN with respect to training, testing, validation and overall values. ANN and ANFIS models both performed well and are compatible for fault detection and able to predict the fault but based on RMSE and R2 of training and testing data, ANN performed better than ANFIS in the proposed experiment. The ANN model has been applied upto 200 epochs and best validation has been received in 150 epochs. Figure 11 shows the best fit value of the proposed model of ANN with respect to training, testing, validation and overall values. shows the best fit value of the proposed model of ANN with respect to training, testing, validation and overall values. Figure 11: Best fit output of ANN model Various previous works have been done for inter turn fault detection of induction motor and motor based pumping system. Till now researches on inter turn fault of induction motor analysed only changes of induction motor parameters after the fault occurs. Figure 11: Best fit output of ANN model K-NN is non-parametric versatile learning algorithm which is also used for classification and regression problem. Instead of learning the discriminative function the algorithm memorizes the training dataset. By minimizing the training set intense based learning helps to avoid error. The disadvantages of K-NN are ample memory storage, long prediction time, and unnecessary sensitivity to irrelevant features. But when the data size is limited, K-NN works better than any other supervised learning algorithms. A decision tree dendritic classification model is used for both classification and regression problems. Breaking in smaller subset, classification process can be done and based on this feature selection can be done. The final structure is like tree branches and each node highlights the feature. Regression analysis provides user equation for graph for the prediction of the data. It always shows the weighted average value for the prediction purpose. Through the statistical analysis it can predict the accurate output. Most elementary statistical courses cover fundamental techniques, like making scatter plots and performing linear regression. Based on overall accuracy rate, prediction speed and training time the most suitable algorithm can be found out. For experiment, features are divided into two categories randomly in the ratio of 70:30. In most of the cases 70 % of data are used for training purposes and 30% are used for testing data for evaluation. For all the algorithms the rule is same. The entire diagnosis is carried through MATLAB pattern recognition app and classification learner app. Based on the evaluation, the accuracy rate of each algorithm is obtained using the formula. With the help of For experiment, features are divided into two categories randomly in the ratio of 70:30. In most of the cases 70 % of data are used for training purposes and 30% are used for testing data for evaluation. For all the algorithms the rule is same. The entire diagnosis is carried through MATLAB pattern recognition app and classification learner app. Based on the evaluation, the accuracy rate of each algorithm is obtained using the formula. With the help of classification learner app the accuracy rate, prediction speed and training time of each algorithm have been analysed and compared. Figure 11: Best fit output of ANN model The final structure is like tree branches and each node highlights the feature. Regression analysis provides user equation for graph for the prediction of the data. It always shows the weighted average value for the prediction purpose. Through the statistical analysis it can predict the accurate output. Most elementary statistical courses cover fundamental techniques, like making scatter plots and performing linear regression. Based on overall accuracy rate, prediction speed and training time the most suitable algorithm can be found out. For experiment, features are divided into two categories randomly in the ratio of 70:30. In most of the cases 70 % of data are used for training purposes and 30% are used for testing data for evaluation. For all the algorithms the rule is same. The entire diagnosis is carried through MATLAB pattern recognition app and classification learner app. Based on the evaluation, the accuracy rate of each algorithm is obtained using the formula. With the help of classification learner app the accuracy rate, prediction speed and training time of each algorithm have been analysed and compared. found that performance of ANN is better than ANFIS model. Similarly the authors also have implemented various supervised ML algorithms like SVM, K-NN, Decision tree, Naïve Bayes, Regression analysis with ANN and ANFIS. Based on accuracy rate, prediction speed and training time, algorithms are compared to find out most suitable algorithm for this experiment. Generally ML algorithms have target variables which are to be predicted from independent variables and these variables generate function for mapping of input to achieve desired output. After that training process should be done for the better achievement and for more accuracy. Until the desired accuracy rate is obtained the training process is going on. No target variable is required for unsupervised learning as it follows the clustering process. SVM is well known pattern recognition algorithm which is mainly used for classification and regression process. SVM has hyperplane and margin by which it separates the dataset and performs the classification task. Optimum hyperplane in the SVM maximizes the width of hyperplane to avoid the overlapping between the classes and this is the process of classification. Margins are classified between hard margin and soft margin. Since the present diagnosis deals with the non-linear classification problem, a soft margin is used. The accuracy of SVM depends on three factors like threshold function, cost function and kernel function. [34]. From the table 4 it is seen that performance of the K-NN and ANN are better for this research. But based on accuracy rate, prediction speed and training time K-NN is more suitable than ANN. Figure 12 picturizes the overall accuracy of the ML algorithms. Figure 11: Best fit output of ANN model As it is Simulink model so high range that is upto 40% of inter turn fault is possible to be created for analysis. Here ANN and ANFIS models have been implemented in the experimental results and it is found that performance of ANN is better than ANFIS model. created for analysis. Here ANN and ANFIS models have been implemented in the experimental results and it is found that performance of ANN is better than ANFIS model. Similarly the authors also have implemented various supervised ML algorithms like SVM, K-NN, Decision tree, Naïve Bayes, Regression analysis with ANN and ANFIS. Based on accuracy rate, prediction speed and training time, algorithms are compared to find out most suitable algorithm for this experiment. Generally ML algorithms have target variables which are to be predicted from independent variables and these variables generate function for mapping of input to achieve desired output. After that training process should be done for the better achievement and for more accuracy. Until the desired accuracy rate is obtained the training process is going on. No target variable is required for unsupervised learning as it follows the clustering process. SVM is well known pattern recognition algorithm which is mainly used for classification and regression process. SVM has hyperplane and margin by which it separates the dataset and performs the classification task. Optimum hyperplane in the SVM maximizes the width of hyperplane to avoid the overlapping between the classes and this is the process of classification. Margins are classified between hard margin and soft margin. Since the present diagnosis deals with the non-linear classification problem, a soft margin is used. The accuracy of SVM depends on three factors like threshold function, cost function and kernel function. [34]. K-NN is non-parametric versatile learning algorithm which is also used for classification and regression problem. Instead of learning the discriminative function the algorithm memorizes the training dataset. By minimizing the training set intense based learning helps to avoid error. The disadvantages of K-NN are ample memory storage, long prediction time, and unnecessary sensitivity to irrelevant features. But when the data size is limited, K-NN works better than any other supervised learning algorithms. A decision tree dendritic classification model is used for both classification and regression problems. Breaking in smaller subset, classification process can be done and based on this feature selection can be done. Conclusion Inter turn fault analysis of induction motor based pumping system is explained in this article and the parameter changes during fault situation in different turn conditions have been shown. The simulation results have been verified through HIL loop based (OP5700) device and it is seen that phase current of motor increases when the fault occurs. Once current increases speed and torque also increases and it affects the pumping system. Speed helps to increase the flow rate of the pump suddenly and it causes huge pressure drop and decrease in head value. If pressure drops drastically, cavitation problem occurs and sudden increase in flow rate causes huge vibration in the pipe which causes water hammering problem. In this research at first by ANN and ANFIS algorithm based models identification and prediction of the fault have been done. Both the techniques are used and it is seen that ANN performs better than ANFIS, based on RMSE and R2 value. Various other research works are also compared with proposed work to find out the new development in the proposed work. It is observed that the proposed research is suitable for industrial application and can easily identify the faulty condition for large amount of data. In future the ANN would have been used for other fault detections in motor and pumping system and for other machineries also and can become a comprehensive diagnosis technique. The authors also compared various ML algorithms with ANN and ANFIS among which based on accuracy rate, prediction speed and training time it is seen that K-NN and ANN can work better for the proposed research. But based on overall accuracy rate K-NN works better than ANN. In addition, the deployment of the developed technique in a laboratory environment is an extension of the present work. More researches are possible through HIL based OP5700 device to verify the simulation results. Author Contributions: Nabanita. Dutta, Palanisamy Kaliannan have done the inter turn fault analysis in induction motor based centrifugal pumping system. Expertise in the field of drives, control and fault analysis in pumping system using VFD and induction motor based pump has been shared by Paramasivam Shanmugam All authors have equally contributed to articulate the research work for its final depiction as a full research paper. Funding: There is no source of funding for this research activity. Figure 11: Best fit output of ANN model (34) Table 4: performance analysis of various algorithms Algorithms Accuracy rate (percentage) Prediction Speed (obs/sec ) Training time (Sec) SVM 98.3 290 0.562 K-NN 100 510 0.063 Naïve Bayes 75.6 410 0.982 Decision Tree 72.7 350 1.567 Regression Analysis 90.9 467 2.987 ANN 99.6 480 0.086 ANFIS 94.6 320 4.236 Table 4: performance analysis of various algorithms Figure 12; Overall accuracy of various ML algorithms Figure 12; Overall accuracy of various ML algorithms Figure 12; Overall accuracy of various ML algorithms References ces 1. Neti, Prabhakar, and Subhasis Nandi. "Stator inter-turn fault analysis of reluctance synchronousmotor." In Canadian Conference on Electrical and Computer Engineering, 2005, pp. 1283-1286. IEEE, 2005. 2. Obeid, Najla Haje, Thierry Boileau, and Babak Nahid-Mobarakeh. "Modeling and diagnostic of incipient interturn faults for a three-phase permanent magnet synchronous motor." IEEE Transactions on Industry Applications 52, no. 5 (2016): 4426-4434. 3. Gao, Fengyang, Guoheng Zhang, Mingming Li, Yunbo Gao, and Shengxian Zhuang. "Inter-turn Fault Identification of Surface-Mounted Permanent Magnet Synchronous Motor Based on Inverter Harmonics." Energies 13, no. 4 (2020): 899. 4. Maraaba, Luqman S., Ssennoga Twaha, Azhar Memon, and Zakariya Al-Hamouz. "Recognition of Stator Winding Inter-Turn Fault in Interior-Mount LSPMSM Using Acoustic Signals." Symmetry 12, no. 8 (2020): 1370. 5. Yassa, Nacera, and M. Rachek. "Modeling and detecting the stator winding inter turn fault of permanent magnet synchronous motors using stator current signature analysis." Mathematics and Computers in Simulation 167 (2020): 325-339. 6. Kim, Byeong-Woo, Kyung-Tae Kim, and Jin Hur. "Simplified impedance modeling and analysis for inter-turn fault of IPM-type BLDC motor." Journal of Power Electronics 12, no. 1 (2012): 10-18. 7. Kalaskar, Chaitali S., and Vitthal J. Gond. "Motor current signature analysis to detect the fault in induction motor." International Journal of Engineering Research and Applications 4, no. 6 (2014): 58-61. 8. Mohanty, Amiya Ranjan, Prasanta Kumar Pradhan, Nitaigour P. Mahalik, and Sabyasachi G. Dastidar. "Fault detection in a centrifugal pump using vibration and motor current signature analysis." International Journal of Automation and Control 6, no. 3-4 (2012): 261-276. 9. Orrù, Pier Francesco, Andrea Zoccheddu, Lorenzo Sassu, Carmine Mattia, Riccardo Cozza, and Simone Arena. "Machine learning approach using MLP and SVM algorithms for the fault prediction of a centrifugal pump in the oil and gas industry." Sustainability 12, no. 11 (2020): 4776. 10. Sołjan, Zbigniew, Grzegorz Hołdyński, and Maciej Zajkowski. "CPC-Based Minimizing of Balancing Compensators in Four-Wire Nonsinusoidal Asymmetrical Systems." Energies 14, no. 7 (2021): 1815. 11. HS, Ranganatha Chakravarthy, Sai Charan Bharadwaj, S. Umashankar, Sanjeevikumar Padmanaban, Nabanita Dutta, and Jens Bo Holm-Nielsen. "Electrical fault detection using machine learning algorithm for centrifugal water pumps." In 2019 IEEE International Conference on Environment and Electrical Engineering and 2019 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I&CPS Europe), pp. 1-6. IEEE, 2019. 12. Dutta, Nabanita, Palanisamy Kaliannan, and Umashankar Subramaniam. Acknowledgment The authors like to express sincere gratitude to Danfoss Advanced Drives Laboratory, VIT University, for making this research work for execution and technical implementation in real-time. This is the industry-academic joint collaborative project for industrial pumping solution. The collaboration is between VIT Vellore and Danfoss Pvt Ltd industry. Conflicts of Interest: The authors declare no conflict of interest. References "Effect of Motor Vibration Problems on Power Quality of Water Pumping at Residency." In IOP Conference Series: Materials Science and Engineering, vol. 937, no. 1, p. 012019. IOP Publishing, 2020. 13. Arun Shankar, V. K., Umashankar Subramaniam, Sanjeevikumar Padmanaban, Jens Bo Holm- Nielsen, Frede Blaabjerg, and S. Paramasivam. "Experimental investigation of power signatures for cavitation and water hammer in an industrial parallel pumping system." Energies 12, no. 7 (2019): 1351. 14. Jain, Geet, and S. Umashankar. "Modelling and simulation of solar photovoltaic fed induction motor for water pumping application using perturb and observer MPPT algorithm." In 2016 International Conference on Energy Efficient Technologies for Sustainability (ICEETS), pp. 250- 254. IEEE, 2016. 15. Nabanita, D., S. Sreedhar, Neha Zade, Onkar Bhalerao, Pratik Potdar, and R. Raja Singh. "Inter turn Fault Analysis on Wound Rotor Induction Machine." In IOP Conference Series: Materials Science and 2019, Engineering, vol. 623, no. 1, p. 012016. IOP Publishing. 16. Hussein, HA Taha, M. E. Ammar, and MA Moustafa Hassan. "Three phase induction motor's stator turns fault analysis based on artificial intelligence." International Journal of System Dynamics Applications (IJSDA) 6, no. 3 (2017): 1-19. 17. Skowron, Maciej, Teresa Orlowska-Kowalska, Marcin Wolkiewicz, and Czeslaw T. Kowalski. "Convolutional neural network-based stator current data-driven incipient stator fault diagnosis of inverter-fed induction motor." Energies 13, no. 6 (2020): 1475. 18. Salih, Hamad Raad, Ali Abdulwahhab Abdulrazzaq, and Basarab Dan Guzun. "Dynamic Modeling of Pump Drive System utilizing Simulink/MATLAB Program." International Research Journal of Engineering and Technology (IRJET) 3, no. 1 (2016): 21-24. 19. Husari, Fatimatelbatoul, and Jeevanand Seshadrinath. "Incipient Inter Turn Fault Detection and Severity Evaluation in Electric Drive System Using Hybrid HCNN-SVM Based Model." IEEE Transactions on Industrial Informatics (2021). 20. Ostojic, Pero, Arijit Banerjee, Dhaval Patl, Wrichik Basu, and Shahid Ali. "Advanced motor monitoring and diagnostics." In Industry Applications Society 60th Annual Petroleum and Chemical Industry Conference, pp. 1-9. IEEE, 2013. 21. Haroun, S., A. Nait Seghir, and S. Touati. "Stator inter turn fault and voltage unbalance detection and discrimination approach for an reactor coolant pump." In 3rd International Conference on Systems and Control, pp. 99-104. IEEE, 2013. 22. Mohamed, Menshawy A., Mohamed A. Moustafa Hassan, Fahad Albalawi, Sherif SM Ghoneim, Ziad M. Ali, and Mostafa Dardeer. "Diagnostic Modelling for Induction Motor Faults via ANFIS Algorithm and DWT-Based Feature Extraction." Applied Sciences 11, no. 19 (2021): 9115. 23. Rajeswari, R., and N. Kamaraj. 35. Irfan, Muhammad, and Adam Glowacz. "Design of a novel electric diagnostic technique for fault analysis of centrifugal pumps." Applied Sciences 9, no. 23 (2019): 5093. References "Diagnosis of inter turn fault in the stator of synchronous generator using wavelet based ANFIS." World Academy of Science, Engineering and Technology 36 (2007): 203-209. 24. Panda, S., D. P. Mishra, and S. N. Dash. "Comparison of ANFIS and ANN Techniques in Fault Classification and Location in Long Transmission Lines." In 2018 International Conference on Recent Innovations in Electrical, Electronics & Communication Engineering (ICRIEECE), pp. 1112- 1117. IEEE, 2018. 25. Esakimuthu Pandarakone, Shrinathan, Yukio Mizuno, and Hisahide Nakamura. "A comparative study between machine learning algorithm and artificial intelligence neural network in detecting minor bearing fault of induction motors." Energies 12, no. 11 (2019): 2105. 26. Duan, Fang, and Rastko Zivanovic. "A model for induction motor with stator faults." In 2012 22nd Australasian Universities Power Engineering Conference (AUPEC), pp. 1-5. IEEE, 2012. 27. Samanta, S., J. N. Bera, and G. Sarkar. "An Intelligent Adaptive Neuro Fuzzy based Fault Diagnosis System for Severity and Phase Detection of Induction Machine." 28. Ghafari, Shahab Hasanzadeh. A fault diagnosis system for rotary machinery supported by rolling element bearings. Waterloo, ON, Canada: University of Waterloo, 2007. 29. Solodkiy, Evgeniy, Dmitry Dadenkov, and Saveliy Salnikov. "Detection of stator inter-turn short circuit in three-phase induction motor using current coordinate transformation." In 2019 26th International Workshop on Electric Drives: Improvement in Efficiency of Electric Drives (IWED), pp. 1-4. IEEE, 2019. 30. Ghorbanzadeh, Omid, Thomas Blaschke, Jagannath Aryal, and Khalil Gholaminia. "A new GIS-based technique using an adaptive neuro-fuzzy inference system for land subsidence susceptibility mapping." Journal of Spatial Science 65, no. 3 (2020): 401-418. 31. Rajamany, Gayatridevi, Sekar Srinivasan, Krishnan Rajamany, and Ramesh K. Natarajan. "Induction motor stator interturn short circuit fault detection in accordance with line current sequence components using artificial neural network." Journal of Electrical and Computer Engineering 2019 (2019). 32. Maraaba, Luqman, Zakariya Al-Hamouz, and Mohammad Abido. "An efficient stator inter-turn fault diagnosis tool for induction motors." Energies 11, no. 3 (2018): 653. 33. Umashankar, S., P. Sanjeevikumar, and S. Paramasivam. "Adaptive neuro-fuzzy inference system (anfis) based direct torque control of pmsm driven centrifugal pump." (2017). 34. Babaa, Fatima, and Ouafae Bennis. "An accurate inter-turn short circuit faults model dedicated to induction motors." International Journal of Electrical and Computer Engineering 11, no. 1 (2021): 9. 35. Irfan, Muhammad, and Adam Glowacz. "Design of a novel electric diagnostic technique for fault analysis of centrifugal pumps." Applied Sciences 9, no. 23 (2019): 5093. 3/3
https://openalex.org/W3199423645	https://europepmc.org/articles/pmc8490748?pdf=render	English	null	Non-Invasive Electrochemical Biosensors for TNF-α Cytokines Detection in Body Fluids	Frontiers in bioengineering and biotechnology	2,021	cc-by	7,359	REVIEW published: 21 September 2021 doi: 10.3389/fbioe.2021.701045 Non-Invasive Electrochemical Biosensors for TNF-α Cytokines Detection in Body Fluids Yang Lu 1, Qingqing Zhou 2 and Lin Xu 2* 1Department of Cardiovascular, The Second Hospital of Jilin University, Jilin University, Changchun, China, 2State Key Laboratory on Integrated Optoelectronics, College of Electronic Science and Engineering, Jilin University, Changchun, China The measurement of pro-inﬂammatory cytokine tumour necrosis factor-alpha (TNF-α), which is an important indicator of the inﬂammatory process, has received increasing attention recently because it is easy to extract from body ﬂuid and serves as an early sign of a serious systemic inﬂammatory disease. Developing fast and simple detection methods to quantify the concentration of TNF-α is essential. Saliva, tears, and urine, which can easily be sampled in a non-invasive way, are considered to be important matrices for monitoring and assessing the physiological status of humans; importantly, they also provide an ideal window for monitoring the concentration of TNF-α. As a fast, accurate, inexpensive, portable, and scalable method, electrochemical biosensors are very promising for biomarker detection in matrices obtained in a non-invasive manner. This review summarises and compares the electrochemical biosensors for the detection of TNF-α in a non-invasive manner and highlights recent advances and future prospects in developing high-performance electrochemical platforms for noninvasive measurement of TNF-α. INTRODUCTION Correspondence: Lin Xu linxu@jlu.edu.cn Correspondence: Lin Xu linxu@jlu.edu.cn Tumour necrosis factor alpha (TNF-α), a pleiotropic cytokine produced by activating monocytes/ macrophages in the inﬂammatory process, is considered an important biomarker of autoimmune diseases (Arya et al., 2007; Ali et al., 2011; Altintas et al., 2014; Kalliolias and Ivashkiv, 2016). It has distinct roles in modulating a series of biological processes, such as in rheumatoid arthritis, human immunodeﬁciency virus infection, Alzheimer’s disease, stroke, diabetic retinopathy (Boss et al., 2017), and colon and lung cancers (Sadighbayan et al., 2019). It is also related to the severity of cardiovascular lesions and is involved in suppressing myocardial contractility, elevating the content of cardiac peroxynitrite and nitric oxide, changes in intracellular calcium homeostasis, and induction of pathological changes in the failing myocardium (Berthonneche et al., 2004). Moreover, the TNF-α cytokine plays a role in preventing bone and cartilage damage and assisting wound healing (Lipsky et al., 2000; Sandborn et al., 2007). Therefore, developing an accurate and fast TNF-α measurement technique is highly desirable for predicting, preventing, assessing, and monitoring inﬂammation. Over the past few decades, evaluating physiological status, detecting illness at an early stage, monitoring disease progression, and assessing therapeutic outcomes through non-invasive methods has become the most ideal and promising approach for clinical diagnosis and research (Bellagambi et al., 2020; Loo and Pui, 2020; Vozgirdaite et al., 2021). Previous studies have revealed that TNF-α can be easily extracted from non-invasively obtained body ﬂuids, such as saliva, tears, and urine Tumour necrosis factor alpha (TNF-α), a pleiotropic cytokine produced by activating monocytes/ macrophages in the inﬂammatory process, is considered an important biomarker of autoimmune diseases (Arya et al., 2007; Ali et al., 2011; Altintas et al., 2014; Kalliolias and Ivashkiv, 2016). It has distinct roles in modulating a series of biological processes, such as in rheumatoid arthritis, human immunodeﬁciency virus infection, Alzheimer’s disease, stroke, diabetic retinopathy (Boss et al., 2017), and colon and lung cancers (Sadighbayan et al., 2019). It is also related to the severity of cardiovascular lesions and is involved in suppressing myocardial contractility, elevating the content of cardiac peroxynitrite and nitric oxide, changes in intracellular calcium homeostasis, and induction of pathological changes in the failing myocardium (Berthonneche et al., 2004). Moreover, the TNF-α cytokine plays a role in preventing bone and cartilage damage and assisting wound healing (Lipsky et al., 2000; Sandborn et al., 2007). Edited by: Edited by: Masoud Mozafari, University of Toronto, Canada Reviewed by: Suzanne Hagan, Glasgow Caledonian University, United Kingdom Sajjad Ashraf, University of Toronto, Canada INTRODUCTION Therefore, developing an accurate and fast TNF-α measurement technique is highly desirable for predicting, preventing, assessing, and monitoring inﬂammation. Specialty section: This article was submitted to Biomaterials, a section of the journal Frontiers in Bioengineering and Biotechnology Received: 27 April 2021 Accepted: 10 August 2021 Published: 21 September 2021 Citation: Lu Y, Zhou Q and Xu L (2021) Non- Invasive Electrochemical Biosensors for TNF-α Cytokines Detection in Body Fluids. Front Bioeng Biotechnol 9:701045 Specialty section: This article was submitted to Biomaterials, a section of the journal Frontiers in Bioengineering and Biotechnology Specialty section: This article was submitted to Biomaterials, a section of the journal Frontiers in Bioengineering and Biotechnology Received: 27 April 2021 Accepted: 10 August 2021 Published: 21 September 2021 Received: 27 April 2021 Accepted: 10 August 2021 Published: 21 September 2021 Keywords: non-invasive, electrochemical biosensor, TNF-α cytokines, cardiovascular disease, diease biomarker Citation: For example, the TNF-α concentration in the saliva of healthy humans is lower than 3 pg ml−1 and increases to about 30 pg ml−1 in the saliva of patients with severe autoimmune diseases (Liu and Duan, 2012; Bellagambi et al., 2017). The concentration of urinary TNF-α was determined to be 4 pg mg−1 in healthy individuals, normalised by the concomitant urinary creatinine level (to compensate for the TNF-α variations in urine ﬂuid) (Bellagambi et al., 2017). Its concentration can be up to 196.9 ± 121.2 pg/ml in tear ﬂuid, and research is beginning to focus on how immune and inﬂammatory diseases affect the content of TNF-α in tear ﬂuid, which has a less complex biological composition (Comoglu et al., 2013; Costagliola et al., 2013). Therefore, methods offering low limits of detection and high sensitivity are of great interest for clinical diagnosis and research. TNF-α are summarised, and their advantages and disadvantages are discussed. The immunosensor technique with different sensing receptors and transducers is considered as a critical method in biomarker screening (Raluca-Ioana et al., 2000; Baraket et al., 2012) and is considered to be the most promising candidate to compete with well-established laboratory techniques. Barhoumi et al. (2018) designed an electrochemical immunosensor on a gold working electrode (Figure 1A). After modifying 4-carboxymethylaniline (CMA) via electrodeposition, the surface of the electrode was activated by ethanolamine, and then the anti-TNF-α antibody was further functionalised through the carboxylic acid groups of CMA. For TNF-α determination, a secondary antibody (Ab- TNF-α-HRP) labelled with horseradish peroxidase (HRP) was employed to utilise a sandwich-type strategy in tetramethylbenzidine solution. Since the concentration of TNF-α showed a good correlation with heart failure and its severity, the sensor was applied to detect different levels of Ag-TNF-α in artiﬁcial saliva. The as-fabricated biosensor exhibited a linear detection range of 1–30 pg ml−1 with a fast response in 5 s to the Ag-TNF-α cytokine (Figure 1B). In addition, it showed good selectivity against other cytokines, such as interleukin (IL)-10 and the hormone cortisol. The above sensing performance of the as-built biosensor enables it to serve as a potential tool for fast and accurate quantiﬁcation of Ag-TNF-α biomarkers. However, the sensing performance of this biosensor was evaluated in artiﬁcial saliva, and further clinical examinations should be performed. Similarly, a bifunctional electrochemical immunosensor for the detection of TNF-α and IL-1β cytokines was reported by Sanchez-Tirado et al. (2017). Citation: In this study, they used screen-printed carbon electrodes (SPCEs) with dual working electrodes as the sensing platform and then introduced CMA as scaffolds for further modiﬁcation of the corresponding antibodies (Figure 1C). Such a dual immunosensor displayed a linearity range of 1–100 pg ml−1 with a limit of detection (LOD) of 0.85 pg ml−1 and 0.5–100 pg ml−1 with an LOD of 0.38 pg ml−1 for TNF-α and IL-1β, respectively, which was sufﬁcient for application in clinical testing. In addition, the cross-talk between the adjacent SPCEs that were modiﬁed with anti-IL or anti-TNF in the dual immunosensor was non-perceptible, and no obvious differences were detected in the amperometric responses of immunosensors when faced with the non-target antigen or without the corresponding target antigen (Figure 1D). When tested in the saliva sample, the results obtained from the dual electrochemical immunosensor showed almost consistent results with the clinical data measured from the ELISA arrays. Note that the reagent consumption was also very low (2.5 μl) compared to that of ELISA (100 μl). Both of these immunosensors used CMA as the conjugated For this purpose, intensive research on TNF-α determination has been conducted, using techniques such as enzyme-linked immunosorbent assay (ELISA), radio immunoassays, chemiluminescence imaging, mass spectrometry, ﬂuorescence immunoassay, and electrochemical immunosensors (Bosnjakovic et al., 2012). Among these techniques, electrochemical biosensors are considered promising platforms that take into account the advantages of easy miniaturisation, small sample volume, simple operation, and low cost (Arya and Bhansali, 2011; Baraket et al., 2016; Povedano et al., 2018). In addition, compared to the ELISA method, which is commonly used in clinical analysis, electrochemical biosensors are more ﬂexible and time-saving, contributing to the rapid evaluation of inﬂammatory TNF-α (Baydemir et al., 2016). This mini- review summarises the recent progress in the development of electrochemical biosensors for TNF-α determination in a non-invasive manner, and highlights recent advances and future prospects in developing high-performance electrochemical platforms for non-invasive determination of TNF-α biomarkers. Citation: Lu Y, Zhou Q and Xu L (2021) Non- Invasive Electrochemical Biosensors for TNF-α Cytokines Detection in Body Fluids. Front. Bioeng. Biotechnol. 9:701045. doi: 10.3389/fbioe.2021.701045 Over the past few decades, evaluating physiological status, detecting illness at an early stage, monitoring disease progression, and assessing therapeutic outcomes through non-invasive methods has become the most ideal and promising approach for clinical diagnosis and research (Bellagambi et al., 2020; Loo and Pui, 2020; Vozgirdaite et al., 2021). Previous studies have revealed that TNF-α can be easily extracted from non-invasively obtained body ﬂuids, such as saliva, tears, and urine September 2021 \| Volume 9 \| Article 701045 Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org Non-Invasive Biosensors for TNF-α Lu et al. (Brennan and Galvin, 2018; Yanez-Sedeno et al., 2019). Monitoring the trace concentration of TNF-α in these body ﬂuids provides a painless, simple, real-time means for cardiovascular disease diagnosis; in particular, it may serve as a sign of inﬂammation, which is important for the early diagnosis of some acute diseases, such as acute heart failure. However, the level of TNF-α in non-invasively obtained body ﬂuids is extremely low. For example, the TNF-α concentration in the saliva of healthy humans is lower than 3 pg ml−1 and increases to about 30 pg ml−1 in the saliva of patients with severe autoimmune diseases (Liu and Duan, 2012; Bellagambi et al., 2017). The concentration of urinary TNF-α was determined to be 4 pg mg−1 in healthy individuals, normalised by the concomitant urinary creatinine level (to compensate for the TNF-α variations in urine ﬂuid) (Bellagambi et al., 2017). Its concentration can be up to 196.9 ± 121.2 pg/ml in tear ﬂuid, and research is beginning to focus on how immune and inﬂammatory diseases affect the content of TNF-α in tear ﬂuid, which has a less complex biological composition (Comoglu et al., 2013; Costagliola et al., 2013). Therefore, methods offering low limits of detection and high sensitivity are of great interest for clinical diagnosis and research. (Brennan and Galvin, 2018; Yanez-Sedeno et al., 2019). Monitoring the trace concentration of TNF-α in these body ﬂuids provides a painless, simple, real-time means for cardiovascular disease diagnosis; in particular, it may serve as a sign of inﬂammation, which is important for the early diagnosis of some acute diseases, such as acute heart failure. However, the level of TNF-α in non-invasively obtained body ﬂuids is extremely low. NON-INVASIVE ELECTROCHEMICAL BIOSENSORS FOR SALIVARY TNF-α Immunosensors for Detection of TNF-α in Saliva electro-addressing of 4-carboxymethylaniline (CMA) on gold working electrode followed by CMA activation, C. biofunctionalisation through incubation with Ab-TNF-α, D. deactivation of the remaining active carboxylic acid groups by incubation in ethanolamine solution, E. incubation of the device in Ag-TNF-α solution, F. subsequent incubation in Ab-TNF-α-HRP solution. (B) Chronoamperometric analyses for different incubation of the immunosensor in real human saliva. Reproduced with permission from Barhoumi et al.(2018). Copyright 2018, Elsevier Science SA. (C) Schematic illustration of the electrochemical immunosensor for simultaneous determination of TNF-α and IL-1β using dual SPEs. (D) Amperometric responses recorded with the dual immunosensor for right: 25 or 0 pg ml−1 IL-1β (green), or 10 pg ml−1 TNF-a (blue) (right), and left: for 10 or 0 pg ml−1 TNF-a (blue), or 25 pg ml−1 IL-1β (green). Reproduced with permission from Sanchez-Tirado et al., 2017). Copyright 2017, Elsevier. (E) Schematic illustration of the immunosensor based on ITO thin ﬁlms covered by P3-conjugated polymer. Reproduced with permission from Aydin et al., 2017). Copyright 2017, Elsevier Advanced Technology. (F) Surface characterisation of the nanostructured electrode by scanning electron microscopy. (G) Schematic illustration of sandwich-type detection for nanostructured ITO electrode-based electrochemical immunosensors. Reproduced with permission from Pruna et al., 2018a). Copyright 2018, Pergamon-Elsevier Science Ltd. in carboxylic acid groups, which was beneﬁcial for the effective immobilisation of the anti-TNF-α antibodies. Using P3 as the conjugated matrix, the linear detection response of the developed immunosensor had an LOD of 3.7 fg ml−1, which was suitable for clinical salivary sample determination. To evaluate the practicability of the P3-based electrochemical immunosensor, it was used to quantify the level of TNF-α in clinical saliva from a local hospital, and it showed recoveries between 98.39 and 105.20%, demonstrating the good potential of the developed immunosensor to be employed as an accurate diagnostic tool for TNF-α determination. Using a similar strategy, the sensing performance of the TNF-α immunosensor could be improved by using a polymer-coated magnetic microparticles matrix for antibody immobilisation (Barhoumi et al., 2019). Although the above polymer-coated immunosensors could realise the improvement of the sensing properties, the employed electrode-modiﬁed strategy still requires polymers or other layers to cover the whole surface (Eletxigerra et al., 2015; Arya and Estrela, 2020; Nessark et al., 2020), which may require additional nanostructures to maintain a larger surface area (Hou et al., 2013; Baraket et al., 2017; Yola and Atar, 2021). For this purpose, Pruna et al. NON-INVASIVE ELECTROCHEMICAL BIOSENSORS FOR SALIVARY TNF-α Immunosensors for Detection of TNF-α in Saliva Both of these immunosensors used CMA as the conjugated and blocking layer; however, the weak electron transfer capacity limits the sensing performance to some extent (Barhoumi et al., 2019). To avoid this problem, Aydin et al. (2017) developed a label-free immunosensor for salivary TNF-α determination by applying a semiconductive polymer (poly (3-thiophene acetic acid, P3)) on an indium tin oxide (ITO) electrode (Figure 1E). Polymer P3 not only showed good conductivity but was also rich As a promising tool for non-invasive detection, the saliva matrix can comprise more than 25% of the proteins in the human serum with strong correlations. Beneﬁting from the characteristics of easy collection and storage as well as good stability, saliva detection can effectively decrease the demand for cost, sophisticated instruments, and professional operators. In this section, recent studies focused on the detection of salivary September 2021 \| Volume 9 \| Article 701045 Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org 2 Non-Invasive Biosensors for TNF-α Lu et al. Lu et al. FIGURE 1 \| (A) Schematic illustration of the immunosensor preparation: A. bare gold, B. electro-addressing of 4-carboxymethylaniline (CMA) on gold working electrode followed by CMA activation, C. biofunctionalisation through incubation with Ab-TNF-α, D. deactivation of the remaining active carboxylic acid groups by ncubation in ethanolamine solution, E. incubation of the device in Ag-TNF-α solution, F. subsequent incubation in Ab-TNF-α-HRP solution. (B) Chronoamperometric analyses for different incubation of the immunosensor in real human saliva. Reproduced with permission from Barhoumi et al.(2018). Copyright 2018, Elsevier Science SA. (C) Schematic illustration of the electrochemical immunosensor for simultaneous determination of TNF-α and IL-1β using dual SPEs. (D) Amperometric esponses recorded with the dual immunosensor for right: 25 or 0 pg ml−1 IL-1β (green), or 10 pg ml−1 TNF-a (blue) (right), and left: for 10 or 0 pg ml−1 TNF-a (blue), or 25 pg ml−1 IL-1β (green). Reproduced with permission from Sanchez-Tirado et al., 2017). Copyright 2017, Elsevier. (E) Schematic illustration of the immunosensor based on ITO thin ﬁlms covered by P3-conjugated polymer. Reproduced with permission from Aydin et al., 2017). Copyright 2017, Elsevier Advanced Technology. (F) Surface characterisation of the nanostructured electrode by scanning electron microscopy. (G) Schematic illustration of sandwich-type detection for nanostructured ITO electrode-based electrochemical immunosensors. Reproduced with permission from Pruna et al., 2018a). Copyright 2018, Pergamon-Elsevier Science Ltd. FIGURE 1 \| (A) Schematic illustration of the immunosensor preparation: A. bare gold, B. NON-INVASIVE ELECTROCHEMICAL BIOSENSORS FOR SALIVARY TNF-α Immunosensors for Detection of TNF-α in Saliva (2018a) constructed a nanostructured ITO electrode with a high surface-to-volume ratio to build opto-electrochemical sensors (Figure 1F). Together with the good optical transparency and electrical conductivity, a nanostructured ITO electrode was fabricated to sandwich the antibody-antigen type immunosensor for TNF-α recognition (Figure 1G). Because of its good linear range of 10–100 pg ml−1, the proposed immunosensor was expected to in carboxylic acid groups, which was beneﬁcial for the effective immobilisation of the anti-TNF-α antibodies. Using P3 as the conjugated matrix, the linear detection response of the developed immunosensor had an LOD of 3.7 fg ml−1, which was suitable for clinical salivary sample determination. To evaluate the practicability of the P3-based electrochemical immunosensor, it was used to quantify the level of TNF-α in clinical saliva from a local hospital, and it showed recoveries between 98.39 and 105.20%, demonstrating the good potential of the developed immunosensor to be employed as an accurate diagnostic tool for TNF-α determination. Using a similar strategy, the sensing performance of the TNF-α immunosensor could be improved by using a polymer-coated magnetic microparticles matrix for antibody immobilisation (Barhoumi et al., 2019). September 2021 \| Volume 9 \| Article 701045 Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org 3 Non-Invasive Biosensors for TNF-α Lu et al. FIGURE 2 \| (A) Schematic illustrations of the chemical surface modiﬁcation and biofunctionalisation process of the immunosensor with anti-TNF-α antibodies. a) Si3N4 surface cleaning and activation, (B) Si3N4 surface functionalisation with TESUD, (C) anti-TNF-α antibodies immobilisation step, d) Si3N4 surface blocking using 1% ethanolamine, and (E) electrochemical TNF-α detection step. (B) Detection sensitivity curves of Ag-TNF-α, IL-10 cytokine, and cortisol in artiﬁcial saliva. Reproduced with permission from Bahri et al. (2020). Copyright 2020, Elsevier.(C) Schematic illustrations of the three electrochemical aptamer-based TNF-α biosensors (A–C) and the optimal (minimum energy) structure of the adopted aptamer (D). (D) Alternating current voltammetry of the A1 sensor in the absence and presence of 100 nM TNF-α in a 50% synthetic urine sample. Reproduced with permission from Mayer and Lai (2018) Copyright 2018, Elsevier.(E) Schematic representation of the immunosensor biofunctionalisation process. Reproduced with permission from Cruz et al. (2019). Copyright 2019, American Chemical Society. FIGURE 2 \| (A) Schematic illustrations of the chemical surface modiﬁcation and biofunctionalisation process of the immunosensor with anti-TNF-α antibodies. NON-INVASIVE ELECTROCHEMICAL BIOSENSORS FOR SALIVARY TNF-α Immunosensors for Detection of TNF-α in Saliva a) Si3N4 surface cleaning and activation, (B) Si3N4 surface functionalisation with TESUD, (C) anti-TNF-α antibodies immobilisation step, d) Si3N4 surface blocking using 1% ethanolamine, and (E) electrochemical TNF-α detection step. (B) Detection sensitivity curves of Ag-TNF-α, IL-10 cytokine, and cortisol in artiﬁcial saliva. Reproduced with permission from Bahri et al. (2020). Copyright 2020, Elsevier.(C) Schematic illustrations of the three electrochemical aptamer-based TNF-α biosensors (A–C) and the optimal (minimum energy) structure of the adopted aptamer (D). (D) Alternating current voltammetry of the A1 sensor in the absence and presence of 100 nM TNF-α in a 50% synthetic urine sample. Reproduced with permission from Mayer and Lai (2018) Copyright 2018, Elsevier.(E) Schematic representation of the immunosensor biofunctionalisation process. Reproduced with permission from Cruz et al. (2019). Copyright 2019, American Chemical Society. trigger a stable response to trace TNF-α in salivary samples from healthy patients. in which the anti-TNF-α antibody was labelled with rhodamine. Mott-Schottky analyses of this capacitance electrochemical biosensor were used for TNF-α quantiﬁcation. The result showed that this biosensor had a sensitivity of 4.4 mV pM−1 with an LOD of 1 pg ml−1 in artiﬁcial saliva (Figure 2B). Furthermore, it displayed good selectivity against cortisol and IL-10. Other Electrochemical Biosensors Apart from the typical amperometric method, electrochemical biosensors based on the principle of capacitance as well as biosensors with aptamers as recognition species have also been developed for TNF-α in saliva. Bahri et al. (2020) studied a capacitance electrochemical biosensor using a silicon nitride (Si3N4) transducer (Figure 2A). After activation by the piranha solution, silanol and silylamine groups were introduced on the surface of the Si3N4 transducer in conjunction with bio/chemical substances. Then, a self- assembled monolayer of aldehyde-silane (11-(triethoxysilyl) undecanal) (TEUSD) was further modiﬁed for anti-TNF-α bonding, and the bonded condition was checked using the micro-contact printing technique. The sandwich antibody- antigen bio-recognition was chosen for TNF-α determination, Aptamers are single-stranded DNA or RNA molecules that can bind to speciﬁc targets in terms of base pairing (Tuerk and Gold, 1990; Ellington and Szostak, 1992). Unlike antibodies utilised in immunosensors, biosensors using aptamers as biorecognition components are more stable under different experimental conditions and are much cheaper than the generation of antibodies. Mayer and Lai. (2018) studied electrochemical aptamer-based TNF-α biosensors using three aptamer probes labelled with methylene blue (Figure 2C). When exposed to TNF-α, the conformation of the chosen aptamer probe changed, which adjusted the electron transfer September 2021 \| Volume 9 \| Article 701045 Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org 4 Non-Invasive Biosensors for TNF-α Lu et al. Lu et al. FIGURE 3 \| Fluorescence optical images of A: anti-TNF-α/TNF-α-modiﬁed working electrode (WE) after incubation in secondary ﬂuorescent TNF-α, B: no functionalised WE, and C: optical image of the fully integrated BioMEMS. WE: working electrode, CE: counter electrode, RE: reference electrode. Reproduced with permission from Longo et al. (2016). Copyright 2016, Elsevier Science BV. (B) Nyquist impedance plots obtained from the standard addition method performed on a real saliva sample (black point) Ab TNF-α for Ag TNF-α (red point) Level 1 (corresponding to an addition of 0 pg ml−1); (blue point) Level 2 (addition of 3 pg ml−1); (green point) Level 3 (addition of 7 pg ml−1). Reproduced with permission from Bellagambi et al. (2017). Copyright 2017, Elsevier Science SA. (C) Manufacturing details of the printed circuit board and the biosensor array. The potentiostatic circuits for the different channels are labelled as 1–4 (top view) and 5–8 (bottom view). Other Electrochemical Biosensors Regarding the rest of the elements, A) is the PIC32 microcontroller; B) is the UART connector for PC communication; C) is the external power supply; D) is the 10-pin connector for the biosensor array; E) is the RJ-11 connector to program the microcontrollers; F) is the data bus; G) is the LM1117 voltage regulator; H) is the MCP1603 B voltage regulator; I) is the ADR4525 stable voltage reference; J) is the PIC24 microcontroller, and K) is the TC962 voltage regulator. Reproduced with permission from Pruna et al. (2018b). Copyright 2018, Elsevier Advanced Technology. (D) KardiaTool concept. (E) High level architecture of the KardiaTool platform. Reproduced with permission Tripoliti et al. (2018). Copyright 2018, IEEE Engineering in Medicine and Biology Society. FIGURE 3 \| Fluorescence optical images of A: anti-TNF-α/TNF-α-modiﬁed working electrode (WE) after incubation in secondary ﬂuorescent TNF-α, B: no functionalised WE, and C: optical image of the fully integrated BioMEMS. WE: working electrode, CE: counter electrode, RE: reference electrode. Reproduced with permission from Longo et al. (2016). Copyright 2016, Elsevier Science BV. (B) Nyquist impedance plots obtained from the standard addition method performed on a real saliva sample (black point) Ab TNF-α for Ag TNF-α (red point) Level 1 (corresponding to an addition of 0 pg ml−1); (blue point) Level 2 (addition of 3 pg ml−1); (green point) Level 3 (addition of 7 pg ml−1). Reproduced with permission from Bellagambi et al. (2017). Copyright 2017, Elsevier Science SA. (C) Manufacturing details of the printed circuit board and the biosensor array. The potentiostatic circuits for the different channels are labelled as 1–4 (top view) and 5–8 (bottom view). Regarding the rest of the elements, A) is the PIC32 microcontroller; B) is the UART connector for PC communication; C) is the external power supply; D) is the 10-pin connector for the biosensor array; E) is the RJ-11 connector to program the microcontrollers; F) is the data bus; G) is the LM1117 voltage regulator; H) is the MCP1603 B voltage regulator; I) is the ADR4525 stable voltage reference; J) is the PIC24 microcontroller, and K) is the TC962 voltage regulator. Reproduced with permission from Pruna et al. (2018b). Copyright 2018, Elsevier Advanced Technology. (D) KardiaTool concept. (E) High level architecture of the KardiaTool platform. Reproduced with permission Tripoliti et al. (2018). Copyright 2018, IEEE Engineering in Medicine and Biology Society. Other Electrochemical Biosensors investigate the practical applicability in clinical samples, the immunosensor was applied to measure the trace concentration of TNF-α in tears, blood serum, and human cerebral spinal ﬂuid. As a result, the immunosensor could detect TNF-α in tears beyond the LOD of ELISA (4 pg ml−1). capacity of the labelled methylene blue. As a result, the corresponding redox current was altered. Among these biosensors, the one employing the aptamer with methylene blue modiﬁed at the distal end (A1 biosensor, as shown in Figure 2C) had the best sensing performance. This biosensor showed an LOD of 100 pM and good reproducibility. Moreover, it exhibited a good response in synthetic ﬂuid with 50% urine and 50% saliva (Figure 2D). Smart Point-of-Care Biosensors With the rapid development of technology, the need for intelligent POC devices in clinical and home self-monitoring is rapidly increasing, especially for patients with cardiovascular disease. Longo et al. (2016) constructed an integrated electrochemical bioMEMS for TNF-α cytokine detection. Because the BioMEMS had eight gold working microelectrodes (integrated reference and counter electrodes on one chip), it allowed for the simultaneous detection of multiple target molecules through the antibody-antigen bio-recognition mode (Figure 3A). Using EIS characterisation, the BioMEMS platform exhibited a linear detection range of 1–15 pg ml−1 in artiﬁcial human saliva against other cytokines, such as IL-1 and IL-8. They further applied the same system for TNF-α determination in In addition to saliva, electrochemical biosensors for TNF-α in other non-invasively obtained body ﬂuids, such as tears, were also studied. Cruz et al. (2019) built a highly sensitive label-free immunosensor for TNF-α screening (Figure 2E). The SPE (screen-printed electrodes) with gold as the working part was modiﬁed with sulfo-LC-SPDP (sulfosuccinimidyl 6-(3’-(2- pyridyldithio) propionamido) hexanoate) as a crosslinker for anti-TNF-α immobilisation. After blocking with bovine serum albumin, the biosensor was further bonded with TNF-α for quantitative determination. The sensing performance of the as-built immunosensor was tested using Nyquist plots. As calculated, it exhibited an LOD of 0.085 pg ml−1 in PBS. To September 2021 \| Volume 9 \| Article 701045 Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org 5 Non-Invasive Biosensors for TNF-α Lu et al. TABLE 1 \| A comparison of the sensing performance of non-invasive immunosensors discussed in the review. Immunosensor structure Sensor type Selectivitya Sensitivity Linear range (pg ml−1) LOD (pg ml−1) References Au WE/CMA/Ab-TNF-α/antigen-Ag- TNF-α/Ab-conjugate-HRP Chronoampemetric >3.5 I (μA) 0.0758C (ppt)-0.2125, R2 0.993 1–30 1 Barhoumi et al. (2018) SPCE/DWCNTs/Phe-Ab-TNF/TNF- α/Biotin-Ab-TNF/poly-HRP Amperometric >5 I (nA) (766.3 ± 0.2) logC (pg mL−1) + (237.8 ± 0.2) R2 0.999 1–200 0.85 Sanchez-Tirado et al. (2017) Si3N4 ISFET/TESUD/Ab-TNF-α Electochemical >2.5 Not mentioned 5–20 5 Vozgirdaite et al. (2021) ELISA plate/PEG/G4-OH/Ab- MAB610 ELISA ∼3 A (O. D.) 0.1502 + 0.0037C (pg mL−1) R 0.996 0–300 0.48 Bosnjakovic et al. (2012) Au WE/CMA/Ab-TNF-α Potentiostat >1.3 Not mentioned 266–666,000 266 Pruna et al. (2018b) ITO/Polymer-P3/NHS-EDC/Ab- TNF-α/BSA Impedimetric >2.5 A (kΩ) 1,381.8C (pg mL−1)+54.811, R2 0.998 0.01–2 0.0037 Aydin et al., 2017 Si3N4 WE/TESUD/Ab-TNF-α/Et- Amine Capacitance >2 Y (mV) 0.004C (pM) +0.058, R2 0.9908 1–30 1 Bahri et al. CONCLUSION human saliva with a linear range of 1–100 pg ml−1 and an LOD of 3.1 pg ml−1 (Figure 3B) (Bellagambi et al., 2017). Miniaturising the potentiostat in POC electrochemical biosensing systems while maintaining their effective characteristics is still a challenge. Using a similar BioMEMS electrode, Pruna et al. (2018b) integrated analogue circuitry, microcontrollers, and digital ﬁlter implementation to avoid the complex structures of the previously reported circuits, which have a total size of 12 × 7 × 2 cm3. This electronic system was further connected to BioMEMS electrodes for TNF-α determination (Figure 3C). This device had a high sensitivity in the randomly chosen range of 266 pg ml−1–666 ng ml−1. This review summarises the progress in electrochemical biosensors for non-invasive detection of TNF-α cytokines, including immunosensors for detection of TNF-α in saliva, electrochemical biosensors based on other principles or other non-invasive ﬂuids (such as tears), and smart POC biosensors. Among these sensors, saliva testing plays an important role in non-invasive TNF-α cytokine determination because of its ability to be easily collected, good stability, and suitable TNF-α concentration within the detection capability of electrochemical biosensors. These electrochemical biosensors for non-invasive detection of TNF-α cytokines have promising prospects for continuous, fast, real-time, and portable monitoring of cardiovascular disease. Moreover, smart POC biomarkers provide a great solution for chronic disease self-management, such as chronic heart failure. To make a clearer comparison, the sensing performance, such as sensitivity and speciﬁcity, of different non-invasive biosensors that discussed in this review are listed in Table 1. As compared, the most immunosensors can response to TNF-α concentration level of pg mL−1. Due to the different electrochemical methods, It’s hard to quantitatively compare the sensitivity. However, the selectivity is generally good, which is related to the well-designed structure and speciﬁc recognition of Ab-TNF-α. Especially, the Au WE/ sulfo-LC-SPDP/Ab-TNF-α immunosensor built by Cruz et al. exhibited a high speciﬁcity to TNF-α, which is at least 6 times higher than interferon gamma (IFNγ) and IL-4. This can be attributed to the introduction of SPDP monolayer which is considered to be effective for Ab-TNF-α immobilization. Tripoliti et al. (2018) presented an integrated device with the name of Kardiatool platform as an example for the non-invasive diagnosis of heart failure (Figure 3D). This device consists of two parts: a portable POC device (KardioPOC) and the corresponding software. Smart Point-of-Care Biosensors (2020) Au WE/sulfo-LC-SPDP/Ab-TNF-α Impedimetric >6 Y(Ω) 0.6199C (log (TNF-α, pg mL−1)) + 0.02926, R2 0.9852 1–25 0.085 Cruz et al. (2019) Au WE/CMA/Ab-TNF-α Impedimetric ∼1 Not mentioned 1–15 — Bellagambi et al. (2017) Au WE/CMA/Ab-TNF-α Impedimetric >5 Not mentioned 1–15 — Longo et al. (2016) ITO/CMA/Ab-TNF-α Impedimetric — Not mentioned 10–100 5 Pruna et al. (2018a) aThe selectivity is deﬁned as the ratio of response to TNF-α to that of other interferences. WE: working electrode, LOD: limit of detection; CMA:4-carboxymethylaniline; HRP: horseradish peroxidase; SPCE: screen-printed carbon electrodes; DWCNTs: double-walled carbon nanotubes; ISFET: ion-sensitive ﬁeld effect transistor TESUD: 11- (triethoxysilyl) undecanal ELISA: enzyme-linked immunosorbent assay; PEG: polyethylene-glycol; G4-OH: poly (amidoamine) dendrimer; NHS: N-hydroxysuccinimide; BSA: bovine serum albumin. A comparison of the sensing performance of non-invasive immunosensors discussed in the review. aThe selectivity is deﬁned as the ratio of response to TNF-α to that of other interferences. WE: working electrode, LOD: limit of detection; CMA:4-carboxymethylaniline; HRP: horseradish peroxidase; SPCE: screen-printed carbon electrodes; DWCNTs: double-walled carbon nanotubes; ISFET: ion-sensitive ﬁeld effect transistor TESUD: 11- (triethoxysilyl) undecanal ELISA: enzyme-linked immunosorbent assay; PEG: polyethylene-glycol; G4-OH: poly (amidoamine) dendrimer; NHS: N-hydroxysuccinimide; BSA: bovine serum albumin. REFERENCES D., Singh, P. K., Pandya, H. K., Tosi, J., Kim, C., Tewari, A., et al. (2017). Assessment of Neurotrophins and Inﬂammatory Mediators in Vitreous of Patients with Diabetic Retinopathy. Invest. Ophthalmol. Vis. Sci. 58, 5594–5603. doi:10.1167/iovs.17-21973 Aydın, E. B., Aydın, M., and Sezgintürk, M. K. (2018). Highly Sensitive Electrochemical Immunosensor Based on Polythiophene Polymer with Densely Populated Carboxyl Groups as Immobilization Matrix for Detection of Interleukin 1β in Human Serum and Saliva. Sensors Actuators B: Chem. 270, 18–27. doi:10.1016/j.snb.2018.05.014 Brennan, D., and Galvin, P. (2018). Flexible Substrate Sensors for Multiplex Biomarker Monitoring. MRS Commun. 8, 627–641. doi:10.1557/mrc.2018.134 Çomoğlu, S. S., Güven, H., Acar, M., Öztürk, G., and Koçer, B. (2013). Tear Levels of Tumor Necrosis Factor-Alpha in Patients with Parkinson’s Disease. Neurosci. Lett. 553, 63–67. doi:10.1016/j.neulet.2013.08.019 Bahri, M., Baraket, A., Zine, N., Ben Ali, M., Bausells, J., and Errachid, A. (2020). Capacitance Electrochemical Biosensor Based on Silicon Nitride Transducer for TNF-α Cytokine Detection in Artiﬁcial Human Saliva: Heart Failure (HF). Talanta 209, 120501. doi:10.1016/j.talanta.2019.120501 Costagliola, C., Romano, V., De Tollis, M., Aceto, F., dell’Omo, R., Romano, M. R., et al. (2013). TNF-alpha Levels in Tears: A Novel Biomarker to Assess the Degree of Diabetic Retinopathy. Mediators Inﬂamm. 2013, 1–6. doi:10.1155/2013/629529 Baraket, A., Lee, M., Zine, N., Sigaud, M., Bausells, J., and Errachid, A. (2017). A Fully Integrated Electrochemical Biosensor Platform Fabrication Process for Cytokines Detection. Biosens. Bioelectron. 93, 170–175. doi:10.1016/ j.bios.2016.09.023 Cruz, A., Queirós, R., Abreu, C. M., Barata, C., Fernandes, R., Silva, R., et al. (2019). Electrochemical Immunosensor for TNFα-Mediated Inﬂammatory Disease Screening. ACS Chem. Neurosci. 10, 2676–2682. doi:10.1021/acschemneuro.9b00036 Baraket, A., Lee, M., Zine, N., Yaakoubi, N., Bausells, J., and Errachid, A. (2016). A Flexible Electrochemical Micro Lab-On-Chip: Application to the Detection of Interleukin-10. Microchim Acta 183, 2155–2162. doi:10.1007/s00604-016- 1847-y Eletxigerra, U., Martinez-Perdiguero, J., and Merino, S. (2015). Disposable Microﬂuidic Immuno-Biochip for Rapid Electrochemical Detection of Tumor Necrosis Factor Alpha Biomarker. Sensors Actuators B: Chem. 221, 1406–1411. doi:10.1016/j.snb.2015.08.026 Baraket, A., Lee, M., Zine, N., Yaakoubi, N., Trivella, M. G., Zabala, M., et al. (2012). Cytokine Detection Using Diazonium Modiﬁed Gold Microelectrodes onto Polyimide Substrates with Integrated Ag/AgCl Reference Electrode. Proced. Eng. 47, 1181–1184. doi:10.1016/j.proeng.2012.09.362 Ellington, A. D., and Szostak, J. W. (1992). Selection In Vitro of Single-Stranded DNA Molecules that Fold into Speciﬁc Ligand-Binding Structures. Nature 355, 850–852. doi:10.1038/355850a0 Barhoumi, L., Baraket, A., Bellagambi, F. G., Karanasiou, G. S., Ali, M. B., Fotiadis, D. I., et al. (2018). REFERENCES Barhoumi, L., Bellagambi, F., Vivaldi, F., Baraket, A., Clément, Y., Zine, N., et al. (2019). Ultrasensitive Immunosensor Array for TNF-α Detection in Artiﬁcial Saliva Using Polymer-Coated Magnetic Microparticles onto Screen-Printed Gold Electrode. Sensors 19, 692. doi:10.3390/s19030692 Ali, M. K. M., Ibrahim, K., Hamad, O. S., Eisa, M. H., Faraj, M. G., and Azhari, F. (2011). Deposited Indium Tin Oxide (ITO) Thin Films by Dcmagnetron Sputtering on Polyethylene Terephthalate Substrate (PET). Rom. J Phys. 56, 730–741. Baydemir, G., Bettazzi, F., Palchetti, I., and Voccia, D. (2016). Strategies for the Development of an Electrochemical Bioassay for TNF-Alpha Detection by Using a Non-immunoglobulin Bioreceptor. Talanta 151, 141–147. doi:10.1016/ j.talanta.2016.01.021 Altintas, Z., Fakanya, W. M., and Tothill, I. E. (2014). Cardiovascular Disease Detection Using Bio-Sensing Techniques. Talanta 128, 177–186. doi:10.1016/ j.talanta.2014.04.060 Bellagambi, F. G., Baraket, A., Longo, A., Vatteroni, M., Zine, N., Bausells, J., et al. (2017). Electrochemical Biosensor Platform for TNF-α Cytokines Detection in Both Artiﬁcial and Human Saliva: Heart Failure. Sensors Actuators B: Chem. 251, 1026–1033. doi:10.1016/j.snb.2017.05.169 Arya, S. K., and Bhansali, S. (2011). Lung Cancer and its Early Detection Using Biomarker-Based Biosensors. Chem. Rev. 111, 6783–6809. doi:10.1021/ cr100420s Bellagambi, F. G., Lomonaco, T., Salvo, P., Vivaldi, F., Hangouët, M., Ghimenti, S., et al. (2020). Saliva Sampling: Methods and Devices. An Overview. Trac Trends Anal. Chem. 124, 115781. doi:10.1016/j.trac.2019.115781 Arya, S. K., and Estrela, P. (2020). Electrochemical ELISA Protein Biosensing in Undiluted Serum Using a Polypyrrole-Based Platform. Sensors 20, 2857. doi:10.3390/s20102857 Berthonneche, C., Sulpice, T., Boucher, F., Gouraud, L., De Leiris, J., O’connor, S. E., et al. (2004). New Insights into the Pathological Role of TNF-α in Early Cardiac Dysfunction and Subsequent Heart Failure after Infarction in Rats. Am. J. Physiology-Heart Circulatory Physiol. 287, H340–H350. doi:10.1152/ ajpheart.01210.2003 Arya, S. K., Prusty, A. K., Singh, S. P., Solanki, P. R., Pandey, M. K., Datta, M., et al. (2007). Cholesterol Biosensor Based on N-(2-aminoethyl)-3-aminopropyl- trimethoxysilane Self-Assembled Monolayer. Anal. Biochem. 363, 210–218. doi:10.1016/j.ab.2007.01.029 Aydın, E. B., Aydın, M., and Sezgintürk, M. K. (2017). A Highly Sensitive Immunosensor Based on ITO Thin Films Covered by a New Semi- conductive Conjugated Polymer for the Determination of TNFα in Human Saliva and Serum Samples. Biosens. Bioelectron. 97, 169–176. doi:10.1016/ j.bios.2017.05.056 Bosnjakovic, A., Mishra, M. K., Han, H. J., Romero, R., and Kannan, R. M. (2012). A Dendrimer-Based Immunosensor for Improved Capture and Detection of Tumor Necrosis Factor-α Cytokine. Analytica Chim. Acta 720, 118–125. doi:10.1016/j.aca.2012.01.017 Boss, J. AUTHOR CONTRIBUTIONS electrochemical biosensors, further optimisation of sensor reliability should be critically investigated, and biosensors that can reliably and accurately determine TNF-α in clinical settings are still in their infancy. Besides the TNF-α cytokine, lactate, 8- isoprostalandin F2α, C-reactive protein, and IL-1β from saliva can also act as biomarkers for cardiovascular disease (Aydın et al., 2018; Vilian et al., 2019; Ghimenti et al., 2020). Designing biosensors that can simultaneously detect multiple biomarkers is necessary to improve accuracy and reliability. Owing to the accessibility of non-invasive body ﬂuids, especially saliva, the design of portable and integrated devices for POC applications in remote and self-monitoring is promising, and related studies are needed. YL and LX conceived the idea. YL and QZ wrote the manuscript and equally contributed in the study with support from LX. FUNDING This work was supported by the National Natural Science Foundation of China (Grant Nos. 61874049, 61775080, 61822506), the Jilin Province Natural Science Foundation of China (No. 20200801017GH), and the Fundamental Research Funds for the Central Universities of China. CONCLUSION The KardioPOC has a ﬁrst layer with magnetic nanoparticles and antibodies for target species recognition, and then an added microﬂuidic system, immunosensor component, and saliva samples for measurement. The third layer is composed of integrated circuit sections, such as the power supply, memory, microcontroller, and communication module (Figure 3E). In addition, it can simultaneously measure four different biomarkers, such as TNF-α, IL-10, N-terminal pro b-type natriuretic peptide (NT-pro BNP), and cortisol. This platform provides a proof of concept for the POC device for heart failure evaluation. The above work is promising for the development of low-cost, portable, and time-saving electrochemical systems for POC TNF-α determination. However, POC platforms for in situ tracking of cardiovascular diseases are not satisfactory, and the development of new types of biosensors is still needed. Although research attention has been rapidly gained in the area of non-invasive TNF-α cytokine detection via September 2021 \| Volume 9 \| Article 701045 Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org 6 Lu et al. Lu et al. Non-Invasive Biosensors for TNF-α REFERENCES Proced. Eng. 168, 97–100. doi:10.1016/j.proeng.2016.11.156 Vilian, A. T. E., Kim, W., Park, B., Oh, S. Y., Kim, T., Huh, Y. S., et al. (2019). Efﬁcient Electron-Mediated Electrochemical Biosensor of Gold Wire for the Rapid Detection of C-Reactive Protein: A Predictive Strategy for Heart Failure. Biosens. Bioelectron. 142, 111549. doi:10.1016/ j.bios.2019.111549 Loo, S. W., and Pui, T.-S. (2020). Cytokine and Cancer Biomarkers Detection: The Dawn of Electrochemical Paper-Based Biosensor. Sensors 20, 1854. doi:10.3390/s20071854 Mayer, M. D., and Lai, R. Y. (2018). Effects of Redox Label Location on the Performance of an Electrochemical Aptamer-Based Tumor Necrosis Factor- Alpha Sensor. Talanta 189, 585–591. doi:10.1016/j.talanta.2018.07.055 Vozgirdaite, D., Ben Halima, H., Bellagambi, F. G., Alcacer, A., Palacio, F., Jaffrezic-Renault, N., et al. (2021). Development of an ImmunoFET for Analysis of Tumour Necrosis Factor-α in Artiﬁcial Saliva: Application for Heart Failure Monitoring. Chemosensors 9, 26. doi:10.3390/ chemosensors9020026 Nessark, F., Eissa, M., Baraket, A., Zine, N., Nessark, B., Zouaoui, A., et al. (2020). Capacitance Polypyrrole-based Impedimetric Immunosensor for Interleukin- 10 Cytokine Detection. Electroanalysis 32, 1795–1806. doi:10.1002/ elan.201900633 Yáñez-Sedeño, P., Campuzano, S., and Pingarrón, J. M. (2019). Pushing the Limits of Electrochemistry toward Challenging Applications in Clinical Diagnosis, Prognosis, and Therapeutic Action. Chem. Commun. 55, 2563–2592. doi:10.1039/c8cc08815b Povedano, E., Vargas, E., Montiel, V. R.-V., Torrente-Rodríguez, R. M., Pedrero, M., Barderas, R., et al. (2018). Electrochemical Afﬁnity Biosensors for Fast Detection of Gene-speciﬁc Methylations with No Need for Bisulﬁte and Ampliﬁcation Treatments. Sci. Rep. 8, 6418. doi:10.1038/s41598-018-24902-1 Yola, M. L., and Atar, N. (2021). Novel Voltammetric Tumor Necrosis Factor- Alpha (TNF-α) Immunosensor Based on Gold Nanoparticles Involved in Thiol-Functionalized Multi-Walled Carbon Nanotubes and Bimetallic Ni/ Cu-MOFs. Anal. Bioanal. Chem. 413, 2481–2492. doi:10.1007/s00216-021- 03203-z Pruna, R., Baraket, A., Bonhommé, A., Zine, N., Errachid, A., and López, M. (2018a). Novel Nanostructured Indium Tin Oxide Electrode for Electrochemical Immunosensors: Suitability for the Detection of TNF-α. Electrochimica Acta 283, 1632–1639. doi:10.1016/j.electacta.2018.07.066 Pruna, R., Palacio, F., Baraket, A., Zine, N., Streklas, A., Bausells, J., et al. (2018b). A Low-Cost and Miniaturized Potentiostat for Sensing of Biomolecular Species Such as TNF-α by Electrochemical Impedance Spectroscopy. Biosens. Bioelectron. 100, 533–540. doi:10.1016/j.bios.2017.09.049 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. Sadighbayan, D., Sadighbayan, K., Tohid-Kia, M. R., Khosroushahi, A. Y., and Hasanzadeh, M. (2019). REFERENCES A Novel Chronoamperometric Immunosensor for Rapid Detection of TNF-α in Human Saliva. Sensors Actuators B: Chem. 266, 477–484. doi:10.1016/j.snb.2018.03.135 Ghimenti, S., Lomonaco, T., Bellagambi, F. G., Biagini, D., Salvo, P., Trivella, M. G., et al. (2020). Salivary Lactate and 8-isoprostaglandin F2α as Potential Non- invasive Biomarkers for Monitoring Heart Failure: a Pilot Study. Sci. Rep. 10, 7441. doi:10.1038/s41598-020-64112-2 September 2021 \| Volume 9 \| Article 701045 Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org 7 Non-Invasive Biosensors for TNF-α Lu et al. Lu et al. Stefan, R. I., van Staden, J., and Aboul-Enein, H. (2000). Design and Use of Electrochemical Sensors in Enantioselective High Throughput Screening Hou, Y., Li, T., Huang, H., Quan, H., Miao, X., and Yang, M. (2013). Electrochemical Immunosensor for the Detection of Tumor Necrosis Factor α Based on Hydrogel Prepared from Ferrocene Modiﬁed Amino Acid. Sensors Actuators B: Chem. 182, 605–609. doi:10.1016/j.snb.2013.03.067 Stefan, R. I., van Staden, J., and Aboul-Enein, H. (2000). Design and Use of Electrochemical Sensors in Enantioselective High Throughput Screening of Drugs. A Minireview. Cchts 3, 445–454. doi:10.2174/ 1386207003331382 of Drugs. A Minireview. Cchts 3, 445–454. doi:10.2174/ 1386207003331382 ctuators B: Chem. 182, 605–609. doi:10.1016/j.snb.2013.03.067 Tripoliti, E. E., Karanasiou, G. S., Ioannidou, P., Toumpaniaris, P., Goletsis, Y., Baussels, J., Lomonaco, T., Pfeiffer, N., Anasthase, R., Tsopela, A., Leekens, B., Naka, K. K., Lourme, J.-C., Salvo, P., Liakopoulos, T., Jordan, J., Gallagher, J., Errachid, A., and Fotiadis, D. I. (2018). “KardiaTool: An Integrated POC Solution for Non-invasive Diagnosis and Therapy Monitoring of Heart Failure Patients,” in Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 3878–3881. doi:10.1109/embc.2018.8513298 Kalliolias, G. D., and Ivashkiv, L. B. (2016). TNF Biology, Pathogenic Mechanisms and Emerging Therapeutic Strategies. Nat. Rev. Rheumatol. 12, 49–62. doi:10.1038/nrrheum.2015.169 Lipsky, P. E., Van Der Heijde, D. M. F. M., St. Clair, E. W., Furst, D. E., Breedveld, F. C., Kalden, J. R., et al. (2000). Inﬂiximab and Methotrexate in the Treatment of Rheumatoid Arthritis. N. Engl. J. Med. 343, 1594–1602. doi:10.1056/ nejm200011303432202 Liu, J., and Duan, Y. (2012). Saliva: a Potential media for Disease Diagnostics and Monitoring. Oral Oncol. 48, 569–577. doi:10.1016/j.oraloncology.2012.01.021 Tuerk, C., and Gold, L. (1990). Systematic Evolution of Ligands by Exponential Enrichment: RNA Ligands to Bacteriophage T4 DNA Polymerase. Science 249, 505–510. doi:10.1126/science.2200121 Longo, A., Baraket, A., Vatteroni, M., Zine, N., Baussells, J., RogerFuocoDi Francesco, F., et al. (2016). Highly Sensitive Electrochemical BioMEMS for TNF-α Detection in Humansaliva: Heart Failure. Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org REFERENCES Development of Electrochemical Biosensors for Tumor Marker Determination towards Cancer Diagnosis: Recent Progress. Trac Trends Anal. Chem. 118, 73–88. doi:10.1016/j.trac.2019.05.014 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. Sánchez-Tirado, E., Salvo, C., González-Cortés, A., Yáñez-Sedeño, P., Langa, F., and Pingarrón, J. M. (2017). Electrochemical Immunosensor for Simultaneous Determination of Interleukin-1 Beta and Tumor Necrosis Factor Alpha in Serum and Saliva Using Dual Screen Printed Electrodes Modiﬁed with Functionalized Double-Walled Carbon Nanotubes. Analytica Chim. Acta 959, 66–73. doi:10.1016/j.aca.2016.12.034 Copyright © 2021 Lu, Zhou and Xu. 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. Sandborn, W. J., Feagan, B. G., Stoinov, S., Honiball, P. J., Rutgeerts, P., Mason, D., et al. (2007). Certolizumab Pegol for the Treatment of Crohn’s Disease. N. Engl. J. Med. 357, 228–238. doi:10.1056/NEJMoa067594 September 2021 \| Volume 9 \| Article 701045 Frontiers in Bioengineering and Biotechnology \| www.frontiersin.org 8
https://openalex.org/W3216770720	https://www.frontiersin.org/articles/10.3389/fpls.2021.777753/pdf	English	null	Origin of Intra-annual Density Fluctuations in a Semi-arid Area of Northwestern China	Frontiers in plant science	2,021	cc-by	7,419	ORIGINAL RESEARCH published: 22 November 2021 doi: 10.3389/fpls.2021.777753 Origin of Intra-annual Density Fluctuations in a Semi-arid Area of Northwestern China Our findings confirmed the hypothesis that the June-July drought induces latewood-like IADFs by limiting the process of cell enlargement in the xylem. Our finding suggests a higher occurrence of IADF in trees of arid and semi-arid climates of continental Asia if the changes to monsoon flows result in more frequent drought events during the earlywood formation in June. Intra-annual density fluctuation (IADF) is a structural modification of the tree ring in response to fluctuations in the weather. The expected changes in monsoon flow would lead to heterogeneous moisture conditions during the growing season and increase the occurrence of IADF in trees of the arid ecosystems of continental Asia. To reveal the timings and physiological mechanisms behind IADF formation, we monitored cambial activity and wood formation in Chinese pine (Pinus tabuliformis) during 2017–2019 at three sites in semi-arid China. We compared the dynamics of xylem formation under a drought event, testing the hypothesis that drought affects the process of cell enlargement and thus induces the production of IADF. Wood microcores collected weekly from April to October were used for anatomical analyses to estimate the timings of cambial activity, and the phases of enlargement, wall thickening, and lignification of the xylem. The first cells started enlargement from late April to early May. The last latewood cells completed differentiation in mid-September. Trees produced IADF in 2018. During that year, a drought in June limited cell production in the cambium, only 36% of the xylem cells being formed in IADF trees, compared to 68% in normal tree rings. IADF cells enlarged under drought in early July and started wall thickening during the rainfall events of late July. The drought restricted cell enlargement and affected wall thickening, resulting in narrow cells with wide walls. Cambium and cell enlargement recovered from the abundant rainfall, producing a new layer with large earlywood tracheids. IADF is a specific adaptation of trees to cope with water deficit events occurring during xylem formation. Our findings confirmed the hypothesis that the June-July drought induces latewood-like IADFs by limiting the process of cell enlargement in the xylem. Our finding suggests a higher occurrence of IADF in trees of arid and semi-arid climates of continental Asia if the changes to monsoon flows result in more frequent drought events during the earlywood formation in June. Keywords: water availability, IADF, xylogenesis, cambial activity, Chinese pine, cell enlargement Origin of Intra-annual Density Fluctuations in a Semi-arid Area of Northwestern China Jiani Gao 1,2,3, Sergio Rossi 3,4 and Bao Yang 1,5,6* 1 Key Laboratory of Desert and Desertification, Northwest Institute of Eco-Environment and Resources, Chinese Academy of Sciences, Lanzhou, China, 2 College of Resources and Environment, University of Chinese Academy of Sciences, Beijing, China, 3 Département des Sciences Fondamentales, Université du Quebec à Chicoutimi, Chicoutimi, QC, Canada, 4 Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems, Guangdong Provincial Key Laboratory of Applied Botany, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou, China, 5 Center for Excellence in Tibetan Plateau Earth Sciences, Chinese Academy of Sciences, Beijing, China, 6 Qinghai Research Center of Qilian Mountain National Park, Academy of Plateau Science and Sustainability and Qinghai Normal University, Xining, China Intra-annual density fluctuation (IADF) is a structural modification of the tree ring in response to fluctuations in the weather. The expected changes in monsoon flow would lead to heterogeneous moisture conditions during the growing season and increase the occurrence of IADF in trees of the arid ecosystems of continental Asia. To reveal the timings and physiological mechanisms behind IADF formation, we monitored cambial activity and wood formation in Chinese pine (Pinus tabuliformis) during 2017–2019 at three sites in semi-arid China. We compared the dynamics of xylem formation under a drought event, testing the hypothesis that drought affects the process of cell enlargement and thus induces the production of IADF. Wood microcores collected weekly from April to October were used for anatomical analyses to estimate the timings of cambial activity, and the phases of enlargement, wall thickening, and lignification of the xylem. The first cells started enlargement from late April to early May. The last latewood cells completed differentiation in mid-September. Trees produced IADF in 2018. During that year, a drought in June limited cell production in the cambium, only 36% of the xylem cells being formed in IADF trees, compared to 68% in normal tree rings. IADF cells enlarged under drought in early July and started wall thickening during the rainfall events of late July. The drought restricted cell enlargement and affected wall thickening, resulting in narrow cells with wide walls. Cambium and cell enlargement recovered from the abundant rainfall, producing a new layer with large earlywood tracheids. IADF is a specific adaptation of trees to cope with water deficit events occurring during xylem formation. Edited by: Ze-Xin Fan, Xishuangbanna Tropical Botanical Garden (CAS), China Reviewed by: Angela Balzano, University of Ljubljana, Slovenia Xin Song, Shenzhen University, China Correspondence: Bao Yang yangbao@lzb.ac.cn Reviewed by: Angela Balzano, University of Ljubljana, Slovenia Xin Song, Shen hen Uni e sit China Correspondence: Bao Yang yangbao@lzb.ac.cn Specialty section: This article was submitted to Functional Plant Ecology, a section of the journal Frontiers in Plant Science Received: 15 September 2021 Accepted: 26 October 2021 Published: 22 November 2021 INTRODUCTION The rise in temperature worldwide, associated with local reduction in precipitation, is expected to increase frequency and severity of warming-induced drought (IPCC. Climate Change, 2014; Wu et al., 2018). Vegetation and fertility of Eastern Asia benefits from the wet spring and summer resulting from precipitation due to the periodic monsoons (Ding et al., 2018). The asymmetric changes in land and ocean temperatures are expected to weaken the monsoon flows, resulting in a drier climate, with serious consequences for the arid Asian regions (Huang et al., 2017). The fluctuations of monsoon intensity would lead to diversified water vapor patterns and frequent extreme climate events, making trees growing in this area experience a more complicated hydrothermal environment. Severe drought events and fluctuations in moisture conditions affect stem growth and increase the production of abnormal tree ring structures and IADFs (Cuny and Rathgeber, 2016). This region is therefore an ideal location to investigate the responses of xylogenesis to drought. Intra-annual density fluctuation (IADF) is a structural modification of the tree ring involving an abrupt change in wood density (Battipaglia et al., 2016), producing a cell layer similar to the boundary of a true tree ring. IADF exhibit either latewood-like cells with thicker cell wall and narrower lumen area in earlywood intra-annual density fluctuations (E- IADFs), or earlywood-like cells with larger lumina and thinner walls in latewood intra-annual density fluctuation (L-IADFs; Campelo et al., 2007). Large tracheids are the key for efficient water conduction but are more vulnerable to cavitation and embolism. Thick cell walls contribute to the increase of wood density (Beeckman, 2016). IADF represents the ability of a species to adapt to the substantial changes in the growing conditions according to changes in xylogenesis, i.e., cambial activity or cell differentiation or both (Eilmann et al., 2011; Deslauriers et al., 2016; De Micco et al., 2016a). The morphology of IADF cells affects the hydraulic properties of wood and tree survival. In addition, an abrupt modification of the tree- ring structure due to IADF affects the physical or mechanical properties of the xylem and reduces wood quality and its potential use (Olivar et al., 2015).l In this study, we collected samples in Chinese pine (Pinus tabuliformis) in three sites located in the semi-arid region lying between the East Asian monsoon zone and the arid region of Northwestern China and investigated the dynamics of xylogenesis and IADF formation. Citation: November 2021 \| Volume 12 \| Article 777753 1 Frontiers in Plant Science \| www.frontiersin.org Intra-annual Density Fluctuation Gao et al. INTRODUCTION The objective of this study was to (i) assess the phenology of IADF and (ii) investigate the climatic drivers of IADF production. We tested the hypothesis that summer drought affects the process of cell enlargement and thus induces IADF. Intra-annual density fluctuations have been studied mainly in Mediterranean ecosystems, characterized by a long growing season with a hot and dry summer, inducing bimodal growth patterns in most conifers. Trees in Mediterranean ecosystems generally exhibit L-IADFs occurring after the summer drought, at the beginning of the autumnal rainfall (Battipaglia et al., 2010; De Micco et al., 2016b; Balzano et al., 2018). E-IADFs are observed in temperate forests, as a result of latewood cells developed within earlywood or at the transition between earlywood and latewood (Edmondson, 2010; Rozas et al., 2011; Gao et al., 2021). E-IADF is linked to drought conditions during the summer. IADFs occur in other environments such as tropical and boreal forests (Marchand and Filion, 2012; Palakit et al., 2015; Venegas-González et al., 2015). Despite the wide literature on L-IADF, knowledge on the mechanisms of the formation on E-IADF remains unclear in ecosystems outside the Mediterranean area, especially in arid inland areas. Based on the available literature, IADF is less investigated in China (Zhang et al., 2020; Gao et al., 2021). Frontiers in Plant Science \| www.frontiersin.org Study Site and Tree Selection Intra-Annual Density Fluctuation FIGURE 1 \| (A) Map of sampling sites, and (B,C) Walter and Lieth climatic diagram at Helan and Hasi during 1951–2019. Red line and blue line represent the monthly mean temperature and precipitation. Values on the left axis are the average maximum temperature of the warmest month and average minimum temperature of the coldest month. The upper right of the diagram shows an annual mean temperature and annual total precipitation. FIGURE 1 \| (A) Map of sampling sites, and (B,C) Walter and Lieth climatic diagram at Helan and Hasi during 1951–2019. Red line and blue line represent the monthly mean temperature and precipitation. Values on the left axis are the average maximum temperature of the warmest month and average minimum temperature of the coldest month. The upper right of the diagram shows an annual mean temperature and annual total precipitation. TABLE 1 \| Characteristics of the trees sampled in three sites in a semi-arid region of Northwestern China. a Trephor (Rossi et al., 2006a), and stored in 50% ethanol solution. Microcores were dehydrated through successive immersions in ethanol solutions and dewaxing agent, embedded in paraffin, cut in sections (8–12 μm thickness) using a rotary microtome, and then stained with a water solution of safranin and astral blue. Sections were examined under visible and polarized light at 200–400 × magnifications to discriminate cells in different stages of xylem differentiation. The number of cells in the cambial zone, enlargement, secondary wall thickening, and mature xylem cells were counted along three radial files. Cambial cells were flat with thin primary walls. Enlarging cells were at least twice the radial length of cambial cells, with thin cell walls. In spring, xylogenesis was considered to have started when at least one horizontal row of enlarging cells was observed (Rossi et al., 2006b). Cell wall thickening and mature cells were blue or light red and dark red, respectively, with thick walls and showed birefringence under polarized light. We consider that xylem formation was complete when all cells of the tree ring were mature. DBH (cm) Height (m) Age (years) Helan1 72 ± 9 7.6 ± 1.8 103 ± 7 Helan2 72 ± 12 10.5 ± 1.5 70 ± 10 Hasi 105 ± 23 12 ± 4 93 ± 25 Study Site and Tree Selection Study Site and Tree Selection We selected three sites in a semi-arid forest in Northwestern China (Figure 1A). Two sites (named as Helan1 and Helan2) were located at two altitudes (2010 m a.s.l. and 2,330 m a.s.l.) at the Helan Mountain national forest park in Ningxia (38°44′ N, 105°54′ E). The third site (named as Hasi) was located 400 km away, on Hasi Mountain (37°03′ N, 104°31′ E, 2410 m a.s.l.). This study area forms the boundary between the arid region in the Northwest and the East Asian monsoon zone. Chinese pine (Pinus tabuliformis) is one of the dominant tree species growing at the northwestern boundary of its distribution. The long-term climate recorded by the national weather stations during 1951–2019 showed similar mean annual temperature between the two regions, with 9.7 and 10°C in Helan and Hasi, respectively (Figures 1B,C). Total precipitation was 195 and 234 mm, with 65 and 55% of the rain falling in June–August, at Helan and Hasi, respectively. The climate is dry in winter and generally wet from spring to autumn. Hasi receives more precipitation in late spring and summer compared with Helan. Precipitation and temperature are the climatic factors most frequently considered in relation to IADF, although late frosts and defoliation may be involved in its formation (De Micco et al., 2016b). The study indicated that 72% of the years showing IADF in the tree ring of black pine had low precipitations in May (Wimmer et al., 2000). The frequency of IADFs varies greatly among trees with different species, ages, and tree ring widths. Different species show a different aptitude to form IADFs (Balzano et al., 2019). It is also suggested that younger trees with wider tree rings are more prone to form IADF than older trees with narrower rings (Schweingruber, 1996). Knowledge on the occurrence, frequency, and mechanisms of the production on IADF remains scarce and deserve more attention. Detailed analyses of xylem phenology at high time resolution can be a tool to characterize IADF phenology and its environment drivers. Sixteen dominant and healthy Chinese pines were chosen (eight trees per site at Helan1 and Helan2, 10 trees at Hasi). Trees at Hasi were bigger, i.e., larger and taller, and older than those at the other two sites (Table 1). The younger trees (70 years old) were located at Helan2. November 2021 \| Volume 12 \| Article 777753 2 Gao et al. Meteorological Datah Three automated weather stations were installed at each site to measure air temperature and precipitation every 30 min, and daily means were calculated from the half-an-hour time series. To quantify drought severity, the standardized precipitation evapotranspiration index (SPEI) was calculated according to the Hargreaves equation, as a monthly difference between precipitation and potential evapotranspiration, using “SPEI” package (Beguería et al., 2017) in R environment (R Core Team, 2017). IADFs Identification and Statistics Wood samples were collected in 2017–2019 weekly, at breast height (1.3 m above ground), from April to September, the main growing season of Chinese pine (Zeng et al., 2018) using We recorded the proportion of trees with IADFs in all sampling trees in each site and year; we compared difference in monthly November 2021 \| Volume 12 \| Article 777753 Frontiers in Plant Science \| www.frontiersin.org 3 Intra-annual Density Fluctuation Gao et al. March and April 2018, which was 4°C and 1.6°C higher than that in 2017 and 2019, respectively. Precipitation of June was lower in 2018 (27 mm) compared to 2017 (100 mm) and 2019 (102 mm). However, rain was more abundant in July and August in 2018 compared to the same period in 2017 and 2019. As a consequence, the climate in June was dry in 2018, with SPEI being <−1. In June 2017 and 2019, SPEI indicated wet conditions. July and August were wetter in 2018 compared to 2017 and 2019. Overall, SPEI of April and May 2018 was higher at Hasi (1.3 and 0.19) compared to Helan1 (−0.25 and − 0.58) and Helan2 (−0.23 and − 0.78). air temperature, precipitation, and SPEI using mixed models to assess which factor affected IADFs occurrence. Year and site were considered as random effects in the model. To identify the timings of IADFs formation, the number of cells in earlywood, IADFs, and latewood were counted and reported in the form of proportion. Tracheids were classified as earlywood or latewood according to the Mork’s formula, which classifies all cells with lumen areas of less than twice the thickness of a double cell wall as latewood (Denne, 1988). The cumulative proportion of (i) mature cells, (ii) mature cell and wall thickening cells, and (iii) total xylem cells (mature, wall thickening, and enlarging cells) in the tree ring for each day in 2018 were estimated using the Gompertz function. The function is defined as: May temperature was significantly higher at the sites where IADF occurred (Figure 5). Lower winter temperature was recorded at sites where IADF occurred. IADF was associated to lower precipitations in June and October, and higher precipitations in July and August. No significant difference in precipitation was observed in the other months. The monthly drought condition between the presence of and absence of IADF was contrary. Xylem Formation in 2018 To test our hypothesis, we focused on the xylogenesis in 2018, the year with IADF formation. The first enlarging cells, corresponding to the onset of xylem cell differentiation, were observed from late April to early May in 2018 (Figure 6). The process of cell enlargement lasted until late August at Hasi, and until mid-September at the two sites in Helan. Compared to Hasi, the number of enlarging cells increased more slowly at Helan1 and Helan2 at the beginning of the growing season. Cell wall thickening began in late May to early June and was completed in mid-September. The first mature xylem cells were observed in mid-June. Xylem differentiation, including cell enlargement and secondary wall formation, lasted for 112, 127 and 130 days at Helan1, Helan2 and Hasi, respectively. The total number of xylem cells increased slowly before DOY 200, with only 40% of the xylem cells being produced at the two sites in Helan. IADF Occurrenceh The sections of all sampled trees during 3 years were identified to investigate the occurrence of IADFs. No IADFs were observed at three sites in 2017 and 2019 (Table 2). A normal tree ring structure exhibits large, thin-walled earlywood cells, followed by small, heavy-walled latewood cells (Figure 2). E-IADFs were observed in earlywood of all trees from Helan1 and Helan2 sampled in 2018. E-IADF appeared in the form of narrow cells with thick walls within the earlywood zone. No IADF was observed at Hasi.i In 2018, 35% of xylem cells were classified as earlywood at Helan1 and Helan2 (Figure 3). IADFs accounted for about 12 and 16% of tree ring at Helan1 and Helan2, and 50% of the tree ring was latewood at the two sites. At Hasi, 68% of the tree ring was earlywood, and 32% was latewood. IADFs Identification and Statistics The climate was dry from March to June and humid from July to September in the IADF group, while it was humid from March to June and July to September in the absence of IADF group. Y t A kt ( ) = - - e e b where Y(t) is the cumulative proportion of cells at time t; A is the upper asymptote parameter; β is the x-axis placement parameter; and k is the rate of change parameter. The cumulative proportion of mature cells, mature cell and wall thickening cells, and total xylem cells represented the timings when the cells produced by the cambium underwent enlargement and wall thickening (Rossi et al., 2003). Frontiers in Plant Science \| www.frontiersin.org Comparing Climate Between Yearsh At the beginning of the growing season, xylem formation progressed slowly under low or absent precipitations at Helan1 and Helan2 (Figure 7), and only 35–37% of xylem cells were formed before DOY 190. IADF cells enlarged between DOY 190 and 210 at the Helan sites. In this period, Helan was warm, and precipitation was absent during the first half of the period when IADF cells enlarged. After DOY 200, the temperature decreased and precipitation recovered. IADF cells underwent wall thickening during DOY 200–220, under abundant rainfall. During IADF formation, the proportion of enlarging cells in the tree ring decreased, and only 12–16% of the tree ring belonging to IADF was formed. After DOY 220, cell number increased again, and more than 50% of the total amount of annual xylem cells was produced within 50 days. p g The years 2017 and 2019 were warmer than 2018. A lower precipitation was recorded in 2019 compared to the other 2 years (Figure 4). The mean air temperature was 6.3°C in Frontiers in Plant Science \| www frontiersin org TABLE 2 \| Occurrence of Intra-annual density fluctuations (IADFs) of the trees sampled in three sites in a semi-arid region of Northwestern China. Values are reported as trees showing IADFs on total sampled trees. 2017 2018 2019 Helan1 0/8 8/8 0/8 Helan2 0/8 8/8 0/8 Hasi 0/10 0/10 0/10 Frontiers in Plant Science \| www.frontiersin.org TABLE 2 \| Occurrence of Intra-annual density fluctuations (IADFs) of the trees sampled in three sites in a semi-arid region of Northwestern China. Values are reported as trees showing IADFs on total sampled trees. 2017 2018 2019 Helan1 0/8 8/8 0/8 Helan2 0/8 8/8 0/8 Hasi 0/10 0/10 0/10 TABLE 2 \| Occurrence of Intra-annual density fluctuations (IADFs) of the trees sampled in three sites in a semi-arid region of Northwestern China. Values are reported as trees showing IADFs on total sampled trees. November 2021 \| Volume 12 \| Article 777753 4 Gao et al. Intra-Annual Density Fluctuation FIGURE 2 \| Wood anatomy in tree-rings sampled at three sites during 2017–2019 in a semi-arid region of Northwestern China. Arrows indicate the IADF. FIGURE 2 \| Wood anatomy in tree-rings sampled at three sites during 2017–2019 in a semi-arid region of Northwestern China. tomy in tree-rings sampled at three sites during 2017–2019 in a semi-arid region of Northwestern China. Arrows indicate the IADF. DISCUSSION Earlywood intra-annual density fluctuation is a response to a drought event occurring during the growing season, mainly in the summer (Campelo et al., 2007, 2015). In this study, we investigated the dynamics of xylogenesis and IADF formation in Chinese pine (Pinus tabuliformis) in a semi-arid region of northwest China and found that IADF cells were produced under a prolonged summer drought and differentiate during the precipitation deficit in June and abundant rainfall in July and August. These findings suggest that summer drought triggers cell enlargement and affects IADF formation, which allowed us to accept our initial hypothesis. FIGURE 3 \| Proportion of earlywood, IADF, and latewood in tree ring in 2018 at three sites in a semi-arid region of Northwestern China. Comparing Climate Between Yearsh earlywood cells had completed cell enlargement at Hasi, which indicated a faster progression of xylem formation compared to Helan. No IADF cells were observed at Hasi, where cell enlargement of latewood cells progressed after DOY 190. FIGURE 3 \| Proportion of earlywood, IADF, and latewood in tree ring in 2018 at three sites in a semi-arid region of Northwestern China. IADF Formation and Its Environmental Control Regarding at Hasi site, precipitation during DOY 100–150 was three times higher compared to Helan, and 68% of the tree ring was produced before DOY 190. At that date, all the Regarding at Hasi site, precipitation during DOY 100–150 was three times higher compared to Helan, and 68% of the tree ring was produced before DOY 190. At that date, all the Latewood-like IADF cells enlarged during the drought event, although some of them completed enlargement at the beginning November 2021 \| Volume 12 \| Article 777753 Frontiers in Plant Science \| www.frontiersin.org 5 Gao et al. Gao et al. Intra-annual Density Fluctuation FIGURE 4 \| Monthly mean air temperature, precipitation and SPEI at three sites during 2017–2019. FIGURE 4 \| Monthly mean air temperature, precipitation and SPEI at three sites during 2017–2019. of the abundant rainfall (DOY 190–210). IADF cells started wall thickening when the precipitations had started again (DOY 200) and matured before mid-August. A significant decrease in the number of cambial and enlarging cells at the end of the drought was also observed, suggesting that drought inhibited cambial division and thus limited cell differentiation. Water availability is the primary climatic factor limiting stem growth in dry and warm regions (Rossi et al., 2014). Prolonged and severe summer droughts or high transpiration rates can disrupt the steady-state of water transfer in xylem cells and cause declines in water potential of the xylem (Bogeat-Triboulot et al., 2007). A low turgor pressure during enlargement prevents differentiating tracheids to increase in diameter, resulting in narrow, latewood-like cells (Steppe et al., 2015). Our finding is consistent with other studies, demonstrating that narrower cells are formed under water deficit occurring during formation (Rossi et al., 2009; Battipaglia et al., 2010). According to our findings, some IADF cells completed enlargement during the rainfall period. This unexpected result may represent the time lag between rainfall and the increase of water potential in xylem cells, or be related to a statistical error due to the weekly sampling used in this study. documented that carbohydrate in the stem, including structural C, such as cellulose and lignin, often accumulates under drought (Muller et al., 2011), suggesting that more carbon is available for the wall thickening process. Furthermore, the amount of materials of wall deposition stays almost constant in most of the cells in a tree ring (Cuny et al., 2014). IADF Formation and Its Environmental Control The reduction in size of IADF cells would lead to more wall materials allocated to wall thickening, producing thicker walls in IADF cells. The availability of soluble sugars regulates the wall deposition process (Cartenì et al., 2018). Our results indicate that climate drivers are uniform for cell enlarging and wall thickening of IADF cells: large IADF cells with thick walls were mainly induced by summer drought. Stangler et al. (2021) demonstrated that the transition from a dry June to a wet July initiated IADF formation in Norway spruce at lower elevations in Germany. Accordingly, we detected the specific IADF phenology under water deficit and showed evidence of the climate driver of enlargement and wall thickening during IADF formation. Frontiers in Plant Science \| www.frontiersin.org Frontiers in Plant Science \| www.frontiersin.org Water Availability and Cell Productionf Water Availability and Cell Production Drought affected both wood anatomy and xylem cell production. Growth dropped once water deficit began. Enlargement of the first earlywood cells occurred during the high precipitation of April-May at Hasi, resulting in 30% of the tree ring. In the same period, only 10% of xylem cells started enlargement at Helan, because of a lack of rain and despite the water provided by the snowmelt in spring. The rainfall in the early growing season at Hasi provided sufficient water availability, thus trees could maintain high photosynthetic rates for sustaining cambium division and xylem growth, processes that are large carbon sinks (Bréda et al., 2006; Zhang et al., 2020). Moreover, the precipitation in April and May might have helped trees to maintain a higher water potential and protect the xylem cells from the negative influence of drought. As a result, xylem cells at Hasi were able to differentiate and mature their secondary walls under dry conditions, thus producing tree rings lacking IADF. Conversely, dry conditions during March–June may limit stem growth and, in case of prolonged lack of rain, accelerate leaf senescence to maintain the water balance. This defense mechanism led to growth losses, as observed by the lower amount of xylem cells produced at Helan, compared to Hasi, before the July precipitations. Our weekly monitoring of xylem formation provides evidence of the limiting effect of drought on cambial activity and xylogenesis at Helan. Dendrochronological studies also demonstrated that water availability of March–July affected stem growth (Li et al., 2007; Cai, 2009). Moreover, the precipitation from late spring to early summer (May–June) played a role in stem growth of Sabina przewalskii in semi-arid northwestern China (Gou et al., 2015; Yang et al., 2019, 2021). FIGURE 5 \| Monthly air temperature, precipitation, and SPEI between absence and presence of IADF during 2017–2019 at three sites. Asterisks represent significant differences (p < 0.05) between the two categories. The Functional Explanation of the IADF Formationi IADF formation is a specific adaptation under summer drought. This modification in the xylem structure protects the trees from irreversible damage by xylem cavitation and vessel embolism at the cost of the reduction of water transfer efficiency IADF cells started secondary wall formation during the rainfall period, producing thick cell walls. It is suggested that wall thickness is encoded during the previous phase of cell enlargement (Cuny et al., 2014; Buttὸ et al., 2019). It is widely November 2021 \| Volume 12 \| Article 777753 6 Intra-Annual Density Fluctuation Gao et al. FIGURE 5 \| Monthly air temperature, precipitation, and SPEI between absence and presence of IADF during 2017–2019 at three sites. Asterisks represent significant differences (p < 0.05) between the two categories. flexible xylogenesis reflects the plasticity of our species, which allows trees to survive under water deficit and concentrate growth during favorable water supply conditions. Our study demonstrates the mechanism of IADF formation under summer drought and highlights the plasticity of xylogenesis to the varied water conditions. CONCLUSION As a structural anomaly of the tree ring, IADF is a suitable indicator of weather fluctuations occurring during the growing season. However, the physiological processes behind IADF formation and the impacts of the climate drivers remain partially unknown. To our knowledge, this is the first time that timings and dynamics of IADF formation have been assessed at high temporal resolutions. We investigated the weekly xylogenesis of Chinese pine at three sites in semi-arid continental China by describing the development of an E-IADF. Prolonged and severe drought in June and early July affected the processes of xylem production and differentiation and induced IADF. Specifically, drought limited enlargement and affected wall thickening, resulting in IADF cells with narrow tracheids and wide walls. Rainfall on April–May provided sufficient water availability for xylogenesis, protecting trees against summer drought. Warming-induced droughts may become more frequent and severe during the summer in the drier regions of the (Martin-Benito et al., 2013; Puchi et al., 2019). The physiological response to water limitation is to protect trees against irreversible damage when stress becomes severe and long (Claeys and Inzé, 2013). In other words, trees ensure their survival when a water stress is prolonged or becomes more severe.h g The rainfall of mid-July improved growing conditions and reactivated cambial activity and cell enlargement. In other words, once water availability is sufficient, xylogenesis recovers the expected activities of cell division and differentiation. During cell enlargement, the vacuoles loaded with sugars increase the turgor pressure and attract water inside the cells, then producing the large earlywood tracheids (Vieira et al., 2020). This mechanism failed under a Mediterranean climate, where a small amount of irrigation in September was unable to trigger a second period of cambial activity (Vieira et al., 2020). The observed November 2021 \| Volume 12 \| Article 777753 7 Gao et al. Gao et al. Intra-annual Density Fluctuation FIGURE 6 \| Proportion of tree ring of cells during different phases of xylem formation at three sites in 2018 in a semi-arid region of Northwestern China. Horizontal bars represent IADFs at Helan1 and Helan2. FIGURE 6 \| Proportion of tree ring of cells during different phases of xylem formation at three sites in 2018 in a semi-arid region of Northwestern China. Horizontal bars represent IADFs at Helan1 and Helan2. REFERENCES Cartenì, F., Deslauriers, A., Rossi, S., Morin, H., De Micco, V., Mazzoleni, S., et al. (2018). The physiological mechanisms behind the earlywood-to-latewood transition: a process-based modeling approach. Front. Plant Sci. 9:1053. doi: 10.3389/fpls.2018.01053 Balzano, A., Battipaglia, G., and De Micco, V. (2019). Wood-trait analysis to understand climatic factors triggering intra-annual density-fluctuations in co-occurring Mediterranean trees. IAWA J. 40, 241–258. doi: 10.1163/22941932-40190220 Claeys, H., and Inzé, D. (2013). The agony of choice: how plants balance growth and survival under water-limiting conditions. Plant Physiol. 162, 1768–1779. doi: 10.1104/pp.113.220921 Balzano, A., Čufar, K., Battipaglia, G., Merela, M., Prislan, P., Aronne, G., et al. (2018). Xylogenesis reveals the genesis and ecological signal of IADFs in Pinus pinea L. and Arbutus unedo L. Ann. Bot. 121, 1231–1242. doi: 10.1093/aob/mcy008 Cuny, H. E., and Rathgeber, C. B. K. (2016). Xylogenesis: coniferous trees of temperate forests are listening to the climate tale during the growing season but only remember the last words! Plant Physiol. 171, 306–317. doi: 10.1104/ pp.16.00037 Battipaglia, G., Campelo, F., Vieira, J., Grabner, M., De Micco, V., Nabais, C., et al. (2016). Structure and function of intra–annual density fluctuations: mind the gaps. Front. Plant Sci. 7:595. doi: 10.3389/fpls.2016.00595 Cuny, H. E., Rathgeber, C. B. K., Frank, D., Fonti, P., and Fournier, M. (2014). Kinetics of tracheid development explain conifer tree-ring structure. New Phytol. 203, 1231–1241. doi: 10.1111/nph.12871 Battipaglia, G., De Micco, V., Brand, W. A., Linke, P., Aronne, G., Saurer, M., et al. (2010). Variations of vessel diameter and δ13C in false rings of Arbutus unedo L. reflect different environmental conditions. New Phytol. 188, 1099–1112. doi: 10.1111/j.1469-8137.2010.03443.x De Micco, V., Balzano, A., Čufar, K., Aronne, G., Gričar, J., Merela, M., et al. (2016a). Timing of false ring formation in Pinus halepensis and Aruburus unedo in southern Italy: outlook from an analysis of xylogenesis and tree- ring chronologies. Front. Plant Sci. 7:705. doi: 10.3389/fpls.2016.00705 Beeckman, H. (2016). Wood anatomy and trait-based ecology. IAWA J. 37, 127–151. doi: 10.1163/22941932-20160127 Beguería, S., Vicente-Serrano, S.M., and Beguería, M.S. (2017). SPEI: Calculation of the Standardised Precipitation-Evapotranspiration Index. Available at: https://CRAN.R-project.org/package=SPEI (Accessed July 7, 2021).h De Micco, V., Campelo, F., De Luis, M., Bräuning, A., Grabner, M., Battipaglia, G., et al. (2016b). Intra-annual density fluctuations in tree rings: how, when, where, and why? IAWA J. 37, 232–259. doi: 10.1163/22941932-20160132i Bogeat-Triboulot, M. B., Brosché, M., Renaut, J., Jouve, L., Thiec, D. L., Fayyaz, P., et al. (2007). REFERENCES Gradual soil water depletion results in reversible changes of gene expression, protein profiles, ecophysiology, and growth performance in Populus euphratica, a poplar growing in arid regions. Plant Physiol. 143, 876–892. doi: 10.1104/pp.106.088708 Denne, M. P. (1988). Definition of latewood according to Mork. IAWA J. 10, 59–62. doi: 10.1163/22941932-90001112h Deslauriers, A., Huang, J., Balducci, L., Beaulieu, M., and Rossi, S. (2016). The contribution of carbon and water in modulating wood formation in black spruce saplings. Plant Physiol. 170, 2072–2084. doi: 10.1104/pp.15.01525 Bréda, N., Huc, R., Granier, A., and Dreye, E. (2006). Temperate forest trees and stands under severe drought: a review of ecophysiological responses, adaptation processes and long-term consequences. Ann. Forest Sci. 63, 625–644. doi: 10.1051/forest:2006042 Ding, Y., Si, D., Liu, Y., Wang, Z., Li, Y., Zhao, L., et al. (2018). On the characteristics, driving forces and inter-decadal variability of the east Asian summer monsoon. Chin. J. Atmos. Sci. 42, 533–558. doi: 10.3878/j.issn.1006-9895.1712.17261hi Edmondson, J. R. (2010). The meteorological significance of false rings in eastern redcedar (Juniperus virginiana L.) from the southern Great Plains, USA. Tree Ring Res. 66, 19–33. doi: 10.3959/2008-13.1 Buttὸ, V., Rossi, S., Deslauriers, A., and Morin, H. (2019). Is size an issue of time? Relationship between the duration of xylem development and cell traits. Ann. Bot. 123, 1257–1265. doi: 10.1093/aob/mcz032 Eilmann, B., Zweifel, R., Buchmann, N., Pannatier, E. G., and Rigling, A. (2011). Drought alters timing, quantity, and quality of wood formation in scots pine. J. Exp. Bot. 62, 2763–2771. doi: 10.1093/jxb/erq443 Cai, Q. (2009). Response of Pinus tabuliformis tree-ring growth to three moisture indices and January to July Walter index reconstruction in Helan Mountain. Marine Ecol. Quaternary. Geol. 29, 131–136. doi: 10.3724/SP.J.1140.2009. 06131 Gao, J., Yang, B., Peng, X., and Rossi, S. (2021). Tracheid development under a drought event producing intra-annual density fluctuations in the semi-arid China. Agric. For. Meteorol. 308-309:108572. doi: 10.1016/j.agrformet.2021. 108572 Campelo, F., Nabais, C., Freitas, H., and Gutiérrez, E. (2007). Climatic significance of tree-ring width and intra-annual density fluctuations in Pinus pinea from a dry Mediterranean area in Portugal. Ann. For. Sci. 64, 229–238. doi: 10.1051/forest:2006107 Gou, X., Deng, Y., Gao, L., Chen, F., Cook, E., Yang, M., et al. (2015). Millennium tree-ring reconstruction of drought variability in the eastern Qilian Mountains, Northwest China. Clim. Dyn. 45, 1761–1770. doi: 10.1007/s00382-014- 2431-y Campelo, F., Vieira, J., Battipaglia, G., De Luis, M., Nabais, C., Freitas, H., et al. (2015). FUNDING This work was jointly supported by the National Natural Science Foundation of China (grant no. 42130511 and 41520104005), and a State Scholarship Fund provided by the China Scholarship Council (202004910757). This work was jointly supported by the National Natural Science Foundation of China (grant no. 42130511 and 41520104005), and a State Scholarship Fund provided by the China Scholarship Council (202004910757). AUTHOR CONTRIBUTIONS We are grateful to the two reviewers for their helpful and constructive comments. We thank Linzhou Xia for the help of samples collection. We thank Alison Garside for English correction of the text. JG, SR, and BY designed the study and revised the manuscript. JG and SR performed the statistical analysis. JG wrote the fluctuations in Pinus pinaster: age or size? Trees Struct. Funct. 29, 237–245. doi: 10.1007/s00468-014-1108-9 DATA AVAILABILITY STATEMENT The original contributions presented in the study are included in the article/supplementary material, and further inquiries can be directed to the corresponding author. CONCLUSION FIGURE 7 \| Xylem formation described by Gompertz functions and weather at three sites in 2018 in a semi-arid region of Northwestern China. Horizontal bars FIGURE 7 \| Xylem formation described by Gompertz functions and weather at three sites in 2018 in a semi-arid region of Northwestern China. Horizontal bars represent IADFs at Helan1 and Helan2. Black and green rectangles correspond to the time windows of enlargement and cell-wall thickening of IADF cells. FIGURE 7 \| Xylem formation described by Gompertz functions and weather at three sites in 2018 in a semi-arid region of Northwestern China. Horizontal bars represent IADFs at Helan1 and Helan2. Black and green rectangles correspond to the time windows of enlargement and cell-wall thickening of IADF cells. November 2021 \| Volume 12 \| Article 777753 8 Gao et al. Intra-Annual Density Fluctuation first version of the manuscript. All authors contributed to the article and approved the submitted version. northern hemisphere, enhancing the occurrence of IADF in trees. The formation of IADF appears to be a defense mechanism of trees against unfavorable environmental conditions, which ensures plasticity to the conducting system of trees and the capacity to tolerate and resist to water deficits. REFERENCES Which matters most for the formation of intra-annual density November 2021 \| Volume 12 \| Article 777753 Frontiers in Plant Science \| www.frontiersin.org 9 Intra-annual Density Fluctuation Gao et al. Huang, J., Li, Y., Fu, C., Chen, F., Fu, Q., Dai, A., et al. (2017). Dryland climate change: recent progress and challenges. Rev. Geophys. 55, 719–778. doi: 10.1002/2016RG000550 Stangler, D. F., Kahle, H.-P., Raden, M., Larysch, E., Seifert, T., and Spiecker, H. (2021). Effects of intra-seasonal drought on kinetics of tracheid differentiation and seasonal growth dynamics of Norway spruce along an elevational gradient. Forests 12:274. doi: 10.3390/f12030274 IPCC. Climate Change 2014: Mitigation of Climate Change: Working Group III Contribution to the IPCC Fifth Assessment Report. Cambridge, UK: Cambridge University Press. Steppe, K., Sterck, F., and Deslauriers, A. (2015). Diel growth dynamics in tree stems: linking anatomy and ecophysiology. Trends Plant Sci. 20, 335–343. doi: 10.1016/j.tplants.2015.03.015 Li, J., Chen, F., Cook, E. R., Gou, X., and Zhang, Y. (2007). Drought reconstruction for North China from tree rings: the value of the palmer drought severity index. Int. J. Climatol. 27, 903–909. doi: 10.1002/joc.1450 Venegas-González, A., von Arx, G., Chagas, M. P., and Tomazello Filho, M. (2015). Plasticity in xylem anatomical traits of two tropical species in response to intra-seasonal climate variability. Trees Struct. Funct. 29, 423–435. doi: 10.1007/s00468-014-1121-z Marchand, N., and Filion, L. (2012). False rings in the white pine (Pinus strobus) of the Outaouais Hills, Québec (Canada), as indicators of water stress. Can. J. For. Res. 42, 12–22. doi: 10.1139/x11-151l Vieira, J., Carvalho, A., and Campelo, F. (2020). Tree growth under climate change: evidence from xylogenesis timings and kinetics. Front. Plant Sci. 11:90. doi: 10.3389/fpls.2020.00090 Martin-Benito, D., Beeckman, H., and Canellas, I. (2013). Influence of drought on tree rings and tracheid features of Pinus nigra and Pinus sylvestris in a Mesic Mediterranean forest. Eur. J. For. Res. 132, 33–45. doi: 10.1007/ s10342-012-0652-3 Wimmer, R., Strumia, G., and Holawe, F. (2000). Use of false rings in Austrian pine to reconstruct early growing season precipitation. Can. J. For. Res. 30, 1691–1697. doi: 10.1139/x00-095 Muller, B., Pantin, F., Génard, M., Turc, O., Freixes, S., Piques, M., et al. (2011). Water deficits uncouple growth from photosynthesis, increase C content, and modify the relationships between C and growth in sink organs. J. Exp. Bot. 62, 1715–1729. doi: 10.1093/jxb/erq438 Wu, X., Liu, H., Li, X., Ciais, P., Babst, F., Guo, W., et al. (2018). REFERENCES Differentiating drought legacy effects on vegetation growth over the temperate northern hemisphere. Glob. Chang. Biol. 24, 504–516. doi: 10.1111/gcb. 13920 J. Exp. Bot. 62, 1715–1729. doi: 10.1093/jxb/erq438 Olivar, J., Rathgeber, C., and Bravo, F. (2015). Climate change, tree-ring width and wood density of pines in Mediterranean environments. IAWA J. 36, 257–269. doi: 10.1163/22941932-20150098l Yang, B., Qin, C., Bräuning, A., Osborn, T. J., Trouet, V., Ljungqvist, F. C., et al. (2021). Long-term decrease in Asian monsoon rainfall and abrupt climate change events over the past 6,700 years. Proc. Natl. Acad. Sci. U. S. A. 118:e2102007118. doi: 10.1073/pnas.2102007118 Palakit, K., Duangsathaporn, K., and Siripatanadilok, S. (2015). Climatic fluctuations trigger false ring occurrence and radial-growth variation in teak (Tectona grandis L.f.). iForest 9, 286–293. doi: 10.3832/ifor1100-008 Yang, B., Wang, J. L., and Liu, J. J. (2019). A 1556 year-long early summer moisture reconstruction for the Hexi corridor, northwestern China. Sci. China Earth Sci. 62, 953–963. doi: 10.1007/s11430-018-9327-1f g Puchi, P. F., Castagneri, D., Rossi, S., and Carrer, M. (2019). Wood anatomical traits in black spruce reveal latent water constraints on the boreal forest. Glob. Chang. Biol. 26, 1767–1777. doi: 10.1111/gcb.14906 Zeng, Q., Rossi, S., and Yang, B. (2018). Effects of age and size on xylem phenology in two conifers of northwestern China. Front. Plant Sci. 8:2264. doi: 10.3389/fpls.2017.02264 R Core Team (2017). R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing. Zhang, J., Alexander, M. R., Gou, X., Deslauriers, A., Fonti, P., Zhang, F., et al. (2020). Extended xylogenesis and stem biomass production in Juniperus przewalskii Kom. During extreme late-season climatic events. Ann. For. Sci. 77:99. doi: 10.1007/s13595-020-01008-1 Rossi, S., Anfodillo, T., and Menardi, R. (2006a). Trephor: a new tool for sampling microcores from tree stems. IAWA J. 27, 89–97. doi: 10.1163/22941932- 90000139 Zhang, J., Alexander, M. R., Gou, X., Deslauriers, A., Fonti, P., Zhang, F., et al. (2020). Extended xylogenesis and stem biomass production in Juniperus przewalskii Kom. During extreme late-season climatic events. Ann. For. Sci. 77:99. doi: 10.1007/s13595-020-01008-1 Rossi, S., Deslauriers, A., and Anfodillo, T. (2006b). Assessment of cambial activity and xylogenesis by microsampling tree species: an example at the alpine timberline. IAWA J. 27, 383–394. doi: 10.1163/22941932-90000161 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. Rossi, S., Deslauriers, A., and Morin, H. Frontiers in Plant Science \| www.frontiersin.org REFERENCES (2003). Application of the Gompertz equation for the study of xylem cell development. Dendrochronologia 21, 33–39. doi: 10.1078/1125-7865-00034 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. Rossi, S., Girard, M. J., and Morin, H. (2014). Lengthening of the duration of xylogenesis engenders disproportionate increases in xylem production. Glob. Chang. Biol. 20, 2261–2271. doi: 10.1111/gcb.12470 Rossi, S., Simard, S., Rathgeber, C. B. K., Deslauriers, A., and De Zan, C. (2009). Effects of a 20-day-long dry period on cambial and apical meristem growth in Abies balsamea seedlings. Trees Struct. Funct. 23, 85–93. doi: 10.1007/s00468-008-0257-0 Copyright © 2021 Gao, Rossi and Yang. 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. Rozas, V., García-González, I., and Zas, R. (2011). Climatic control of intra- annual wood density fluctuations of Pinus pinaster in NW Spain. Trees Struct. Funct. 25, 443–453. doi: 10.1007/s00468-010-0519-5 Schweingruber, F.H. (1996). Tree Rings and Environment: Dendroecology. Berne: Paul Haupt AG Bern. November 2021 \| Volume 12 \| Article 777753 Frontiers in Plant Science \| www.frontiersin.org Frontiers in Plant Science \| www.frontiersin.org 10
https://openalex.org/W2110604315	https://jwcn-eurasipjournals.springeropen.com/counter/pdf/10.1186/1687-1499-2011-87	English	null	Power allocation, bit loading and sub-carrier bandwidth sizing for OFDM-based cognitive radio	EURASIP Journal on wireless communications and networking	2,011	cc-by	20,915	© 2011 Thumar et al; licensee Springer. 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 The function of the Radio Resource Management module of a Cognitive Radio (CR) system is to evaluate the available resources and assign them to meet the Quality of Service (QoS) objectives of the Secondary User (SU), within some constraints on factors which limit the performance of the Primary User (PU). While interference mitigation to the PU spectral band from the SU’s transmission has received a lot of attention in recent literature; the novelty of our work is in considering a more realistic and effective approach of dividing the PU into sub-bands, and ensuring that the interference to each of them is below a specified threshold. With this objective, and within a power budget, we execute the tasks of power allocation, bit loading and sizing the sub-carrier bandwidth for an orthogonal frequency division multiplexing (OFDM)-based SU. After extensively analyzing the solution form of the optimization problems posed for the resource allocation, we suggest iterative algorithms to meet the aforementioned objectives. The algorithm for sub-carrier bandwidth sizing is novel, and not previously presented in literature. A multiple SU scenario is also considered, which entails assigning sub-carriers to the users, besides the resource allocation. Simulation results are provided, for both single and multi-user cases, which indicate the effectiveness of the proposed algorithms in a CR environment. Keywords: cognitive radio, OFDM, interference mitigation, power allocation, bit loading, sub-carrier bandwidth sizing itive radio, OFDM, interference mitigation, power allocation, bit loading, sub-carrier bandwidth sensing, spectrum allocation, spectrum sharing and spectrum mobility, one of the key functions of CR nodes in spectrum-aware xG networks is spectrum utilization. The spectrum utilization function entails efficient Radio Resource Management (RRM), the aim of which is to evaluate the available resources (power, time slots, band- width, etc) and assign them to meet the QoS objectives of the SU, within some constraints on factors (typically interference) which limit the performance of the PU [3]. Furthermore, for optimum spectrum utilization it is necessary to be adaptive to, one or more, time-varying characteristics of the system, such as the wireless chan- nel state, number of users, QoS requirements, etc. Vinay Thumar1, Taskeen Nadkar1, Tej Gopavajhula1, Uday B Desai2 and Shabbir N Merchant1 Vinay Thumar1, Taskeen Nadkar1, Tej Gopavajhula1, Uday B Desai2 and Shabbir N Merchant1 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Open Access Open Access I. Introduction A new paradigm, called Cognitive Radio (CR), has emerged in the field of wireless communication, to alle- viate the imbalance between spectrum allocation and its use [1,2]. CR entails the temporary usage of unused por- tions of the spectrum (spectrum holes or white spaces), owned by the licensed users (Primary Users–PUs), to be accessed by unlicensed users (Secondary Users–SUs). Built on the platform of software-defined radio (SDR), a CR node is rendered reconfigurable: the SDR allows the operating parameters such as frequency range, modula- tion type or output power to be reconfigured in soft- ware, without making any alteration in the hardware [2]. It is anticipated that the Next-Generation (xG) commu- nication networks will be based on CR [2]. These net- works will provide high bandwidth to mobile users via heterogenous wireless architectures and dynamic spec- trum access techniques. Besides the tasks of spectrum OFDM is a widely-deployed multi-carrier modulation technology for various wireless application segments, besides being a popular choice for CR. Other than its ability to handle multi-path fading and inter-symbol interference, it offers flexibility of resource allocation (power, constellation size and bandwidth) of its indivi- dual sub-carriers. The two main impairments in OFDM are inter-symbol interference (ISI) and inter-carrier * Correspondence: vinay_thumar@ee.iitb.ac.in 1Indian Institute of Technology, Bombay, 400076, India Full list of author information is available at the end of the article * Correspondence: vinay_thumar@ee.iitb.ac.in 1Indian Institute of Technology, Bombay, 400076, India Full list of author information is available at the end of the article * Correspondence: vinay_thumar@ee.iitb.ac.in 1Indian Institute of Technology, Bombay, 400076, India Full list of author information is available at the end of the article Figure 1 Resource allocation for OFDM-based cognitive radio. bands is separately constrained. In case of both single and multi-user scenarios, the optimization problems are difficult to solve due to either non-linearity of equations or their combinatorial nature. A rigorous examination of their solution form motivates the development of computationally simple, sub-optimum algorithms for the problems posed. The proposed strategies for power allo- cation and bit loading outperform those which have been previously presented in literature; while those for adaptive sub-carrier sizing for CR, are novel and have not been proposed earlier. (We would like to note here that the titles of some works of literature on CR suggest adaptive sub-carrier bandwidth/allocation [11,23,24], which actually refers to the assignment of sub-carriers to users in a multiple SU scenario, and not sub-carrier bandwidth sizing.) To detail the proposed scheme, the paper has been organized as follows: Section 30 presents related litera- ture. Section 31 describes the system model and com- munication scenario for a single SU. Sections 33, 35, VI and VII describe the power allocation, bit loading, sub- carrier bandwidth sizing and combined optimization problems, respectively. Likewise, SectionsVIII-XII are dedicated to the corresponding multiple SU situation. It is followed by a complexity analysis of each of the pro- posed algorithms, in Section XIII. Section XIV presents exhaustive simulation results and their discussion, while Section XV concludes the paper. The contribution of this paper is in developing a hol- istic resource allocation scheme for an OFDM-based CR, which includes power allocation, bit loading and sub-carrier bandwidth sizing. First, we address each of these issues as independent problems; the objective being - maximization of the SU’s throughput under a power budget and an interference constraint for the PU spectral band. Then, a joint optimization problem is for- mulated, which encompasses the aforementioned indivi- dual problems (Figure 1). In each case, we consider a realistic and efficient strategy, wherein the PU is divided into sub-bands, and the interference to each of its sub- Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 2 of 24 interference (ICI) [4]. ISI is mitigated by the addition of a guard interval (GI) which should be longer than that delay spread of the channel (also known as the cyclic prefix, since it is a cyclic copy of the original symbol). The loss of orthogonality between the sub-carriers of OFDM due to its sensitivity to frequency offsets results in ICI. Frequency errors which occur due to local oscil- lator errors can be easily compensated by frequency tracking, while those due to Doppler spread are poorly compensated for. Figure 1 Resource allocation for OFDM-based cognitive radio. Figure 1 Resource allocation for OFDM-based cognitive radio. p In conventional OFDM systems, optimum power allo- cation that maximizes the channel capacity under a total power budget is water-filling [4]. However, when OFDM is used for the SU system in a CR scenario, it’s side- lobes causes interference to the PUs, limiting their per- formance. The Federal Communications Commission’s (FCC) Spectrum Policy Task Force has recommended a metric called the interference temperature which when exceeded causes harmful interference to the PU band. The issue of interference mitigation in the PU band is receiving increasing attention in recent literature [5-14]. In an OFDM-based SU system, the amount of interfer- ence to the PU band depends on the SU’s sub-carrier parameters (power and bandwidth), the spectral distance between the SU’s sub-carriers and the PU band, as well as the channel between the SU and PU. Bit loading (or constellation sizing or modulation) for CR imposes an additional condition that a given performance should be achieved in every sub-carrier. The SNR gap is used to measure the reduction of SNR (signal to noise ratio) with respect to the capacity; it depends on the target error probability required in every sub-carrier when it carries log2(M ) bits per symbol, either QAM (quadra- ture amplitude modulation) or PSK (phase shift keying) modulated [15]. The sub-carrier bandwidth selection in OFDM is a trade-off between increasing the sub-carrier bandwidth to decrease the ICI, and reducing the band- width to mitigate ISI [16,22]. In CR, the interference to the PU band is a function of the SU sub-carrier band- width; the optimum sub-carrier bandwidth is, therefore, the one that maximizes the SU throughput while miti- gating the PU interference. C. Sub-carrier bandwidth sizing h f l The most significant literature on sub-carrier bandwidth sizing is summarized in this section. Das et al. [16,17] have proposed an approach for adaptive bandwidth for sub-carriers for single user OFDM and a multi-user sce- nario [18]. Zhang and Ma [19] have also proposed the implementation of variable sub-carrier bandwidth for a multi-user OFDM down-link scenario. Steendam and Moeneclaey [20], Harvatin and Ziemer [21], and Tufves- son and Maseng [22] have demonstrated the impact of varying the sub-carrier bandwidth on the system perfor- mance in a time and frequency-selective channel (either in terms of interference power or in terms of BER), but do not discuss the gains from dynamically adjusting the bandwidth. Power allocation for multiple SUs in the CR scenario has also received considerable attention in recent litera- ture. Chengshi et al. [9] have performed multi-user water-filling for CR. More recently, Shaat et al. [10] and Bansal et al. [11], have presented a Lagrangian formula- tion for maximizing the sum capacity of multiple SUs subject to a power budget and PU interference con- straints. Since the combinatorial optimization problem is computationally complex, both references have pro- posed sub-optimal schemes. First the users are allocated SU sub-carriers based on the best channel conditions, and the interference constrained maximum power limit on each SU sub-carrier is computed; then a cap-limited water-filling is executed [10]. On the other hand, the users are allocated sub-carriers based on the channel-to- noise ratio (CNR), and the Lagrangian formulation is used to maximize the sum capacity of the SUs under the PU interference constraints [11]. We infer from our analysis of the aforementioned works in literature, that most of the power allocation algorithms for CR have considered the entire PU band as one, for characterizing the interference. This is not as effective as the proposed strategy of dividing the PU into sub-bands, and separately mitigating the interfer- ence to each of them. While the authors of [10] have characterized the interference to each PU sub-band, in their problem solution, only the spectrally closest PU band is considered for the interference constraint. Moreover, the channel gain from different SUs to each PU sub-band has been ignored in their formulation. In [11], a brute-force combinatorial approach is executed for power allocation, which has high computational complexity. A. Power allocation Weiss et al. [5] have characterized the mutual interfer- ence between the PU and SU in an OFDM-based CR. Bansal et al. [6] have formulated the power allocation problem for a single SU with the objective of maximiz- ing it’s throughput while maintaining the interference to the entire PU band below a threshold, however, without a total power constraint. The model of Wang et al. [7] considers a single SU and multiple PUs; the system bandwidth is divided into sub-channels, and different PUs co-exist with the SU on each sub-channel. A path- Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 3 of 24 Page 3 of 24 greedy method in order to obtain optimum integer bit allocation results ([31-38]). Bit loading for a multi-user OFDM scenario has been addressed by Wong et al. [23] and Huang et al. [39] for MM and RM problems, respectively. loss model is used between the SU and PU to determine the peak power constraint of each sub-channel. Addi- tionally, a total power constraint is included and the objective is to maximize the SU’s capacity. The algo- rithm used is called iterative partitioned water-filling. The system model of Wang et al. [8] is similar to that of [7], however, they have additionally considered the side-lobe power in each SU sub-channel, contributed by the neighboring sub-carriers, in the optimization problem. In the CR context, the following work exists in litera- ture: Tang et al. [12] have formulated a bit loading pro- blem for multiple SUs, which is based on maximizing total system throughput under interference power con- straint to PUs, individual data rate constraints for the SUs and total transmission power constraint. Cheng et al. [13] have used a game-theoretic approach to formu- late a transmit power control game for CR, which jointly solves the bandwidth allocation, bit loading and power allocation problems. Budiarjo et al. [14] have used the Fischer and Huber algorithm [37] for bit-loading for a single SU, followed by Raised Cosine windowing to miti- gate the side lobe interference to the PU. The situation for multiple SUs is more challenging, since it involves allotment of sub-carriers to users, besides power allocation, under the specified constraints. Münz et al. [25] and Jang et al. [26] have suggested stra- tegies for multi-user power allocation with the objective of maximizing the total data rate. Shen et al. C. Sub-carrier bandwidth sizing h f l In the proposed power allocation algorithm, we have jointly considered interference mitigation to each PU sub-band, within the power budget, while max- imizing the throughput of the single SU, or the sum throughput in case of multiple SUs. The approach attempts to strike a balance between performance A. Power allocation [27] have proposed power allocation with proportional fairness among the users. Wong et al. [23] and Kivanc et al. [24] have provided bit-loading and power allocation algo- rithms to minimize the total transmit power in the multi-user scenario. B. Bit loading Two main classes of bit loading problems are: rate max- imization (RM)–maximizing the data rate within a power budget; and margin maximization (MM)–mini- mizing power consumption given a target data rate [28]. The implementation of bit loading algorithms in litera- ture fall into two broad categories. The first category of algorithms use numerical methods that employ Lagran- gian optimization, resulting in real numbers for the bit loading ([23,29,30]). However, for practical constellation sizing, the number of bits allocated per sub-carrier is restricted to integer values, which imposes a combina- torial structure in the loading optimization problem. The second category of algorithms employ a discrete Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 4 of 24 Page 4 of 24 The maximum achievable throughput of the SU, in bits/sec, is given by [16] optimization and computational complexity. Similar considerations are applied for PU interference mitigation in the proposed bit loading and sub-carrier bandwidth sizing algorithms. C = 1 Tg + 1 B Ns i=1 log2 1 + Pihi σ 2 i (1) (1) III. System Model and Communication Scenario: Single SU in which B is the sub-carrier bandwidth, Tg is the dura- tion of the guard interval, and Pi is the power allocated to the ith SU sub-carrier. σ 2 i = σ 2 + Ji, where s2 is the Addi- tive White Gaussian Noise (AWGN) variance, and Ji is the interference from the PU on the ith SU subcarrier. Ji depends on the power spectral density (PSD) of the PU and the channel gain between the PU Tx and SU Rx. g In the current model, a single SU transceiver is consid- ered, and a PU exists in its radio range (Figure 2). OFDM is the communication technology of the SU, the use of which divides the available bandwidth into fre- quency-flat sub-carriers. When the PU claims a portion of the spectrum, the SU nulls the corresponding sub- carriers. Let Ns be the number of active sub-carriers for the SU. The transmission opportunity is detected by the SU in the spectrum sensing phase of its cognitive cycle [1]. The channel power gain of the ith sub-carrier on the link between the SU transmitter (Tx) and receiver (Rx) is denoted by hi. To efficiently control the interfer- ence to the PU, the PU spectrum is divided into Np sub-bands of equal width, and the gain of the jth sub- band from the SU Tx to the PU Rx is given by gj. In the present work, we have considered an immobile SU, resulting in no Doppler spread. It is assumed that the frequency offset due to any other source is compensated [40], and consequently we ignore the effect of ICI. The mutual interference model between the PU and SU is assumed [5]. The interference from the SU on the jth PU sub-band is formulated as Ij = gj Ns i=1 Pi jthPUband Sinc2[(f −fi)Ts′] (2) (2) where Ts′ = Ts + Tg. Ts′ is the total length of the symbol after adding the guard interval, Ts is the length of the sym- bol without the guard interval, and fi represents the center frequency of the ith subcarrier. Sinc(x) is the mathematical function commonly defined by Sin(πx)/(πx). IV. Power allocation In the power allocation problem, our objective is to maximize the SU throughput under a total node power constraint Pt, in such a way that the interference to the jth PU sub-band is less than a threshold Ij th. Ij th = Tj thBWj, where Tj th is the interference temperature limit for the jth PU sub-band and BWj is its bandwidth. For simplicity of representation, we assume that the interference thresh- old is the same for all PU sub-bands and is denoted by Ith. The optimization problem can stated as Resource allocation strategies in CR require that the channel state information (CSI) be known to the SU Tx. It is assumed that the SU Rx estimates the channel by measuring the received power of the pilot signals sent by the transmitter, and the CSI is fed back to the trans- mitter [41]. A robust and low-complexity protocol can be used for the feedback. A block fading propagation channel is assumed where the channel remains constant during the resource allocation and transmission process. The channel sensing and feedback is done once per coherence time. Estimating the channel between the PU Tx and SU Rx, as well as that between the SU Tx and PU Rx, is more challenging, and entails the use of blind estimation techniques [41]. Problem P1 obj = max C Pi (3) Problem P1 obj = max C Pi (3) subject to Ij ≤Ith ∀j (4) Ns i=1 Pi ≤Pt (5) Pi ≥0 (6) Problem P1 Generally, the jth PU sub-carrier which receives the maximum interference will be responsible for the binding constraint; and the solution looks like P∗ i = max 1 λjgjQj,i −σ 2 i hi , 0 (13) (13) P∗ i = max ⎛ ⎝ 1 Np j=1 λjgjQj,i + μ −σ 2 i hi , 0 ⎞ ⎠ (9) in which Qj,i = jthPUband Sinc2[(f −fi)Ts′] (10) λj ≥0, μ ≥0 (11) P∗ i = max ⎛ ⎝ 1 Np j=1 λjgjQj,i + μ −σ 2 i hi , 0 ⎞ ⎠ (9) To make it a general water-filling solution with a con- stant water-level, we can multiply by gjQj, i, to get (9) ϑi = max( 1 λj −Qj,igjσ 2 hi , 0) (14) in which (14) Qj,i = jthPUband Sinc2[(f −fi)Ts′] (10) (10) and the power allocation is and the power allocation is P∗ i = ϑi Qj,igj (15) P∗ i = ϑi Qj,igj (15) λj ≥0, μ ≥0 (11) If we consider the above solution as the peak power on each SU sub-carrier i.e. Pmax i , under the PU interfer- ence constraint (as in [10]), and then execute water-fill- ing, it is referred to as cap-limited water-filling. The solution takes the form Though the above solution looks like water-filling, it is different from the conventional water-filling technique in the fact that each SU sub-carrier has a different water level. We would like to note here that the problem formula- tion in [7] and [8] appear similar to the above problem (P1). However, the system model of the current work and that of the aforementioned references are significantly dif- ferent–while the former considers the system bandwidth to be frequency division multiplexed by the PU and SU, the latter assumes the two entities to be spatially separate but occupying the same spectrum. In the problem formu- lation of [7], the inequality constraints are decoupled, making the problem simpler to solve using either an exhaustive search-based approach or an iterative parti- tioned water-filling. On the other hand, in the formulation of [8], the inequality constraints are coupled by the use of dependent variables. Its solution involves segregating the equality (binding) and inequality (non-binding) constraints for the given power budget using a search-based approach and computing the optimal solution from the equality constraints. This technique has a high computationally complexity. Problem P1 The proposed method attempts to find a low- complexity sub-optimum solution after a detailed analysis of the solution form. P∗ i = min(max 1 μ −σ 2 i hi , 0), Pmax i (16) (16) If the power budget is neither too high nor too low, the solution will take the form given by (9). On substi- tuting P∗ i in the constraint of (4), we get gj Np k=1 P∗ i Qj,i = Ith ∀j (17) (17) The solution to the above Np equations cannot be obtained directly, and we propose an iterative algorithm (Algorithm 1) to achieve the objective of P1, given the interference constraints on each PU sub-band and the power budget. Compute the total power allocated as Ps = ∑Pi Calculate the interference caused to each PU sub- band, Ij, as given by (2). Problem P1 obj = max C Pi (3) (3) subject to Ij ≤Ith ∀j (4) Ij ≤Ith ∀j (4) Ns i=1 Pi ≤Pt (5) Pi ≥0 (6) SU Tx PU SU - SU link SU I th g 1 g 2 g 3 h 1 h 2 h 3 … … SU - PU link SU PU SU Rx Figure 2 System model for a single secondary user. (5) Pi ≥0 (6) (6) Pi ≥0 The Lagrangian for the above is formulated as The Lagrangian for the above is formulated as L(Pi, λj, μ, βi) = Ns i=1 log2 1 + Pihi σ 2 i − Np j=1 λj(Ij −Ith) (7) Figure 2 System model for a single secondary user. Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 5 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 5 of 24 Page 5 of 24 Page 5 of 24 be a binding constraint and all interference constraints are non-binding; the corresponding Lagrange multipliers (lj) are zero and the solution looks like that of conven- tional water-filling with a constant water level: −μ Ns i=1 Pi −Pt + Ns i=1 βiPi (8) (8) The multiplicative factor 1/(Tg + 1/B) in the expres- sion for C, is a constant in the power optimization pro- blem, and is ignored in the above expression and all the subsequent analysis in this section. lj, μ and bi are the Lagrangian multipliers. The problem is a convex optimi- zation problem, and Karush-Kuhn-Tucker (KKT) condi- tions [42] are applied to find the optimum solution. Also, since we require Pi ≥0, bi is substituted as 0, due to the complementary slackness condition [42]. The optimum power allocation is given by [43] (which refers to our own previous work) P∗ i = max( 1 μ −σ 2 i hi , 0) (12) (12) If the power budget is very high, then only the inter- ference constraint will be binding. Algorithm 1 1) Initialize all lj and μ. 2) Compute Pi by substituting the above lj and μ in (9). As the optimization problem (P1) is convex with lin- ear constraints, at the optimum point some constraints are binding, while the others are non-binding. If the power budget of the SU (Pt) is too small, then that will Compute the total power allocated as Ps = ∑P Calculate the interference caused to each PU sub- band, Ij, as given by (2). Calculate the interference caused to each PU sub- band, Ij, as given by (2). Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 6 of 24 3) For each PU sub-band calculate the difference between the interference generated and the threshold, as diffj = Ij-Ith. Calculate the difference between the total power allocated and the power budget, as diffp = Ps-Pt. 4) F h PU b i If(diff 0) For non-rectangular QAM signal constellations (bi Î 3, 5, 7, ...), the SNR gap is given by (19) without the equality [45]. In the case of BPSK, the SNR gap is approximated by [Q-1(Pe/4)]2/2, which is slightly larger than the right hand side of (19). However, for simplicity and practicality, (19) with the equality sign is used to approximate the SNR gap for bi Î ℤ+ [45]. 4) For each PU sub-carrier, If(diffj > 0) lj = lj + aj * diffj end If If(diffp > 0) μ = μ + b * diffp end If The optimization problem for bit-loading can stated as Problem P2 obj = max bi Ns i=1 bi (21) 5) If (diffj > 0) or (diffp > 0) (21) Goto Step2. Else End Algorithm end If subject to subject to gj Ns i=1 (2bi −1) αi Qj,i ≤Ith ∀j (22) (22) In the first step of the algorithm, we initialize all lj and μ, such that the resultant power allocation violates one or all of the constraints. In the subsequent steps, we update the Lagrange multipliers lj and μ in proportion to diffj and diffp respectively. aj and b are the step sizes; aj = diffj/max(diffj) and b = 1/Ns. The process is itera- tively repeated until all the constraints are satisfied. Ns i=1 2bi −1 αi ≤Pt Ns i=1 2bi −1 αi ≤Pt (23) (23) bi ∈Z+ (24) bi ∈Z+ (24) where αi = hi σ 2 i . (22) and (23) represent the interfer- ence and power budget constraints respectively. The constraint of (24) represents the integer constraint for practical constellation sizing. It turns out that the above problem (P2) is a combinatorial optimization problem [28]; to make it tractable, the integer constraint is relaxed to 3) Compute the interference caused to each PU sub- band, Ij, using (2). 3) Compute the interference caused to each PU sub- band, Ij, using (2). j 4) While {Ij > Ith ∀j } Do Compute the reduced interference in the jth PU sub- band due to removal of one bit from every ith SU sub-carrier as ΔIj, i = gjΔPiQj, i, which is a vector of size Np × Ns. j 4) While {Ij > Ith ∀j } Do j 4) While {Ij > Ith ∀j } Do Compute the power saved in removing one bit from the ith SU subcarrier as Pi = 1 αi 2bi−1. Compute the power saved in removing one bit from the ith SU subcarrier as Pi = 1 αi 2bi−1. Compute the reduced interference in the jth PU sub- band due to removal of one bit from every ith SU sub-carrier as ΔIj, i = gjΔPiQj, i, which is a vector of size Np × Ns. Compute the maximum element of the vector ΔIj, i, max{ΔIj, i}, and remove a bit from the sub-carrier identified by the corresponding column index i. i Compute the reduced interference in the jth PU sub- band due to removal of one bit from every ith SU sub-carrier as ΔIj, i = gjΔPiQj, i, which is a vector of size Np × Ns. i Compute the reduced interference in the jth PU sub- band due to removal of one bit from every ith SU sub-carrier as ΔIj, i = gjΔPiQj, i, which is a vector of size Np × Ns. Update the bit allocation profile bi and the corre- sponding power allocation profile Pi. Compute the maximum element of the vector ΔIj, i, max{ΔIj, i}, and remove a bit from the sub-carrier identified by the corresponding column index i. While {Ps > Pt } Do { Compute the power saved in removing one bit from the ith SU subcarrier as Pi = 1 αi 2bi−1. f y p g Update the bit allocation profile bi, the corresponding power allocation profile Pi, and the interference caused to each PU sub-band, Ij Remove one bit from the sub-carrier that corresponds to the highest ΔPi. Update the bit allocation profile bi, and the corre- sponding power allocation profile Pi. The execution of two passes can be further condensed to a single loop, which executes till both the power and interference constraints are met. 3) Compute the interference caused to each PU sub- band, Ij, using (2). This is rendered possi- ble in Algorithm 4, by the introduction of a new metric, viz, power weighted by the spectral distance from the PU band. Compute the total power allocated as Ps = ∑Pi. Motivated by the need to reduce the computational complexity associated with Algorithm 2 (due to the iterative power allocation process of Algorithm 1 in its Step 1), we also propose a simple greedy bit allocation process with two passes. In the first pass bit-loading is executed till the power constraint is met; and in the second pass, bit-removal is performed till the interfer- ence constraint is satisfied. The algorithm is as follows: V. Bit loading The power allocation and bit-loading problems are closely related. However, in this section we treat bit-loading as an independent problem, and address the issue of practical constellation sizing with integer number of bits per sym- bol, under a power budget and PU interference constraint for an OFDM-based CR. The number of bits that can be transmitted on the ith OFDM sub-carrier is given by [44] bi ≥0 bi ≥0 (25) (25) bi = log2 1 + Pihi σ 2 i (18) and the following substitution is made (18) 2bi −1 αi = Pi (26) (26) where Γ is the SNR gap calculated according to the gap approximation formula [44,15], based on the target probability of error (Pe). M-ary QAM (M-QAM) is a preferred choice of modulation, because it is more energy efficient than M-ary PSK (M-PSK) while retain- ing the same bandwidth-efficiency. When rectangular M-QAM is deployed (bi Î 2, 4, 6, ...), we can write [45] The problem is now equivalent to the single user power allocation problem (P1), and the solution to it is characterized the way it has been done in Section IV. We propose a few iterative algorithms, with varying degrees of trade-off between optimality of solution and computational complexity. ≥1 3 Q−1(Pe/4) 2 (19) The first of the proposed bit loading algorithms com- prises two steps; to start with, the power allocation Pi is computed using Algorithm 1, and the corresponding bit- load bi is obtained from (18). These are, however, real values. The next step, is to round the real values to the nearest higher integer, for practical constellation sizing. This may cause the interference or power constraint, or both to be violated. Therefore, a greedy bit-removal is (19) where Q -1 is the inverse of the well-known Q func- tion given by Q = 1 √ 2 ∞ x e−t2/2dt (20) (20) Page 7 of 24 Page 7 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 executed till both the constraints are met. The complete algorithm operates as follows: executed till both the constraints are met. The complete algorithm operates as follows: Algorithm 3 1) Initialize the bits allocated to each sub-carrier bi to zero. Compute the corresponding power allocation Pi using (25), and the total power allocation as Ps = ∑Pi. Compute the corresponding power allocation Pi using (25), and the total power allocation as Ps = ∑Pi. 1) Compute the transmit power Pi using Algorithm 1, and the corresponding bit-load bi using (18) 1) Compute the transmit power Pi using Algorithm 1, and the corresponding bit-load bi using (18) p g g ( ) 2) bi = ceil(bi), where ceil() represents rounding to the nearest higher integer. 2) While {Ps < Pt } Do 2) While {Ps < Pt } Do 3) Calculate the transmit power Pi corresponding to the quantized bi using (25), and the interference caused to each PU sub-band, Ij, using (2). 3) Calculate the transmit power Pi corresponding to the quantized bi using (25), and the interference caused to each PU sub-band, Ij, using (2). { Compute the power required to add one bit to the ith SU subcarrier as Pi = 1 αi 2bi. Add one bit to the sub-carrier that corresponds to the lowest ΔPi. Update the bit allocation profile bi and the corre- sponding power allocation profile Pi. Compute the total power allocated as Ps = ∑Pi. } { Compute the power required to add one bit to the ith SU subcarrier as Pi = 1 αi 2bi. Compute the total power allocated as Ps = ∑Pi Add one bit to the sub-carrier that corresponds to the lowest ΔPi. 4) If {(Ps > Pt) OR (Ij > Ith (for any j))} Update the bit allocation profile bi and the corre- sponding power allocation profile Pi. Compute the total power allocated as Ps = ∑Pi. } { While {Ij > Ith ∀j } Do { Compute the power saved in removing one bit from the ith SU subcarrier as Pi = 1 αi 2bi−1. Compute the corresponding power allocation Pi, and the total power allocation as Ps = ∑Pi. Compute the interference caused to each PU sub- band, Ij, using (2). Problem P3 Compute the metric ΔWPi = ΔPi/di, which represents the power saved in removing one bit from the ith SU subcarrier weighted by the distance of the ith sub- carrier from the PU band. obj = max B C (27) subject to Add one bit to the sub-carrier that corresponds to the lowest ΔWPi. (28) Ij ≤Ith ∀j Update the bit allocation profile bi, the corresponding power allocation profile Pi, and the interference caused to each PU sub-band, Ij (29) Compute the total power allocated as Ps = ∑Pi. 0 ≤B ≤ fc (30) 0 ≤B ≤ fc (30) The proposed algorithms have been compared on the basis of their computational complexity and perfor- mance in Section XIII and XIV, respectively. Intuitively, we can expect Algorithm 2 to give the best performance, since its solution is obtained from the optimization problem. But it is associated with high complexity. Algorithm 3 entails bit-removal till the PU interference constraint is met without any compensatory bit-addition in some other sub-carrier to improve the throughput. Consequently its performance will be inferior to Algorithm 2. Algorithm 4, though computational the simplest, will result in poorer performance as compared to the previous two algorithms because of weighting ΔPi with di, which may not always give the desired result. For instance, if ΔPi is very small and di is small, it may result in an overall low value of the metric causing a bit to be added on that sub-carrier at the cost of increased PU interference. The first two constraints are the same as those of Equations 4 and 5, but are repeated for completeness. Δfc is the coherence bandwidth of the channel. Since presently mobility is not considered, the bandwidth is lower bounded by 0 (in the case of mobile SUs, the bandwidth B should be greater than the Doppler spread of the channel). To solve the above problem for the optimum bandwidth B, the sub-carrier power is consid- ered to be uniform, i.e. Pi = Pt /Ns. However, it is possi- ble that none of the values of bandwidth satisfy the PU interference constraint within the given power budget, and consequently the solution to the above problem does not exist. Only if the power budget is very small, some value of bandwidth may satisfy the interference constraint. 2) While { (Ps < Pt) AND (Ij < Ith ∀j) } Do Problem P3 obj = max B C (27) subject to Ij ≤Ith ∀j (28) Ns i=1 Pi ≤Pt (29) 0 ≤B ≤ fc (30) Problem P3 obj = max B C (27) subject to Ij ≤Ith ∀j (28) Ns i=1 Pi ≤Pt (29) 0 ≤B ≤ fc (30) Problem P3 Therefore, both the sub-carrier bandwidth and power need to be varied to arrive at an optimum OFDM configuration which meets the interference con- straint, within the power budget, while maximizing the achievable throughput. The problem entails solving for B and P∗ i , and can be posed as VI. Sub-carrier bandwidth sizing The OFDM sub-carrier bandwidth should be greater than the Doppler spread of the channel and less than the coherence bandwidth. An increase in the bandwidth results is a corresponding increase in the throughput (1) unto a certain point, after which the throughput falls due to a drop in the bandwidth efficiency. In a CR sce- nario, the sub-carrier bandwidth also impacts the PU interference. Increasing the bandwidth implies decreas- ing the number of sub-carriers, and thereby, the node power is distributed among lesser sub-carriers; a higher power in each sub-carrier generates higher side-lobe interference in the PU band. Consequently, as the band- width increases, the interference to the PU band increases, within a fixed power budget. This has been observed during simulation study and the results are plotted in Sect. XIV. Algorithm 4 1) Initialize the bits allocated to each sub-carrier bi to zero. Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 8 of 24 Page 8 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 the SU throughput under a power budget and PU interference constraint. It can be posed as follows: the SU throughput under a power budget and PU interference constraint. It can be posed as follows: 2) While { (Ps < Pt) AND (Ij < Ith ∀j) } Do Problem P4 Problem P4 obj = max B,Pi C (31) obj = max B,Pi C (31) (31) subject to the same constraints as those of problem P3, and additionally Pi ≥0 (32) Pi ≥0 (32) Here the number of SU sub-carriers is a function of the bandwidth B, as follows: Ns = 2 ∗BW B −1 (33) (33) where BW is the total system bandwidth. where BW is the total system bandwidth. The objective function (31) is concave since its Hessian is positive semi-definite [42], and the problem (P4) has a combination of linear and non-linear In the optimum sub-carrier bandwidth sizing problem for an OFDM-based CR, the objective is to maximize Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 9 of 24 Page 9 of 24 4) Ns = Ns-s. (polynomial in B) constraints. It has been analyzed to form a convex optimization problem (though the proof has not been included). Its Lagrangian will look like Calculate the throughput for the number of sub- carriers Ns, Ns+1, Ns-1 and represent them CNs(Pi), CNs+1(Pi), CNs-1(Pi), respectively, using Algorithm 1 and (1). L(B, Pi, κj, χ, ω, ψi) = 1 Tg + 1 B Ns i=1 log2 1 + Pihi σ 2 i −(34) Np j=1 κj(Ij −Ith) −χ Ns i=1 Pi −Pt −ω(B − fc) + Ns i=1 ψiPi (35) L(B, Pi, κj, χ, ω, ψi) = 1 Tg + 1 B Ns i=1 log2 1 + Pihi σ 2 i −(34) 5) While {(CNs(Pi) < (CNs+1(Pi)) OR (CNs(Pi) <CNs-1 (Pi))} Do 5) While {(CNs(Pi) < (CNs+1(Pi)) OR (CNs(Pi) <CNs-1 (Pi))} Do Np j=1 κj(Ij −Ith) −χ Ns i=1 Pi −Pt −ω(B − fc) + Ns i=1 ψiPi (35) where j, c, ω and ψi are the Lagrangian multipliers. where j, c, ω and ψi are the Lagrangian multipliers. Applying KKT conditions to solve the problem results in complex non-linear equations (as discussed in Appendix A), which cannot be solved directly. A graphical, as well as intuitive analysis of the variation of sub-carrier bandwidth (in discrete steps of Ns) with corresponding power allocation uniformly (Pi = Pt/Ns), by water-filling, and by Algorithm 1, reveals its relation with the achievable throughput. 6) Nsopt = Ns, and the corresponding sub-carrier bandwidth Bopt is obtained using (33). 6) Nsopt = Ns, and the corresponding sub-carrier bandwidth Bopt is obtained using (33). 1) Initialize the sub-carrier bandwidth to its maximum value, i.e. B = Δfc. 1) Initialize the sub-carrier bandwidth to its maximum value, i.e. B = Δfc. subject to the PU interference constraint (4), power budget (23), the integer bit granularity (24), and bounds on the sub-carrier bandwidth (30). subject to the PU interference constraint (4), power budget (23), the integer bit granularity (24), and bounds on the sub-carrier bandwidth (30). 2) Calculate the corresponding number of sub carriers as Nsmin, using (33). 3) Initialize Cprev(Pi)=Cnew(Pi)=0 (where C(Pi) represents the achievable throughput obtained from (1)). The proposed algorithm first computes the power allocation and sub-carrier bandwidth using the strategy discussed in the previous section. The corresponding bit load are real values, which are rounded to the nearest higher integer, and a greedy bit removal is executed till the power and PU interference constraint are met. Initialize the number of sub-carriers Ns = Nsmin While {Cprev(Pi) <= Cnew(Pi)} Do Initialize the number of sub-carriers Ns = Nsmin While {Cprev(Pi) <= Cnew(Pi)} Do Initialize the number of sub-carriers Ns = Nsmin While {Cprev(Pi) <= Cnew(Pi)} Do While {Cprev(Pi) <= Cnew(Pi)} Do While {Cprev(Pi) <= Cnew(Pi)} Do { Cprev(Pi)=Cnew(Pi) Problem P5 (36) 1) Compute the optimum power allocation Pi and sub-carrier bandwidth B using Algorithm 5. 2) Compute the corresponding bit load bi using (18). 3) Execute Step 2 onwards of Algorithm 2. Increment the number of sub-carriers with some suitable step-size s, i.e. Ns = Ns+s. Find the power allocation Pi using Algorithm 1. Calculate throughput Cnew(Pi) using (1). 3) Execute Step 2 onwards of Algorithm 2. VII. Sub-carrier power allocation, bandwidth sizing and bit loading After having addressed the power allocation, bit loading and bandwidth sizing individually, we formulate the problem of doing all the three together, for an OFDM- based CR, with the objective of maximizing the SU throughput. It is as follows: Problem P5 obj = ma B,Pi xC (36) Problem P4 Unto a certain point, an increase in bandwidth results in a corresponding increase in throughput; after which, any further increase results in the symbol duration becoming relatively smal- ler than the guard interval, and the bandwidth efficiency reduces. The proposed iterative algorithm is motivated by this discussion; it is a search-based approach, in which, initially the throughput is computed in larger steps of Ns, with the power allocation at every point obtained from Algorithm 1 (which ensures the PU inter- ference constraint being met within the power budget). Then a finer search is executed to look for the global optima. Ns (number of sub-carriers) is the preferred choice of variable, as compared to B, due to its integer granularity. The two are related as given by (33). The algorithm is as follows: f If {CNs(Pi) <CNs-1(Pi)} end If end If f Calculate throughput for the number of sub-carriers Ns, Ns+1, Ns-1 and represent them as CNs(Pi), CNs+1 (Pi), CNs-1(Pi), respectively, using Algorithm 1 and (1). } Algorithm 6 Increment the number of sub-carriers with some suitable step-size s, i.e. Ns = Ns+s. Find the power allocation Pi using Algorithm 1. Calculate throughput Cnew(Pi) using (1). Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 10 of 24 IX. Power allocation (Multiple SUs) To formulate the power allocation problem for the multi-user CR scenario, (38) is re-written as In this scenario, we assume that there are K SU transceivers, and the PU is in the radio range of all of them (Figure 3). The assumptions on the propagation channel are the same as in the single user case (Sect. III). The multi-user scenario is more complex than the single user situation, since it involves assigning sub-car- riers to users, besides allocating power under the given constraints. The throughput of the kth user on the ith sub-carrier is defined as Cm = 1 Tg + 1 B K k=1 Ns i=1 ρk,ic( ζk,i ρk,i ) (39) (39) where ζk, i = pk, i * rk, i ρk,i = 1 if the ith sub carrier is allocated to kth user; 0 if the ith sub carrier is not allocated to kth user. (40) ρk,i = 1 if the ith sub carrier is allocated to kth user; 0 if the ith sub carrier is not allocated to kth user. (40) (40) Our objective is to maximize the sum throughput, given the total power budget on all users Pt, and the interference constraint on each PU sub-band. The pro- blem is posed as ck,i(pk,i) = log2 1 + pk,ihk,i σ 2 (37) (37) where pk, i is the power allocated to the ith sub carrier assigned to the kth user, and hk, i is channel power gain of kth user on ith sub carrier. roblem P6 obj = max Cm (41) VIII. System Model and Communication Scenario: Multiple SUs IX. Power allocation (Multiple SUs) Problem P6 EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 11 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 11 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 11 of 24 ch 6) ljold = lj ∀j and μold = μ 6) ljold = lj ∀j and μold = μ in which Hk,i(λj, μ) = log( hk,i ( Np j=1 λjgk,jQj,i + μ)σ 2 )− (50) (1 − ( Np j=1 λjgk,jQj,i + μ)σ 2 hk,i ) (51) Hk,i(λj, μ) = log( hk,i ( Np j=1 λjgk,jQj,i + μ)σ 2 )− (50) (1 − ( Np j=1 λjgk,jQj,i + μ)σ 2 hk,i ) (51) If(max(diffj) < 0) and (diffp < 0) lj = (ljold+lj)/2 ∀j μ = (μold+μ)/2 Goto Step3. end If 7) For each PU sub carrier If(diff > (51) 7) For each PU sub-carrier, If(diffj > 0) From the above analysis, we infer that there are two main steps in solving the multi-user power allocation problem within the power budget and the PU interfer- ence constraint. In the first step, we allocate sub- carriers to the users. This can be done by assigning sub-carrier i to that user k that will maximize the function Hk, i, i.e. lj = lj + aj * diffj end If If(diffp > 0) μ = μ + b * diffp end If 8) If (diffj > 0) or (diffp > 0) ρ∗ k,i = 1 Hk′,i > Hk,i ∀k; 0 otherwise. (52) (52) Goto Step3. Else End Algorithm end If Next, we compute the power on each SU sub-carrier using (48). This looks like a water-filling solution with different water levels, as in the case of a single user. But it can be inferred from (48)-(51), that for multiple SUs, the sub-carrier assignment and power allocation are not independent of each other and the solution to the equa- tions cannot be obtained directly. Hk, i is proportional to the ratio of the channel gain of the kth user on the ith sub-carrier to the cumulative interference of all the Np PU bands, weighted by the corresponding Lagrangian multipliers lj. Hk, i will be used as the metric to assign sub-carriers to users in the proposed power allocation algorithm (Algorithm 7). X. Bit loading (Multiple SUs) The objective of the bit loading problem is the same as the corresponding single user case, i.e. Problem P2, addi- tionally requiring the sub-carriers to be assigned to the K users. The problem is posed as Problem P7 obj = max K k=1 Ns i=1 εk,i (53) (53) where εk, i = bk, i * rk, i (rk, i is described in (40)) where εk, i = bk, i * rk, i (rk, i is described in (40)) Problem P6 The proposed algorithm is devised to iteratively assign sub-carriers and allocate the powers till neither the interference or power constraints are violated. The step sizes aj and b are the same as those defined in Algorithm 1. j 2) Initialize ljold and μold to zero. j 3) Assign each sub-carrier i to that user k that will maximize the function Hk, i. Problem P6 obj = max Cm (41) The Ns active SU sub-carriers will be assigned to the various users, while optimizing the sum through- put under a power budget and an interference con- straint on each PU sub-band. The sum throughput is given by subject to K k=1 Ns i=1 gk,jζk,iQj,i ≤Ith ∀j (42) (42) Cm = 1 Tg + 1 B K k=1 Ns i=1 ck,i(pk,i) (38) where gk, j is the channel power gain between kth SU and jth primary band. (38) K k=1 Ns i=1 ζk,i ≤Pt (43) All the CSI estimated at the receivers, is now required to be sent to a centralized controller, which is responsi- ble for coordinating the resource allocation in the multi-user CR network. A centralized mode involves considerable signaling overheads, especially in fast fading environments. In a slow fading environment as is assumed in this work, the centralized architecture will compensate for the overheads with near-optimum solutions. (43) K k=1 ρk,i = 1 ∀i (44) (44) ζk,i ≥0 ∀k, i ζk,i ≥0 ∀k, i (45) (45) The Lagrangian for the above is formulated as Note: To avoid complexity of notations, we have used the same variables (for the Lagrangian multipliers) for the single and multi-user cases. Their values will, however, depend on the specific problem. The Lagrangian for the above is formulated as L(ζk,i, ρk,i, λj, μ, γi, βk,i) = C − Np j=1 λj( K k=1 Ns i=1 gk,jζk,iQj,i −Ith) (46) (46) −μ( K k=1 Ns i=1 ζk,i −Pt) − Ns i=1 γi( K k=1 ρk,i −1 + K k=1 Ns i=1 βk,i(ζk,i) (47) SU1 Tx PU Shared by SUs Ith PU SU1 Rx SU2 Tx SU2 Rx SU centralized controller Figure 3 System model for a single secondary user. SU1 Tx PU Shared by SUs Ith PU SU1 Rx SU2 Tx SU2 Rx SU centralized controller Figure 3 System model for a single secondary user. On applying KKT conditions to solve the convex optimization problem, we get (details in Appendix XV) ζ ∗ k,i = max([( 1 Np j=1 λjgk,jQj,i + μ −σ 2 hk,i )ρ∗ k,i], 0) (48) (48) and ρ∗ k,i = 0 if γi ≥Hi,k(λj, μ); 1 if γi < Hi,k(λj, μ). (49) (49) Figure 3 System model for a single secondary user. Thumar et al. Algorithm 7 K k=1 Ns i=1 gk,j (2εk,i −1) αk,i Qj,i ≤Ith ∀j (54) 1) Initialize all lj and μ. (54) j 3) Assign each sub-carrier i to that user k that will maximize the function Hk, i. j 3) Assign each sub-carrier i to that user k that will maximize the function Hk, i. where gk, j is the channel power gain between kth SU and jth PU band. where gk, j is the channel power gain between kth SU and jth PU band. 4) Compute ξk, i by substituting the above lj and μ in (48). K k=1 Ns i=1 (2εk,i −1) αk,i ≤Pt (55) where αk,i = hk′i σ 2 i K k=1 Ns i=1 (2εk,i −1) αk,i ≤Pt (55) (55) Compute the total power allocated as Ps = ∑k ∑i ζk, i Calculate the interference caused to each PU sub- band, Ij (from left hand side of (42)). Compute the total power allocated as Ps = ∑k ∑i ζk, i Calculate the interference caused to each PU sub- band, Ij (from left hand side of (42)). where αk,i = hk′i σ 2 i where αk,i = hk′i σ 2 i K k=1 ρk,i = 1 ∀i 5) For each PU sub-band calculate the difference between the interference generated and the threshold, as diffj = Ij-Ith. Calculate the difference between the total power allocated and the power budget, as diffp = Ps-Pt. K k=1 ρk,i = 1 ∀i (56) (56) εk,i ∈Z+ (57) εk,i ∈Z+ (57) Page 12 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 (58) (59) user user e at Compute the power saved in removing one bit from the ith SU subcarrier as ζk′,i = 1 αk′,i 2εk′,i−1. Remove one bit from the sub-carrier that corresponds to the highest Δζk’, i. Update the bit allocation profile, εk’, i and the corre- sponding power allocation profile, ζk’, i. Compute the total power allocated as Ps = ∑i ζk’, i. } } end If. We relax the integer constraint to εk,i ≥0 (58) εk,i ≥0 (58) and make the following substitution and make the following substitution (2εk,i −1) αk,i = ζk,i (59) (59) The problem becomes equivalent to the multi-user power allocation problem (P6). Compute the corresponding power allocation pk’, i, and the total power allocation as Ps = ∑i pk’, i. Compute the corresponding power allocation pk’, i, and the total power allocation as Ps = ∑i pk’, i. 3) Calculate the transmit power, ζk’, i, corresponding to the quantized bk’, i using (59), and the interference caused to each PU sub-band, Ij, using the left hand side of (42). 3) Calculate the transmit power, ζk’, i, corresponding to the quantized bk’, i using (59), and the interference caused to each PU sub-band, Ij, using the left hand side of (42). 4) While { Ps < Pt } Do Algorithm 7 Similar to the single user bit-loading, we propose iterative algorithms to arrive at the optimum integer bit allocation for practical constel- lation sizing. The first such algorithm (Algorithm 8) computes the power allocation using Algorithm 7 and rounds the corresponding bit load to the nearest higher integer. Then a greedy bit-removal is executed till both the power and interference constraints are met. The next algorithm (Algorithm 9), on the other hand, involves a greedy bit allocation which reduces the computational complexity. In its first pass, bit-loading is executed till the power constraint is met; and in the sec- ond pass, bit-removal is performed till the interference constraint is satisfied. The algorithm is as follows: 2) εk’, i = ceil(εk’, i), where ceil() represents rounding to the nearest higher integer. 2) εk’, i = ceil(εk’, i), where ceil() represents rounding to the nearest higher integer. 2) εk’, i = ceil(εk’, i), where ceil() represents rounding to the nearest higher integer. Algorithm 9 g 1) Compute the transmit power, ζk’, i, using Algorithm 7. 1) Compute the transmit power, ζk’, i, using Algorithm 7. Compute the total power allocated as Ps = ∑i ζk’, i Compute the power required to add one bit to the ith SU subcarrier as pk′,i = 1 αk′,i 2bk′,i. 4) If {(Ps > Pt) OR (Ij > Ith)} , Add one bit to the sub-carrier that corresponds to the lowest Δpk’, i. { While {Ij > Ith ∀j } Do { Update the bit allocation profile, bk’, i and the corre- sponding power allocation profile, pk’, i. { Compute the power saved in removing one bit from the ith SU subcarrier as ζk′,i = 1 αk′,i 2εk′,i−1, where αk′,i = hk′,i σ 2 i Compute the reduced interference in the jth PU sub-band due to removal of one bit from every ith SU sub-carrier as ΔIj, i = gk’, j Δζk’, i Qj, i, which is a vector of size Np × Ns. { Compute the power saved in removing one bit from the ith SU subcarrier as ζk′,i = 1 αk′,i 2εk′,i−1, where Compute the total power allocated as Ps = ∑i pk’, i. } αk′,i = hk′,i σ 2 i Compute the reduced interference in the jth PU sub-band due to removal of one bit from every ith SU sub-carrier as ΔIj, i = gk’, j Δζk’, i Qj, i, which is a vector of size Np × Ns. αk′,i = hk′,i σ 2 i Compute the reduced interference in the jth PU sub-band due to removal of one bit from every ith SU sub-carrier as ΔIj, i = gk’, j Δζk’, i Qj, i, which is a vector of size Np × Ns. 5) Compute the interference caused to each PU sub-band, Ij, using left hand side of (42). 6) While { Ij > Ith ∀j } Do Compute the maximum element of the vector ΔIj, i, max{ΔIj, i}, and remove a bit from the sub-carrier identified by the corresponding column index i. { Compute the power saved in removing one bit from the ith SU subcarrier as pk′,i = 1 αk′,i 2bk′,i−1. Compute the power saved in removing one bit from the ith SU subcarrier as pk′,i = 1 αk′,i 2bk′,i−1. Update the bit allocation profile, εk’, i and the corre- sponding power allocation profile, ζk’, i. Update the bit allocation profile, εk’, i and the corre- sponding power allocation profile, ζk’, i. While {Ps > Pt } Do { 1) Compute the transmit power, ζk’, i, using Algorithm 7. 1) Compute the transmit power, ζk’, i, using Algorithm 7. 1) Compute the metric h(k, i)/∑j g(k, j), which is a vector of size K × Ns. 1) Compute the metric h(k, i)/∑j g(k, j), which is a vector of size K × Ns. 1) Compute the metric h(k, i)/∑j g(k, j), which is a vector of size K × Ns. Compute the corresponding bit-load εk’, i using (59). The subscript k’ indicates the optimum user assign- ment on the ith subcarrier using rk, i. 2) Identify the maximum element of each column, and corresponding row index k’ denotes the assignment of that user to the ith sub-carrier. 3) Initialize the bits allocated to each ith sub-carrier, bk’, i to zero. Compute the corresponding power allocation pk’, i, and the total power allocation as Ps = ∑i pk’, i. Compute the corresponding power allocation pk’, i, and the total power allocation as Ps = ∑i pk’, i. Compute the interference caused to each PU sub- band, Ij, using the right hand side of (42). Compute the corresponding power allocation pk’, i, and the total power allocation as Ps = ∑i pk’, i. Compute the interference caused to each PU sub- band, Ij, using the right hand side of (42). −χ( K k=1 Ns i=1 ζk,i −Pt) − Ns i=1 γi( K k=1 ρk,i −1) (67) (67) 4) While { (Ps < Pt) AND (Ij < Ith)} Do 4) While { (Ps < Pt) AND (Ij < Ith)} Do −ω(B − fc) + K k=1 N i=1 ψk,i(ζk,i) (68) (68) Compute the metric ΔW Pk’, i = Δpk’, i /di, which represents the power spent in adding one bit to the ith SU subcarrier weighted by the distance of the ith sub-carrier from the PU band. The non-linearity of the equations associated with the problem solution (as explained for the corresponding single user case in Appendix XV), coupled with the task of assigning sub-carriers to the multiple users, make the problem extremely complex. To make the problem tractable, a search-based algorithm is executed, in dis- crete steps of Ns, to identify that the value of the sub- carrier bandwidth, which will maximize the net through- put of the SUs, within the power budget, while mitigat- ing the PU interference. The algorithm is as follows: Add one bit to the sub-carrier that corresponds to the lowest ΔW Pk’, i. Update the bit allocation profile, bk’, i, the corre- sponding power allocation profile, pk’, i, and the inter- ference caused to each PU sub-band, Ij j Compute the total power allocated as Ps = ∑i pk’, i. } Algorithm 11 1) Initialize the sub-carrier bandwidth to its maximum value, i.e. B = Δfc. Problem P8 Problem P8 obj = max B,ζk,i Cm (60) (60) subject to subject to K k=1 Ns i=1 gk,jζk,iQj,i ≤Ith ∀j (61) K k=1 Ns i=1 gk,jζk,iQj,i ≤Ith ∀j (61) (61) The execution of the two passes can be further con- densed to a single loop, which executes till both the power and interference constraints are met. Algorithm 10 achieves that by using the power weighted by the dis- tance from the PU band, as its metric. K k=1 Ns i=1 ζk,i ≤Pt (62) K k=1 Ns i=1 ζk,i ≤Pt (62) K k=1 ρk,i = 1 ∀i (63) (63) Algorithm 10 1) Compute the metric h(k, i)/∑j g(k, j), which is a vector of size K × Ns. Algorithm 10 1) Compute the metric h(k, i)/∑j g(k, j), which is a vector of size K × Ns. ζk,i ≥0 ∀k, i (64) ζk,i ≥0 ∀k, i (64) 2) Identify the maximum element of each column, and corresponding row index k’ denotes the assignment of that user to the ith sub-carrier. 2) Identify the maximum element of each column, and corresponding row index k’ denotes the assignment of that user to the ith sub-carrier. 0 ≤B ≤ fc (65) 0 ≤B ≤ fc (65) 3) Initialize the bits allocated to each ith sub-carrier, bk’, i to zero. The Lagrangian for the above is formulated as The Lagrangian for the above is formulated as L(ζk,i,ρk,i, κj, χ, γi, ω, ψk,i) = Cm − Np j=1 κj( K k=1 Ns i=1 gk, jζk,iQj,i −Ith) (66) −χ( K k=1 Ns i=1 ζk,i −Pt) − Ns i=1 γi( K k=1 ρk,i −1) (67) −ω(B − fc) + K k=1 N i=1 ψk,i(ζk,i) (68) L(ζk,i,ρk,i, κj, χ, γi, ω, ψk,i) = Cm − Np j=1 κj( K k=1 Ns i=1 gk, jζk,iQj,i −Ith) (66) Compute the corresponding power allocation pk’, i, and the total power allocation as Ps = ∑i pk’, i. Compute the corresponding power allocation pk’, i, and the total power allocation as Ps = ∑i pk’, i. Compute the interference caused to each PU sub- band, Ij, using the right hand side of (42). 3) Initialize Cmprev(pk,i) = Cmnew(pk,i) = 0 (where Cm (pk, i) represents the achievable throughput obtained from (38)). Compute the total power allocated as Ps = ∑i ζk’, i , Compute the reduced interference in the jth PU sub-band due to removal of one bit from every ith SU sub-carrier as as ΔIj, i = gk’, j Δpk’, i Qj, i, which is a vector of size Np × Ns. , Compute the reduced interference in the jth PU sub-band due to removal of one bit from every ith SU sub-carrier as as ΔIj, i = gk’, j Δpk’, i Qj, i, which is a vector of size Np × Ns. While {Ps > Pt } Do { Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 13 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Compute the maximum element of the vector ΔIj, i, max{ΔIj, i}, and remove a bit from the sub-carrier identified by the corresponding column index i. Update the bit allocation profile, bk’, i, the corre- sponding power allocation profile, pk’, i, and the inter- ference caused to each PU sub-band, Ij } A. Single User Algorithms 4) Ns = Ns-s. The computational complexity of single-user water-fill- ing for conventional OFDM is O(N log N), where N is the number of sub-carriers that are sorted in the order of their channel gain-to noise ratio [28]. The proposed power allocation algorithm for a single SU in the CR scenario (Algorithm 1) is an iterative process which starts from some initial values of the Lagrangian multi- pliers μ and lj, and converges when both the PU interference constraint and power budget are satisfied. Let C1 and C2 represent the initial total power (∑Pi) and initial maximum interference among all the PU sub- bands (max Ij), respectively. Then, since each iteration entails Np × Ns computations, the complexity of this algorithm is given by Calculate the throughput for the number of sub-car- riers Ns, Ns+1, Ns-1 and represent them as CmNs(pk,i), CmNs−1(pk,i), CmNs−1(pk,i), respectively, using Algorithm 7 and (38). Calculate the throughput for the number of sub-car- riers Ns, Ns+1, Ns-1 and represent them as CmNs(pk,i), CmNs−1(pk,i), CmNs−1(pk,i), respectively, using Algorithm 7 and (38). 5) While {(CNs(pk, i) <CNs+1(pk, i)) OR (CNs(pk, i) <CNs-1 (pk, i))} Do 5) While {(CNs(pk, i) <CNs+1(pk, i)) OR (CNs(pk, i) <CNs-1 (pk, i))} Do { s = ceil(s/2) If {CNs(pk, i) <CNs+1(pk, i)} Ns = Ns+s. end If If {CNs(pk, i) <CNs-1(pk, i)} Ns = Ns-s. end If Calculate throughput for the number of sub-carriers Ns, Ns+1, Ns-1 and represent them as CmNs(pk,i), CmNs−1(pk,i), CmNs−1(pk,i), respectively, using Algorithm 7 and (39). } O(max(C1 −Pt, C2 −Ith)NpNs) (70) (70) The first of the proposed bit loading algorithms for CR (Algorithm 2) obtains non-real values of the bit allocation using Lagrangian multiplier optimization, followed by rounding to the nearest higher integer and then a greedy bit removal till both the PU interference constraint and power budget are satisfied. Each greedy bit-removal from Ns SU sub-carriers involves Np × Ns computations. Thus, the computational complexity is given by 6) Nsopt = Ns, and the corresponding sub-carrier band- width Bopt is obtained using (33). O(\|max(C1 −Pt, C2 −Ith)NpNs\| + \|NpN2 s \|) (71) (71) XIII. Complexity Analysis In this section, we analyze the worst-case computational complexity of each of the proposed algorithms. Calculate throughput Cmnew(pk,i) using (38). Calculate throughput Cmnew(pk,i) using (38). Algorithm 12 1) Compute the optimum power allocation pk, i and sub-carrier bandwidth B using Algorithm 11. 1) Compute the optimum power allocation pk, i and sub-carrier bandwidth B using Algorithm 11. 2) Compute the corresponding bit load bk, i using (59), and ε k, i = bk, i * rk, i. 2) Compute the corresponding bit load bk, i using (59), and ε k, i = bk, i * rk, i. Increment the number of sub-carriers with a suitable step-size s, i.e. Ns = Ns+s. 3) Execute Step 2 onwards of Algorithm 8. 3) Execute Step 2 onwards of Algorithm 8. Find the assignment of sub-carriers to users and the power allocation using Algorithm 7. Find the assignment of sub-carriers to users and the power allocation using Algorithm 7. XI. Sub-carrier bandwidth sizing (Multiple SUs) The sub-carrier bandwidth sizing problem for multiple SUs entails computing the assignment of sub-carriers to the users, and the optimum power and bandwidth that will maximize the sum throughput, subject to a net power budget and PU interference constraint. The pro- blem is formulated as 2) Calculate the corresponding number of sub carriers as Nsmin using 33. 3) Initialize Cmprev(pk,i) = Cmnew(pk,i) = 0 (where Cm (pk, i) represents the achievable throughput obtained from (38)). Page 14 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Initialize the number of sub-carriers Ns = Nsmin While {Cmprev(pk,i) <= Cmnew(pk,i)} Do {Cmprev(pk,i) = Cmnew(pk,i) XII. Sub-carrier power allocation, bandwidth sizing and bit loading (Multiple SUs) The complexity of greedy bit-addition and bit-removal for conventional OFDM are given as O(BtarN ) and O ((Bmax - Btar)N ), respectively. In the proposed greedy bit-addition algorithm for CR (Algorithm 3), we start with an all zeros allocation and add bits till the power budget is met. This involves log(Pt × CNR) iterations, where CNR represents the maximum channel-to-noise ratio among all the SU sub-carriers. Then, the algorithm executes a greedy bit-removal till the PU interference constraint is met. The complexity of Algorithm 3 is given by The joint problem of power allocation, bit loading and bandwidth sizing for the multiple SU scenario, is posed as follows: obj = max Cm (69) (69) obj = max Cm obj = max Cm subject to the PU interference constraint (42), power budget (55), the integer bit granularity (57), bounds on the sub-carrier bandwidth (30), and the integer variable for assigning sub-carriers to users (44). The iterative algorithm which is meant to solve for optimum power allocation, bit load and sub-carrier bandwidth, operates as follows: O(\|log(Pt × CNR)Ns\| + \|log(Ith Pt × CNR)NpNs\|) (72) O(\|log(Pt × CNR)Ns\| + \|log(Ith Pt × CNR)NpNs\|) Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 15 of 24 Page 15 of 24 As compared to Algorithm 3, Algorithm 4 has condensed the two passes into a single pass which exe- cutes till both the constraints are satisfied. Its complexity is given by Algorithm 6 uses Algorithm 5 for power allocation and bandwidth sizing, which is followed by greedy bit-removal using Algorithm 2. Thus, its computational complexity is given by O(\|X\| + \|Y\| + \|NpN2 sopt\|) (81) O(min(log(Pt × CNR), log(Ith × CNR)) × NpNs) (73) (81) Out of the three proposed bit loading algorithms, Algorithm 2 has the highest computational complexity. This is followed by Algorithm 4 and Algorithm 3 in that order. It can be attributed to the following facts: Algorithm 2 involves calculation Pi using Algorithm 1, and hence the overall number of iterations are very high. Algorithm 3 initially entails a conventional greedy bit-addition pass which has complexity O(Ns) in each iteration, and is followed by a greedy bit-removal pass with a complexity O(NpNs) in each iteration. The sec- ond pass will involve very few iterations, unless the power budget is very high. On the other hand, in Algo- rithm 4, each essential iteration requires a complexity O(NpNs). B. Multiple User Algorithms The multi-user power allocation algorithm (Algorithm 7) entails assigning sub-carriers to the K SUs, besides the power allocation. Hence, its complexity is given by O(min(C −Pt, C −Ith)K NpNs) (82) (82) The complexity of the bit allocation algorithm (Algorithm 8) is given by O(\| max(C1 −Pt, C2 −Ith)K NpNs\| + \|NpN2 s \|) (83) (83) The complexity of Algorithm 9 and Algorithm 10 will be the same as the corresponding single-user algorithms (Algorithm 3 and Algorithm 4). As such, their complex- ities are given by (72) and (73), respectively. This is because, in these two algorithms assigning sub-carriers to users is not an iterative process, rather is computed once at the beginning of the algorithms. The sub-carrier bandwidth sizing algorithm (Algorithm 5) involves two passes: in the first pass a crude search is conducted with a relatively larger step size of Ns, followed by a fine search to look for the global optima. The computational complexity of the complete algorithm is therefore given by O(\|X\| + \|Y\|) (74) (74) O(\|X\| + \|Y\|) Sub-carrier bandwidth sizing for multiple SUs (Algorithm 11) also involves assigning sub-carriers to the K SUs, and its complexity is given by (referring to the corresponding single-user algorithm complexity in (74)) in which X and Y represent the complexities of the crude search and fine search respectively. The algorithm involves computing the power allocation for each Ns using Algorithm 1. If we denote Nsmax−Nsmin s as SN, then O(\|K X\| + \|K Y\|) (84) (84) O(\|K X\| + \|K Y\|) The complexity of Algorithm 12, which involves power allocation, bandwidth sizing and bit loading for multiple SUs is given by (referring to the corresponding single- user algorithm complexity in (81)) X = max(C1 −Pt, C2 −Ith) × Np × (Nsmin + (Nsmin + s) (75) +(Nsmin + 2s) + . . . (Nsmin + SNs)) (76) +(Nsmin + 2s) + . . . (Nsmin + SNs)) (76) (76) O(\|K X\| + \|K Y\| + \|NpN2 sopt\|) (85) (85) which can be simplified as X = max(C1 −Pt, C2 −Ith)NpSNNsmax (77) (77) XII. Sub-carrier power allocation, bandwidth sizing and bit loading (Multiple SUs) where Nsopt represents the number of sub-carriers cor- responding to the optimum bandwidth. XIV. Simulation Results and Discussion A. Single user Power Allocation The fine search involves computing the power alloca- tion three times corresponding to Ns, Ns + 1 and Ns - 1 in each iteration. Thus, we get For the single user case, we assume that the total system bandwidth for the PU and SU is 6 MHz wide, of which the SU occupies a contiguous band of 5 MHz, while the PU occupies 1 MHz bandwidth. The SU transceiver uses 32 sub-carrier OFDM for communication. The channels undergo Rayleigh multi-path fading, defined in the time domain by L−1 l=0 hlδ(t −lT) where hl is the com- plex amplitude of path l and L is the number of channel Y = max(C1 −Pt, C2 −Ith) × Np × (Nsmax + (Nsmax + s 2)+ (78) (Nsmax + s 2 + s 4) + (Nsmax + s 2 + s 4 + s 8) + · · · · (Nsmax + · · · · + s 2log s )) (79) which can be simplified as Y = max(C1 −Pt, C2 −Ith) × Np × log s × (Nsmax + ( s 2)log s) (80) Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 16 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 16 of 24 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.5 2 2.5 3 3.5 4 4.5 5 5.5 6 Primary power budget ( Pt ) in Watts Throughput in Bits/sec/Hz Proposed scheme (Algo 1) Waterfillingíbased allocation Uniform allocation Capílimited scheme Figure 5 Secondary user throughput versus power budget. taps. The lth channel coefficient is distributed as N(0, s2), and the frequency domain channel is given by its Fourier Transform. The AWGN variance is assumed to be s2 = 1e-4 and the guard interval length is Tg = 1 μsec. The PU band is divided into 8 sub-bands, and we attempt to mitigate the interference to each of them. We set the interference temperature, Tth = 1e-5 W/Hz for each PU sub-band. Without loss of generality, it is assumed that the interference induced by the PU to the SU is negligible. The power allocation profile of the SU is shown in Figure 4. For the result of Figure 4a, the power budget Pt = 100 mW. XIV. Simulation Results and Discussion A. Single user Power Allocation We have also included for comparison, the power allocation scheme which consid- ers the entire PU as a single band for mitigating 0 4 8 12 16 20 24 28 32 0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Subícarrier Index Power in mW Primary Band 5 MHz 1 MHz (a) 0 4 8 12 16 20 24 28 32 0 10 20 30 40 50 60 Subícarrier Index Power in mW Primary Band (b) Figure 4 Power profile: (a) Pt = 100 mW; (b) Pt = 1 W. 1 2 3 4 5 6 7 8 0 0.2 0.4 0.6 0.8 1 1.2 1.4 x 10 í4 Primary subíband Index Interference power in watts/Hz Proposed scheme ( Algo 1 ) Capílimited scheme Waterfilling based allocation Uniform allocation Considering single PU band Ith Figure 6 Primary user interference profile. 0 4 8 12 16 20 24 28 32 0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Subícarrier Index Power in mW Primary Band 5 MHz 1 MHz (a) 0 4 8 12 16 20 24 28 32 0 10 20 30 40 50 60 Subícarrier Index Power in mW Primary Band (b) Figure 4 Power profile: (a) Pt = 100 mW; (b) Pt = 1 W. 0 4 8 12 16 20 24 28 32 0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 Subícarrier Index Power in mW Primary Band 5 MHz 1 MHz (a) We have also provided the interference profile to the 8 PU sub-bands on execution of the various power allo- cation schemes (Figure 6). The proposed algorithm maintains the interference to each PU sub-band under the threshold. The cap-limited scheme is successful in keeping only the interference to the closest PU sub- band under the threshold. XIV. Simulation Results and Discussion A. Single user Power Allocation This value being very small, the interfer- ence constraint is non-binding, and it is observed (though the channel gains have not been plotted) that the solution closely resembles that of conventional water-filling: better channels are allocated higher powers as compared to the poorer ones. As the power budget is increased to 1 W (Figure 4b), the interference constraint becomes binding. The graph tapers towards the PU band because lesser power is allocated in the SU sub- carriers spectrally closer to the PU. Figure 5 Secondary user throughput versus power budget. power allocation. The cap-limited scheme [10] is only partially interference constrained by the closest PU sub-band. On the other hand, the proposed scheme (Algorithm 1) considers the interference threshold to each PU sub-band. The SU throughput achieved is the optimum result within the given power budget and interference constraints. It is observed that when the power budget is very low, the solution is very close to that of conventional water-filling since only the power constraint is binding. Furthermore, after a certain power budget (Pt = 700 mW), the throughput hardly increases, since only the interference constraint is binding. Any further increase in the power budget, cannot increase the SU sub-carrier power allocation without violating the interference constraint. These results have been averaged over 100 independent reali- zations of the channel. The SU throughput versus power budget is plotted in Figure 5. Conventional water-filling gives the highest SU throughput, since it is unconstrained by the PU interference threshold. It is closely followed by uniform binding. Any further increase in the power budget, cannot increase the SU sub-carrier power allocation without violating the interference constraint. These results have been averaged over 100 independent reali- zations of the channel. We have also provided the interference profile to the 8 PU sub-bands on execution of the various power allo- cation schemes (Figure 6). The proposed algorithm maintains the interference to each PU sub-band under the threshold. The cap-limited scheme is successful in keeping only the interference to the closest PU sub- band under the threshold. XIV. Simulation Results and Discussion A. Single user Power Allocation We have also included for comparison, the power allocation scheme which consid- ers the entire PU as a single band for mitigating (a) 0 4 8 12 16 20 24 28 32 0 10 20 30 40 50 60 Subícarrier Index Power in mW Primary Band (b) gure 4 Power profile (a) P 100 mW (b) P 1 W 1 2 3 4 5 6 7 8 0 0.2 0.4 0.6 0.8 1 1.2 1.4 x 10 í4 Primary subíband Index Interference power in watts/Hz Proposed scheme ( Algo 1 ) Capílimited scheme Waterfilling based allocation Uniform allocation Considering single PU band Ith Figure 6 Primary user interference profile. 1 2 3 4 5 6 7 8 0 0.2 0.4 0.6 0.8 1 1.2 1.4 x 10 í4 Primary subíband Index Interference power in watts/Hz Proposed scheme ( Algo 1 ) Capílimited scheme Waterfilling based allocation Uniform allocation Considering single PU band Ith Figure 6 Primary user interference profile. (b) Figure 6 Primary user interference profile. Figure 4 Power profile: (a) Pt = 100 mW; (b) Pt = 1 W. Figure 4 Power profile: (a) Pt = 100 mW; (b) Pt = 1 W. Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 17 of 24 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 9 Subícarrier Index Number of Bits Primary Band (a) 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 9 Subícarrier Index Number of Bits Primary Band (b) 0 5 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 9 Subícarrier Index Numbers of Bits Primary Band (c) Figure 8 Bit profile for Algorithm 2: (a) After Lagrangian; (b) rounding; (c) bit removal. interference, as is done in literature [6]. It can be observed that it is not as effective as dividing the PU into sub-bands, as is proposed in Algorithm 1. These results are reported for a power budget Pt = 1 W, and a single instance of the channel. Bit loading (a) The simulation parameters for bit loading are the same as those assumed for power allocation. The SNR gap is computed considering an error probability Pe = 10-6. 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 9 Subícarrier Index Number of Bits Primary Band (b) The result of Algorithm 2 is depicted in Figure 8. Fig- ure 8a demonstrates the results of executing the Lagran- gian optimization problem, which yields real numbers for the bit allocation. This is followed by integer quanti- zation to the nearest higher integer (Figure 8b), causing the interference to the PU to increase beyond the speci- fied threshold (Ith), or the power constraint (Pt) to be violated. A greedy bit-removal is executed, resulting in the final allocation of bits (Figure 8c). It can be observed that the interference constraint causes the profile to taper towards the PU band. A power budget Pt of 1 W is considered for these graphs. (b) 0 5 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 9 Subícarrier Index Numbers of Bits Primary Band (c) The bit allocation profile achieved with the proposed algorithms, i.e. Algorithm 2, Algorithm 3 and Algorithm 4 are shown together in Figure 9; while those obtained with the schemes proposed in literature are depicted in Figure 10. A power budget Pt of 1 W is considered in each of these cases. Though much cannot be concluded from Figure 9, the SU throughput for various algo- rithms, averaged over 100 independent realizations of the channel (Figure 11), reveals more about their (c) Figure 7 Power profile for different positions of the primary user. Figure 8 Bit profile for Algorithm 2: (a) After Lagrangian; (b) rounding; (c) bit removal. 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits Proposed scheme ( Algo 2 ) Proposed scheme ( Algo 3 ) Proposed Scheme ( Algo 4 ) Primary Band Figure 9 Bit profile (Algorithms 2-4). 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits Proposed scheme ( Algo 2 ) Proposed scheme ( Algo 3 ) Proposed Scheme ( Algo 4 ) Primary Band Figure 9 Bit profile (Algorithms 2-4). XIV. Simulation Results and Discussion A. Single user Power Allocation 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 9 Subícarrier Index Number of Bits Primary Band (a) The proposed power allocation algorithm is verified for five different positions of the PU within the given system bandwidth of 6 MHz, and a power budget of 1 W in each case. As observed in Figure 7, each profile tapers towards the PU band. Bit loading Bit loading Figure 7 Power profile for different positions of the primary user. Figure 7 Power profile for different positions of the primary user. Figure 9 Bit profile (Algorithms 2-4). Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 18 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 18 of 24 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits Literature ( Capílimited ) Literature ( Nulling ) Primary Band Figure 10 Bit profile (literature). 1 2 3 4 5 6 7 8 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 x 10 í5 Primary subíband Index Interference Power in Watts/Hz Literature ( Capílimited ) Proposed scheme ( Algo 4 ) Proposed sheme ( Algo 2 ) Literature ( Nulling ) Proposed scheme ( Algo 3 ) Ith Figure 12 Primary user interference profile (all bit loading schemes). 1 2 3 4 5 6 7 8 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 x 10 í5 Primary subíband Index Interference Power in Watts/Hz Literature ( Capílimited ) Proposed scheme ( Algo 4 ) Proposed sheme ( Algo 2 ) Literature ( Nulling ) Proposed scheme ( Algo 3 ) Ith Figure 12 Primary user interference profile (all bit loading schemes). 1 2 3 4 5 6 7 8 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 x 10 í5 Primary subíband Index Interference Power in Watts/Hz Literature ( Capílimited ) Proposed scheme ( Algo 4 ) Proposed sheme ( Algo 2 ) Literature ( Nulling ) Proposed scheme ( Algo 3 ) Ith Figure 12 Primary user interference profile (all bit loading schemes). 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits Literature ( Capílimited ) Literature ( Nulling ) Primary Band Figure 10 Bit profile (literature). Figure 12 Primary user interference profile (all bit loading schemes). performance: the highest throughput is achieved by the cap-limited scheme proposed in literature [10], since it only mitigates interference to the closest PU band. It is followed closely by that obtained from the proposed Algorithm 2. Algorithm 3 and Algorithm 4 follow, in that order. Bit loading The scheme from literature which is based on sub-carrier nulling [14] gives the lowest throughput. sτ is the rms delay spread, and assumed to be 1 μ sec). The power budget at the SU Tx is Pt = 1 W. p g In Figure 13, we analyze the SU throughput, while increasing the sub-carrier bandwidth unto the coher- ence bandwidth Δfc. Although not plotted, it is expected that the SNR will increase with an increase in the sub-carrier bandwidth, however, the same cannot be said about the throughput. It is observed (Figure 13) that unto a certain point, an increase in bandwidth results in a corresponding increase in throughput; after which, any further increase results in the symbol dura- tion becoming relatively smaller than the guard inter- val, and the throughput reduces. Initially, a crude search was conducted by varying Ns in a step size of 10 as indicated by the markers in Figure 13. Then a fine search was conducted to look for the global optima. The execution of Algorthm 5 yielded Nsopt as 101 and the corresponding Bopt as 99.01 Khz. The optimum SU sub-carrier bandwidth should also main- tain the interference to the PU band below the speci- fied threshold. To understand the effect of varying sub-carrier bandwidth on the PU interference, we have plotted Figures 14 and 15 by dividing the PU band into 4 and 8 sub-bands, respectively, and allocating the power uniformly. In both cases, the following The interference profile to the 8 PU sub-bands on execution of the aforementioned bit allocation schemes is depicted in Figure 12. While the proposed algorithms, Algorithm 2-4, are successful in mitigating the interfer- ence to each of the PU sub-bands, the cap-limited scheme only does so for the spectrally closest PU sub- band. The subcarrier-nulling scheme, however, generates the lowest interference profile, since it nulls sub-carriers till interference to each band is mitigated. Sub-carrier bandwidth sizing The simulation parameters are the same as those described for the power allocation and bit loading pro- blems. However, the 5 MHz SU bandwidth is not divided into 32 sub-carriers anymore. Instead, the pro- blem entails searching for that sub-carrier bandwidth which will maximize the SU throughput, while mitigat- ing the interference to the PU band. The coherence bandwidth is Δfc = 200 KHz (Δfc = 1 = 5sτ [46], where 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 7 8 9 10 Power budget ( Pt ) in Watt Throughput in Bits/sec/Hz Proposed scheme ( Algo 2 ) Proposed scheme ( Algo 3 ) Proposed scheme ( Algo4 ) Literature ( Capílimited) Literature ( Nulling ) Figure 11 Secondary user throughput versus power budget (all bit loading schemes). 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 5 3 4 5 6 7 8 9 X: 9.901e+004 Y: 8.732 Subícarrier Bandwidth in Hz Throughput in Bits/sec/Hz Nopt = 101 Bopt Figure 13 Secondary user throughput vs. sub-carrier bandwidth. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 5 3 4 5 6 7 8 9 X: 9.901e+004 Y: 8.732 Subícarrier Bandwidth in Hz Throughput in Bits/sec/Hz Nopt = 101 Bopt Figure 13 Secondary user throughput vs. sub-carrier bandwidth. 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 7 8 9 10 Power budget ( Pt ) in Watt Throughput in Bits/sec/Hz Proposed scheme ( Algo 2 ) Proposed scheme ( Algo 3 ) Proposed scheme ( Algo4 ) Literature ( Capílimited) Literature ( Nulling ) Figure 11 Secondary user throughput versus power budget (all bit loading schemes). 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 7 8 9 10 Power budget ( Pt ) in Watt Throughput in Bits/sec/Hz Proposed scheme ( Algo 2 ) Proposed scheme ( Algo 3 ) Proposed scheme ( Algo4 ) Literature ( Capílimited) Literature ( Nulling ) Figure 13 Secondary user throughput vs. sub-carrier bandwidth. Figure 11 Secondary user throughput versus power budget (all bit loading schemes). Page 19 of 24 Page 19 of 24 Thumar et al. Sub-carrier bandwidth sizing EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 1 2 3 4 0.8 1 1.2 1.4 1.6 1.8 2 2.2 x 10 5 0 1 2 3 4 5 6 x 10 í4 Secondary subícarrier bandwidth in Hz Primary subíband Index Interference power in Watts/Hz (a) 1 2 3 4 0.8 1 1.2 1.4 1.6 1.8 2 2.2 x 10 5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 x 10 í4 Secondary subícarrier bandwidth in Hz Interference power in Watts/Hz Primary subíband Index (b) Figure 14 Interference to four primary user sub-bands: (a) Without PU gains; (b) with PU gains. 1 2 3 4 0.8 1 1.2 1.4 1.6 1.8 2 2.2 x 10 5 0 1 2 3 4 5 6 x 10 í4 Secondary subícarrier bandwidth in Hz Primary subíband Index Interference power in Watts/Hz (a) observations are made: (1) as the SU sub-carrier band- width decreases, the interference to the PU sub-bands decreases. This is because, decreasing the bandwidth implies increasing the number of sub-carriers, and thereby, the power budget is now distributed among a relatively large number of sub-carriers; a lower power in each sub-carrier generates lower interference power to the PU band. (2) As expected, the interference decreases with the increasing spectral distance. The same results will be obtained for water-filling-based allocation. (a) Figures 14a and 14b (resp. Figures 15a and 15b), represent the interference profile of the PU, without and with consideration of the SU to PU channel gains, respectively, for 4 PU sub-bands (resp. 8 PU sub-bands). 1 2 3 4 0.8 1 1.2 1.4 1.6 1.8 2 2.2 x 10 5 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 x 10 í4 Secondary subícarrier bandwidth in Hz Interference power in Watts/Hz Primary subíband Index (b) y The interference profile to 8 PU sub-bands with the optimum SU sub-carrier bandwidth is reported in Figure 16, with uniform power allocation, water-filling and the proposed algorithm (Algorithm 5). Only Algorithm 5 is successful in keeping the interference to the PU band below the specified threshold. Sub-carrier power allocation, bandwidth sizing and bit- loading Sub-carrier power allocation, bandwidth sizing and bit- loading On executing Algorithm 6, first the optimum bandwidth is obtained with the corresponding power allocation. Figure 17a reports the optimum number of sub-carriers as Nsopt = 78 and the bandwidth Bopt = 126.6 KHz, while Figure 17b depicts the power allocation profile. The bit loading profile is as shown in Figure 17c. (b) Figure 14 Interference to four primary user sub-bands: (a) Without PU gains; (b) with PU gains. 1 2 3 4 5 6 7 8 0.8 1 1.2 1.4 1.6 1.8 2 2.2 x 10 5 0 0.2 0.4 0.6 0.8 1 x 10 í4 Secondary subícarrier bandwidth in Hz Primary subíband Index Interference Power in Watts/Hz (a) 1 2 3 4 5 6 7 8 1 1.5 2 x 10 5 0 2 4 6 x 10 í5 Secondary subícarrier bandwidth in Hz Primary subíband Index Interference power in Watts/Hz (b) Figure 15 Interference to eight primary user sub-bands: a without PU gains and; b with PU gains. 1 2 3 4 5 6 7 8 0.8 1 1.2 1.4 1.6 1.8 2 2.2 x 10 5 0 0.2 0.4 0.6 0.8 1 x 10 í4 Secondary subícarrier bandwidth in Hz Primary subíband Index Interference Power in Watts/Hz (a) Power Allocation The simulation parameters are the same as those of the single user case. 3 SUs have been assumed, which con- tend for the 5 MHz bandwidth, which is divided into 32 OFDM sub-carriers. Figure 18 demonstrates the power allocation profile and the assignment of sub-carriers to the users, with a total power budget Pt = 1W. Due to the interference constraint, the profile tapers towards the PU band. Figure 19 depicts the sum throughput the users, with a total power budget Pt 1W. Due to the interference constraint, the profile tapers towards the PU band. Figure 19 depicts the sum throughput 7 8 1.8 2 2.2 Secondary subícarrier bandwidth in Hz Primary subíband Index (a) 1 2 3 4 5 6 7 8 1 1.5 2 x 10 5 0 2 4 6 x 10 í5 Secondary subícarrier bandwidth in Hz Primary subíband Index Interference power in Watts/Hz (b) Figure 15 Interference to eight primary user sub-bands: a without PU gains and; b with PU gains. 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 x 10 í5 Primary subíband Index Interference power in Watts/Hz Waterfilling based allocation Uniform allocation Proposed scheme ( Algo 5 ) Ith Figure 16 Primary user interference profile (Algorithm 5). (a) 1 2 3 4 5 6 7 8 1 1.5 2 x 10 5 0 2 4 6 x 10 í5 Secondary subícarrier bandwidth in Hz Primary subíband Index Interference power in Watts/Hz (b) 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 x 10 í5 Primary subíband Index Interference power in Watts/Hz Waterfilling based allocation Uniform allocation Proposed scheme ( Algo 5 ) Ith Figure 16 Primary user interference profile (Algorithm 5). 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 x 10 í5 Primary subíband Index Interference power in Watts/Hz Waterfilling based allocation Uniform allocation Proposed scheme ( Algo 5 ) Ith Figure 16 Primary user interference profile (Algorithm 5). (b) (b) Figure 15 Interference to eight primary user sub-bands: a without PU gains and; b with PU gains. Figure 16 Primary user interference profile (Algorithm 5). Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 20 of 24 Thumar et al. Power Allocation 0.1 0.2 0.3 0.4 0.5 0.6 0.7 4 5 6 7 8 9 10 Power budget ( Pt ) in Watts Throughput in Bits/sec/Hz Proposed Uniform a Capílimi Waterfilli Figure 19 Secondary user throughput versu (multiple SUs). 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 4 5 6 7 8 9 10 Power budget ( Pt ) in Watts Throughput in Bits/sec/Hz Proposed scheme ( Algo 7 ) Uniform allocation Capílimited scheme Waterfilling based allocation Figure 19 Secondary user throughput versus power budget (multiple SUs). 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 4 5 6 7 8 9 10 Power budget ( Pt ) in Watts Throughput in Bits/sec/Hz Proposed scheme ( Algo 7 ) Uniform allocation Capílimited scheme Waterfilling based allocation Figure 19 Secondary user throughput versus power budget (multiple SUs). 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 10.5 X: 1.266e+005 Y: 10.05 Subícarrier Bandwidth in Hz Throughput in Bits/sec/Hz Bopt Nopt = 78 (a) Figure 19 Secondary user throughput versus power budget (multiple SUs). (a) 0 10 20 30 40 50 60 70 80 90 0 2 4 6 8 Subícarrier Index Power in mW Primary Band Nopt = 78 (b) graphs of the SU, averaged over 100 independent reali- zations of the channel, and Figure 20 represents the PU interference profile. Both these graphs have the same interpretations as the corresponding single user results. Bit loading Figure 21 demonstrates the bit loading profile and the assignment of sub-carriers to the users on execution of Algorithm 8. Figure 21a, b and 21c depict the result of the Lagrangian optimization, rounding and bit removal, respectively. Figure 22 represents the comparative sum SU throughput of the various bit loading schemes, viz., Algorithm 8, Algorithm 9, Algorithm 10, the cap-limited scheme and sub-carrier nulling, averaged over 100 inde- pendent realizations of the channel. The highest throughput is achieved by the cap-limited scheme, since it only mitigates interference to the closest PU band. It is followed closely by that obtained from the proposed Algorithm 8. Algorithm 9 and Algorithm 10 follow in that order. The sub-carrier nulling scheme gives the lowest throughput. Figure 23 depicts the PU interfer- ence profile. Power Allocation (b) 0 10 20 30 40 50 60 70 80 90 0 1 2 3 4 5 6 7 Subícarrier Index Number of Bits Primary Band Nopt = 78 (b) Power Allocation EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 20 of 24 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 10.5 X: 1.266e+005 Y: 10.05 Subícarrier Bandwidth in Hz Throughput in Bits/sec/Hz Bopt Nopt = 78 (a) 0 10 20 30 40 50 60 70 80 90 0 2 4 6 8 Subícarrier Index Power in mW Primary Band Nopt = 78 (b) 0 10 20 30 40 50 60 70 80 90 0 1 2 3 4 5 6 7 Subícarrier Index Number of Bits Primary Band Nopt = 78 (b) Figure 17 Joint resource allocation: a Optimum bandwidth computation, b Power profile and c Bit profile. graphs of the SU, averaged over 100 ind zations of the channel, and Figure 20 re interference profile. Both these graphs interpretations as the corresponding sing Bit loading Figure 21 demonstrates the bit loading assignment of sub-carriers to the users Algorithm 8. Figure 21a, b and 21c dep the Lagrangian optimization, rounding a respectively. Figure 22 represents the co SU throughput of the various bit loadin Algorithm 8, Algorithm 9, Algorithm 10, scheme and sub-carrier nulling, averaged pendent realizations of the channel throughput is achieved by the cap-limite it only mitigates interference to the clos is followed closely by that obtained from Algorithm 8. Algorithm 9 and Algorith that order. The sub-carrier nulling sc lowest throughput. Figure 23 depicts t ence profile. Sub-carrier power allocation, bandwidth siz and bit-loading On executing Algorithm 12, first bandwidth is obtained with the corres allocation. Figure 24a reports the optim 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 5 5.5 6 6.5 7 7.5 8 8.5 9 9.5 10 10.5 X: 1.266e+005 Y: 10.05 Subícarrier Bandwidth in Hz Throughput in Bits/sec/Hz Bopt Nopt = 78 (a) 0 10 20 30 40 50 60 70 80 90 0 2 4 6 8 Subícarrier Index Power in mW Primary Band Nopt = 78 (b) 0 10 20 30 40 50 60 70 80 90 0 1 2 3 4 5 6 7 Subícarrier Index Number of Bits Primary Band Nopt = 78 (b) Figure 17 Joint resource allocation: a Optimum bandwidth computation, b Power profile and c Bit profile. Sub-carrier power allocation, bandwidth sizing and bit-loading 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits Primary Band User 1 User 2 User 3 (a) 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits User 1 User 2 User 3 Primary Band (b) 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits User 1 User 2 User 3 Primary Band (c) Figure 21 Bit profile for Algorithm 8 (multiple SUs): (a) After Lagrangian, (b) Rounding and (c) Bit removal. 0. 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 5 6 7 8 9 Power budget ( Pt ) in Watt Throughput in Bits/sec/Hz Proposed scheme ( Algo 8 ) Proposed scheme ( Algo 9 ) Proposed scheme ( Algo 10 ) Literature ( Capílimited ) Literature ( Nulling ) Figure 22 Secondary user throughput versus power budget (all bit loading schemes for multiple SUs). 0. 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 5 6 7 8 9 Power budget ( Pt ) in Watt Throughput in Bits/sec/Hz Proposed scheme ( Algo 8 ) Proposed scheme ( Algo 9 ) Proposed scheme ( Algo 10 ) Literature ( Capílimited ) Literature ( Nulling ) Figure 22 Secondary user throughput versus power budget (all bit loading schemes for multiple SUs). 0. 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 5 6 7 8 9 Power budget ( Pt ) in Watt Throughput in Bits/sec/Hz Proposed scheme ( Algo 8 ) Proposed scheme ( Algo 9 ) Proposed scheme ( Algo 10 ) Literature ( Capílimited ) Literature ( Nulling ) Figure 22 Secondary user throughput versus power budget (all bit loading schemes for multiple SUs). 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits Primary Band User 1 User 2 User 3 (a) (a) Figure 22 Secondary user throughput versus power budget (all bit loading schemes for multiple SUs). Sub-carrier power allocation, bandwidth sizing and bit-loading (b) On executing Algorithm 12, first the optimum bandwidth is obtained with the corresponding power allocation. Figure 24a reports the optimum number of Figure 17 Joint resource allocation: a Optimum bandwidth computation, b Power profile and c Bit profile. 0 4 8 12 16 20 24 28 32 0 5 10 15 20 25 30 35 40 Subícarrier Index Power in mW User 1 User 2 User 3 Primary Band Figure 18 Power profile (multiple SUs). 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 x 10 í5 Primary subíband Index Interference power in Watts/Hz Uniform allocation Waterfillingíbased alocation Capílimited scheme Proposed scheme ( Algo ííí) Figure 20 Primary user interference profile (multiple SUs). 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 x 10 í5 Primary subíband Index Interference power in Watts/Hz Uniform allocation Waterfillingíbased alocation Capílimited scheme Proposed scheme ( Algo ííí) Figure 20 Primary user interference profile (multiple SUs). 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 x 10 í5 Primary subíband Index Interference power in Watts/Hz Uniform allocation Waterfillingíbased alocation Capílimited scheme Proposed scheme ( Algo ííí) Figure 20 Primary user interference profile (multiple SUs). 0 4 8 12 16 20 24 28 32 0 5 10 15 20 25 30 35 40 Subícarrier Index Power in mW User 1 User 2 User 3 Primary Band Figure 18 Power profile (multiple SUs). Figure 20 Primary user interference profile (multiple SUs). Figure 20 Primary user interference profile (multiple SUs). Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 21 of 24 Thumar et al. Sub-carrier power allocation, bandwidth sizing and bit-loading 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits User 1 User 2 User 3 Primary Band (b) issues of power allocation, bit loading and sub-carrier bandwidth sizing are addressed individually, and then as a joint problem, for both single and multiple SU scenarios; the objective being-maximization of the SU’s throughput within a power budget and PU interference constraints. The PU spectral band is divided into sub- bands, and the proposed algorithms effectively mitigate the interference to each of them. For bit loading, mul- tiple algorithms are suggested. The algorithms for sub- carrier bandwidth sizing for the SU are novel, and effective in mitigating the PU interference. The com- putational complexity of all the algorithms is analyzed. Exhaustive simulation results are provided with rigor- ous interpretations of each of the graphs, and compari- sons with techniques from literature, wherever possible. The performance results are encouraging, and motivate the deployment of the suggested strategies in practical CR networks. While the proposed algorithms are mainly for stationary SUs and may be applicable to walking speeds, the resource allocation for medium/ high speed mobile SUs, is an issue we intend to tackle in the near future. (b) 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits User 1 User 2 User 3 Primary Band (c) (c) Figure 21 Bit profile for Algorithm 8 (multiple SUs): (a) After Lagrangian, (b) Rounding and (c) Bit removal. 1 2 3 4 5 6 7 8 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 x 10 í5 Primary subíband Index Interference power in Watts/Hz Literature (Capílimited ) Proposed scheme ( Algo 9 ) Literature ( Nulling ) Proposed scheme ( algo 8 ) Proposed scheme ( Algo 10 ) Figure 23 PU interference profile (all bit loading schemes for multiple users). sub-carriers as Nsopt = 144 and the bandwidth Bopt = 68.97KHz, while Figure 24b depicts the power allocation profile and the assignment of sub-carriers to the 3 users. The bit loading profile is as shown in Figure 24c. The results of Algorithm 11 have not been included, since they would be similar to the results of bandwidth sizing, power allocation and the assignment of sub-carriers to users, as those demonstrated in Figures 24a and 24b. Sub-carrier power allocation, bandwidth sizing and bit-loading EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 21 of 24 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits Primary Band User 1 User 2 User 3 (a) 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits User 1 User 2 User 3 Primary Band (b) 0 4 8 12 16 20 24 28 32 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits User 1 User 2 User 3 Primary Band (c) Figure 21 Bit profile for Algorithm 8 (multiple SUs): (a) After Lagrangian, (b) Rounding and (c) Bit removal. issues of power allocation, bit loading and sub-carrier bandwidth sizing are addressed individually, and then as a joint problem, for both single and multiple SU scenarios; the objective being-maximization of the SU’s throughput within a power budget and PU interference constraints. The PU spectral band is divided into sub- bands, and the proposed algorithms effectively mitigate the interference to each of them. For bit loading, mul- tiple algorithms are suggested. The algorithms for sub- carrier bandwidth sizing for the SU are novel, and effective in mitigating the PU interference. The com- putational complexity of all the algorithms is analyzed. Exhaustive simulation results are provided with rigor- ous interpretations of each of the graphs, and compari- sons with techniques from literature, wherever possible. The performance results are encouraging, and motivate the deployment of the suggested strategies in practical CR networks. While the proposed algorithms are mainly for stationary SUs and may be applicable to walking speeds, the resource allocation for medium/ high speed mobile SUs, is an issue we intend to tackle in the near future. XV. Conclusion y Figure 23 PU interference profile (all bit loading schemes for multiple users). y Figure 23 PU interference profile (all bit loading schemes for multiple users). The major contribution of the paper is towards Radio Resource Management in an OFDM-based CR. The Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 22 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 22 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 22 of 24 Page 22 of 24 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 5 8 8.5 9 9.5 10 10.5 11 11.5 12 12.5 X: 6.897e+004 Y: 12.2 Subícarrier bandwidth in Hz Throughput in Bits/sec/Hz Bopt Nopt = 144 (a) 0 15 30 45 60 75 90 105 120 135 150 0 2 4 6 8 10 12 Subícarrier Index Power in mW User 1 User 2 User 3 Primary Band (b) 0 15 30 45 60 75 90 105 120 135 150 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits User 1 User 2 User 3 Primary Band (c) Figure 24 Joint resource allocation (multiple SUs): (a) Optimum bandwidth computation, (b) Power profile and (c) Bit profile. XV. Conclusion − Np j=1 κjgj 2BW B −1 i=1 Pi jthPUband Q′ j,idf −ω (87) 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 x 10 5 8 8.5 9 9.5 10 10.5 11 11.5 12 12.5 X: 6.897e+004 Y: 12.2 Subícarrier bandwidth in Hz Throughput in Bits/sec/Hz Bopt Nopt = 144 (a) (87) where Q′ j,i = {2Sinc[(f −fi)(Tg + 1 B)] × (f −fi)(−1 B2 )× (88) where Q′ j,i = {2Sinc[(f −fi)(Tg + 1 B)] × (f −fi)(−1 B2 )× (88) (88) (cos[(f −fi)(Tg + 1 B)] [(f −fi)(Tg + 1 B)] −Sin[(f −fi)(Tg + 1 B)] [(f −fi)(Tg + 1 B)] 2 )} (89) (89) Differentiating (35) wrt Pi, (a) ( ∂L ∂Pi )B∗P∗ i = 1 (Tg + 1 B)(Pi + σ 2 i hi ) (90) (90) 0 15 30 45 60 75 90 105 120 135 150 0 2 4 6 8 10 12 Subícarrier Index Power in mW User 1 User 2 User 3 Primary Band (b) − NP j=1 κjgj jthPUband Sinc2[(f −fi)(Tg + 1 B)]df −χ (91) (91) Appendix B pp Differentiating (47) wrt ζk, i and applying KKT condi- tions, ( ∂L ∂ζk,i )ζ ∗ k,iρ∗ k,i = c′ k,i( ζ ∗ k,i ρ∗ k,i ) − Np j=1 λjgk,jQj,i −μ + βk,i = 0 (92) (92) (b) The above equation is expressed as (with bk, i substi- tuted as 0, due to the complementary slackness condi- tion [42]) 0 15 30 45 60 75 90 105 120 135 150 0 1 2 3 4 5 6 7 8 Subícarrier Index Number of Bits User 1 User 2 User 3 Primary Band (c) c′ k,i( ζ ∗ k,i ρ∗ k,i ) − Np j=1 λjgk,jQj,i −μ < 0 if ζ ∗ k,i = 0; = 0 if ζ ∗ k,i > 0. (93) (93) From (92) and (93) we can write From (92) and (93) we can write ζ ∗ k,i = max([c−1′ k,i ( Np j=1 λjgk,jQj,i + μ)ρ∗ k,i], 0) (94) (94) (c) where c’−1 k,i is the inverse of derivative of function ck, i. Differentiating (47) wrt rk, i and applying KKT condi- tions, where c’−1 k,i is the inverse of derivative of function ck, i. where c’−1 k,i is the inverse of derivative of function ck, i. Differentiating (47) wrt rk, i and applying KKT condi- tions, where c’−1 k,i is the inverse of derivative of function ck, i. Figure 24 Joint resource allocation (multiple SUs): (a) Optimum bandwidth computation, (b) Power profile and (c) Bit profile. , Differentiating (47) wrt rk, i and applying KKT condi- tions, ( ∂L ∂ρk,n )ζ ∗ k,i,ρ∗ k,i = ck,i( ζ ∗ k,i ρ∗ k,i ) − ζ ∗ k,i ρ∗ k,i c′ k,i( ζ ∗ k,i ρ∗ k,i ) −γi > 0 if ρ∗ k,i = 1; = 0 if ρ∗ k,iε(0, 1). (95) (95) Further, Further, Appendix A Substituting Ns from (33) in (35), and differentiating it wrt B, ρ∗ k,i = 0 if γi ≥Hi,k(λj, μ); 1 if γi < Hi,k(λj, μ). (96) (96) ( ∂L ∂B)B∗P∗ i = 1 B2(Tg + 1 B) 2 2BW B −1 i=1 log2(1 + Pihi σ 2 i ) (86) j where Hk,i(λj, μ) = ck,i( ζ ∗ k,i ρ∗ k,i ) − ζ ∗ k,i ρ∗ k,i c′ k,i( ζ ∗ k,i ρ∗ k,i ) (97) ( ∂L ∂B)B∗P∗ i = 1 B2(Tg + 1 B) 2 2BW B −1 i=1 log2(1 + Pihi σ 2 i ) (86) (86) where Hk,i(λj, μ) = ck,i( ζ ∗ k,i ρ∗ k,i ) − ζ ∗ k,i ρ∗ k,i c′ k,i( ζ ∗ k,i ρ∗ k,i ) (97) = 1 B2(Tg + 1 B) 2 B i=1 log2(1 + Pihi σ 2 i ) (86) where Hk,i(λj, μ) = ck,i( ζ ∗ k,i ρ∗ k,i ) − ζ ∗ k,i ρ∗ k,i c′ k,i( ζ ∗ k,i ρ∗ k,i ) (97) (86) (97) Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Page 23 of 24 Page 23 of 24 Page 23 of 24 From (94) and (97) we can write From (94) and (97) we can write Hk,i(λj, μ) = ck,i(c ′−1 k,i ( Np j=1 λjgk,jQj,i + μ)) (98) − ⎛ ⎝ Np j=1 λjgk,jQj,i + μ ⎞ ⎠c ′−1 k,i ⎛ ⎝ Np j=1 λjgk,jQj,i + μ ⎞ ⎠ (99) Hk,i(λj, μ) = ck,i(c ′−1 k,i ( Np j=1 λjgk,jQj,i + μ)) (98) ⎛ Np ⎞ ⎛ Np ⎞ (98) + μ)) (98) 110–115 (2008) 10. M Shaat, F Bader, Computationally Efficient Power Allocation Algorithm in Multicarrier-Based Cognitive Radio Networks: OFDM and FBMC Systems, EURASIP Adv. Signal Process (2010). Article ID 528378 g 11. Appendix A G Bansal, Z Hasan, MJ Hossain, VK Bhargava, Subcarrier and power adaptation for multiuser OFDM-based cognitive radio systems, in Proc National Conference on Communications, 1–5 (2010) − ⎛ ⎝ Np j=1 λjgk,jQj,i + μ ⎞ ⎠c ′−1 k,i ⎛ ⎝ Np j=1 λjgk,jQj,i + μ ⎞ ⎠ (99) (99) On substituting ck, i from (37) in (94) and (99), we get, respectively ζ ∗ k,i = max([( 1 Np j=1 λjgk,jQj,i + μ −σ 2 hk,i )ρ∗ k,i], 0) (100) Hk,i(λj, μ) = log( hk,i ( Np j=1 λjgk,jQj,i + μ)σ 2 )− (101) (1 − ( Np j=1 λjgk,jQj,i + μ)σ 2 hk,i ) (102) ζ ∗ k,i = max([( 1 Np j=1 λjgk,jQj,i + μ −σ 2 hk,i )ρ∗ k,i], 0) (100) Hk,i(λj, μ) = log( hk,i ( Np j=1 λjgk,jQj,i + μ)σ 2 )− (101) ζ ∗ k,i = max([( 1 Np j=1 λjgk,jQj,i + μ −σ 2 hk,i )ρ∗ k,i], 0) (100) 16. SS Das, E De Carvalho, R Prasad, Performance analysis of OFDM systems with adaptive sub carrier bandwidth. IEEE Trans Wireless Commun. 7(4), 1117–1122 (2008) (101) 17. SS Das, E De Carvalho, R Prasad, Dynamically adaptive bandwidth for sub carriers in OFDM based wireless systems, in Proc IEEE Wirel Commun Networking Conference, 1378–1383 (2007) 17. SS Das, E De Carvalho, R Prasad, Dynamically adaptive bandwidth for sub carriers in OFDM based wireless systems, in Proc IEEE Wirel Commun Networking Conference, 1378–1383 (2007) (102) g 18. SS Das, E De Carvalho, R Prasad, Variable sub-carrier bandwidth in OFDM framework. Electron. Lett. 43(1), 46–47 (2007). doi:10.1049/el:20072920 18. SS Das, E De Carvalho, R Prasad, Variable sub-carrier bandwidth in OFDM framework. Electron. Lett. 43(1), 46–47 (2007). doi:10.1049/el:20072920 19. X Zhang, Ma Ni, An adaptive scheme to determine the sub-carrier spacing for multi-carrier systems. International Patent No. W0 2010/015102 A1, 11 Feb 2010 (2007) 19. X Zhang, Ma Ni, An adaptive scheme to determine the sub-carrier spacing for multi-carrier systems. International Patent No. W0 2010/015102 A1, 11 Feb 2010 (2007) Competing interests p g The authors declare that they have no competing interests. The authors declare that they have no competing interests. 24. D Kivanc, G Li, H Liu, Computationally efficient bandwidth allocation and power control for OFDMA. IEEE Trans on Wirel Commun. 2(6), 1150–1158 (2003). doi:10.1109/TWC.2003.819016 24. D Kivanc, G Li, H Liu, Computationally efficient bandwidth allocation and power control for OFDMA. IEEE Trans on Wirel Commun. 2(6), 1150–1158 (2003). doi:10.1109/TWC.2003.819016 Received: 14 February 2011 Accepted: 5 September 2011 Received: 14 February 2011 Accepted: 5 September 2011 Published: 5 September 2011 Published: 5 September 2011 25. G Münz, S Pfletschinger, J Speidel, An Efficient Waterfilling Algorithm for Multiple Access OFDM, in Proc IEEE Global Telecommunications Conference, 681–685 (2002) 25. G Münz, S Pfletschinger, J Speidel, An Efficient Waterfilling Algorithm for Multiple Access OFDM, in Proc IEEE Global Telecommunications Conference, 681–685 (2002) References 1. S Haykin, Cognitive radio: brain-empowered wireless communications. IEEE Trans Sel Areas in Comm. 23, 201–220 (2005) 1. S Haykin, Cognitive radio: brain-empowered wireless communications. IEEE Trans Sel Areas in Comm. 23, 201–220 (2005) 26. J Jang, KB Lee, Transmit power adaptation for multiuser OFDM systems. IEEE J Sele Areas Commun. 21(2), 171–178 (2003). doi:10.1109/ JSAC.2002.807348 2. IF Akyildiz, W Leea, MC Vuran, S Mohantya, NeXt generation/dynamic spectrum access/cognitive radio wireless networks: A survey. Computer Networks. 50(13), 2127–2159 (2006). doi:10.1016/j.comnet.2006.05.001 27. Z Shen, JG Andrews, BL Evans, Optimal power allocation in multiuser OFDM systems, in Proc IEEE Global Telecommun Conference, 337–341 (2003) 3. V Corvino, L Giupponi, AP Neira, V Tralli, R Verdone, Cross-layer radio resource allocation: The journey so far and the road ahead, in Proc of 2nd International Workshop on Cross Layer Design (2009) 28. N Papandreou, T Antonakopoulos, Bit and Power Allocation in Constrained Multicarrier Systems: the Single-User Case. EURASIP J Adv Signal Process (2008) 4. D Tse, P Vishwananth, Fundamentals of Wireless Communication, (Cambridge University Press, Cambridge, 2005) 4. D Tse, P Vishwananth, Fundamentals of Wireless Communication, (C b id U i it P C b id 2005) 29. B Krongold, K Ramchandran, D Jones, Computationally efficient optimal power allocation algorithms for multicarrier communication systems. IEEE Trans Comm. 48, 23–27 (2000). doi:10.1109/26.818869 (Cambridge University Press, Cambridge, 2005) 5. T Weiss, J Hillenbrand, A Krohn, FK Jondral, Mutual interference in OFDM- based spectrum pooling systems, in Proc 59th IEEE Vehicular Technology Conference, 1873–1877 (2004) 30. H Ko, K Lee, S Oh, C Kim, Fast Optimal Discrete Bit-Loading Algorithms for OFDM-Based Systems, in Proc 18th International Conference on Computer Communications and Networks, pp. 1–6 (2009) 6. G Bansal, MJ Hossain, VK Bhargava, Adaptive power loading for OFDM- based cognitive radio systems, in Proc IEEE Int Conference on Comm, 5137–5142 (2007) 31. D Hughes-Hartogs, Ensemble modem structure for imperfect transmission media, 4 679 227 (July 1987), 4 731 816 (March 1988) and 4 833 796 (May 1989). 7. P Wang, M Zhao, L Xiao, S Zhou, J Wang, Power allocation in OFDM-Based cognitive radio systems, in Proc of IEEE Global Telecommunications Conference, 4061–4065 (2007) 32. RV Sonalkar, RR Shively, Efficient bit-loading algorithm for DMT applications. IEEE Commun Letts. 4(3), 80–82 (2000). doi:10.1109/4234.831031 8. Acknowledgements 20. H Steendam, M Moeneclaey, Analysis and optimization of the performance of OFDM on frequency-selective time-selective fading channels. IEEE Trans Comm. 47(12), 1811–1819 (1999). doi:10.1109/26.809701 20. H Steendam, M Moeneclaey, Analysis and optimization of the performance of OFDM on frequency-selective time-selective fading channels. IEEE Trans Comm. 47(12), 1811–1819 (1999). doi:10.1109/26.809701 This work has been supported by the Ministry of Communication and Information Technology, Government of India, New Delhi, and also been supported by Microsoft Corporation and Microsoft Research India under the Microsoft Research India PhD Fellowship Award 2009. 21. DT Harvatin, RE Ziemer, Orthogonal frequency division multiplexing performance in delay and Doppler spread channels, in Proc IEEE Veh Tech Conf, 1644–1647 (1997) 21. DT Harvatin, RE Ziemer, Orthogonal frequency division multiplexing performance in delay and Doppler spread channels, in Proc IEEE Veh Tech Conf, 1644–1647 (1997) Author details 1 22. F Tufvesson, T Maseng, Multiaccess, mobility and teletraffic - advances in wireless networks, (Dordrecht, Kluwer Academic Publishers, 1998) 22. F Tufvesson, T Maseng, Multiaccess, mobility and teletraffic - advances in wireless networks, (Dordrecht, Kluwer Academic Publishers, 1998) 1Indian Institute of Technology, Bombay, 400076, India 2Indian Institute of Technology, Hyderabad, 502205, India 23. CY Wong, RS Cheng, KB Lataief, RD Murch, Multiuser OFDM with adaptive subcarrier, bit, and power allocation. IEEE J Sele Areas Commun. 17(10), 1747–1758 (1999). doi:10.1109/49.793310 23. CY Wong, RS Cheng, KB Lataief, RD Murch, Multiuser OFDM with adaptive subcarrier, bit, and power allocation. IEEE J Sele Areas Commun. 17(10), 1747–1758 (1999). doi:10.1109/49.793310 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 34. A Fasano, On the optimal discrete bit loading for multicarrier systems with constraints, in Proc 57th IEEE Semiannual Vehicular Technology Conference, 915–919 (2003) 35. A Leke, JM Cioffi, A Maximum rate loading algorithm for discrete multitone modulation systems, in Proc IEEE Global Telecommunications Conference, 1514–1518 (1997) 36. AM Wyglinski, F Labeau, P Kabal, An Efficient Bit Allocation Algorithm for Multicarrier Modulation, in Proc IEEE Wirel Commun Networking Conference, 1194–1199 (2004) 37. R Fischer, J Huber, A New Loading Algorithm for Discrete Multitone Transmission, in Proc IEEE Global Telecommunications Conference, 724–728 (1996) 38. PS Chow, JM Cioffi, J Bingham, Practical discrete multitone transceiver loading algorithm for data transmission over spectrally shaped channels. IEEE Transactions on Communications. 43, 773–775 (1995). doi:10.1109/ 26.380108 39. D Huang, Z Shen, C Miao, C Leung, Resource Allocation in MU-OFDM Cognitive Radio Systems with Partial Channel State Information. EURASIP Journal on Wireless Comm (2010) 40. T Ycek, Channel, Spectrum and Waveform Awareness in OFDM-Based Cognitive Radio Systems, (Ph.D. Dissertation, University of South Florida, 2007) 40. T Ycek, Channel, Spectrum and Waveform Awareness in OFDM-Based Cognitive Radio Systems, (Ph.D. Dissertation, University of South Florida, 2007) 41. B Muquet, M de Courville, P Duhamel, Subspace-based blind and semi- blind channel estimation for OFDM systems. IEEE Transactions on Signal Processing. 50(7), 1699–1712 (2002). doi:10.1109/TSP.2002.1011210 42. S Boyd, L Vandenberghe, Convex Optimization, (Cambridge University Press, Cambridge, 2004) 43. GPS Tej, T Nadkar, VM Thumar, UB Desai, SN Merchant, Power Allocation in Cognitive Radio: Single and Multiple Secondary Users, in Proc of IEEE Wireless Communications and Networking Conference (2010) g 44. JM Cioffi, A multicarrier primer. ANSI Contribution T1E1.4/91-157 (Nov. 1991) 44. JM Cioffi, A multicarrier primer. ANSI Contribution T1E1.4/91-157 (Nov. 1991) 45. C Lee, GJ Jeon, An Efficient Adaptive Modulation Scheme for Wireless OFDM Systems. ETRI Journal. 29, 445–451 (2007). doi:10.4218/ etrij.07.0106.0246 45. C Lee, GJ Jeon, An Efficient Adaptive Modulation Scheme for Wireless OFDM Systems. ETRI Journal. 29, 445–451 (2007). doi:10.4218/ etrij.07.0106.0246 46. TS Rappaport, Wireless communications: Principles & Practice, Prentice Hall, New Jersey (1996) doi:10.1186/1687-1499-2011-87 Cite this article as: Thumar et al.: Power allocation, bit loading and sub- carrier bandwidth sizing for OFDM-based cognitive radio. EURASIP Journal on Wireless Communications and Networking 2011 2011:87. References P Wang, X Zhong, L Xiao, S Zhou, J Wang, A general power allocation algorithm for ofdm-based cognitive radio systems, in Proc of the IEEE International Conference on Communications Workshops (2009) 33. E Baccarelli, M Biagi, Optimal integer bit-loading for multicarrier ADSL systems subject to spectral-compatibility limits. Signal Process. 84(4), 729–741 (2004). doi:10.1016/j.sigpro.2003.12.004 33. E Baccarelli, M Biagi, Optimal integer bit-loading for multicarrier ADSL systems subject to spectral-compatibility limits. Signal Process. 84(4), 729–741 (2004). doi:10.1016/j.sigpro.2003.12.004 Page 24 of 24 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 Thumar et al. EURASIP Journal on Wireless Communications and Networking 2011, 2011:87 http://jwcn.eurasipjournals.com/content/2011/1/87 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
https://openalex.org/W4320810259	https://zenodo.org/records/7481531/files/4-8%20(2).pdf	English	null	INTERPRETATION OF BABUR'S PERSONALITY IN ENGLISH, UZBEK AND INDIAN LITERATURE	Zenodo (CERN European Organization for Nuclear Research)	2,022	cc-by	1,702	GREAT BRITAIN GREAT BRITAIN INTERPRETATION OF BABUR’S PERSONALITY IN ENGLISH, UZBEK AND INDIAN LITERATURE Khayrullayeva Kamola Ravshanovna School Counselor of Bukhara District Specialized School Independent Resercher of Bukhara State University Khayrullayeva Kamola Ravshanovna School Counselor of Bukhara District Specialized School Independent Resercher of Bukhara State University Khayrullayeva Kamola Ravshanovna School Counselor of Bukhara District Specialized School Independent Resercher of Bukhara State University Abstract: This article is based on the interpretation of Zahiriddin Muhammad Babur's personality in English, Indian and Uzbek literature. During the researches literary works by authors, such as Pirimkul Kadyrov, Khayriddin Sultanov, Flora Anne Steel and Muni La’l are widely analyzed. Description of Babur’s personality as a son and brother, husband and father are displayed in a comparative way. Parts of the literary works are included where we can identify personal features of Zahiriddin Muhammad Babur. Key words: Zahiriddin Muhammad Babur, Amir Temur, Kabul, Mirza, Emperor, Afghanistan, Motherland, Samarkand, Andijan, Khojand. (3rd international scientific and practical conference) Babur tried to learn from his grandfather Amir Temur in life. This is expressed in the work of Muni Lal as follows: “Babur always believed that Allah had chosen him to save the state founded by Timur.” For this reason, even when he faced great failures, he did not despair. On the contrary, his failures gave him more energy, encouraged him to move forward, to be patient with all those who are destined for his fate. In fact, Babur, who considered himself the true successor of Amir Temur, tried to be stable at all times. Because he believed that his becoming king was not accidental, but chosen to save the state founded by Temur. That's why he was never disappointed. He considered it a sign of fate. H. Sultanov's images are also in line with this: “While wandering around the places where the footsteps of his illustrious ancestor Temurbek were touched or where his immortal legacy spread, he always communicates absently with the owner, seeks comfort from his teachings, and writes down every historical 4 GREAT BRITAIN information related to him in his heart notebook. respectfully noted. While evaluating the political and moral behavior of the statesmen of his time or those who passed before or after him, His Holiness Babur always takes the activity of Amir Temur as the main criterion. Based on this criterion, he expresses his judgments...”. The above picture also confirms that Babur looked at his great-grandfather Amir Temur with great devotion. Even when he was in trouble, he sought solace from Amir Temur's teachings. As a writer Babur was able to impressively reflect the warm and pleasant moments in his heart and soul. Khayrullayeva Kamola Ravshanovna School Counselor of Bukhara District Specialized School Independent Resercher of Bukhara State University Because according to his interpretation, despite receiving this title, Babur did not give in to ambition, and on the contrary, he worked on himself, continued military training, and during the day was busy only with exercises in the military department. Historical facts also confirm this. In the last years of his life, Babur becomes very depressed, he has almost no interest in all the pleasures of life. Muni Lal describes such situations in the psyche of the king-poet as follows: "In the winter days of Agra, he wandered the streets in search of something he did not know, he felt alone and shed tears for his sins. He wanted to go back to Kabul and Samarkand. He missed the melons and grapes of Samarkand, the snowy peaks of Afghanistan." Ever since he lost his country, Babur lived his whole life with love for the Motherland. Homesickness tormented him, especially in the last years of his life. He wanted to visit the cities of Andijan, Samarkand and Kabul, which he ruled for 20 years, to taste the sweet sugar melons and grapes of Samarkand, and to walk on the snowy peaks of Afghanistan. In fact, Muni Lal Babur's emotional experiences were impressively expressed. In these images, the patriotism of the king- poet was clearly demonstrated. Khayriddin Sultanov expressed this reality as follows: "Babur used to sit alone at the foot of the garden, on the huge mountain porch that rose high among the oleander and laurel trees. In the last years of his life, he had cooled down from the parties and performances in the palace, full of cheerfulness, and the cruel deeds of the Indian dancers, and the charming verses of the nightingale-like poets in the sky of glory, did not bring pleasure to his heart anymore. He spent most of his time out of sight, alone in a secluded corner, only occasionally, when he felt like it, he would call Yusufi and order roses. It is known that in the last years of his life, Babur spent a lot of time in solitude. As it was mentioned above, he was no longer interested in the parties in the palace, the performances of the dancers and the poems of the poets did not give him pleasure. He often spent his time in solitude thinking. In fact, these dreams lead him to the Motherland, he lived with this dream. Khayrullayeva Kamola Ravshanovna School Counselor of Bukhara District Specialized School Independent Resercher of Bukhara State University That is, it can be seen in such things as his communication with the master in absentia, seeking comfort from his teachers, writing the historical information related to him in his notebook, and considering his work as a criterion for himself. It is known from historical sources that Zahiriddin Muhammad, after establishing his rule in Kabul, ordered that he be called "king" as a descendant of Temurbek. This reality is interpreted by the Indian writer Muni Lal as follows: "Shaibani Khan and the Mongol invasions were suppressed, and now there was no force left to threaten Babur's rule in Kabul. Now he decides to take the title of "King". At that time, the princes from Temur's generation were called "Mirza" even when they took over the government. Babur, who received the title of "King", was continuing the work of his ancestors. Adib emphasizes the fact that Babur's great ancestor Amir Temur continued his work. In fact, unlike the contemporary Timurids, Babur orders them to call themselves "King" and not "Mirza". Unlike other Timurids, he founded a new Empire. For this reason, he was a person worthy of being called "King" in every way. (3rd international scientific and practical conference) The English writer Flora Steele interprets this reality a little differently: "Babur, who was very happy with the birth of his son, was deeply engrossed in the new title of "Emperor" given to him and was busy spending his energetic life force on gathering his troops and fighting battles. His whole day was spent in the military department. Anna Steele justifies Babur's great position in the Western world by calling him "Emperor". Moreover, this word is characteristic of Western culture and mentality. Adiba attributes Babur's assumption of this title to the birth of his first son, Humayun. 5 GREAT BRITAIN The writer's deep respect for Babur is also noticeable. Because according to his interpretation, despite receiving this title, Babur did not give in to ambition, and on the contrary, he worked on himself, continued military training, and during the day was busy only with exercises in the military department. Historical facts also confirm this. The writer's deep respect for Babur is also noticeable. (3rd international scientific and practical conference) SCIENTIFIC REVIEW OF THE PROBLEMS AND PROSPECTS OF MODERN SCIENCE AND EDUCATION GREAT BRITAIN Reference Khayrullayeva Kamola Ravshanovna School Counselor of Bukhara District Specialized School Independent Resercher of Bukhara State University Author realistically reflected Babur's mental state in moments of loneliness. (3rd international scientific and practical conference) 6 References: 1. Khayrullayeva, K. R. (2020). Description of Zahiriddin Babur’s achievements in various fields in the works of Uzbek and world authors. ISJ Theoretical & Applied Science, 09 (89), 8-11. Soi: http://s-o-i.org/1.1/TAS-09-89-2 Doi: https://dx.doi.org/10.15863/TAS.2020.09.89.2 2. Ravshanovna, Khayrullayeva Kamola, and Hakimova Muhabbat Alimovna. "THE EXEMPLARY LIFE OF BABUR MIRZA IN THE INTERPRETATION OF STEPHEN MEREDITH." 3. Kamolov Ikhtiyor Nigmatullayevich, Khayrullayeva Kamola Ravshanovna and Quvvatova Dilrabo Khabibovna THE IMAGE OF BABUR IN THE INTERPRETATION OF HAROLD LAMB. Journal of Contemporary Issues in Business and Government Vol. 27, No. 4,2021. P-ISSN: 2204-1990; E-ISSN: 1323- 6903. DOI: 10.47750/cibg.2021.27.04.015. https://cibg.org.au/. 4. Khayrullayeva Kamola Ravshanovna. INTERPRETATION OF ZAHIRIDDIN MUHAMMAD BABUR’S IMAGE IN UZBEK AND WORLD LITERATURE. EPRA International Journal of Research and Development (IJRD), Volume: 5 \| Issue: 5 \| May 2020. SJIF Impact Factor: 7.001\| ISI I.F.Value:1.241\| Journal DOI: 10.36713/epra2016. ISSN: 2455-7838(Online). 5. Ravshanovna, Khayrullayeva Kamola. "The imageof babur mirza in the interpretation of stephen meredith." ACADEMICIA: An International Multidisciplinary Research Journal 11.11 (2021): 239-244. 6. Khayrullayeva Kamola Ravshanovna. (2022). THE IMAGE OF BABUR AS KING AND LEGATE IN WESTERN ENGLISH AND UZBEK PROSE. JournalNX - A Multidisciplinary Peer Reviewed Journal, 8(2), 134–138. https://doi.org/10.17605/OSF.IO/PNJZD 7. Khayrullayeva, K. (2022). Similarities and differences in the plot of the works by Flora Anna Steele and Pirimkul Kadyrov. Science and Education, 3(3), 515- (3rd international scientific and practical conference) 519. Retrieved from https://www.openscience.uz/index.php/sciedu/article/view/2794 7 GREAT BRITAIN SCIENTIFIC REVIEW OF THE PROBLEMS AND PROSPECTS OF MODERN SCIENCE AND EDUCATION G GREAT BR 8. Khayrullayeva Kamola Ravshanovna. (2022). TYPOLOGICAL SIMILARITIES IN FLORA ANNA STEELE'S "KING-ERRANT" AND PIRIMKUL KADYROV'S "STARRY NIGHTS" IN THE INTERPRETATION OF THE KING AND COMMANDER BABUR. Eurasian Journal of Academic Research, 2(3), 105– 108. https://doi.org/10.5281/zenodo.6368804 9. Khayrullayeva Kamola. (2022). TYPOLOGY IN DESCRIPTION OF BABUR’S IMAGE AND PLOT OF THE WORKS IN ENGLISH AND UZBEK LITERATURE. Involta Scientific Journal, 1(6), 264–272. Retrieved from https://involta.uz/index.php/iv/article/view/194 GREAT BRITAIN 8. Khayrullayeva Kamola Ravshanovna. (2022). TYPOLOGICAL SIMILARITIES IN FLORA ANNA STEELE'S "KING-ERRANT" AND PIRIMKUL KADYROV'S "STARRY NIGHTS" IN THE INTERPRETATION OF THE KING AND COMMANDER BABUR. Eurasian Journal of Academic Research, 2(3), 105– 108. https://doi.org/10.5281/zenodo.6368804 (3rd international scientific and practical conference) 8 9. Khayrullayeva Kamola. (2022). TYPOLOGY IN DESCRIPTION OF BABUR’S IMAGE AND PLOT OF THE WORKS IN ENGLISH AND UZBEK LITERATURE. Involta Scientific Journal, 1(6), 264–272. Retrieved from https://involta.uz/index.php/iv/article/view/194 9. Khayrullayeva Kamola. (2022). TYPOLOGY IN DESCRIPTION OF BABUR’S IMAGE AND PLOT OF THE WORKS IN ENGLISH AND UZBEK LITERATURE. Involta Scientific Journal, 1(6), 264–272. Retrieved from https://involta.uz/index.php/iv/article/view/194 (3rd international scientific and practical conference) (3rd international scientific and practical conference) 8 8
https://openalex.org/W4214918013	https://europepmc.org/articles/pmc5553089?pdf=render	English	null	Arkas: Rapid reproducible RNAseq analysis	F1000Research	2,017	cc-by	11,439	SOFTWARE TOOL ARTICLE Arkas: Rapid reproducible RNAseq analysis [version 2; referees: 2 approved] Anthony R. Colombo, Timothy J. Triche Jr, Giridharan Ramsingh Jane Anne Nohl Division of Division of Hematology and Center for the Study of Blood Diseases, Keck School of Medicine of University of Southern California, Los Angeles, CA, 90033, USA 27 Apr 2017, :586 (doi: ) First published: 6 10.12688/f1000research.11355.1 21 Jun 2017, :586 (doi: ) Latest published: 6 10.12688/f1000research.11355.2 v2 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 Open Peer Review Discuss this article (0) Comments SOFTWARE TOOL ARTICLE Arkas: Rapid reproducible RNAseq analysis [version 2; referees: 2 approved] Anthony R. Colombo, Timothy J. Triche Jr, Giridharan Ramsingh Jane Anne Nohl Division of Division of Hematology and Center for the Study of Blood Diseases, Keck School of Medicine of University of Southern California, Los Angeles, CA, 90033, USA Abstract The recently introduced Kallisto pseudoaligner has radically simplified the quantification of transcripts in RNA-sequencing experiments. We offer cloud-scale RNAseq pipelines , and Arkas-Quantification Arkas-Analysis available within Illumina’s BaseSpace cloud application platform which expedites Kallisto preparatory routines, reliably calculates differential expression, and performs gene-set enrichment of REACTOME pathways Due . to inherit inefficiencies of scale, Illumina's BaseSpace computing platform offers a massively parallel distributive environment improving data management services and data importing. deploys Arkas-Quantification Kallisto for parallel cloud computations and is conveniently integrated downstream from the BaseSpace (SRA) Sequence Read Archive import/conversion application titled . annotates the SRA Import Arkas-Analysis Kallisto results by extracting structured information directly from source FASTA files with per-contig metadata, calculates the differential expression and gene-set enrichment analysis on both coding genes and transcripts. The Arkas cloud pipeline supports ENSEMBL transcriptomes and can be used downstream from the facilitating raw sequencing importing, SRA SRA Import FASTQ conversion, RNA quantification and analysis steps. This article is included in the Container collection. Virtualization in Bioinformatics Referee Status: Invited Referees version 2 published 21 Jun 2017 version 1 published 27 Apr 2017 1 2 report report report , Stanford University, Harold Pimentel USA 1 , University of Iowa, USA Ted Abel , University of Iowa, USA Marie Gaine 2 27 Apr 2017, :586 (doi: ) First published: 6 10.12688/f1000research.11355.1 21 Jun 2017, :586 (doi: ) Latest published: 6 10.12688/f1000research.11355.2 v2 Abstract The recently introduced Kallisto pseudoaligner has radically simplified the quantification of transcripts in RNA-sequencing experiments. We offer cloud-scale RNAseq pipelines , and Arkas-Quantification Arkas-Analysis available within Illumina’s BaseSpace cloud application platform which expedites Kallisto preparatory routines, reliably calculates differential expression, and performs gene-set enrichment of REACTOME pathways Due . to inherit inefficiencies of scale, Illumina's BaseSpace computing platform offers a massively parallel distributive environment improving data management services and data importing. deploys Arkas-Quantification Kallisto for parallel cloud computations and is conveniently integrated downstream from the BaseSpace (SRA) Sequence Read Archive import/conversion application titled . annotates the SRA Import Arkas-Analysis Kallisto results by extracting structured information directly from source FASTA files with per-contig metadata, calculates the differential expression and gene-set enrichment analysis on both coding genes and transcripts. The Arkas cloud pipeline supports ENSEMBL transcriptomes and can be used downstream from the facilitating raw sequencing importing, SRA SRA Import FASTQ conversion, RNA quantification and analysis steps. , Stanford University, Harold Pimentel USA 1 , Stanford University, Harold Pimentel USA 1 This article is included in the Container collection. Virtualization in Bioinformatics (0) Comments Page 1 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 Anthony R. Colombo ( ), Timothy J. Triche Jr ( ) Corresponding authors: anthonycolombo60@gmail.com tim.triche@gmail.com : Conceptualization, Data Curation, Formal Analysis, Methodology, Software, Validation, Visualization, Writing – Author roles: Colombo AR Original Draft Preparation, Writing – Review & Editing; : Conceptualization, Data Curation, Formal Analysis, Investigation, J. Triche Jr T Methodology, Software, Supervision, Validation, Visualization, Writing – Original Draft Preparation; : Conceptualization, Project Ramsingh G Administration, Resources, Supervision, Writing – Original Draft Preparation, Writing – Review & Editing Competing interests: No competing interests were disclosed. Colombo AR, J. Triche Jr T and Ramsingh G. How to cite this article: Arkas: Rapid reproducible RNAseq analysis [version 2; referees: 2 2017, :586 (doi: ) approved] F1000Research 6 10.12688/f1000research.11355.2 © 2017 Colombo AR . This is an open access article distributed under the terms of the , Copyright: et al Creative Commons Attribution Licence which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. This project was funded by grants from Leukemia Lymphoma Society-Quest for Cures (0863-15), Illumina (San Diego), STOP Grant information: Cancer and Tower Cancer Research Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction In collaboration with Illumina (San Diego, USA) the available BaseSpace platform was already well-suited for parallel transforma- tion of raw sequencing data into analytical results. BaseSpace has an available application SRA Import which automates SRA import- ing and FASTQ conversion pre-processing steps. The application SRA Import is simple requiring the SRA accession number and limits imports to 25gb per application call. Arkas can ingest success- fully imported samples avoiding all raw data handling. For exam- ple, if one were interested in re-analyzing an experiment from SRA with reads totaling 141.3GB, Arkas facilitates SRA processing and state-of-the-art pseudoalignment by reducing raw sequencing data to summary quantifications totaling 1.63GB and includes an exten- sive analysis report of less than 10MB. The total data reduction exceeds 4 orders of magnitude with little or no loss of user-visible information. Moreover, the untouched original data is never dis- carded unless the user explicitly demands it. The appropriate placement of Arkas applications adjacent to the origin of sequenc- ing data removes cumbersome data relocation costs and greatly facilitates sequencing archive re-analysis using state-of-the-art pseudoalignment. High-performance computing based bioinformatic workflows have three main subfamilies: in-house computational packages, virtual- machines (VMs), and cloud based computational environments. The in-house approaches are substantially less expensive when raw hardware is in constant use and dedicated support is available, but internal dependencies can limit reproducibility of computational experiments. Specifically, “superuser’” access needed to deploy container-based, succinct code encapsulations (often referred to as “microservices” elsewhere) can run afoul of normal permissions, and the maintenance of broadly usable sets of libraries across nodes for users can lead to shared code dynamically linking to differ- ent libraries under various user environments. By contrast, mod- ern cloud-based approaches and parallel computing are forced by necessity to offer a user-friendly platform with high availability to the broadest audience. Platform-as-a-service approaches take this one step further, offering controlled deployment and fault tolerance across potentially unreliable instances provided by third parties such as Amazon Web Service Elastic Compute Cloud (AWS EC2) and enforcing a standard for encapsulation of developers’ serv- ices such as Docker. Within this framework, the user or developer cedes some control of the platform and interface, in exchange for the platform provider handling the details of workflow distribution and execution. This has provided the best compromise of usability and reproducibility when dealing with general audiences. Abstract 27 Apr 2017, :586 (doi: ) First published: 6 10.12688/f1000research.11355.1 Page 2 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 A major bottleneck in RNAseq analysis is the processing steps for importing raw data. The majority of RNAseq analysis pipelines consist of read preparation steps, followed by computationally expensive alignment against a reference. Software for calculating transcript abundance and assembly can surpass 30 hours of compu- tational time1. If known or putative transcripts of defined sequences are the primary interest, then pseudoalignment, which is defined as near-optimal RNAseq transcript quantification, is achievable in minutes on a standard laptop using Kallisto software1. Arkas was developed using a simple framework, yet massively parallel, for RNAseq transcript quantification that would allow users to expedite pseudoalignment on arbitrary datasets, and significantly reduce the amount of required preparatory routines. Amendments from Version 1 REVISED This revised manuscript eliminates previously unrelated discussion points such as, an in-depth examination of Docker and its role in data sharing. This version 2 is concise, explicitly stating motivations for developing the Docker software. It includes updated Figure 1, which has swapped the order of images from the previous version Figure 1. In addition, Supplementary Figure S1 and Supplementary Figure S2 were added showing the application interface. Further, this revised manuscript discusses more relevant topics such as comparing cloud computing platforms. See referee reports See referee reports Introduction In this regard, the lightweight-container approach exemplified by Docker lead to rapid development and deployment compared to VMs. Combined with versioning of deployments, it is feasible for users to reconstruct results from an earlier point in time, while simultaneously re-evaluating the generated data under state-of-the-art implementations. Arkas, encapsulates Kallisto, automates the construction of com- posite transcriptomes from, quantifies transcript abundances, and implements reproducible rapid differential expression analysis coupled with gene set enrichment analysis. The Arkas workflow is versionized into Docker containers and publicly deployed within Illumina’s BaseSpace platform which ingests raw RNA sequencing data and completes a full analysis in approximately 2 hours. Arkas-Quantification implementation Arkas is a two-step cloud pipeline. Arkas-Quantification is the first step, which reduces the computational steps required to quantify and annotate large numbers of samples against large catalogs of transcriptomes. Arkas-Quantification calls Kallisto for on-the-fly transcriptome indexing and quantification recursively for numerous sample directories. Kallisto quantifies transcript abundance from input RNAseq reads by using pseudoalignment, which identifies the read-transcript compatibility matrix1. The compatibility matrix is formed by counting the number of reads with the matching align- ment; the equivalence class matrix has a much smaller dimension compared to matrices formed by transcripts and read coverage. Computational speed is gained by performing the Expectation Maximization (EM) algorithm over a smaller matrix. Subsequent analyses import the data structure from Summarize- dExperiment and creates a sub-class titled KallistoExperiment that preserves the S4 structure and is convenient for handling assays, phenotypic and genomic data. KallistoExperiment includes Genom- icRanges6, preserving the ability to handle genomic annotations and alignments, supporting efficient methods for analyzing high- throughput sequencing data. The KallistoExperiment sub-class serves as a general-purpose container for storing feature genomic intervals and pseudoalignment quantification results against a ref- erence genome called by Kallisto. By default KallistoExperiment couples assay data such as the estimated counts, effective length, estimated median absolute deviation, and transcript per million count where each assay data is generated by a Kallisto run; the stored feature data is a GenomicRanges object from 6, storing tran- script length, GC content, and genomic intervals. For RNAseq projects with many sequenced samples, Arkas- Quantification encapsulates expensive transcript quantification preparatory routines, while uniformly preparing Kallisto execu- tion commands within a versionized environment encouraging reproducible protocols. The quantification step automates the index caching, annotation, and quantification associated while run- ning the Kallisto pseudoaligner integrated within the BaseSpace environment. For users interested in quality control checks, BaseSpace offers an independent application FastQC which per- forms fastqc on sequencing data. The first step in the pipeline can process raw reads into transcript and pathway collection results within Illumina’s BaseSpace cloud platform, quantifying against default transcriptomes such as ERCC spike-ins, ENSEMBL non-coding RNA, or cDNA build 88 for both Homo sapiens and Mus musculus; further the first step supports user uploaded FASTA files for customized analyses. Arkas-Quantification can support microRNAs (miRNA), however we encourage users to analyze miRNAs separately because pseudoalignment requires reducing k-mer size in the Target-DeBruijn Graph (TDBG) to miRNA sequence lengths (ranging from 16–22) which can increase path ambiguities. Methodsi The first step in the Arkas pipeline requires Arkas-Quantification to transform all the raw RNA sequencing data of an entire experiment into Kallisto pseudoaligned quantification output data. The second step, Arkas-Analysis, requires the pseudoaligned data to be input with respect to a comparison and control group, and returns a com- prehensive analysis including differential expression, and gene-set enrichment. Docker offers advantages for reproducible research practices, and also is the principal infrastructure to leading platforms such as Illumina’s BaseSpace platform, Google Genomics, Galaxy and SevenBridges. Cloud computational ecosystems preserve develop- mental environments using Docker containerization framework, and improves bioinformatic validation. Containerized cloud appli- cations form part of the global distributive effort and are favorable over local in-house computational pipelines because they offer rapid access to numerous public workflows, easy interfacing to archived read databases, and accelerate the upholding process of raw data. If the user selects the defaults, Arkas-Quantification will complete pseudoalignment in approximately 43–60 minutes. Arkas- Quantification completion time is independent of the number of samples input, but is restricted to node availability (AWS EC2 node If the user selects the defaults, Arkas-Quantification will complete pseudoalignment in approximately 43–60 minutes. Arkas- Quantification completion time is independent of the number of samples input, but is restricted to node availability (AWS EC2 node Page 3 of 21 Page 3 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 defined by the External RNA Control Consortium4, are detected then Arkas-Analysis will calculate Receiver Operator Characteris- tic (ROC) plots using ‘erccdashboard’5. The ERCC analysis reports average ERCC Spike amount volume, comparison plots of ERCC volume amount, and normalized ERCC counts (Figure 1). availability is fairly high). Arkas-Analysis, will complete using a sin- gle node in approximately 1–1.5 hours for moderate sample group sizes (N ≤ 20), and under 2–2.5 hours for much larger designs. Arkas-Quantification implementation Arkas-Quantification is packaged into a Docker container and is publicly available as a cloud application within BaseSpace. Given a KallistoExperiment containing the Kallisto sample abun- dances, principal component analysis (PCA) is performed7 on trimmed mean of M-value (TMM) normalized counts8 (Figure 2A). Differential expression (DE) is calculated on the library normalized transcript expression values, and the aggregated transcript bundles of corresponding coding genes using limma/voom linear model9 (Figure 3A). In addition to library normalization, we wished to add an optional data driven normalization. In the analysis pipeline, an unsupervised normalization method would not require more than a two group experimental design which was favorable due to its simplicity. Alternatively, supervised data driven normalization is a specialized task which requires users to define batch groups, and/or additional experimental groups. Further, the adjusted data must be evaluated in the context of the experiment. In-silico normalization, using factor analysis, effectively removes unwanted variation driven entirely by data10. The analysis report returns comprehensive visualization results. PCA and DE analysis of both transcripts and coding genes is per- formed with easily interpretable images (Figure 2B, Figure 3B, Figure 3C). In each DE analysis FDR filtering method is defaulted to ‘Benjamini-Hochberg’, if there are no resultant DE genes/ transcripts the FDR methods is switched to ‘none’. Arkas-Analy- sis consumes the Kallisto data output from Arkas-Quantification, and automates DE analysis using TMM normalization and in-silico normalization on both transcript and coding gene expression in a defaulted two group experimental design, allowing customized selections. One must examine and compare the PCA sample clus- tering, and sample boxplots between the two methods to determine the improvement of in-silico normalization (Figure 2). If RUV improves the PCA clustering within the context of an experiment, and reduces the number of outliers observed in boxplots then it is likely that the normalization weights are useful. Arkas-Analysis implementationi Previous work2 has revealed that filtering transcriptomes to exclude lowly-expressed isoforms can improve statistical power, while more-complete transcriptome assemblies improve sensitivity in detecting differential transcript usage. Based on earlier work by Bourgon et al.3, we included this type of filtering for both gene- and transcript-level analyses within Arkas-Analysis. The analysis pipeline automates annotations of quantification results, resulting in more accurate interpretation of coding and transcript sequences in both basic and clinical studies by just-in-time annotation and visualization. Arkas-Analysis integrates quality control analysis for experiments that include Ambion spike-in controls, multiple normalization selections for both coding gene and transcript differential expres- sion analysis, and differential gene-set analysis. If ERCC spike-ins, Gene set differential expression, which includes gene-gene corre- lation inflation corrections, is calculated using Qusage11. Qusage calculates the variance inflation factor, which corrects the inter-gene Page 4 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 Figure 1. Arkas-Analysis ERCC spike-in Controls Report. A) The Receiver Operator Characteristic plot of (detected) ERCC ratios in gene expression experiments. The X-axis shows the False Positive Rate, the Y-axis shows True Positive Rate. B) and C) shows the dynamic range of abundances of ERCC RNA amounts with a linear model fit, and ERCC RNA counts. D) shows a dispersion of mean transcript abundances and the estimated dispersion. Figure 1. Arkas-Analysis ERCC spike-in Controls Report. A) The Receiver Opera Figure 1. Arkas-Analysis ERCC spike-in Controls Report. A) The Receiver Operator Characteristic plot of (detected) ERCC ratios in gene expression experiments. The X-axis shows the False Positive Rate, the Y-axis shows True Positive Rate. B) and C) shows the dynamic range of abundances of ERCC RNA amounts with a linear model fit, and ERCC RNA counts. D) shows a dispersion of mean transcript abundances and the estimated dispersion. Page 5 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 2. Arkas-Analysis Normalization Report: Normalization Analysis Using TMM and RUV. A) TMM normalization is perform data and depicts the sample quantiles on normalized sample expression, PCA plot, and histogram of the adjusted p-values calc e DE analysis. Orange is the comparison group and green is the control group. B) A similar analysis is performed with RUV in zation. kas Analysis Normalization Report: Normalization Analysis Using TMM and RUV A) TMM normalization Figure 2. Arkas-Analysis Normalization Report: Normalization Analysis Using TMM and RUV. Arkas-Analysis implementationi A) TMM normalization is performed on sample data and depicts the sample quantiles on normalized sample expression, PCA plot, and histogram of the adjusted p-values calculated from the DE analysis. Orange is the comparison group and green is the control group. B) A similar analysis is performed with RUV in-silico normalization. Figure 2. Arkas-Analysis Normalization Report: Normalization Analysis Using TMM and RUV. A) TMM normalization is performed on sample data and depicts the sample quantiles on normalized sample expression, PCA plot, and histogram of the adjusted p-values calculated from the DE analysis. Orange is the comparison group and green is the control group. B) A similar analysis is performed with RUV in-silico normalization. Page 6 of 21 Page 6 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 Figure 3. Arkas-Analysis Differential Expression Report: DE using TMM and RUV. A) DE analysis using TMM normalization. The X-axis is the sample names (test data), the Y-axis are Gene symbols (HUGO). Expression values are plotted in log10 1+TPM. B) Similar analysis using RUV normalization. C) The design matrix with the RUV adjusted weights. The sample names are test data used in demonstrating the general analysis report output. Page 7 of 21 Figure 3. Arkas-Analysis Differential Expression Report: DE using TMM and RUV. A) DE analysis using TMM normalization. The X-axis is the sample names (test data), the Y-axis are Gene symbols (HUGO). Expression values are plotted in log10 1+TPM. B) Similar analysis using RUV normalization. C) The design matrix with the RUV adjusted weights. The sample names are test data used in demonstrating the general analysis report output. F1000Research 2017, 6:586 Last updated: 30 AUG 2017 correlation that results in high type 1 errors using pooled or non-pooled variances between experimental groups. The gene set enrichment is conducted using Reactome pathways constructed using ENSEMBL transcript/gene identifiers (Figure 4 and Table 1); REACTOME gene sets are not as large as other databases, so Arkas-Analysis outputs DE analysis in formats compatible with more exhaustive databases such as Advaita. The DE files are compatible as a cus- tom upload into Advaita iPathway guide, which offers an extensive Figure 4. Arkas-Analysis Gene-Set Enrichment Plot. Gene-Set enrichment output report, each point represents the differential mean activity of each gene-set with 95% confidence intervals. The X-axis are individual gene-sets. The Y-axis is the log2 fold change. Figure 4. Arkas-Analysis Gene-Set Enrichment Plot. Arkas-Analysis implementationi Gene-Set enrichment output report, each point represents the differential mean activity of each gene-set with 95% confidence intervals. The X-axis are individual gene-sets. The Y-axis is the log2 fold change. Table 1. Arkas-Analysis Gene-Set Enrichment Statistics. The columns represent the Reactome pathway name corresponding to the depicted pathways in Figure 4, the log2fold change, p-value, adjusted FDR, and an active link to the Reactome website with visual depictions of the gene/transcript pathway. Arkas-Analysis will output a similar report testing transcript-level sets. Pathway name Log fold change P.value FDR Gene URL R-HAS-1989781 -0.87 0.0008 0.06 http://www.reactome.org/PathwayBrowser/#/R-HSA-1989781 R-HAS-2173796 -0.51 0.007 0.217 http://www.reactome.org/PathwayBrowser/#/R-HSA-2173796 R-HAS-6804759 -1.62 0.009 0.217 http://www.reactome.org/PathwayBrowser/#/R-HSA-6804759 R-HAS-381038 -0.43 0.013 0.226 http://www.reactome.org/PathwayBrowser/#/R-HSA-381038 R-HAS-2559585 -0.4 0.032 0.341 http://www.reactome.org/PathwayBrowser/#/R-HSA-2559585 R-HAS-4086398 -0.95 0.033 0.341 http://www.reactome.org/PathwayBrowser/#/R-HSA-4086398 R-HAS-4641265 -0.95 0.033 0.341 http://www.reactome.org/PathwayBrowser/#/R-HSA-4641265 R-HAS-422085 -1.17 0.04 0.361 http://www.reactome.org/PathwayBrowser/#/R-HSA-422085 R-HAS-5467345 -0.56 0.069 0.389 http://www.reactome.org/PathwayBrowser/#/R-HSA-5467345 R-HAS-6804754 -0.57 0.07 0.389 http://www.reactome.org/PathwayBrowser/#/R-HSA-6804754 R-HAS-6803204 -1.19 0.081 0.389 http://www.reactome.org/PathwayBrowser/#/R-HSA-6803204 Table 1. Arkas-Analysis Gene-Set Enrichment Statistics. The columns represent the Reactome pathway name corresponding to the depicted pathways in Figure 4, the log2fold change, p-value, adjusted FDR, and an active link to the Reactome website with visual depictions of the gene/transcript pathway. Arkas-Analysis will output a similar report testing transcript-level sets. Table 1. Arkas-Analysis Gene-Set Enrichment Statistics. The columns represent the Reactome pathway name corresponding to the depicted pathways in Figure 4, the log2fold change, p-value, adjusted FDR, and an active link to the Reactome website with visual depictions of the gene/transcript pathway. Arkas-Analysis will output a similar report testing transcript-level sets. Pathway name Log fold change P.value FDR Gene URL R-HAS-1989781 -0.87 0.0008 0.06 http://www.reactome.org/PathwayBrowser/#/R-HSA-1989781 R-HAS-2173796 -0.51 0.007 0.217 http://www.reactome.org/PathwayBrowser/#/R-HSA-2173796 R-HAS-6804759 -1.62 0.009 0.217 http://www.reactome.org/PathwayBrowser/#/R-HSA-6804759 R-HAS-381038 -0.43 0.013 0.226 http://www.reactome.org/PathwayBrowser/#/R-HSA-381038 R-HAS-2559585 -0.4 0.032 0.341 http://www.reactome.org/PathwayBrowser/#/R-HSA-2559585 R-HAS-4086398 -0.95 0.033 0.341 http://www.reactome.org/PathwayBrowser/#/R-HSA-4086398 R-HAS-4641265 -0.95 0.033 0.341 http://www.reactome.org/PathwayBrowser/#/R-HSA-4641265 R-HAS-422085 -1.17 0.04 0.361 http://www.reactome.org/PathwayBrowser/#/R-HSA-422085 R-HAS-5467345 -0.56 0.069 0.389 http://www.reactome.org/PathwayBrowser/#/R-HSA-5467345 R-HAS-6804754 -0.57 0.07 0.389 http://www.reactome.org/PathwayBrowser/#/R-HSA-6804754 R-HAS-6803204 -1.19 0.081 0.389 http://www.reactome.org/PathwayBrowser/#/R-HSA-6803204 Page 8 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 Gene Ontology (GO) pathway analysis. Pathway enrichment analysis can be performed from the BaseSpace cloud system downstream from parallel differential expression analysis and can integrate with other pathway analysis software tools. Supplementary Table 2). Arkas-Analysis implementationi We calculated the standardized mean dif- ferences, and the variation of the differences between the same 5 samples from Kallisto (setting bootstraps = 42) versions 0.43 and 0.43.1 (Supplementary Table 2), and found large variation of differences between raw values generated by differing Kallisto versions, signifying the importance of version analysis of Kallisto results. Arkas cloud pipeline: modern and simple Arkas cloud pipeline: modern and simple Recent developments for virtualized operating systems, such as Docker, allow for local software environments to be preserved, whereas cloud platforms deploy the preserved software. Docker allows users to build layers of read/write access files, creating a portable operating system which exhaustively controls software versions and data, while systematically preserving the pipeline software. Currently, Docker is the principal infrastructure for cloud bioinformatic computational software platforms such as Illumina’s BaseSpace platform, Google Genomics, SevenBridges, and Galaxy. The Arkas-Analysis instructions are provided within the BaseSpace environment. The input for the analysis app is the Arkas-Quantification sample data, and the output files are separated into corresponding folders. The analysis also depicts figures for each respective analysis (Figure 1–Figure 4) and the images can be downloaded as a HTML format. The Google Cloud Platform supports popular languages such as Python, Node, and Ruby with services related to computing, and storage. Google Genomics Platform has a steeper learning curve recommending familiarity with services such as Compute Engine, and Cloud Storage. Google Genomics hosts cloud storage transfer services for importing source data to storage buckets from HTTP/ HTTPS locations. Data management services outside the Google Genomics platform, such as Globus, serves SRA database which can interact with Google Genomics applications reducing the bottleneck of SRA downloads. Discussion Complete transcriptomes enrich annotation information, improving downstream analysesi Discussion Complete transcriptomes enrich annotation information, improving downstream analyses The choice of catalog, and the type of quantification per- formed, influence the results of sequencing analysis. ENSEMBL reference genomes are provided to GENCODE as a merged database from Havana’s manually curated annotations with ENSEMBL’s automatic curated coordinates. AceView, UCSC, RefSeq, and GENCODE have approximately twenty thousand protein coding genes, however AceView and GENCODE have a greater number of protein coding transcripts in their databases. RefSeq and UCSC references have less than 60,000 protein coding transcripts, whereas GENCODE has 140,066 protein coding loci. AceView has 160,000 protein coding transcripts, but this database is not manually curated. GENCODE is anno- tated with special attention given to long non-coding RNAs (lncRNAs) and pseudogenes, improving annotations and coupling automated labeling with manual curating. The database selected for protein coding transcripts can influence the amount of anno- tation information returned when querying gene/transcript level databases. Supplementary Table 2 shows that there is large variation of the differences of Kallisto data calculated between differing versions. Figure 5 depicts the standardized mean differences, i.e. errors, between Kallisto versions fitted to a theoretical normal dis- tribution. The quantile-quantile plots show that the errors are marginally normal, with a consistent line centered near 0 but also large outliers (Figure 5). As expected, containerizing analysis pipelines will enforce versionized software, which benefits reproducible analyses. The Dockerization of Arkas BaseSpace applications versionizes the Kallisto reference index to enforce that the Kallisto software versions are identical, and further documents the Kallisto ver- sion used in every cloud analysis. The enforcement of reference versions and Kallisto software versions prevents errors when comparing experiments. Data variance between software versions We wished to show the importance of enforcing matching versions of Kallisto when quantifying transcripts because there is deviation of data between versions. Due to updated versions and improve- ments of Kallisto software, there obviously exists variation of data between algorithm versions (Figure 5, Supplementary Table 1, Supplementary Table 1 shows the variation of the errors of the raw values such as estimated counts, effective length, and estimated median absolute deviation using the same Kallisto version 0.43.0. Figure 5. Quantile-Quantile Plots of Data Variation Comparing Differences in Kallisto Data from Versions 0.43.1 and 0.43.0. The X-axis depicts the theoretical quantiles of the standardized mean differences. The Y-axis represents the observed quantiles of standardized mean differences. Figure 5. Quantile-Quantile Plots of Data Variation Comparing Differences in Kallisto Data from Versions 0.43.1 and 0.43.0. The X-axis depicts the theoretical quantiles of the standardized mean differences. The Y-axis represents the observed quantiles of standardized mean differences. Page 9 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 KallistoExperiment class for 5 samples was 23.551 seconds (Supplementary Table 4). KallistoExperiment class for 5 samples was 23.551 seconds (Supplementary Table 4). As expected, Kallisto data generated by the same Kallisto version had very low variation of errors within the same version 0.43.0 for every transcript across all samples. However, upon compar- ing Kallisto version 0.43.1 to version 43.0 using the raw data such as estimate abundance counts, effective length, estimated median absolute deviation, and transcript per million values, we found, as expected, large variation of data. Operation Although previously overlooked, lncRNAs have been shown to share features and alternate splice variants with mRNA, revealing that lncRNAs play a central role in metastasis, cell growth and cell invasion12. LncRNA transcripts have been shown to be functional and are associated with cancer prognosis. Arkas’ default transcrip- tomes include ENSEMBL (build 88) cDNA and non-coding RNA reference sequences. Arkas-Quantification instructions are provided within BaseSpace (details for new users can be found here). Arkas is a web style format, but can also be launched using the command line using BaseSpace Command Line Interface. The inputs are RNA sequenc- ing samples, which may include SRA imported reads, and the out- puts include the Kallisto data, .tar.gz files of the Kallisto sample data, and a report summary (Supplementary Figure 1 and Supplementary Figure 2). Users may select for species type (Homo sapiens or Mus musculus), optionally correct for read length bias, and optionally select for the generation of pseudoBAMs. More sig- nificantly, users have the option to use the default transcriptome (ENSEMBL build 88) or to upload a custom FASTA of their choosing. For users that wish for local analysis, they can download the sample .tar.gz Kallisto files and analyze the data locally. Annotation of coding genes and transcripts The extraction of genomic and functional annotations directly from FASTA contig comments, eliding sometimes-unreliable dependencies on services such as BioMart, are calculated rapidly. The annotations were performed with a run time of 2.336 seconds (Supplementary Table 3) which merged the previous Kallisto data from 5 samples, creating a KallistoExperiment class with feature data containing a GenomicRanges11 object with 213782 ranges and 9 metadata columns. The system runtime for creating a merged Google offers very cost effective means for analyzing data, but requires expensive preparatory routines. Tatlow et al. employed Kallisto to pseudoalign 12,307 RNA-sequencing samples by renting Page 10 of 21 Page 10 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 Data availability Data used in testing variation between versions Controls: SRR1544480 Immortal-1 SRR1544481 Immortal-2 SRR1544482 Immortal-3 Comparison: SRR1544501 Qui-1 SRR1544502 Qui-2 Software availability Latest source code: https://github.com/RamsinghLab/Arkas-RNASeq Archived source code as at the time of publication: DOI: 10.5281/zenodo.54565421 License: MIT license Data availability Data used in testing variation between versions Controls: SRR1544480 Immortal-1 SRR1544481 Immortal-2 SRR1544482 Immortal-3 Data availability Data used in testing variation between versions Controls: SRR1544480 Immortal-1 SRR1544481 Immortal-2 SRR1544482 Immortal-3 Comparison: SRR1544501 Qui-1 SRR1544502 Qui-2 preemptible VMs from Google Cloud Platform for as little as $0.09 per sample. Tatlow et al. pseudoaligned 1,811 breast carcinoma samples completing on average in 101 minutes, and 934 Cancer Cell Line Encyclopedia (CCLE) BAM files completing in 84.7 minutes on average. However large scale efforts require spe- cialized knowledge for controlling containers (e.g. Kubernetes), manage resources, and queues. Although Tatlow et al. skillfully employed a cost-effective implementation of RNA-Seq analysis of massive databases, they mention a critical need for reducing the preprocessing routines involved in cloud computing13. preemptible VMs from Google Cloud Platform for as little as $0.09 per sample. Tatlow et al. pseudoaligned 1,811 breast carcinoma samples completing on average in 101 minutes, and 934 Cancer Cell Line Encyclopedia (CCLE) BAM files completing in 84.7 minutes on average. However large scale efforts require spe- cialized knowledge for controlling containers (e.g. Kubernetes), manage resources, and queues. Although Tatlow et al. skillfully employed a cost-effective implementation of RNA-Seq analysis of massive databases, they mention a critical need for reducing the preprocessing routines involved in cloud computing13. Competing interests Competing interests p g No competing interests were disclosed. Conclusion Arkas integrates the Kallisto pseudoalignment algorithm into the BaseSpace cloud computation ecosystem that can implement large- scale parallel ultra-fast transcript abundance quantification. We reduce a computational bottleneck by freeing inefficiencies from utilizing rapid transcript abundance calculations and connect- ing accelerated quantification software to the Sequencing Read Archive. We remove the second bottleneck because we reduce the necessity of database downloading; instead we encourage users to download aggregated analysis results. We also expand the range of common sequencing protocols to include an improved gene- set enrichment algorithm, Qusage, and allow for exporting into an exhaustive pathway analysis platform, Advaita, over the AWS EC2 field in parallel. License: MIT license Reference FASTA annotation files For Homo-sapiens and Mus-musculus ENSEMBL FASTA files were downloaded here for release 88. Reference FASTA annotation files For Homo-sapiens and Mus-musculus ENSEMBL FASTA files were downloaded here for release 88. BaseSpace offers other RNAseq tools and another analysis pipeline RNAExpress which reduces preparatory routines. RNAExpress runs DESeq2 and can be used to cross validate Arkas-Analysis. DESeq2 uses a negative binomial distribution to model differential expres- sion, whereas Arkas implements limma/voom empirical Bayes analysis pipeline. RNAExpress completed in 109 minutes compar- ing 4 controls and 4 comparison samples. Using the same samples, Arkas-Quantification completed in 42 minutes, and Arkas-Analysis completed in 54 minutes. Illumina’s BaseSpace catalog of modern, yet simple, tools are attractive for users wishing share sessions, and to rapidly (re)analyze entire experiment(s). ERCC sequences The ERCC sequences are provided in a SQL database format located here Author contributions AC wrote the manuscript, and developed the web-application and related software. TJ developed software, and helped the project design. GR wrote the manuscript and contributed to the develop- ment of software. Annotation of coding genes and transcripts Comparison: SRR1544501 Qui-1 SRR1544502 Qui-2 Software availability Latest source code: https://github.com/RamsinghLab/Arkas-RNASeq Software availability Latest source code: https://github.com/RamsinghLab/Arkas-RNASeq Galaxy offers shared workflows and analytical pipelines but is limited in the services related to storage due to the usage of pub- lic servers. In this light private storage platforms can flexibly store experimental data, although the range of analysis tools is not as wide compared to open-source platforms. Galaxy offers many usa- ble tools with a wide range of visualization pipelines. In contrast, BaseSpace offers tools to accomplish specific tasks at the expense of lowering the learning curve, which may be attractive for researchers interested in immediate, and verifiable, results. Archived source code as at the time of publication: DOI: 10.5281/zenodo.54565421 License: MIT license Archived source code as at the time of publication: DOI: 10.5281/zenodo.54565421 Archived source code as at the time of publication: DOI: 10.5281/zenodo.54565421 License: MIT license References data. BMC Bioinformatics. 2011; 12: 480. PubMed Abstract \| Publisher Full Text \| Free Full Text 8. Robinson MD, McCarthy DJ, Smyth GK: edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics. 2010; 26(1): 139–140. PubMed Abstract \| Publisher Full Text \| Free Full Text 9. Ritchie ME, Phipson B, Wu D, et al.: limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015; 43(7): e47. PubMed Abstract \| Publisher Full Text \| Free Full Text 10. Risso D, Ngai J, Speed TP, et al.: Normalization of RNA-seq data using factor analysis of control genes or samples. Nat Biotechnol. 2014; 32(9): 896–902. PubMed Abstract \| Publisher Full Text \| Free Full Text 11. Yaari G, Bolen CR, Thakar J, et al.: Quantitative set analysis for gene expression: a method to quantify gene set differential expression including gene-gene correlations. Nucleic Acids Res. 2013; 41(18): e170. PubMed Abstract \| Publisher Full Text \| Free Full Text 12. Mitra SA, Mitra AP, Triche TJ: A central role for long non-coding RNA in cancer. Front Genet. 2012; 3: 17. PubMed Abstract \| Publisher Full Text \| Free Full Text 13. Tatlow PJ, Piccolo SR: A cloud-based workflow to quantify transcript- expression levels in public cancer compendia. Sci Rep. 2016; 6: 39259. PubMed Abstract \| Publisher Full Text \| Free Full Text 1. Bray NL, Pimentel H, Melsted P, et al.: Near-optimal probabilistic RNA-seq quantification. Nat Biotechnol. 2016; 34(5): 525–527. PubMed Abstract \| Publisher Full Text 2. Soneson C, Matthes KL, Nowicka M, et al.: Isoform prefiltering improves performance of count-based methods for analysis of differential transcript usage. Genome Biol. 2016; 17: 12. PubMed Abstract \| Publisher Full Text \| Free Full Text 3. Bourgon R, Gentleman R, Huber W: Independent filtering increases detection power for high-throughput experiments. Proc Natl Acad Sci U S A. 2010; 107(21): 9546–9551. PubMed Abstract \| Publisher Full Text \| Free Full Text 4. Baker SC, Bauer SR, Beyer RP, et al.: The External RNA Controls Consortium: a progress report. Nat Methods. 2005; 2(10): 731–734. PubMed Abstract \| Publisher Full Text 5. Munro SA, Lund SP, Pine PS, et al.: Assessing technical performance in differential gene expression experiments with external spike-in RNA control ratio mixtures. Nat Commun. 2014; 5: 5125. PubMed Abstract \| Publisher Full Text 6. Grant information This project was funded by grants from Leukemia Lymphoma Society-Quest for Cures (0863-15), Illumina (San Diego), STOP Cancer and Tower Cancer Research Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Supplementary material Supplementary material Supplementary Table 1: Data variation with matching Kallisto versions. This shows the variation of mean differences between data using the matching Kallisto version 0.43.0. The rows represent the samples from the first run using version 0.43.0. The columns represent the samples from an additional run with version 0.43.0. Click here to access the data. Page 11 of 21 Page 11 of 21 Page 11 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 Supplementary Table 2: Data variation with non-matching Kallisto versions. Variation of mean differences between non-matching Kallisto versions and a randomly selected run previously generated (Supplementary Table 1). The rows are samples run using version 0.43.0, the columns are runs using version 0.43.1. Click here to access the data. Supplementary Table 3: Annotation runtime. System runtime for full annotation of a merged KallistoExperiment (seconds). The columns represent system runtime, the Elapsed Time is the total runtime. Click here to access the data. Supplementary Table 4: KallistoExperiment formation runtime. System runtime for the creation of a merged KallistoExperiment (seconds). The columns are similar to Supplementary Table 3. Click here to access the data. Supplementary Figure S1: Arkas-Quantification Web-Style user interface. The input field for A) Arkas-Quantification and B) Arkas- Analysis demonstrating SRA re-quantification. The control and comparison samples were obtained using BaseSpace SRA Import application, and were input into the Arkas pipeline. Click here to access the data. Supplementary Figure S2: Arkas-Quantification output directory. A) Depicts the Arkas-Quantification output directory which includes sample folders containing Kallisto data. B) Depicts the contents of a specific folder output by Arkas-Quantification. : Arkas-Quantification output directory. A) Depicts the Arkas-Quantification output directory which include Kallisto data. B) Depicts the contents of a specific folder output by Arkas-Quantification. upplementary Figure S2: Arkas-Quantification output directory. A) Depicts the Arkas-Quantification outpu ample folders containing Kallisto data. B) Depicts the contents of a specific folder output by Arkas-Quantificat Click here to access the data. 1. Bray NL, Pimentel H, Melsted P, et al.: Near-optimal probabilistic RNA-seq quantification. Nat Biotechnol. 2016; 34(5): 525–527. PubMed Abstract \| Publisher Full Text Open Peer Review Current Referee Status: References Lawrence M, Huber W, Pagès H, et al.: Software for computing and annotating genomic ranges. PLoS Comput Biol. 2013; 9(8): e1003118. PubMed Abstract \| Publisher Full Text \| Free Full Text 7. Risso D, Schwartz K, Sherlock G, et al.: GC-content normalization for RNA-Seq Page 12 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 doi:10.5256/f1000research.12854.r23689 Harold Pimentel Department of Genetics, Stanford University, Stanford, CA, USA Hello Colombo , et al. Firstly: I am so very sorry for such a late review. Anyway, the new manuscript looks much better. Thanks for the revisions. I just have one nitpick: in Figure 2 you show the p-value distribution over the range (0, 0.05). Perhaps I missed something, but I'm not sure I completely understand the value of showing over this interval rather than the whole interval (0, 1). Does it have to do with the normalization adjusting p-values specifically in this range? Regardless -- nice work and congrats! No competing interests were disclosed. Competing Interests: Referee Expertise: RNA-Seq analysis methods and data analysis I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Harold Pimentel Department of Genetics, Stanford University, Stanford, CA, USA Hello Colombo , et al. Firstly: I am so very sorry for such a late review. Anyway, the new manuscript looks much better. Thanks for the revisions. I just have one nitpick: in Figure 2 you show the p-value distribution over the range (0, 0.05). Perhaps I missed something, but I'm not sure I completely understand the value of showing over this interval rather than the whole interval (0, 1). Does it have to do with the normalization adjusting p-values specifically in this range? Regardless -- nice work and congrats! No competing interests were disclosed. Competing Interests: Referee Expertise: RNA-Seq analysis methods and data analysis I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Version 2 07 August 2017 Referee Report Version 1 24 May 2017 Referee Report doi:10.5256/f1000research.12258.r22616 doi:10.5256/f1000research.12258.r22616 , Ted Abel Marie Gaine Iowa Neuroscience Institute, University of Iowa, Iowa, USA This paper introduces a RNA-Seq analysis pipeline, Arkas, which combines currently available tools typically used in RNA-Seq studies. The novelty of this pipeline is the encapsulation of tools needed to prepare the data, run quality control checks, analyze the data and perform secondary analyses. This is , Ted Abel Marie Gaine Iowa Neuroscience Institute, University of Iowa, Iowa, USA This paper introduces a RNA-Seq analysis pipeline, Arkas, which combines currently available tools typically used in RNA-Seq studies. The novelty of this pipeline is the encapsulation of tools needed to prepare the data, run quality control checks, analyze the data and perform secondary analyses. This is This paper introduces a RNA-Seq analysis pipeline, Arkas, which combines currently available tools typically used in RNA-Seq studies. The novelty of this pipeline is the encapsulation of tools needed to prepare the data, run quality control checks, analyze the data and perform secondary analyses. This is Page 13 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 especially beneficial for investigators new to RNA-Seq analysis with little experience navigating through computational tools. The authors take care to outline the rationale behind creating an easy-to-use interface and how this will increase reproducibility and consistency across RNA-Seq studies. They emphasize the importance of consistency with versions by showing differing results between two Kallisto versions. especially beneficial for investigators new to RNA-Seq analysis with little experience navigating through computational tools. The authors take care to outline the rationale behind creating an easy-to-use interface and how this will increase reproducibility and consistency across RNA-Seq studies. They emphasize the importance of consistency with versions by showing differing results between two Kallisto versions. However, there are some minor limitations also found in this study: It would be beneficial to include quality control checks at the beginning of the pipeline to generate data regarding the inputted sequencing files. It would be beneficial to include quality control checks at the beginning of the pipeline to generate data regarding the inputted sequencing files. It would be interesting to see more processing time information to show the benefit of using this pipeline compared to similar methods. It would be interesting to see more processing time information to show the benefit of using this pipeline compared to similar methods. Is the rationale for developing the new software tool clearly explained? Yes Is the description of the software tool technically sound? Yes Are sufficient details of the code, methods and analysis (if applicable) provided to allow replication of the software development and its use by others? Yes Is sufficient information provided to allow interpretation of the expected output datasets and any results generated using the tool? Yes Are the conclusions about the tool and its performance adequately supported by the findings presented in the article? Yes No competing interests were disclosed. Competing Interests: Referee Expertise: Molecular neuroscience We have read this submission. We believe that we have an appropriate level of expertise to confirm that it is of an acceptable scientific standard. Version 1 As is discussed, the inclusion of lncRNAs increases the amount of potentially interesting results from this pipeline. However, the authors have chosen to ignore microRNAs, an important regulator of cellular function. The inclusion of microRNAs as a default option in this pipeline would provide even more potentially interesting results. The normalization steps and Figure 2 should be discussed in more detail. Specifically, expand on the reasons for choosing these two methods and the differences between the methods and their outputs. In addition, a note about how a user should select a normalization type would help new users. Whilst the authors suggest that the integration of Docker will help produce reproducible research methods, the in-depth look into Docker is unnecessary, as no data has been provided to show its benefit above other options. , University of Southern California, USA Giridharan Ramsingh , University of Southern California, USA Giridharan Ramsingh Thank you very much Dr. Abel for your insightful review. The revised manuscript removed the in-depth discussion of Docker because it was too broad. The revised version included a discussion section that compares processing times between Google Genomics, and another BaseSpace application. Your comments helped address the analysis of microRNAs. For example, Kallisto can process smaller FASTA sequences, however this invokes limitations to the construction of the Target DeBruijn Graph by increasing the path ambiguity of longer read sequences. The revised manuscript now addressed this limitation, and suggested that users analyze microRNAs separately. This analysis feature is not yet a default, but would be a great future addition. We further address details in regard to normalization motivation and selection. As suggested by the first reviewer Dr. Pimentel, we have significantly reduced the broad discussion section, and explicitly described the motivation for the development of . We have Arkas additionally revised the 'Methods' section to provide a brief overview of the applications, and clearer descriptions of the interface style that included Supplementary Figures depicting both interfaces. "This paper introduces a RNA-Seq analysis pipeline, Arkas, which combines currently available tools typically used in RNA-Seq studies. The novelty of this pipeline is the encapsulation of tools needed to prepare the data, run quality control checks, analyze the data and perform secondary analyses. This is especially beneficial for investigators new to RNA-Seq analysis with little experience navigating through computational tools. The authors take care to outline the rationale behind creating an easy-to-use interface and how this will increase reproducibility and consistency across RNA-Seq studies. They emphasize the importance of consistency with versions by showing differing results between two Kallisto versions. However, there are some minor limitations also found in this study: It would be beneficial to include quality control checks at the beginning of the pipeline to generate data regarding the inputted sequencing files." Thank you for this suggestion. Analyzing read quality will guide users into the important decision to filter low quality reads, however was not designed to address this. In the revised Arkas manuscript, we have now mentioned another independent BaseSpace application which FastQC can assess read quality. For users interested in manually uploading sequencing data to BaseSpace, each read must pass a quality filter. Is the description of the software tool technically sound? Yes Is the description of the software tool technically sound? Yes Page 14 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 , University of Southern California, USA Giridharan Ramsingh This quality filter will automatically reject poor quality reads, and for this we designed with the assumption that sequenced reads input were Arkas of good quality. "It would be interesting to see more processing time information to show the benefit of using this pipeline compared to similar methods." Thank you very much for addressing processing times. The revised manuscript significantly reduced the discussion section to comparisons of processing times. Your remarks inspired the addition of processing times of We’ve included further information comparing the Arkas. Page 15 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 processing time to another BaseSpace application Further, we added processing RNAExpress. time information of a different Kallisto analysis pipeline implemented over Google Genomics Platform. The discussion section now is far more concise with greater relevance toward the functionality of our developed software. processing time to another BaseSpace application Further, we added processing RNAExpress. time information of a different Kallisto analysis pipeline implemented over Google Genomics Platform. The discussion section now is far more concise with greater relevance toward the functionality of our developed software. "As is discussed, the inclusion of lncRNAs increases the amount of potentially interesting results from this pipeline. However, the authors have chosen to ignore microRNAs, an important regulator of cellular function. The inclusion of microRNAs as a default option in this pipeline would provide even more potentially interesting results." Including microRNAs is a very great idea. can quantify microRNAs, but we decided not Arkas include microRNAs as default yet. In the revised manuscript we address that the small sequence sizes are a potential limitation to quantification of cDNAs/ncRNAs because it may increase path ambiguities during the construction of the Target DeBruijn graphs. Hence, we suggest that users analyze microRNAs separately and locally. This would be a great additional feature for the next version of Arkas. "The normalization steps and Figure 2 should be discussed in more detail. Specifically, expand on the reasons for choosing these two methods and the differences between the methods and their outputs. In addition, a note about how a user should select a normalization type would help new users." Thank you for addressing this. The revised manuscript has now explicitly stated how end-users may decide a selection of the normalization type. We further provide a brief explanation to why unsupervised normalization was selected. , University of Southern California, USA Giridharan Ramsingh "Whilst the authors suggest that the integration of Docker will help produce reproducible research methods, the in-depth look into Docker is unnecessary, as no data has been provided to show its benefit above other options." We agree that the discussion of Docker was too broad, and the revised discussion is focused on comparative performance from other cloud platforms. We agree that the discussion of Docker was too broad, and the revised discussion is focused on comparative performance from other cloud platforms. None. Competing Interests: None. Competing Interests: 18 May 2017 Referee Report doi:10.5256/f1000research.12258.r22282 doi:10.5256/f1000research.12258.r22282 doi:10.5256/f1000research.12258.r22282 Harold Pimentel Department of Genetics, Stanford University, Stanford, CA, USA Note: I am a co-author of the kallisto tool, one of the tools that is used in this pipeline. Colombo et al. describe Arkas, a tool that takes raw RNA-Seq data and produces several d of downstream analyses. Arkas leverages existing analysis tools (e.g. kallisto and limma) a (Illumina BaseSpace) to create an easy to use, fast, and reproducible pipeline. A very usefu Harold Pimentel Department of Genetics, Stanford University, Stanford, CA, USA Note: I am a co-author of the kallisto tool, one of the tools that is used in this pipeline. Note: I am a co-author of the kallisto tool, one of the tools that is used in this pipeline. ote: I am a co-author of the kallisto tool, one of the tools that is used in this pipeline. Note: I am a co-author of the kallisto tool, one of the tools that is used in this pipeline Colombo et al. describe Arkas, a tool that takes raw RNA-Seq data and produces several different types of downstream analyses. Arkas leverages existing analysis tools (e.g. kallisto and limma) and platforms (Illumina BaseSpace) to create an easy to use, fast, and reproducible pipeline. A very useful (unique?) Page 16 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 feature is that it documents software versions and enforces consistent software versions allowing users to see the potential differences with different software versions. This is made explicit in the "Results" section. feature is that it documents software versions and enforces consistent software versions allowing users to see the potential differences with different software versions. This is made explicit in the "Results" section. Having all of these tools together greatly reduces the time to setup analyses and also reduces the complexity for RNA-Seq novices who might have no idea where to start. Arkas makes all of the typical figures one might make in a standard RNA-Seq analysis. It also provides gene-set analyses which are often excluded from other pipelines. In my experience, gluing together analyses from differential expression to gene-set analyses can often be an annoyance due to inconsistencies and annotations and versions of these annotations. Arkas nicely solves this problem. While I think the idea is very good and the tool seems comprehensive, I feel the manuscript needs a bit of work. doi:10.5256/f1000research.12258.r22282 Here are a few points: - There are a few areas where the scope seems too broad. In general, I feel that the manuscript can be shortened to be more clear as well as more precise. In particular, the Docker section in the discussion is too broad and the role of Arkas seems lost. I strongly recommend shortening this section and discussing the role of Docker in Arkas more clearly. - There are a few areas where the scope seems too broad. In general, I feel that the manuscript can be shortened to be more clear as well as more precise. In particular, the Docker section in the discussion is too broad and the role of Arkas seems lost. I strongly recommend shortening this section and discussing the role of Docker in Arkas more clearly. - While the abstract and introduction provide a description of Arkas in RNA-Seq analysis, they do not provide a motivation. It is sort of hinted in several sections in the paper, but it is not explicit. The motivation of building another pipeline should be explicit. - While the abstract and introduction provide a description of Arkas in RNA-Seq analysis, they do not provide a motivation. It is sort of hinted in several sections in the paper, but it is not explicit. The motivation of building another pipeline should be explicit. - How does this pipeline compare to other pipelines such as Galaxy, DNANexus, etc.? Should probably be noted in the introduction/discussion. - How does this pipeline compare to other pipelines such as Galaxy, DNANexus, etc.? Should probably be noted in the introduction/discussion. - Perhaps I missed it, but the interface of Arkas does not appear to be described. There is a short subsection "Operation" that doesn't describe the type of interface. It appears to be available on Illumina BaseSpace, but does this make it a commandline tool or an online web form style tool? A short description of this interface and possibly supplementary figures (if it is a web form style) should be provided. This is unclear to folks who are not familiar with BaseSpace. - Perhaps I missed it, but the interface of Arkas does not appear to be described. There is a short subsection "Operation" that doesn't describe the type of interface. doi:10.5256/f1000research.12258.r22282 It appears to be available on Illumina BaseSpace, but does this make it a commandline tool or an online web form style tool? A short description of this interface and possibly supplementary figures (if it is a web form style) should be provided. This is unclear to folks who are not familiar with BaseSpace. - It should be greater emphasized how this tool can be used to reanalyze existing SRA data with relative ease. In my opinion this is a very strong argument as to why one might want a tool like this. Areas that can be shortened: - "Data variance between software versions" can be shortened as some of this is repeated - "Data variance between software versions" can be shortened as some of this is repeated in "Results." - "Complete transcriptomes enrich annotation information..." Specifics of annotations can probably be removed/condensed. It is probably sufficient to say that some are 3x times larger which can change results drastically. - "Complete transcriptomes enrich annotation information..." Specifics of annotations can probably be removed/condensed. It is probably sufficient to say that some are 3x times larger which can change results drastically. - "Docker as a cornerstone of reproducible research" The role of Docker in general can probably be shortened and how Arkas leverages it should be made more clear. - "Docker as a cornerstone of reproducible research" The role of Docker in general can probably be shortened and how Arkas leverages it should be made more clear. More minor points: - A short sentence at the beginning of "Methods" should give an overview of the two-step p - The Galaxy Project (https://usegalaxy.org/) should probably be cited even though the sco different. - Figure 1a: "Receiver Operator Characteristic plot" of what? This is stated in the main text, but should also the stated in the figure caption. p g - It seems like BaseSpace sessions can easily be shared? If so, this is an additional strong point of using BaseSpace in Arkas. Page 17 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 Overall, I'm very excited to see this comprehensive tool exist and be described in this paper. Is the rationale for developing the new software tool clearly explained? Partly Is the description of the software tool technically sound? Author Response 08 Jun 2017 Author Response 08 Jun 2017 doi:10.5256/f1000research.12258.r22282 Yes Are sufficient details of the code, methods and analysis (if applicable) provided to allow replication of the software development and its use by others? Yes Is sufficient information provided to allow interpretation of the expected output datasets and any results generated using the tool? Yes Are the conclusions about the tool and its performance adequately supported by the findings presented in the article? Yes I am a co-author of the kallisto tool, one of the tools that is used in this pipeline. Competing Interests: Referee Expertise: RNA-Seq analysis methods and data analysis I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. Author Response 08 Jun 2017 , University of Southern California, USA Giridharan Ramsingh Thank you very much Dr. Pimentel for your thorough review. We have significantly reduced the broad discussion section, and narrowed the manuscript to the most important features. The 'Abstract' and 'Introduction' section was reduced to explicitly state the motivations for the design of In the revised manuscript, the 'Methods' section provides a brief overview of the Arkas. applications, and the 'Operation' section describes the interface style and includes Supplementary Figures depicting both apps. "Note: I am a co-author of the kallisto tool, one of the tools that is used in this pipeline. "Note: I am a co-author of the kallisto tool, one of the tools that is used in this pipeline. Colombo et al. describe Arkas, a tool that takes raw RNA-Seq data and produces several different types of downstream analyses. Arkas leverages existing analysis tools (e.g. kallisto and limma) and platforms (Illumina BaseSpace) to create an easy to use, fast, and reproducible pipeline. A very useful (unique?) feature is that it documents software versions and enforces consistent software versions allowing users to see the potential differences with different software versions. This is made explicit in the "Results" section. Having all of these tools together greatly reduces the time to setup analyses and also reduces the complexity for RNA-Seq novices who might have no idea where to start. Arkas makes all of the typical figures one might make in a standard RNA-Seq analysis. It also provides gene-set analyses which are often excluded from other pipelines. In my experience, gluing together analyses from differential expression to gene-set analyses can often be an annoyance due to inconsistencies and annotations and versions of these annotations. Arkas nicely solves this problem. While I think the idea is very good and the tool seems comprehensive, I feel the manuscript needs a bit of work. Here are a few points: - There are a few areas where the scope seems too broad. In general, I feel that the manuscript can be shortened to be more clear as well as more precise. In particular, the Docker section in the discussion is too broad and the role of Arkas seems lost. I strongly recommend shortening this " section and discussing the role of Docker in Arkas more clearly. - There are a few areas where the scope seems too broad. In general, I feel that the manuscript can be shortened to be more clear as well as more precise. In particular, the Docker section in the discussion is too broad and the role of Arkas seems lost. I strongly recommend shortening this " section and discussing the role of Docker in Arkas more clearly. Thank you very much for your input. In the revised manuscript, we have narrowed the Docker discussion section to the scope of BaseSpace platform, and have described ' relationship to Arkas Docker as an applied infrastructure to this platform. , University of Southern California, USA Giridharan Ramsingh , University of Southern California, USA Giridharan Ramsingh Thank you very much Dr. Pimentel for your thorough review. We have significantly reduced the broad discussion section, and narrowed the manuscript to the most important features. The 'Abstract' and 'Introduction' section was reduced to explicitly state the motivations for the design of In the revised manuscript, the 'Methods' section provides a brief overview of the Arkas. applications, and the 'Operation' section describes the interface style and includes Supplementary Figures depicting both apps. The second reviewer Dr. Abel also suggested that the in-depth discussion of Docker was too broad. The revised version includes a discussion section that is compares processing times between Google Genomics, and another BaseSpace application. We also have now included brief points in regard to Galaxy. Your helpful comments helped the manuscript become much more concise. In addition to your remarks, we have addressed important features regarding microRNAs on behalf of the second reviewer. Kallisto can process smaller FASTA sequences, however we have now addressed that users can analyze microRNAs, but we suggest a separate analysis for this. Page 18 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 We thank you very much for your revisions and appreciate your thoughtful remarks. We believe that addressing your remarks the manuscript is greatly elevated. Below are point-by-point responses to your questions. "Note: I am a co-author of the kallisto tool, one of the tools that is used in this pipeline. The previous version of the manuscript detailed the role of Docker in the broad concept of reproducible research. We have omitted these details. The revised manuscript describes the interdependent relationship between Arkas and Docker in the context of BaseSpace. For example, Arkas containerized Node.js and R to parse the BaseSpace JSON input information relating to BaseSpace’s input fields. The new manuscript explained that Docker and Arkas are not independent entities, and pertain specifically to BaseSpace. "- While the abstract and introduction provide a description of Arkas in RNA-Seq analysis, they do not provide a motivation. It is sort of hinted in several sections in the paper, but it is not explicit. The " motivation of building another pipeline should be explicit. Thank you for this suggestion. We have now explicitly provided the motivation for ’ Arkas development by mentioning bottlenecks in RNA-sequencing such as sequencing importing and pre-processing steps, and how rectifies those bottlenecks. In the revised version, we Arkas illustrate how was developed downstream from BaseSpace to greatly reduce Arkas SRA Import importing and conversion steps. Also, we now explicitly stated the motivation for Page 19 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 such that Kallisto was implemented in parallel, which now scales Arkas-Quantification quantification speed to the Amazon AWS EC2 cluster node availability rate. In addition, the revised manuscript explicitly stated the motivation for which provides a Arkas-Analysis, comprehensive analysis. "- How does this pipeline compare to other pipelines such as Galaxy, DNANexus, etc.? Should " probably be noted in the introduction/discussion. "- How does this pipeline compare to other pipelines such as Galaxy, DNANexus, etc.? Should " probably be noted in the introduction/discussion. Thank you for this suggestion. In the revised discussion section, we now compare features of other cloud platforms, and other BaseSpace RNA-Seq applications. The revised discussion now included processing times of a large scale RNA-seq analysis that implemented Kallisto using Google Genomics Platform. In addition to Goolgle Genomics, the revised manuscript briefly compares features offered by Galaxy to BaseSpace. Further we compare to other Arkas BaseSpace RNA-Seq applications. "- Perhaps I missed it, but the interface of Arkas does not appear to be described. There is a short subsection "Operation" that doesn't describe the type of interface. It appears to be available on Illumina BaseSpace, but does this make it a commandline tool or an online web form style tool? A short description of this interface and possibly supplementary figures (if it is a web form style) " should be provided. This is unclear to folks who are not familiar with BaseSpace. Thank you again for this suggestion. We have included a description explicitly stating that Arkas is a web form style. In addition, we included two Supplementary Figures to address the web input forms. Supplementary Figure 1 shows the input form for both web style apps, and Supplementary Figure 2 shows the output folder directory of the . Arkas-Quantification "- It should be greater emphasized how this tool can be used to reanalyze existing SRA data with " relative ease. In my opinion this is a very strong argument as to why one might want a tool like this. Thank you for addressing reanalysis of SRA data. In the updated manuscript, we now mention that design was motivated by the BaseSpace application The revised introduction Arkas' SRA Import. now explicitly stated that is SRA compatible and we have provided citations for readers Arkas interested in utilizing this SRA application. "Areas that can be shortened: - "Data variance between software versions" can be shortened as some of this is repeated in "Results." '" - "Data variance between software versions" can be shortened as some of this is repeated "Results." '" We combined the “Data variance between software versions” and “Results” section into an appropriate concise section. We combined the “Data variance between software versions” and “Results” section into an appropriate concise section. "- How does this pipeline compare to other pipelines such as Galaxy, DNANexus, etc.? Should " probably be noted in the introduction/discussion. - "Complete transcriptomes enrich annotation information..." Specifics of annotations can probably be removed/condensed. It is probably sufficient to say that some are 3x times larger which can " change results drastically. - "Complete transcriptomes enrich annotation information..." Specifics of annotations can probably be removed/condensed. It is probably sufficient to say that some are 3x times larger which can " change results drastically. We reduced this discussion to brief specifics of database sizes. While obvious, we believe that a brief overview provides motivation for the default transcriptomes chosen by In the revised Arkas. Page 20 of 21 F1000Research 2017, 6:586 Last updated: 30 AUG 2017 - "Docker as a cornerstone of reproducible research" The role of Docker in general can probably be " shortened and how Arkas leverages it should be made more clear. - "Docker as a cornerstone of reproducible research" The role of Docker in general can probably be " shortened and how Arkas leverages it should be made more clear. Thank you again for this comment. We agree that this broad discussion went off topic and may distract future readers. The manuscript is greatly improved with the removal of the discussion about democratization of research efforts, and biotechnology. We significantly revised the discussion to a comparison of differing cloud platforms and corresponding processing times of other cloud applications. "More minor points: - A short sentence at the beginning of "Methods" should give an overview of the two-step p We provided an overview of in the section described. Arkas "- The Galaxy Project (https://usegalaxy.org/) should probably be cited even though the scope is a " bit different. Galaxy is briefly mentioned in the discussion. The revised manuscript reviewed and compared processing times of Google Genomics Platform and another RNAseq application within BaseSpace. "- Figure 1a: "Receiver Operator Characteristic plot" of what? This is stated in the main text, but should also the stated in the figure caption. " - Swap Figure 1d and 1c. Thank you for pointing this out. The revised Figure 1a now states that the Receiver Operator Characteristic plot is for ratios of detected and actual spiked ERCC sequences. We have swapped Figure1d and Figure 1c. "- It seems like BaseSpace sessions can easily be shared? If so, this is an additional strong point of using BaseSpace in Arkas." We now mention this brief point in the discussion. "- How does this pipeline compare to other pipelines such as Galaxy, DNANexus, etc.? Should " probably be noted in the introduction/discussion. We now mention this brief point in the discussion. " "Overall, I'm very excited to see this comprehensive tool exist and be described in this paper. Thank you very much Dr. Pimentel. None Competing Interests: " "Overall, I'm very excited to see this comprehensive tool exist and be described in this paper. Page 21 of 21
https://openalex.org/W2796368986	https://europepmc.org/articles/pmc5803918?pdf=render	English	null	Semi-automated ultrasound guidance applied to nasogastrojejunal tube replacement for enteral nutrition in critically ill adults	BioMedical engineering online	2,018	cc-by	4,778	BioMedical Engineering OnLine Li et al. BioMed Eng OnLine (2018) 17:21 https://doi.org/10.1186/s12938-018-0452-1 Open Access RESEARCH © The Author(s) 2018. 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/publicdo- main/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Semi‑automated ultrasound guidance applied to nasogastrojejunal tube replacement for enteral nutrition in critically ill adults Ying Li1, Yu Ye2, Yang Mei1, Haiying Ruan1 and Yuan Yu1 Correspondence: l1i2y3y4y5@163.com 2 Department of Neurosurgery, Longgang Central Hospital of Shenzhen, Shenzhen 518116, People’s Republic of China Full list of author information is available at the end of the article BioMedical Engineering OnLine BioMedical Engineering OnLine BioMedical Engineering OnLine Background g Nutrition support is one major development of clinical medicine in the twentieth cen- tury, and has become an indispensable constituent part in the treatment of critically ill patients, in order to alleviate the nutritional deficit [1, 2]. Enteral nutrition has achieved significant advances in decades, and is beneficial for the patients who have functional guts but can not meet their nutritional requirements via normal diet, on account of can- cer, HIV, stroke, multiple sclerosis, dementia, etc. [3–5]. This kind of enteral feeding can be delivered by means of various approaches, including nasogastric tube, percutaneous endoscopic gastrostomy, jejunostomy. In general, during the period of enteral nutri- tion, the providers also need to assess the nutritional status, and evaluate the nutritional requirements of patients [6]. Besides, the development of enteral nutrition also requires multidisciplinary teams, such as the extended roles for dietitians and nurses, etc. [7]. In addition, more and more serious aging society, various diseases mentioned above, the swallowing difficulties and malnutrition resulted from various complications, are all the main reasons why rapidly increasing enteral nutrition is needed [8, 9]. At present, the enteral nutrition approaches via nose and duodenum (or nose and jejunum) are the preferred method of nutritional support in medical engineering, given the superiority of in line with physiological processes and no serious complication [10, 11]. In this study, the authors adopted saline as the acoustic window, and gave enteral nutrition support treatment to critically ill patients, via the nasogastrojejunal approach guided by semi-automated ultrasound. These above patients benefited a lot from this kind of nutrition support, and we reported the detailed information as followed. Abstract Background and objective: At present, the enteral nutrition approaches via nose and duodenum (or nose and jejunum) are the preferred method of nutritional sup- port in the medical engineering field, given the superiority of in line with physiologi- cal processes and no serious complication. In this study, the authors adopted saline as the acoustic window, and gave enteral nutrition support to critically ill patients, via the nasogastrojejunal approach guided by semi-automated ultrasound. These above patients benefited a lot from this kind of nutrition support treatment, and we aimed to report the detailed information. Methods: Critically ill patients (n = 41) who had been treated with hospitalized intestine nutrition were identified. The Apogee 1200 ultrasonic diagnostic apparatus, and nasogastrojejunal tubes were adopted to carry out intestine nutrition treatment guided by semi-automated ultrasound. In order to confirm the specific positions of car- dia, gastric body, antrum of stomach, and pylorus, the semi-automated ultrasound was utilized to probe the stomach cavity. And then, the ultrasonic probe was placed in the cardia location, and the nasogastrojejunal tube was slowly inserted through the metal thread. After operation, the nursing service satisfaction of patients and mean operation time were calculated, respectively. Results: All the patients were treated with enteral nutrition via nasogastrojejunal tube, and the whole procedure was under the guidance of semi-automated ultrasonog- raphy. The end of the feeding tube is attached to the surface of the stomach with a greater curvature, which can be bent on account of a no gastric peristalsis squeeze function, and thereby were prevented from entering into the antrum and pylorus loca- tions. After this procedure, the mental thread was taken out, and the tube was pushed forward by a “drift” approach in order to allow it to enter into the intestine. The total nursing service satisfaction of patients was 90.24%, and the total incidence of adverse reactions was 17.07%. Conclusions: In summary, the application of saline can be taken as sound window, and the metal wire as the tracking target, the bedside nasogastrojejunal tube guided by semi-automated ultrasound is an effective feeding tube placement method, with relatively good clinical application value in medical engineering. Keywords: Enteral nutrition, Nasogastrojejunal tube, Semi-automated ultrasound, Feeding tube placement Li et al. BioMed Eng OnLine (2018) 17:21 Page 2 of 9 Clinical data of the identified patients This research was approved beforehand by the institution ethics committee in our hospi- tal. According to the relevant regulations of ethics, the informed consent of patients had been obtained before investigation. 41 critically ill patients who would be treated with hospitalized intestine nutrition were identified in our department, from February 2015 to January 2017. In detail, 34 males and 7 females, the age ranged from 21 to 86 years old. 15 cases suffered from primary acute severe pancreatitis, 21 cases suffered from chronic obstructive pulmonary disease and respiratory failure, 3 cases suffered from cer- ebral hemorrhage, 1 case suffered from erosive gastritis, 1 case suffered from multiple organ failure, as shown in Table 1 and Fig. 1. And, Fig. 2 illustrated the case number per year from 2015 to 2017. Table 1 General information of patients identified in this study Diseases Number Gender (male/female) Average age Primary acute severe pancreatitis 15 13/2 48.20 ± 7.80 Chronic obstructive pulmonary disease 21 18/3 59.22 ± 7.77 Cerebral hemorrhage 3 1/2 68.21 ± 6.11 Erosive gastritis 1 1/0 35 Multiple organ failure 1 1/0 62 Table 1 General information of patients identified in this study Li et al. BioMed Eng OnLine (2018) 17:21 Page 3 of 9 Instrument and methods The Apogee 1200 portable ultrasonic diagnostic apparatus (Japan), and Ch 12 nasogas- trojejunal tubes (Manufacturer: Bai Tong in China) were adopted to carry out intestine Fig. 1 Distribution of primary diseases involved in this research Fig. 2 Case number per year from 2015 to 2017 Fig. 1 Distribution of primary diseases involved in this research Fig. 1 Distribution of primary diseases involved in this research Fig. 1 Distribution of primary diseases involved in this research Fig. 1 Distribution of primary diseases involved in this research Fig. 2 Case number per year from 2015 to 2017 Instrument and methodsh The Apogee 1200 portable ultrasonic diagnostic apparatus (Japan), and Ch 12 nasogas- trojejunal tubes (Manufacturer: Bai Tong in China) were adopted to carry out intestine nutrition treatment guided by semi-automated ultrasound. These included patients fasted for 6–8 h, and then received gastrointestinal decompression treatment for 1 h in order to vent the gas and food residues in the stomach, so that the images could be clearly displayed. The patients kept sober and took semi-recumbent, in order to make them coordinate work during treatment. The routine abdomen ultrasound was utilized to probe the stomach cavity, in order to confirm the specific positions of cardia, gastric body, antrum of stomach, and pylorus. And then, the ultrasonic probe was placed in the cardia location, and the nasogastrojejunal tube was slowly inserted through the metal thread. The ultrasonic diagnostic apparatus could capture the image of metal thread, and the intubationist slowly injected pre-heated physiological saline 300–400 ml. After- wards, the ultrasonic probe was used to perform fan-shaped scan, with the purpose of Page 4 of 9 Li et al. BioMed Eng OnLine (2018) 17:21 Fig. 3 Captured image when the nasogastrojejunal tube entered into pylorus and duodenal bulb Fig. 3 Captured image when the nasogastrojejunal tube entered into pylorus and duodenal bulb knowing the specific location of nasogastrojejunal tube and the status of gastric motility. The intubationist slowly rotated the guide wire to make it follow the gastric motility, and at the same time slowly pushed the tube. The nasogastrojejunal tube straightly went into the pylorus, and was forwarded smoothly. When the intubationist felt empty, the tube entered into duodenum. The location of ultrasonic probe was adjusted, the duodenal bulb could be observed, and the ultrasonic echo of metal thread was stronger. And then the thread was pushed forward to another 110–120 cm. The bedside abdomen radio- graphs were obtained to confirm whether the tube position was proper, and then the guide wire was took out. And then, the operator made sure that the patient was in a good condition, and then another 20 ml saline was injected again to assure the tube was unob- structed, the gas over the water tone could be heard by umbilical auscultation. At last, the enteral nutrition pump was connected with the tube. After operation, the nursing service satisfaction of patients and mean operation time were calculated, respectively. Average operation time of intubation The average operation time of intubation in this study was displayed in Table 2. 15 cases suffered from primary acute severe pancreatitis, and the average operation time was 48.20 ± 12.50 min. 21 cases suffered from chronic obstructive pulmonary disease, and the average operation time was 26.80 ± 6.52 min. 3 cases suffered from Cerebral hem- orrhage, and the average operation time was 44.21 ± 4.59 min. 1 patients with Erosive gastritis, and 1 case failed. General information All the patients were treated by using nasogastrojejunal tubes via guidance of semi-auto- mated ultrasonography, and the status of the patients during operations were all good. 40 cases were successfully implanted, 1 case failed. The failed case suffered from mul- tiple organ failure, and the gastric motility disappeared. The end of the feeding tube is attached to the surface of the stomach with a greater curvature, which can be bent on account of a no gastric peristalsis squeeze function, and thereby prevented from enter- ing into the antrum and pylorus locations. After this procedure, the mental thread was taken out, and the tube was pushed forward by a “drift” approach in order to allow it to enter into the intestine. Nevertheless, the operation time lasted longer, and the nasogas- trojejunal tube was placed into the intestine after 2 h. And, Fig. 3 illustrated the captured image when the nasogastrojejunal tube entered into pylorus and duodenal bulb. Page 5 of 9 Li et al. BioMed Eng OnLine (2018) 17:21 Table 2 Average operation time of intubation in this study Diseases Number Average operation time (min) Primary acute severe pancreatitis 15 48.20 ± 12.50 Chronic obstructive pulmonary disease 21 26.80 ± 6.52 Cerebral hemorrhage 3 44.21 ± 4.59 Erosive gastritis 1 32 Multiple organ failure 1 Failed Table 2 Average operation time of intubation in this study Table 3 Nursing service satisfaction of patients identified in this study a Great satisfaction; b satisfaction; c dissatisfaction Diseases Number 1a 2b 3c Total satisfactory (%) Primary acute severe pancreatitis 15 6 8 1 93.33 Chronic obstructive pulmonary disease 21 7 12 2 90.48 Cerebral hemorrhage 3 1 2 0 100 Erosive gastritis 1 0 1 0 100 Multiple organ failure 1 0 0 1 0 In total 41 14 23 4 90.24 Table 3 Nursing service satisfaction of patients identified in this study Patients’ satisfaction degree Besides, we made another questionnaire survey to investigate the patients’ satisfaction degree of nasogastrojejunal tube, and the result was illustrated in Table 3 and Fig. 4. The total nursing service satisfaction of patients was 90.24%. The satisfactory degree of patients with primary acute severe pancreatitis was 93.33%, the satisfactory degree of patients with chronic obstructive pulmonary disease was 90.48%, the satisfactory degree of patients with cerebral hemorrhage and erosive gastritis both were 100%. In general, the total nursing service satisfaction appeared high, more than 90%, which suggested that this kind of technique could be acceptable very well. Nevertheless, several serious complications occurred, on account of the diseases or the operators, and the nursing service satisfaction of these patients decreased. Discussion With the rapid development of medical engineering, the medical care of patients with critical illness has become increasingly complicated [12–14]. And the nutrition support has long been recognized as supportive therapy in critical care. In recent decades, the clinical nutrition has been evolving all the time, and it has been deemed as a form of therapy, rather than just a way of delivering nutrition [15–17]. Because several nutri- tion might exert therapeutic effects via immunomodulation or liver protection, etc. For instance, among cancer patients, the enteral nutrition contributes to decrease the post- operative complications and prolong survival time [18]. At present, the application of immune-enhancing enteral nutrition is an integral part of surgical guidelines. On the basis of latest ESICM clinical practice guidelines, the initiating enteral nutrition has been recommended to be provided within the first 24–48 h after intensive care unit admission, if the patients can not eat by themselves [19, 20].h The main approaches of enteral nutrition are various, but the commonly used and noninvasive method is nasogastrojejunal tube. The traditional method of intubation tube always depends on X ray or gastroscope, or only relies on the operator experi- ence. When the intubationist just operates by his own experience, the success rate of intubation appears lower [21, 22]. When the intubationist operates under the guidance of X ray, the patients have to be moved, but the majority of patients are critically ill in the intensive care unit and are hard to be carried, especially for the patients who need mechanical ventilation [23, 24]. Nasogastrojejunal tube insertion is based on minimally invasive catheterization procedure that is combined with ultrasound guidance. This semi-automated device pertains to a class of medical imaging assisted equipment that can help patients in terms of enteral nutrition [25]. On the other hand, the superiorities of intubation guided by semi-automated ultra- sound are as follows. (1) The operation by the bed is permitted, without moving the patient. And the operator can observe the location of nasogastrojejunal tube. (2) The operation guided by semi-automated ultrasound do not bring about radioactive injury, and the painful feeling of patients was smaller when receiving this noninvasive method. (3) It is cost-effective method, when compared with digital subtraction angiography and gastroscope. In addition, there are several points need to be addressed during imple- mentation. Incidence of adverse drug reactions In addition, in terms of incidence of adverse drug reactions among the identified patients in this study, we calculated them and listed in Table 4 and Fig. 5. The main adverse drug reactions in this research were stomachache, headache, nausea and vomiting. The inci- dence of adverse reactions of primary acute severe pancreatitis was 6.67%, the inci- dence of chronic obstructive pulmonary disease was 14.29%, the incidence of cerebral Li et al. BioMed Eng OnLine (2018) 17:21 Page 6 of 9 Fig. 4 Nursing service satisfaction of patients suffering from different diseases Fig. 4 Nursing service satisfaction of patients suffering from different diseases Table 4 Incidence of adverse reactions among the identified patients in this study a Stomachache; b headache; c nausea and vomiting Diseases Number 1a 2b 3c Total rate (%) Primary acute severe pancreatitis 15 0 0 1 6.67 Chronic obstructive pulmonary disease 21 0 0 3 14.29 Cerebral hemorrhage 3 0 1 1 66.67 Erosive gastritis 1 0 0 0 0 Multiple organ failure 1 0 0 1 100 In total 41 0 1 6 17.07 Fig. 5 Adverse reactions resulting from different diseases Table 4 Incidence of adverse reactions among the identified patients in this study a Stomachache; b headache; c nausea and vomiting Diseases Number 1a 2b 3c Total rate (%) Primary acute severe pancreatitis 15 0 0 1 6.67 Chronic obstructive pulmonary disease 21 0 0 3 14.29 Cerebral hemorrhage 3 0 1 1 66.67 Erosive gastritis 1 0 0 0 0 Multiple organ failure 1 0 0 1 100 In total 41 0 1 6 17.07 Fig. 5 Adverse reactions resulting from different diseases Table 4 Incidence of adverse reactions among the identified patients in this study a Stomachache; b headache; c nausea and vomiting Diseases Number 1a 2b 3c Total rate (%) Primary acute severe pancreatitis 15 0 0 1 6.67 Chronic obstructive pulmonary disease 21 0 0 3 14.29 Cerebral hemorrhage 3 0 1 1 66.67 Erosive gastritis 1 0 0 0 0 Multiple organ failure 1 0 0 1 100 In total 41 0 1 6 17.07 Table 4 Incidence of adverse reactions among the identified patients in this study Fig. 5 Adverse reactions resulting from different diseases Fig. 5 Adverse reactions resulting from different diseases Li et al. BioMed Eng OnLine (2018) 17:21 Page 7 of 9 hemorrhage was 66.67%, the incidence of erosive gastritis was 0%. Incidence of adverse drug reactions At last, the total inci- dence of adverse reactions was 17.07%. Conclusion To sum up, the application of saline can be taken as sound window, and the metal wire as the tracking target, the bedside nasogastrojejunal tube guided by semi-automated ultra- sound is an effective feeding tube placement method, with relatively good clinical appli- cation value in medical engineering. Ethics approval and consent to participate All the clinical information were obtained with informed consent (or a formal waiver of consent) with approval by the Research Ethics Boards in our hospital. And, the informed consents of patients were obtained beforehand [26, 27]. Authors’ contributions Performed the literature review: YL; Carried out research: YL, YM, HR, and YY; Gave advice for setup: YY; Checked the valid- ity of data: YL, YY. All authors read and approved the final manuscript. Discussion (1) Before operation, the abrosia for 6–8 h is needed, as well as gastrointes- tinal decompression for 1 h, in order to exclude interference resulted from food residue and gas in the stomach, and thereby to improve the image quality, to clearly display the tube position, and to reduce the failure rate of operation. (2) When the tube enters into gastral cavity, another 300–400 ml saline should be injected into tube. The saline should be placed in the incubator and be heated to 37 °C, to avoid unnecessary stimulation. The injected saline forms an acoustic window, so that the nasogastrojejunal can be clearly displayed. (3) The tube should be carried forward, with the help of gastric peristalsis, and this is also the pivotal point. The stomach cavity has two physiological curves, namely, stomach angle and pyloric region. The metal wire always remains extended position, and it is difficult to go through the two bending angles. When the peristaltic wave gets Li et al. BioMed Eng OnLine (2018) 17:21 Page 8 of 9 through the two physiological bending, the bending angles disappear, and the extrusion force pushes forward the tube, until entering into duodenum. The only one failure case attributed to the disappearance of gastric motility in the patient suffering from multiple organ failure. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliatio Received: 13 December 2017 Accepted: 27 January 2018 Received: 13 December 2017 Accepted: 27 January 2018 Author details 1 Author details 1 Department of Critical Care Medicine, Second People’s Hospital of Shenzhen, Shenzhen 518035, People’s Republic of China. 2 Department of Neurosurgery, Longgang Central Hospital of Shenzhen, Shenzhen 518116, People’s Republic of China. References 1. Bloomer MJ, Clarke AB, Morphet J. Nurses’ prioritization of enteral nutrition in intensive care units: a national survey. Nurs Crit Care. 2017. https://doi.org/10.1111/nicc.12284. 1. Bloomer MJ, Clarke AB, Morphet J. Nurses’ prioritization of enteral nutrition in intensive care units: a national survey. Nurs Crit Care. 2017. https://doi.org/10.1111/nicc.12284. p g 2. Boland K, Maher N, O’Hanlon C, O’Sullivan M, Rice N, Smyth M, Reynolds JV. Home enteral nutrition recipient ti t ti t i i li ti d ti f ti F tli G t t l 2017 8 79 84 2. Boland K, Maher N, O’Hanlon C, O’Sullivan M, Rice N, Smyth M, Reynolds JV. Home enteral nutrition recip patient perspectives on training, complications and satisfaction. Frontline Gastroenterol. 2017;8:79–84. p p p g, p 3. Buyukcoban S, Akan M, Koca U, Eglen MY, Ciceklioglu M, Mavioglu O. Comparison of two different enteral nutriti 3. Buyukcoban S, Akan M, Koca U, Eglen MY, Ciceklioglu M, Mavioglu O. Compa protocol in critically ill patients. Turk J Anaesthesiol Reanim. 2016;44:265–9. y g g g p protocol in critically ill patients. Turk J Anaesthesiol Reanim. 2016;44:265–9. y 4. Chen ZH, Lin SY, Dai QB, Hua J, Chen SQ. The effects of pre-operative enteral nutrition from nasal feeding tubes on gastric outlet obstruction. Nutrients. 2017;9:E373. 5. Heerasing N, Thompson B, Hendy P, Heap GA, Walker G, Bethune R, Mansfield S, Calvert C, Kennedy NA, Ahmad T, Goodhand JR. Exclusive enteral nutrition provides an effective bridge to safer interval elective surgery for adults with Crohn’s disease. Aliment Pharmacol Ther. 2017;45:660–9. 6. Klek S, Scislo L, Walewska E, Choruz R, Galas A. Enriched enteral nutrition may improve short-term survival in stage IV gastric cancer patients: a randomized, controlled trial. Nutrition. 2017;36:46–53. 7. Ma L, Luo GY, Ren YF, Qiu B, Yang H, Xie CX, Liu SR, Liu SL, Chen ZL, Li Q, Fu JH, Liu MZ, Hu YH, Ye WF, Liu H. Concur- rent chemoradiotherapy combined with enteral nutrition support: a radical treatment strategy for esophageal squamous cell carcinoma patients with malignant fistulae Chin J Cancer 2017 36 8 q p gi 8. Ojo O, Brooke J. Recent advances in enteral nutrition. Nutrients. 2016;8:E709. i Ojo O, Brooke J. Recent advances in enteral nutrition. Nutrients. 2016 j 9. O’Leary-Kelley C, Bawel-Brinkley K. Nutrition support protocols: enhancing delivery of enteral nutrition. Crit Care Nurse. 2017;37:e15–23. 10. Acknowledgements Acknowledgements The ICU research team leaded by Prof. Yongming Yao in the First Affiliated Hospital of PLA General Hospital gave the financial support for this research (Shenzhen’s Sanming Project, Grant No. SZSM201612011). Competing interests Competing interests The authors declare that they have no competing interests. Competing interests The authors declare that they have no competing interests. Ethics approval and consent to participate References Reintam Blaser A, Starkopf J, Alhazzani W, Berger MM, Casaer MP, Deane AM, Fruhwald S, Hiesmayr M, Ichai C, Jakob SM, Loudet CI, Malbrain ML, Montejo Gonzalez JC, Paugam-Burtz C, Poeze M, Preiser JC, Singer P, van Zanten AR, De Waele J, Wendon J, Wernerman J, Whitehouse T, Wilmer A, Oudemans-van Straaten HM. Early enteral nutrition in critically ill patients: ESICM clinical practice guidelines. Intensive Care Med. 2017;43:380–98. y p p g 11. Shu XL, Kang K, Gu LJ, Zhang YS. Effect of early enteral nutrition on patients with digestive tract surgery: a me analysis of randomized controlled trials. Exp Ther Med. 2016;12:2136–44. y 11. Shu XL, Kang K, Gu LJ, Zhang YS. Effect of early enteral nutrition on patient analysis of randomized controlled trials. Exp Ther Med. 2016;12:2136–44. Page 9 of 9 Li et al. BioMed Eng OnLine (2018) 17:21 12. Fletcher J. Giving nutrition support to critically ill adults. Nurs Times. 2015;111:12–6. y 13. Kiss N. Nutrition support and dietary interventions for patients with lung cancer: current insights. Lung Cancer (Auckl). 2016;7:1–9. 14. Lheureux O, Preiser JC. Role of nutrition support in inflammatory conditions. Nutr Clin Pract. 20 O, Preiser JC. Role of nutrition support in inflammatory conditions. 4. Lheureux O, Preiser JC. Role of nutrition support in inflammatory l 15. Marshall S, Agarwal E, Young A, Isenring E. Role of domiciliary and family carers in individualised nutrition support l 15. Marshall S, Agarwal E, Young A, Isenring E. Role of domiciliary and family carers in individualised nutrition for older adults living in the community. Maturitas. 2017;98:20–9. 15. Marshall S, Agarwal E, Young A, Isenring E. Role of domiciliary and fa for older adults living in the community. Maturitas. 2017;98:20–9. for older adults living in the community. Maturitas. 2017;98:20–9. Mulherin DW, Cogle SV. Updates in nutrition support for critically I 16. Mulherin DW, Cogle SV. Updates in nutrition support for critically Ill adult patients. Hosp Pharm. 2017 y 17. Sajeev M, Cohen J, Wakefield CE, Fardell JE, Cohn RJ. Decision aid for nutrition support in pediatric oncology: a pilot study. JPEN. 2017;41:1336–47. y 18. Wei J, Chen W, Zhu M, Cao W, Wang X, Shi H, Dong B, Sun J, Chen H, Zhou Y, Zhou S, Xu J. Guidelines for parentera and enteral nutrition support in geriatric patients in China. Asia Pac J Clin Nutr. 2015;24:336–46. pp g p 19. References Zhang H, Wang Y, Jiang ZM, Kondrup J, Fang H, Andrews M, Nolan MT, Mu SY, Zhang J, Yu K, Lu Q, Kang WM. Impact of nutrition support on clinical outcome and cost-effectiveness analysis in patients at nutritional risk: a prospective cohort study with propensity score matching. Nutrition. 2017;37:53–9. pp g p 19. Zhang H, Wang Y, Jiang ZM, Kondrup J, Fang H, Andrews M, Nolan MT, Mu SY, Zhang J, Yu K, Lu Q, Kang WM. Impact of nutrition support on clinical outcome and cost-effectiveness analysis in patients at nutritional risk: a prospective pp g p 19. Zhang H, Wang Y, Jiang ZM, Kondrup J, Fang H, Andrews M, Nolan MT, M of nutrition support on clinical outcome and cost-effectiveness analysis cohort study with propensity score matching. Nutrition. 2017;37:53–9. of nutrition support on clinical outcome and cost-effectiveness analys cohort study with propensity score matching. Nutrition. 2017;37:53–9. y p p y g 20. Holzinger U, Brunner R, Miehsler W, Herkner H, Kitzberger R, Fuhrmann V, Metnitz PG, Kamolz LP, Madl C. Jejunal tube placement in critically ill patients: a prospective, randomized trial comparing the endoscopic technique with the electromagnetically visualized method. Crit Care Med. 2011;39:73–7. 20. Holzinger U, Brunner R, Miehsler W, Herkner H, Kitzberger R, Fuhrmann V, Metnitz tube placement in critically ill patients: a prospective, randomized trial comparin the electromagnetically visualized method. Crit Care Med. 2011;39:73–7. 20. Holzinger U, Brunner R, Miehsler W, Herkner H, Kitzberger R, Fuhrmann V, Metnitz PG, Kamolz LP, Madl C. Jejunal tube placement in critically ill patients: a prospective, randomized trial comparing the endoscopic technique with the electromagnetically visualized method Crit Care Med 2011;39:73–7 21. Li J, Gu Y, Zhou R. Rhubarb to facilitate placement of nasojejunal feeding tubes in patients in the intensive care unit. Nutr Clin Pract. 2016;31:105–10. 22. Li RH, Li DC, Lv XP, Quan B. Radiophotographically controlled nasointestinal intubation to treat recurrent postopera- tive adhesive ileus. Cell Biochem Biophys. 2015;72:509–14. p y ; 23. Li RH, Li DC, Tian QZ, Wu P, Zhang XH. Diagnosis of small bowel obstruction using targeted enterography during l d C ll B h B h p y 23. Li RH, Li DC, Tian QZ, Wu P, Zhang XH. Diagnosis of small bowel obstruction using targeted enterography during nasointestinal decompression. Cell Biochem Biophys. 2015;72:833–7. nasointestinal decompression. Cell Biochem Biophys. 2015;72:83 p p y 24. References Stimac D, Poropat G, Hauser G, Licul V, Franjic N, Valkovic Zujic P, Milic S. Early nasojejunal tube feeding versus nil-by- h i i i d i d li i l i l P l 2016 16 523 8 p p y 24. Stimac D, Poropat G, Hauser G, Licul V, Franjic N, Valkovic Zujic P, Milic S. Early nasojejuna mouth in acute pancreatitis: a randomized clinical trial. Pancreatology. 2016;16:523–8. mouth in acute pancreatitis: a randomized clinical trial. Pancreatology. 2016;16:523–8. 25. Wong KKL, Tu JY, Sun Z, Dissanayake DW. Methods in research and development of biomedical devices. Singapore: World Scientific Publishing Co.; 2013. i 26. Wong KKL, Hui SCN. Ethical principles and standards for the conduct of biomedical research and publication. Aus- tralas Phys Eng Sci Med. 2015;38(3):377–80. y g 27. Wong KKL, Chu WCW. Ethics policies and procedures in imaging and interventional radiology. Australas Phys Eng Sci Med. 2015;38(2):375–6. y g 27. Wong KKL, Chu WCW. Ethics policies and procedures in imaging and interventional radiology. Australas Phys Eng Sci Med. 2015;38(2):375–6. • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to ﬁnd the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit Submit your next manuscript to BioMed Central and we will help you at every step:
W2753124020.txt	http://journal.umy.ac.id/index.php/FTL/article/download/3074/4	en		The Challenges and Positive Effects in Implementing Strategies in Teaching Tenses	Journal of foreign language teaching & learning/Journal of foreign language teaching and learning	2,018	cc-by-sa	5,681	JOURNAL OF FOREIGN LANGUAGE TEACHING & LEARNING Volume 3, No. 1, January 2018 Maryam Sorohiti graduated from English Departmeny, Faculty of Le�ers, Dadjah Mada University in 1994 for her undergraduate degree. She accomplished her Master’s Degree in TESL at interna�onal Islamic University Malaysia in 2005. Since 2010, she has been teaching at English Educa�on Department of Faculty of language Educa�on, Universitas Muhammadiyah Yogyakarta. She can be reached at maryamsorohi�@umy.ac.id. The Challenges and Positive Effects in Implementing Strategies in Teaching Tenses Milla Farrihatul Ahna was born in Jepara in 1995. She received her Bachelor Degree in English Language Educa�on of Universitas Muhammadiyah Yogyakarta, Indonesia on July 2017. She was an academic staﬀ at the Interna�onal Program of Government Aﬀairs and Administra�on, Universitas Muhammadiyah Yogyakarta. She had par�cipated in the Interna�onal Seminar on Development Studies 2018 in Malaysia, presen�ng her research en�tled “Development Studies: Students’ Par�cipa�on in the Use of Edmodo in Learning Pla�orm”. She is now self-employed. Her research interests involve grammar, teaching strategy, development studies, and teaching policy. 30-41 ABSTRACT Implementing certain strategies in teaching tenses often brings certain challenges as well as positive effects. This research aimed to explore the challenges faced by English teachers and the positive effects they experienced in implementing strategies they chose in teaching tenses based on teachers’ perception. This qualitative research was conducted at a Language Training Center (LTC) of a private university in Yogyakarta, Indonesia. Interviews were administered to three female and one male English teachers with two to seven years of teaching experience to explore their experiences in implementing strategies in teaching tenses. The findings revealed that technical problems, unsuccessful responses, time constrain for preparing the strategies and proper material selection were the challenges the teachers faced in implementing the strategies. Meanwhile, the increased students’ motivation, improved students’ attitude and understanding towards tenses, accomplished teachers as well as student’s awareness of the importance of tenses were claimed as the positive effects of implementing the strategies in teaching tenses. Keywords: tenses, teaching strategies, teaching challenges, positive effects 31 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 INTRODUCTION Lack of various teachers’ strategies in teaching tenses may lead to the lack of students’ mastery of tenses. According to Vijaya and Viswananth (2010), when teachers do not apply appropriate ways to facilitate students in learning tenses, the students will likely get confused to use proper tenses and mix up the tenses in the wrong way. Their confusion is potential to fail them to master tenses. Therefore, a certain strategy to make students understand tenses better is needed in teaching tenses. Iskandarwassid and Sunendar (2009) stated that a strategy is a useful way used by teachers to facilitate students in teaching process and ease the teachers to develop students’ competence in learning process. They further explained that related to teaching tenses, a strategy refers to ways of teaching chosen by a teacher to deliver tenses materials to be meaningful and understandable. They are purposefully selected or determined by teachers to make their students understand the materials. Therefore, when teachers are able to determine appropriate strategies in teaching tenses, students will more likely understand the materials better. In implementing certain strategies in teaching tenses, teachers might experience choosing strategies that at first are considered appropriate, but then turn out to be inappropriate ones. The strategies purposefully chosen and planned to facilitate the teaching and learning process might not work effectively as well as efficiently as expected. In other words, English teachers may face various challenges. As an example, Ludescher (2006) illustrated that when teachers were explaining the tenses materials by using videos, it was found that not all students were interested in watching the video. This means, when teachers teach tenses using videos, there is a challenge when the students do not pay attention to the videos and care to what the teachers are explaining or reviewing. Similarly, when teachers are trying to use more various technologies to teach tenses, instead of driving students’ excitement to learn tenses, using technology is potential to bring problems in the classroom. Moreover, there are always students who do not follow the teachers’ instruction (Young, 2008). Therefore, a careful anticipation of the challenges likely to happen when applying certain strategies in teaching tenses is required. Despite the challenges, the implementation of teaching strategies can bring positive effects to both teachers and students. As an example, the implementation of teaching strategy of Inquiry-Based Learning provides advantages to students and teachers (Guido, 2017). For students, it promotes a deeper understanding of content, makes learning rewarding and builds initiative and direction. It also benefits teachers as it offers differentiated instruction as well as can be applied in any classrooms regardless of students’ different skill levels. Another example can be seen in the implementation of a contextual teaching and learning strategy. According to Ovalna (2010), when English teachers use a contextual teaching and learning strategy for teaching tenses, the teaching process is easier for teachers since students will understand the materials more by seeing the context around them directly. At the same time, students also improve their tenses mastery and their understanding of the current tenses issue through relating it to the real context (Ovalina, 2010). Thus, implementing the strategies in teaching tenses is beneficial for both English teachers and their students. 32 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 Since the challenges and the positive effects in implementing the strategies in teaching tenses are closely related, English teachers need to be aware of the possible challenges which lay ahead when they are trying to implement a certain strategy and realize the promising positive effects expected to happen. Based on the researchers’ informal observation at a Language Training Centre in a private university in Yogyakarta, not all English teachers were aware of particular challenges. In addition, not all English teachers in this LTC realized the positive effects of implementing strategies in teaching tenses. As a teachers’ reflection, the teachers’ perception on the challenges and positive effects in the implementation of tenses teaching strategies needed to be explored. Based on the explanations above, the researchers conducted research focusing on the teachers’ perception on the challenges and positive effects the teachers experienced when implementing certain strategies they chose in teaching tenses. Understanding the challenges can make teachers anticipate the problems and negative responses that may appear during the implementation, so the teacher can prepare more solutions. Meanwhile, realizing the positive effects the teachers experienced during the implementation may motivate them to apply the same strategies in different occasion. Therefore, the findings of this research can be used especially by novice teachers in taking some consideration and making decision in their teaching practices LITERATURE REVIEW Teaching Method Most of teachers think about how to teach their students using the best way to facilitate and succeed the teaching and learning process. A lot of teaching procedures appropriate to students’ need, aims, and students’ learning style are implemented. For example, when the students want to learn tenses well and to write using correct grammar, the teachers may teach them using structural approach. If English teachers want to focus on teaching tenses, they use a structure approach as this approach discusses the grammatical structure or is grammar-based. The structure approach can be suitable for students, since the teachers can facilitate them to focus on learning tenses. Based on Geyser (2006), a structural approach focuses on “grammatical structures and vocabulary items that will form the primary focus of English language instruction” (p.33). Geyser (2006) said that the characteristics of a structural approach are teacher-centered, grammar-based, lots of drill, and controlled and predictable learning. The example of this approach is an “Audio-Lingual Method” (Geyser, 2006, p.33). Meanwhile, if students want to speak English fluently in daily life, a communicative approach might be the most beneficial approach. Geyser (2006) stated that a communicative approach focuses on “meaningful communication to be the primary focus of language instruction” (p.33). Also, Geyser stated that the characteristics of a communicative approach are “students-centered, communication based, lots of students’ interaction (pairs, groups, whole class), and variable rate of acquisition” (p.33). Total Physical Response (TPR) is an example of a communicative approach. From determining the teaching method used, English teachers might develop the strategy in teaching tenses. . 33 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 Strategies in Teaching Tenses. In applying a teaching method, English teachers implement teaching strategies to make teaching and learning process successful. For example, English teachers can teach tenses using various teaching media, such as texts, songs, poems, stories, games, audios, and videos (Hayat, 2011; Ludescher, 2006; Robertson & Acklam, 2000; Witchukriangkai, 2011; Yassaei, 2012). An illustration of teaching tenses using text can be seen in the classroom in which the teacher gives her/his students texts containing past tense and past progressive tense. The students are asked to identify the grammar rules and the sentence structure in the texts. Meanwhile, Hayat (2011) and Ludescher (2006) said that English teachers can use songs, poems, stories, and games to teach tenses to gain students’ interest. By using songs, poems, stories, and games, the teachers can create classroom activities, such as singing songs, reciting poems, retelling stories, and playing games. Those activities are specially designed by the teachers to facilitate their students to learn tenses. The teachers look for the songs, poems, and stories that contain the tenses they are going to teach that day. Hayat (2011) stated that the teachers also use games such as scrabble word games that involve the students to indirectly apply the proper tenses. Besides, the teachers use audios and videos when teaching tenses (Robertson & Acklam, 2000 and Yassaei, 2012). The students are asked to listen to the audios and to watch the videos. Then, they have to analyze the tenses used. In addition, the teachers use technology such as the internet and Microsoft Office (Wang, 2012). Teachers present the tenses materials with the support of technology, such as using Microsoft Power Point to deliver the teaching materials. The teachers also use social networking platform like Edmodo when holding virtual class. All of those teaching strategies implemented are to make the students understand the materials and to make the teachers deliver the materials more easily. The Challenges in Implementing the Strategies in Teaching Tenses It is common for English teachers to face challenges in implementing certain strategies in teaching tenses. The challenges faced by the teachers in implementing the strategies in teaching tenses may contribute to the unsuccessful responses. Students might not give expected good responses on the teaching strategies (Ludescher, 2006; Witchukriangkrai, 2011; Young, 2008). They might not follow the procedures and pay attention to the teachers. When the teachers teach present tenses using videos or movie clips to get the students’ interest, the teachers expect that the students enjoy the videos and can learn the tenses used in the videos or movie clips, so they can apply the use of tenses into real context. However, there is possibility that some of the students do not pay attention to the videos or movie clips and they talk to other students instead. The next possible challenge can occur when the teachers explain the tenses materials using Microsoft Power Point, some students might not pay attention and might not follow the teachers’ instructions. Furthermore, selecting proper materials for preparing the strategies may also become a potential challenge in implementing the strategies in teaching tenses. The teachers are required to consider the content, students’ need, and students’ level when choosing the materials (Solak & Bayar, 2015). It becomes teachers’ concern to see the content first whether the materials presented for example in the videos or movie clips are appropriate to the students’ level, students’ need, and also the students’ 34 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 background before playing them in the classroom. Another potential challenge in implementing the strategies in teaching tenses is that preparing the strategy takes time (Ovalina, 2010). In a different time, when the teachers use games for teaching tenses, they have to provide more time to prepare the equipment, search the appropriate tenses materials, and design the concept (Hayat, 2011). The Positive Effects in Implementing the Strategies in Teaching Tenses Although there are challenges in implementing the strategies in teaching tenses, based on teachers’ perception, the strategies give positive effects to teachers as well as students. Ludescher (2006) argued that there is an improvement of students’ motivation and students’ attitude in learning tenses. Ludescher added that the students enjoy the class, feel happy, feel interested, and are participative. Those reactions occur when the teachers implement the use of selected songs, stories, and poems which are interesting. The procedure involves the students being asked to pay attention to songs, stories, or poems and then analyze the tenses used in those songs, stories, or poems. After that, they are asked to apply the tenses in their own sentences. Ludescher further asserted that when the teachers teach tenses using the interesting strategies, such as using songs, stories, and poems specifically chosen based on the students’ favorite, the students will feel unthreatened since the teachers wrap the classroom activities interestingly. In addition, by implementing teaching strategies, students get better understanding of tenses (Ovalina, 2010; Witchukriangkrai, 2011; Xin (2010). In explaining the concept of present tense, for example, the use of pictures will ease the students in understanding the concept. The pictures showing someone’s activity in the morning will help the students in describing the activities using present tense. The pictures help the students understand the context of the tense used. Another important thing is that successful strategies in teaching tenses give positive feelings to the teachers in teaching process (Ovalina, 2010). The teachers feel that they are successful in their teaching because they see the students understand what they have taught and explained. METHODOLOGY In conducting the research, the researchers used a qualitative research method. This research was conducted to discover the challenges and the positive effects in implementing the strategies in teaching tenses based on teachers’ perception. This research included the description of the participants’ view through words. The research design was the descriptive qualitative research. Through descriptive qualitative research, the researchers explained the findings related to the previous studies. The research was conducted at a Language Training Center (LTC) of a private university in Yogyakarta. This Language Training Center was selected because teaching tenses is included in the teaching materials set by the institution. The English teachers in this language centre teach tenses every semester. Moreover, in this LTC the teachers teach university students from various majors except English language education major. Therefore, the challenges and the positive effects in implementing the strategies in teaching tenses might variously occur. The researchers selected four English teachers of LTC. The criteria to be the participants were that they had experience in teaching tenses at least for two years, so they really knew the context of the chal 35 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 lenges and the positive effects in implementing the strategies. The four participants were three female English teachers and one male English teacher. All of them have experienced in teaching English at LTC for two to seven years. Two of the participants teach tenses in the basic level and two of them teach tenses in the intermediate level. The researchers used an interview as the data collection method. Before doing the interview, the researchers involved expert judgment to validate the interview guideline as the instrument. Based on the expert judgement, there was no significant revison. The points of the interview guideline cover the questions asking the participants what strategies they used in teaching tenses, the challenges faced in implementing the strategies, and the positive effects they experienced when implementing the strategies. The researchers used Bahasa Indonesia in conducting the interview to prevent misunderstanding. The researchers interviewed the participants face to face. Member checking was administered to ensure what was transcribed from the recording was what the participants said during the interview. The result of the member checking was that all participants agreed that the data transcribed were exactly the same as they stated in the interview, meaning there was no manipulation of the data. The strategies in teaching tenses In the teaching and learning process at LTC, all the participants applied certain strategies in teaching tenses. The strategies they used included the use of technology, traditional drills, texts, stories, songs, poems, group work and giving exercises. The first participant mentioned, “I use power point to explain the materials.” The similar answers were also revealed by the other three participants. In addition, participant two mentioned, “Sometimes, I use online exercises,” while participant three said, “I once used video from Funny English website”. A participant also mentioned, “I applied physical activities (a game).” These applied strategies in teaching tenses were perceived by the participants to have brought both the challenges and positive effects. The challenges in implementing strategies in teaching tenses The first research question was “what are the challenges faced by English teachers when implementing the strategies in teaching tenses?” Based on the participants’ experience in teaching tenses, it was revealed that the challenges they faced were related to several conditions. They included technical problem, unsuccessful responses, managing time for preparing the strategies, and selecting the materials. FINDINGS Finding 1: Technical problem In this part, the researcher provides the findings of the research. In addition, the researcher also presents the discussion supporting the findings of the research. The findings reveal the teachers’ perception on the challenges and positive effects from the implementation of teachers’ strategies in teaching tenses. The strategies involving technology used by teachers to support their teaching brought some challenges to the teachers. The source of the challenges came from the laptop and the internet. This challenge was experienced by two participants. The first participant argued, “My laptop suddenly could not work (when I was teaching).” In 36 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 addition, the second participant said, “The students sometimes cannot access the internet.” This challenge sometimes arises although the teachers have already prepared to use computer or the internet. There was a time when the teacher’s laptop was broken while the materials needed were in the laptop. When the teachers showed the movie clip or the video relating to tenses used, the laptop suddenly could not work, so it could offend the teaching process. Another challenge was there was no internet connection when needed. The students could not access the internet for online teaching and learning platform. The finding was in line with the Brändström’s study. One of Brändström’s findings was about disadvantages of using technology in teaching tenses. Brändström (2011) claimed that the technical problem was a challenge in using technology, especially when the teachers want to use technology as their strategy in teaching tenses. Sometimes, the teachers had lack of internet access. When teachers used the internet to teach students in the class, there was no internet connection. When students were asked to search for materials, they did not have internet connection. Moreover, Wang (2012) stated that sometimes, English teachers got a problem with their electronic files while they were teaching tenses in the classroom. Wang’s statement was proven in the teacher’s experience in this research when the teacher presented the past tenses using Microsoft power point, in the middle of presentation the teacher’s laptop could not work and the learning audio played did not work. Finding 2: Unsuccessful responses Ideally, English teachers used strategies to teach tenses in order to succeed their aims and objectives. However, the strate- gies sometimes did not get successful responses from students. This finding was mentioned by three of four participants. The second participant argued, “Sometimes, we do pair work or group work to discuss tenses, but it did not work well because the students were not active or their level of vocabulary acquisition was low.” In addition, the second participant stated, “When I applied physical activities (a game), the class became noisy.” Moreover, the fourth participant also mentioned, “When the videos were played, there were some students who still talked to their friends.” From the participants’ statements, the challenge faced by the teachers was that the teacher received unsuccessful responses, such as the students refused to participate in the group discussion; the students did not follow the teachers’ instruction when doing a game; the students did not pay attention either when the teachers showed the learning videos of tenses. According to Ludescher (2006), some students were not interested in following the strategies being applied by the teacher; they did not pay attention to the teacher. They preferred to talk to their other friends to pay attention to the teacher’s explanation. Those Ludescher’s statements were reflected in the findings of this research which revealed the similar situation in which the students did not respond well as expected as seen when the teachers conducted group work, applied a game, and played a video. It showed that the teachers received unsuccessful responses from the students. Witchukriangkrai ( 2011) mentioned that the cause of unsuccessful response from students was that there were some low learners who could not follow up other fast learners, so they became demotivated and they were not interested to participate in classroom activities. Witchukriangkrai also added that since 37 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 students who were low learners were not able to follow the classroom activity, they distracted the attention. For example, they made the class become noisy and they talked to their other friends. Finding 3: Time Constrain for preparing the strategies When implementing the particular strategies in teaching tenses, the English teachers needed more time for preparation, but sometimes the teachers did not have much time to prepare everything. This challenge was claimed by three of four participants. The third participant admitted, “Since I must prepare the tenses materials that are appropriate with different students’ characteristics, it takes time.” In addition, the second participant mentioned, “When using Edmodo, it needs much time to prepare, so I sometimes have no time (for preparing).” The fourth participant also had the same statement. She said, “I must spend extra time to look for materials or video that is appropriate with the considerations.” The finding justified that when preparing the strategies to teach tenses the teachers needed much time. Ovalina (2010) revealed that teachers required much time for preparation. For example, teachers had to prepare the lesson plan and the materials which were appropriate with different characteristics of students. Moreover, according to Young (2008), the challenge of using technology was that “many teachers found it hard to find time in their overloaded work schedules to attend courses and to practice the new skill” (p.37). Finding 4: Considering proper materials The researcher found out that English teachers should consider the context when selecting the materials. It was pointed out by the fourth participant. She stated, “When we use the movie segment clips, we must choose the movie clips carefully whether the content and the tenses materials are appropriate to be used in our university because the (western) movie clips are culturally different from us.” It means that the teachers were required to think about the content of the materials in teaching tenses. Moreover, the level of the materials whether it was appropriate with the students as well as the institution became the teachers’ concern in teaching tenses. In other words, it also became the challenge faced by the teacher, since they need to consider the students’ level in addition to the content. This finding was in line with Solak and Bayar (2015). They gave suggestion that “the materials such as course book, video, and internet should be chosen carefully according to the students’ interest and level” (p.114). This means that teachers must determine whether the materials fitted students’ language level or not because it might be difficult for them. Thus, there was a consideration when selecting proper materials and it was challenging. The positive effects in implementing the strategies Based on the teachers’ perception, during the implementation of certain strategies in teaching tenses, such as using text, stories, poems, pictures, novels, games, videos/audios, and technology, the teachers and the students got positive effects. In this part, the researcher provided five findings of the positive effects from implementing the strategies. The positive effects were related to students’ motivation, students’ 38 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 attitude, students’ understanding, teachers’ self-accomplishment, and significance for students. Finding 1: The increase of students’ motivation to learn tenses The finding showed that students’ motivation could increase after the English teachers implemented the strategies in teaching tenses, such as using songs and stories. This effect was highlighted by three of four participants. The second participant argued, “I use songs to introduce the verb form, such as the present and past forms.” Further she claimed, “students’ thought about tenses can change, so they do not think that tenses are difficult.” The third participant added, “I use One Republic’s songs and John Meyer’s songs for teaching tenses. Students are happy and excited.” In addition, the fourth participant also mentioned, “If the topic is narrative, I use a story and finally they identify the tenses in the story. Students look interested in learning the materials and they can look for what is required from the task easily.” Those statements proved that students were being motivated. For example, students did not think that tenses were difficult. Students were happy, excited, and interested. The finding showed that the students were being more motivated when the teachers implemented the strategy in teaching tenses. This finding was related to the previous study from Ludescher (2006). Ludescher found out that implementing strategy in teaching tenses, such as using stories, songs, and poems brought positive effect. As an illustration, students enjoyed when teachers used stories as the teaching strategy to teach tenses. The students were interested in analyzing tenses used in the stories. Besides, Ludescher also added that using songs was easy for students to memorize and the songs made the students enjoy their learning tenses process. Students also did not feel threatened since they enjoyed the learning process and they did what they liked. Hence, using stories and songs as one of the strategies in teaching tenses could increase students’ motivation in learning tenses. Finding 2: The improvement of students’ attitude toward tenses Besides, there was also the improvement of students’ attitude toward tenses. This finding was claimed by three of four participants. The second participant said, “All members of the class can participate in doing a task actively.” The third participant added, “I create a game for students to make them not bored and not sleepy.” Moreover, the fourth participant mentioned, “Because the strategy used visual things, the strategy makes students easier to look for verb form based on the event.” From the statements, it seemed that students participated actively; students did not feel bored and sleepy; and students wanted to find out verb form through visual media, like videos and pictures. From the finding above, the students had the improvement of attitude after the teacher applied the strategies. The improvement of the attitudes was that the students participated collaboratively with the teacher since they were not bored and sleepy. It was related to Ludescher (2006) who asserted useful effect of using strategies in teaching tenses. Ludescher stated that the implementation of strategies could encourage all students to participate actively in the classroom. The students were then involved in the whole activities. Another study also mentioned that the students felt interested to follow the teaching and learn 39 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 ing tenses because the teachers used visual tools like a picture and a video to teach tense (Krčelić & Matijević, 2015). Finding 3: The improvement of students’ understanding Students got better understanding after English teachers taught tenses by implementing the strategies in teaching tenses, such as using texts, pictures, and videos. The four participants encountered the finding. The first participant stated, “Students can do self-correction although they are still guided.” Moreover, the second participant said, “They can differentiate simple present and present perfect tense.” The third participant added, “Using reading text, students might understand the context of tenses and how to use the tenses.” Moreover, the fourth participant also mentioned, “The use of a video with subtitle makes them understand the use of the tenses.” It could be concluded that students could use the proper tenses in appropriate context, and they could differentiate between one tense and the other tenses. The finding showed that the strategies created the improvement of the students’ understanding. The improvement of the students’ understanding means that the students’ knowledge about tenses improved. Based on the teachers’ perception, they could understand how to use tenses properly by considering the time needed, such as present, past, and future. The finding was related to Krčelić and Matijević (2015). They highlighted that those visual tools, such as pictures, video, and power point, could be used to give a better understanding of tenses explained. They also mentioned that it helped to lighten the materials. After teaching tenses through using various teaching media, students could use the proper tenses in appropriate context (Ludescher, 2006; Ovalina, 2010; Witchukriangkrai, 2011). Finding 4: Teachers’ self-accomplishment English teachers also felt the positive effects after implementing the strategies, such as using movie clips, contextual teaching learning, and games. It was stated by three of four participants. The first participant admitted, “When I explained tenses to students and they understood, I felt satisfied or got self-accomplishment.” The third participant also added, “Sometimes, from the activities, I find out new ideas, so it improves my teaching development.” Moreover, the fourth participant mentioned, “There are many materials that discuss tenses in movie clips, so it is easy for me to teach tenses.” The positive effects were that the teachers felt satisfied, and their teaching development improved. Ovalina (2010) argued that the strength of contextual teaching learning strategy was that teachers taught tenses to be easier in achieving the goals of teaching process. That means that the teachers could develop their teaching, because they tried to learn more for preparing the materials or finding new strategies used to teach. Finding 5: Student’s awareness of the importance of tenses Based on the teachers’ point of view, when students could understand how to use tenses correctly, they might use the proper tenses in the daily conversation and it is useful for developing their speaking skill. This finding was highlighted by the first participant. She stated, “If students can use the correct tenses in appropriate context, it will be significant for them.” 40 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 The finding was related to Ovalina (2010). Ovalina pointed out that students could improve their communication skill and enhance their understanding about current issues that were related to their lives. Ludescher also added that using authentic text could show how the item was used in real context. Thus, it was significant for the students. CONCLUSION Based on the findings, the certain strategies applied by the LTC teachers in teaching tenses brought both the challenges and the positive effects. The challenges in implementing the strategies in teaching tenses were related to technical problem, unsuccessful responses, time constrain for preparing the strategies and proper materials selection. Meanwhile, the positive effects were the improvement of students’ motivation, the improvement of students’ attitude toward tenses, the improvement of students’ understanding, teachers’ self-accomplishment, and significance for students. As the findings revealed the challenges dealing with technical problems as well as non-technical problems, the teachers are accordingly required to manage their time wisely to prepare the strategies in teaching tenses especially when using games and using technology. Teachers also need to consider students’ need, students’ level, and students’ context when implementing the strategies in teaching tenses, so that the positive effects can be successfully achieved. The teachers’ skills in dealing with the challenges are accordingly needed to gain positive effects deriving from the implementation of teaching tenses. REFERENCES Brändström, C. (2011). Using the internet in education-strengths and weaknesses: A qualitative study of teachers’ opinions on the use of the internet planning and instruction. University of Gavle, Gavle. Retrieved from www.diva-portal.org Creswell, J. W. (2012). Educational research: Planning, conducting, and evaluating quantitative and qualitative research. Boston: Pearson. Geyser, J.P. (2006). English to the world: Teaching methodology made easy. Malaysia: Sdn Bhd Hayat, E. N. (2011). Teaching the present perfect tense by using game (Find Someone Who). Retrieved from http://repository.uinjkt.ac.id. Iskandarwassid & Sunendar, D. (2009). Strategi pembelajaran bahasa [Language teaching strategy]. Bandung: PT. Remaja Rosdakarya. Krčelić, P., & Matijević, A. S. (2015). A pictures and a thousand words: Visual tools in ELT. The International Language Conference, 110-114. Retrieved from http://bib.irb.hr. Ludescher, F. (2006). How to teach grammar. Viktorsberg: International Modul. Ovalina. (2010). Improving students' ability in using the simple past tense through contextual teaching learning. Universitas Islam Negeri Syarif, Hidayatullah, Yogyakarta, Indonesia. Robertson, C., & Acklam, R. (2000). Action plan for teachers. London: British Broadcasting Corporation. Solak, E., & Bayar, A. (2015). Current challenges in English language learning in Turkish EFL context. Participatory Educational Research (PER), 2(1), 106-115. Retrieved from http://dx.doi.org/10.17275/per.15.09.2.1. 41 Journal of Foreign Language Teaching & Learning Vol.3 No. 1, January 2018 Vijaya, K., & Viswanath, L. (2010). The problems of teaching/learning tenses. Language in India: Strength for Today and Bright Hope for Tomorrow, 10, 96-101. Retrieved from www.languageinindia.com Wang, P. (2012). Teaching and learning English verb tenses in a Taiwanese University. English Linguistics Research, 1 (1), 18-34. doi:10.5430/elr.v1n1p18. Witchukriangkrai, T. (2011). Teaching tenses through text and drills: Comparison of text sores. Language Testing in Asia , 1(4), 120-143. Yassaei, S. (2012). Using original video and sound effects to teach English. English Teaching Forum, (1) 12-16. Retrieved from www.eric.ed.gov. Young, R. (2008). Using technology tools in the public school classroom. University of Wisconsin-Stout, Menomonie, United State of America. Retrieved from https://minds.wisconsin.edu.
W2010075709.txt	https://zenodo.org/records/2507537/files/article.pdf	de		�ber die Konturen optischer Bilder	Naturwissenschaften	1,921	public-domain	2,248	liett 4~ I M i t t e i l u n g e n aus v e r s c h i e d e n e n Gebioten. 2. 12. 1921J eion benutzt~r P r i s m e n gesteUt werdon, l~ube~m und Miche~ ~nbschieden sieh d~her ftfr die /Lttfnahme van Isochr.omaten des sel~warsen K~rpers, un~ z ~ a r bei Wel.len]~Lngen von 4, 5, 7, 9, 12, 16, ~2 u ~ 52 ~ u~cl 0el Temperaturen zwisehen tier der ftfissigen Luft und 1400 o C. Sie waren a~f d4ese Weise ~m~tande, eipe P~rlif~ung des StrahIung~g~setzes in 4era ga~a~en, nach Nerns$ und Wulf von tier a-Korrektion ibetroffene~ Bereieh yon x-Werten vorzunehmen. Die We~llenl~ng~ yon 22 un~ 52 p~ wurden als Reststrahlen yon FtuSspath n n d Stein~alz, d,ie kfirzeren VCelMnl~Lngen 4~re~l prismatisehe Zer~legung mittels Prismen ~us Flul~path, Steinsalz oder Syl~'in her,gestellt. ~l.s Strahhmg~quellen d.ienten, je naeh dem Temperaturbere~eh, vier schw~rze Kfirper verschiedener Konstr,uktion. D~e E~nergiemessung ges~h~h ,mit einem M,ikror~diometer, die Temper~turmezsuag der schwarz~n KSrper m i t einem Widerst~ndsthermometer bzw. emg~b~tten Tl~ermoelementen, welche~von der P. T. R. a ~ f da~ gena~esbe geeie~t w~rd~n w~re~. Auf F A n - - t e n der Versueksanor4aung, wele~e sieh gxam~st~zlieh in kei~ae2 W e L ~ ~on ~ yon /gube~a ~oersits frfi~tmr be~ nutz~en untemeheide~, k~mn hier nieh¢ eingega~g~n werden. SelbsC.verstgn4lieh i st, dab Mle n ~ r denkb a r e n Fehterqttellen, ~n& eeien sie noeh so geringffigig, ~ufgesucht un~ auag~schaltet wurden, z. B. Proportionalit~tsa~bweichungen ~er Ausschl~ge des M.ikroracli~meters, ,Schw~nkungen der Empf, indliehkeit tier MeBanordnun~ (z. B. in~o~ge van Anderu~g der Ab~ ~orption der Strah'l~ng in de~ Zimmerluft), spektrale Unreinhei¢ 6er ~Strahlung, T e m p e r a t u r g e ~ l l $ im Innern des sehwar~en K6rpers, welches eine Korrektion der gemessenen Temper~tur niitig machO, Erw~trmu~g yon Bienden u n 4 Klappschirmen usw. Die Auswertung der MeB~eihen ging .in falgender Wei~e vet eich: Innerh~lb derselben Isochromate mu~ bei Giiltigkeit des Plancksehen Ges~t~es die OrSf~e O = E (e~ - - 1) k o n s t a n t eei n, wenn E der bei der Temperatur T beoba~htet~ Aussch~ag d.e~ Mikror~dtiometers (al~ r e l a tives Energiemai]), x 7-- k e- ~h ist Dagegen mfitlte na~h Nerrwf und Wu~f die Gr6Be C' - - - -C- k o n s t a n t sein, wobei ~ der T elle 1 za entnehmen i~t. Es ergib~ eich u u n bei allen a e h t Isochromaten m i t yeller ~IrAarheit da~ ~isiehe Result~t: Die O-Werte sehwanken in allen Mei}rei.hen inner.h~b tier Fehlergrenzen yon _+ 1,25 ~ vS~tig .unre~elmgl~ig u m einen Mittelw~rt, ohne ei,nen Gun,g m i t w e r k e n a e n s u lassen. (7 ist ~lso in der Tat, wie es das P~anek~che Gesetz verlangt, a ~ k o n s t a n t anzu~ehen. I)age~en zeigen 4ie" C~-Werte a~snahmslos einen, yon tier jeweiligen (~r6g~ ~er ~-Korrektion ~bh~tngigen, st~rken Gang m i t x, w~hren~ ~ie lmi Gfiltigkeit tier ¢t-Korrektion k o n s t ~ n t sein sollt~n. Als Beis~iel sei dn TabetRe 2 ~ie Isochromate 9W (genau 8,994~) w.~e~er.gegeben. Das verset~iedene Verhalten der C- un4 C'-Wer~e, wie es besondem dureh ~ie Abweiehu~gen ~) C u~d ~t ~' veto ~.i~telwert dargestellt ist, ist evi.dent. Das R ~ u l t a t i~¢ also d u e v6llige Be~t~ie~ng des Planeb,ehe~ Ge~ setzee innexhalb (Set he.to erreiehbwren Meflgenaulgk~it. W i r dtirfen uns also &es P l a n c k ~ h e n Strahlungsgesetzes yon neuem, u n ~ noch ~nehr als bisher, als eines a~Berordentlieh fe~t gesieherten Be~itzes uns~rer Wi~sensehaft freuen. Die Kri~ik ~ h a t w ~ t e r e~i~real die sehSns~ ihrer Aufga.ben erfiillt: sta.tt ~xteder- Tabs. 377 476 577 635 678 740 844 923 1034 t126 ~235 1332 t437 1533 1653 981 Ta,belle 2 (Lsochromate 9~). g C ~C Cr 10,72 73,41 ~ 56 69,39 26,62 73,90 ~ 7 69,21 49,62 74,28 + 31 69,36 65,68 74,70 + 73 69,69 78,07 74,73 + 76 69,78 96,54 74,12 + 15 69,53 130,51 73,71 ~ ~6 69,93 158,49 73,75 - - 22 70,57 201,08 74,28 + 31 71,70 235,51 73,88 ~ 9 71,73 279,56 7~,32 + 35 72,51 318,25 73,87 ~ t0 73,31 359,0873,36 - - 61 72,0(} 492,13 73,91 -- 6 72,75 449,41 73,30 - - 67 72,29 5C' - - 153 - - 171 - - 166 --- 123 - - 114 ~ 139 ~ 99 ~ 35 + 78 + 81 + 159 + 239 + 114 + 183 + 137 zurei~en, hat ale $eholfen aufzubatmn u n 4 zu feetige~. Ohne die a m Pl~ncksehen Gese~z ge~bte K r i t i k wttr¢ uns &ie sehiine expe~imentelle RestRtigung d,i.eses G~setzes vermutlich z u n ~ s t nioht geschenkt worden, die h~u¢~ Ms Muster e i n e r : ~ u f das sorgfRlti~ste un& scharfs i n n i g s t e durch~gefiihrten Pr~isionsarJbeit das tIerz je&es Experimentalphysikers erfreuen muB. W. Westphal. t~ber d i e K o ~ t u r e n optische~ mlde~. Be~ allen opti~chen P r~isionsmessungen hat m~n bieher m i t d~r VonsteLl~n~ yon scharfen geometrieehop~i~chen Bildr~ndern ~earbeitet, obgieieh (lie Beu~ungstheori~ l~ngs~ erwiesen h~t, da~ es eine e i ~ n t lithe Begrenzungslinie ~n Ol~i~cl~en Bildern nieh~ gibt~ Es bleibt ~l~o die F r a s e zu b ~ n t w o r t e n na~h dem eilgentllchen W,e~en dessen, ~a~ dem A~ge als B41dbegrenzun~ erscheint ~ eine Fr~ge, (lie schon yon W. ~truvel), 8trehl ~) .un~ ~ d e r s yon H e r ~ 9 s) angesehniC~ten wurde, ~ber fiber ~ie qua~itati.o A n t w o r t ¢lureh Hering8 ttinweie ~uf den ,,Grenzkontraet" hi~nau~ nich~ gef~rdert wevden konnte. B e w e i s e ~ un~ quantitativ fruehtbar lltB~ ~ieh .~ie A n t w o r t er~t g ~ s~a~ten, wenn m~a at~f &ie UnC.ersuchungen Maeh#) trod 8eellgevs~) fiber 4ie phy~iologische ~5~,rk~zig rtLumlich verteilter Liehtreize ~uf der Net~zhau~ zur~Ic~grei~ft. Maeh formuliert seine Experiment~lergebnt~se fiber die von ihm zuerst entdeekten Konbrasterseheinun~en a n stetig verl~aufel~&en L i c h t v e r t e i h ~ g e n dahin, da~ alas Aag~e jede &bweichung tier Lichtstl~rke eines F l g e h e n p u n k ~ s vom Mitte~ ~er ngclmt ~ d o n I n t e n s i t ~ t e n beson&ers her~uahebt, in6em es Stetlen mi~ e i n e r ?3berschu~inten~itgt erheb~ieh ~e21er empfinder ul~ ihrer obj,ek~iven Int~nsititt zukommt, Stellea m i t Intensitltts~nterbila~z dagegen ztt dtmkel sieht. D~ rise Abweiehung d e r Inten~igt J i~n P t m k t e x, y yon ,dem l~i~tel d~r ngchs~ umgebead~n Intensi.t~ten ~proportional d~m Aus&ruck d~ d d~ J A J ( x y ) _-- - d ~ - + d y . ~ ist, den mh i,n diesem Z,~ammenha~g ~I~ Konrat- [unktio~ bezeiehne, so k a n n m~n s~gen, d~B ~ eiaer Inten~itg~svert~ihm$ imm,er d ~ r t hel,ie oder d~mkle t) W. Struve, ~3,ber d. Einfl. d. Diffraktion a n Fernrohren auf Liohtsehei~ben. ~) K. Strehl, Theo~ie des Fernrohrs. a) Her~ng, Grundzfige tier Lehre v. Lieht~inn, Berlin, Sprin~er, 1920, § 32 u. f. ~) Maeh, Die physiM. Wirk~. rltuml, ver teilt~r Li~htreize a~ cL Netzhvut, W i e n e r Sitzung~ber. 1865 hie 1868. ~) Soe~iger, Die Vergr6gerung des Erdsohattens bei M<md~in~i~erni~en, Abh. d. k. b. AlmcL d.. W. Mfinellen 1896. 982 Mitteilungen aus verschiedenen Gebieten. [ Die Natu~[wissensehaften Kontra,~tsi~reifen @esehen werden, ~vo die KontraetH~hese von dem Einfl,ul3 d~r Ern~hrung auf El~to hmktioa besonders starke negative oder positive ~icklung uad Gestaltuag ,d~r Pflanze, B~flt. z. bet. Werte erreieht. In ~ler Vernmtun~g, d~aB al'le geseh,enen Cen~rbl. 38, 1921, A bt. I) beleuchtet Lako~ die optisehen Bil4rttmler ~uf solche Kontrasbstrei'fen zu- C~oethesche ~ e ~ n o r p b ~ e n l ~ r e v o ~ S t a m l p u n k t der rtl~kgehen, babe ich .durch Vergl~eh .der yon geeligera) m o d e r n e n P h y s i o l o g i e . Man h ~ ~ i s h e r ~ n diesem Werke meier in zieml,ich einseitiger Weise die morphoa~g~estellten Mod~llmessungen mit dem Verlauf der d~zu berechneten Kontras~f, unkt~i'onen zungchat festl~gi~che Hypo~hese yon der Metamorphose pflanzLieh~r stel,l.en k~{innen, (la~ Begren~un~ti,nien sowohl mit den Orgome gewiirdigt. In neuerer Zei.~ hat 5ann Hanson Grit- an.el T~llin~en als mit ,d~n N.u.Ri~ien der K~ndar0.ttf ~aing~wiesen, ,&aB ~ Goethesche Schrift auoh trastfunktion zusaxnmenf~l~en /~6~nen. Hieraus and recht be,~chtenswerte kau~lphysiologisehe Gesich~sau's dem ~l,lgemei~en Charak%er der Kontrastfunkt~ion punkte zuv ErklKrung d~er Metamorphose enthglt. Diese an Beuguagehii;4ern folgt, .daB ~ wean ,di~ 'KontrastDingo v~erden yon Lakon elner eingehenden Analyse lin,ien fur ~len Bil~r~n4 ma~geben~ sind - - d,ie A~s- ua~erzogen, and er gelangt 5abei~ zu dem SchluB, da~ m~ssung ei,nes optischan Brides Resultate ergeben kann, ~ieh bei Goethe die Wurzeln ztt go~uz modernen Anwelche je nach 2~tt~be un.d Beobach~tta~sbed'i~ngungen sehauungen tiber .die K~usalitgt d'er Org~nbildung mn bestimmte BetrRge grSl~er oder klei~er eiact als f i~d.en. des geometri,sch-olatisehe BiI&, @e~egentlieh ~ aber auch Wie 5ekannt ist, hat Savhs den ~Staa~ltmnkt vermit ibm ti.bereinstimmen k~nnen. Besonders intertreten, dab fiir die Anl~ge bestimmter Pflanzenorgane ess~ut wird d,ie .Kontra~ttheorie da~'urch, dab aus ~er (Blgtter, Blii,ten usw.) des Vorhandensein spezifischer, numeri~chen Berechnung tier Kontras~f~nktion sehr organ,bildender Stoffe notwe~dig ist, die mtr in genahe ~beiein~nder tiegen,der Bi'kler, wie sie bei ~ikroringen Mengen anwesend zu 6eda br~uchen, also nach mebermessungen vorkommen, hater Um,stgn.&en erhebA r t der Ferment~ wirken. Dliese A~ffassung ist 5ann liche VerlagerunF~n ~er KonCa~stli~tien gegentiber den von Goebel und' Klebs 5ahin m~difiziert worden, dab es iso~ierten Bitdern folgen, w.elehe iza altgemei~en der nicht auf bestimmte Stoffe, vielmehr auf alas Verht~l~Verlagerung tier I~opho~en entgege~gesetzt si~d,. Da his tier Ntth.rs~l,zezu den organischen Substanzen aanun fast alle MJkrometermesstmgen a~f Ein, sCel~lurnig kommt. So bd~ingt nach Goebel bei der r,un<Iblgt~rigen tier Bfldr~tn.d.er beru~en, ~o werden hieraus eine Reihe Gl~cken.blume (Campanula rotun~lifolia) fJbersehuB aza yon lYIessun@sfehlern numer,isel~ ~ ableitbar, deren tatN~hrsal~en die Bildung yon PrimgrbIRttern, Cberseh~B slieh~iches Vorhansansein als Beweis fttr die Kontrastan organischen Su.bstanzen die Bilduag yen h~her 4iftheori,e @elten muB. N~ch den hisher d~rch~erechferenzierten Fol@ebl~ttern. In entsprech~nd'er W ei~e neten Beispi'elen gel~ng die wamer£s~he D~rstellang tier konnte Klebs fllr andere Objekte dartun, ,dab m~t tier yon Aube~ g~gebenen Messu~sreihen tier positi~en relaidve~ Zunahme der organischen St~)ffe ein Oberun.d negaLiven Irradiation an sehma~en heHen un~l gang yon rein veget~tivem Gedei~hen ~ Anl~.~e yon d~antd,e~ Streifen, d,ie Erkl~rtm~ der Unabhgngigkeit La~bbIRttern -- zur Pro~uktion yon BlliVen sta~tfindet. d:er schei~baren Pl~neten,du.rchmesser yon tier 0hjektivIn dleser l~i~hbung ~bewegen sieh nun aueh .die Anii~fnung tier Fer~rohre, des AbetoB~mgsfehlers bei seh~u~ngen Goe6hes. ,,Seiner Betrachttmg llegt der engen visuellen Doppelst~rnen, w.ie des ghn~iehen yon Ge4anke zugrunde, ~laB der. Vegetationspunkt befgl~i~t Kostins~y ent~eckten Fehlers bei ~ahe benaehbarten ist, s~imtlicheBlattformen der Spezies eowie ~lieBltitenphot0~oa~phischen Bil.dern; d~e Ableitu~,g der Differenz teile hervorzubringen, und 5aB die Entsoheidung darzw~i,schen den Monddurchmesserwerten, welche aus fiber, welche Blattform jeweils gelyildet w,ird, yon tier Stern.bedeckung~en am hellen ~a~ ~unklen Mon~dr~nd Besehaffenheit der dem Vegetationslmnkte zus~r~men erhal~en ei~,d~ die Darstell~ag der E rscheinung des den ,S~fte abhgnb~." Goethe sprleht yon ,,wRsserigten" Sehwar~en Tropfens bei Planefe~c~urchgltng~n vor and ,,verfeinerten" Sgften, die wKsser.igten hearken tier Sonnenscheibe ~ en<llieh d~ie Able~ung der Abdie Anlage yon primftiven, 41e vexfeinerten eine solche weichungvn unter den Venusdurchmes,serwerten, welehe yon komplizierteren Blgttern und 5ann yon Bliitenau,s Heli~metermes~un@e~ an ,der hellen Pl~neteno~ganen -- je nach dem Grade der Verfeinerung. Diese scheihe, a~s Messungen an ,der d~mklen vor der genre ,,Verfeinerung" let ,a~f den Einf~u~ yon Licht und Luft stehen~en l~lanefenscheibe uh~l aus Messun~en an der zuriickzufilhren. Es IgBt eich der Nachweis erbr,ingen, nahe bed der Sonne stehenden ~ehmalen Siehel erhal4en d~B 6~oethe hierbei an <lie ,Kohlens~ureassimi~ta~ion, die si~I~). ja ~nter Mitwirkung des Liehts ~nd unter Verwer~ung Damit seheint die Branchb~rkeit der Defirnfflen der Kohlensgure der Luft organisehe Substanz schafft, eptiseher Bil.&b~renztmg auf Gvund, der Kontr~stdenkt. So entspricht ,denn der Gegensatz yon wgsse6heorie hinreieher~d erwiesen und an ,der Notrlgten und verfeinerten SRften offen~l~r ~lem Be~griffswen,~igkeit Earer B extteksiehtigung An der Theerie tier paar: N~hr~salze ~und organische Substanzen. Es lieg~ Prgzi~ionsmess~n~en ist nieht~ gtt zweifeln. Aueh ftir an dem ~amali~en Stande tier For~hung, wenn @ie physi~)Ioglsche Optik seheint sie yon Bedeu~ng, sieh ~bei, Goethe An der weitaren Ausgestaltung d~eses in.sofern als sie ~tts ver.sehiedenen hier nleh.t zu er~ed~nkens einige unklare Momente ei~nsehleiehen, wenn iirter~d,en Griind.en ein ~uB,erst pr~zises Prltfmittel beispielsweise ftir die V.erfeiner~ng ,der ~gfte eine Filfi~.r n~here Erfor schung der Netzhxutf.unktionen dartl'~tion dureh die Gefit~e mit h,erangezogen wird. stell~ A. Kiihl. Sehen wir aber yon diesen historiseh notwendigen UnI n eiaer kurzen Abhandlung (Goethes physiologische zulgngliehkeiten ab,-dann tr.itt uns hier d4e moderne Erkl~rung der Pflanzenmetamorphose als moderne Anselmuung sehon in deutli~h greifbarer Form vet Augenelner yon jenen zahlreiehen F~len, we a) Zahlenangaben A. Ki~hl, Wesen ~L VergnderllichGoethe mit sichevem Takt Folgerungen aus zeitg~niiskeit d. Konturen opt. Bik~er, Vortr. a. d. Vers. d. siechen Ergebniesen gezogen hat, d.ie yon der Wisseninternat. Astr. Ges. Pots., Aug. 1921, Centr. Ztg. f. schaf.t erst vi.el spttter anerkannt worden sind. Optik ~. l~.ech~nik, 1921, Heft 25. Deutsche Opt. Woehenschr. 1921, Heft 36. Filr die Redaktion verantwortlieh : Drl Arno/d B~rl~ner, Berlin V¢ 9. Ver]ag yon Julius Springer in Berlin W 9. -- Druek yon H. S. Hermann & CO. in Berlin SW
https://openalex.org/W3194730144	https://boris.unibe.ch/161696/1/41598_2021_Article_96298.pdf	English	null	Engineering and functional characterization of a proton-driven β-lactam antibiotic translocation module for bionanotechnological applications	Scientific reports	2,021	cc-by	8,380	Engineering and functional characterization of a proton‑driven β‑lactam antibiotic translocation module for bionanotechnological applications OPEN Mirko Stauffer, Zöhre Ucurum, Daniel Harder & Dimitrios Fotiadis* Novel approaches in synthetic biology focus on the bottom-up modular assembly of natural, modified natural or artificial components into molecular systems with functionalities not found in nature. A possible application for such techniques is the bioremediation of natural water sources contaminated with small organic molecules (e.g., drugs and pesticides). A simple molecular system to actively accumulate and degrade pollutants could be a bionanoreactor composed of a liposome or polymersome scaffold combined with energizing- (e.g., light-driven proton pump), transporting- (e.g., proton-driven transporter) and degrading modules (e.g., enzyme). This work focuses on the engineering of a transport module specific for β-lactam antibiotics. We previously solved the crystal structure of a bacterial peptide transporter, which allowed us to improve the affinity for certain β-lactam antibiotics using structure-based mutagenesis combined with a bacterial uptake assay. We were able to identify specific mutations, which enhanced the affinity of the transporter for antibiotics containing certain structural features. Screening of potential compounds allowed for the identification of a β-lactam antibiotic ligand with relatively high affinity. Transport of antibiotics was evaluated using a solid-supported membrane electrophysiology assay. In summary, we have engineered a proton-driven β-lactam antibiotic translocation module, contributing to the growing toolset for bionanotechnological applications. In synthetic biology, components and systems involved in biological processes are optimized or repurposed using engineering approaches to address specific challenges in a wide range of fields such as diagnostics, bio- technology and research1,2. This can be achieved by simplifying and/or manipulating existing biological systems for specific purposes (top-down approach)1,3. In contrast, bottom-up approaches combine natural, modified natural and artificial components with specific functions, called modules, to build completely new biological systems, which mimic natural processes or exhibit functionalities not found in nature1,4,5. One problem com- monly addressed with synthetic biology principles3,5 is the growing threat of contamination of environments by small organic molecules such as drugs and pesticides, which cannot be removed by currently used mechanical or biological wastewater disposal techniques6. Especially antibiotics are a growing body of concern, as they lead to a higher occurrence of antibiotic resistant genes in prokaryotic organisms7,8, which contributes to the formation of multiresistant pathogen strains. A simple molecular system to take up and degrade such pollut- ants could be a bionanoreactor (Fig. 1) made up of a scaffold module (e.g., lipids or block copolymers, Fig. www.nature.com/scientificreports www.nature.com/scientificreports Institute of Biochemistry and Molecular Medicine, University of Bern, 3012 Bern, Switzerland. *email: dimitrios. fotiadis@ibmm.unibe.ch Scientific Reports \| (2021) 11:17205 Engineering and functional characterization of a proton‑driven β‑lactam antibiotic translocation module for bionanotechnological applications OPEN 1, in orange), an energy-providing module (e.g., light-driven proton pumps, Fig. 1, in red), a transporting module (e.g., proton-driven membrane transporters for the target pollutant, Fig. 1, in blue), and a degrading module (e.g., enzymes or chemical catalysts, which are able to degrade the pollutants, Fig. 1, in brown)4. Potential pollutants for a proof-of-concept of such a molecular system represent β-lactam antibiotics. As mentioned before, such a system requires a transport protein to efficiently take up β-lactam antibiotics into bionanoreactors, preferably a proton-driven one to be able to energize it using a light-driven proton pump (Fig. 1). p g g g p p p g A promising transporting module with the previously described requirements would be a peptide transporter from the proton-driven oligopeptide transporter (POT) family9–12 (also known as the peptide transporter family \| https://doi.org/10.1038/s41598-021-96298-4 Scientific Reports \| (2021) 11:17205 www.nature.com/scientificreports/ Figure 1. Schematic representation of a molecular system for the uptake and degradation of small organic molecules. A light-driven proton pump (red) is activated by light (h.v) generating a proton gradient (indicated by different sizes of [H+]) across a vesicular structure (e.g., liposome, in orange). A proton-driven transporter (in blue) accumulates the target substrate (S), e.g., a β-lactam antibiotic, inside the vesicle using the established proton gradient. Enzymes (in brown) entrapped inside the vesicle degrade the substrate to a non-active compound (P). Depicted modules are based on structures of proteorhodopsin (in red, PDB ID code: 2L6X), YePEPT (in blue, PDB ID code: 4W6V) and TEM1 β-lactamase (in brown, PDB ID code: 1BTL). Figure 1. Schematic representation of a molecular system for the uptake and degradation of small organic molecules. A light-driven proton pump (red) is activated by light (h.v) generating a proton gradient (indicated by different sizes of [H+]) across a vesicular structure (e.g., liposome, in orange). A proton-driven transporter (in blue) accumulates the target substrate (S), e.g., a β-lactam antibiotic, inside the vesicle using the established proton gradient. Enzymes (in brown) entrapped inside the vesicle degrade the substrate to a non-active compound (P). Depicted modules are based on structures of proteorhodopsin (in red, PDB ID code: 2L6X), YePEPT (in blue, PDB ID code: 4W6V) and TEM1 β-lactamase (in brown, PDB ID code: 1BTL). (PTR)10,13), which belong to the major facilitator superfamily (MFS)14,15. They are secondary active symporters16,17 and use the inwardly-directed electrochemical gradient to actively transport and accumulate their substrates across cellular membranes11. Engineering and functional characterization of a proton‑driven β‑lactam antibiotic translocation module for bionanotechnological applications OPEN In human metabolism, the high-capacity, low-affinity peptide transporter PEPT1 (SLC15A1) is responsible for the uptake of dietary di- and tripeptides by epithelial cells in the small intestine, while the low-capacity, high-affinity peptide transporter PEPT2 (SLC15A2) reabsorbs di- and tripeptides in the kidney18. Beside those metabolic functions, both transporters and bacterial homologues were shown to also bind and transport a wide range of peptidomimetic drugs and prodrugs19 such as (β-lactam) antibiotics20–25 (e.g., cefadroxil, chloramphenicol), antivirals24,26–28 (e.g., valaciclovir, valganciclovir), protease inhibitors24,29 (e.g., bestatin) and Parkinson medications24,26 (e.g., l-dopa-l-Phe, L-dopa).fi g p p To be suitable for application in a bioremediation system, a high-affinity transporter like PEPT2 is needed to be able to take up the target molecule at low concentrations. In addition, components to be used in bionano- technological applications, have to show good thermal stability2 and have to be producible in large quantities, i.e., in milligram amounts. Unfortunately, human proteins often show poor thermal stability and low yields when expressed in eukaryotic expression systems. Production using conventional bacterial expression systems is oftentimes not possible due to the lack of complex post-translational modifications needed for proper function of eukaryotic proteins30. Furthermore, reconstitution of human membrane proteins into liposomes and polymer- somes is challenging, because of poor stability when solubilized in detergents and potential requirements to the membrane environment (i.e., specific phospholipid and sterol composition31). In contrast, bacterial homologues are more suitable for such applications, as they are easier to overexpress, more stable and less demanding in terms of lipid environment. We previously solved the ligand-free inward-facing crystal structure of the prokaryotic proton-driven peptide transporter YePEPT from Yersinia enterocolitica at 3 Å resolution32. Furthermore, we established a bacterial uptake assay to evaluate the specificity and affinity of YePEPT for different compounds by transport inhibition of the reporter radioligand [3H]Ala-Ala, as well as a solid-supported membrane (SSM) electrophysiology assay to assess transport of potential substrates.h p y gy y p p This work focuses on the development of a molecular module from YePEPT, which is able to transport certain β-lactam antibiotics for application in a bioremediation system as described above (Fig. 1) and as a contribu- tion for the growing toolset of modules, which can be used for molecular systems engineering. By combin- ing structure-based mutagenesis with functional characterization, we were able to identify specific mutations, which enhance the affinity of YePEPT for certain β-lactam antibiotics into the low micromolar range. Results and discussion β l ibi i if β‑lactam antibiotic specificity determination of YePEPTWT. To determine the β-lactam antibiotic specificity of wild-type YePEPT (YePEPTWT), uptake inhibition of the reporter radioligand [3H]Ala-Ala by a set of six common and structurally diverse β-lactam antibiotics (Fig. 2a) at a concentration of 5 mM was evalu- ated in Escherichia coli cells overexpressing the transporter (Fig. 2b). This set consists of four penicillin- and two cephalosporin antibiotics. For the group of aminopenicillins (penicillins containing an α-amino group) the prototypical ampicillin (Fig. 2a 1) and its analogue amoxicillin (Fig. 2a 2), which contains a hydroxyphenyl- instead of a phenyl group, were selected. To evaluate the influence of the α-amino group, which corresponds to the N-terminal amino group of peptide substrates33, penicillin G (Fig. 2a 3) and carbenicillin (Fig. 2a 4) were included. In those compounds, the α-amino group is substituted with a hydrogen atom (penicillin G) or a carboxyl group (carbenicillin), respectively. All penicillins contain a penam ring structure consisting of a four- membered β-lactam ring and a five-membered thiazolidine ring fused together. Cephalosporins, in contrast, contain a cephem ring structure, in which the four-membered β-lactam ring is fused with a six membered thiazine ring. To evaluate the influence of those core-structures, cefalexin (Fig. 2a 5) and cefadroxil (Fig. 2a 6) were included, which are the cephalosporin analogues of ampicillin and amoxicillin, respectively. In addition to the selected antibiotics, a positive control without competitor and a negative control with unlabelled Ala-Ala as competitor were included in every screen. A time course analysis of [3H]Ala-Ala uptake by YePEPTWT was conducted to ensure that all functional data is acquired in the linear regime (Supplementary Fig. S1). Th b h b fil f TWT ( b) h d [3 ] l l k h b f d q g pp y g The antibiotic inhibition profile of YePEPTWT (Fig. 2b) showed an [3H]Ala-Ala uptake inhibition of around 50% for both aminopenicillins, with a slight preference for a phenyl- (i.e., ampicillin) over a hydroxyphenyl group (i.e., amoxicillin). Both non-aminopenicillins (i.e., penicillin G and carbenicillin) did not show any inhibition at all, indicating that the α-amino group is necessary for binding to the protein. The aminocephalosporins showed only a weak uptake inhibition, which points to a preference of a penam- over a cephem core structure. In con- trast to the aminopenicillins, a slight preference for a hydroxyphenyl- (i.e., cefadroxil) over a phenyl group (i.e., cefalexin) could be observed. Engineering and functional characterization of a proton‑driven β‑lactam antibiotic translocation module for bionanotechnological applications OPEN Using electrophysiology, we could demonstrate transport of identified compounds and therefore the suitability of the engineered YePEPT version as a β-lactam antibiotic translocation module. Scientific Reports \| (2021) 11:17205 \| https://doi.org/10.1038/s41598-021-96298-4 www.nature.com/scientificreports/ Figure 2. Specificity of YePEPT variants for β-lactam antibiotics and dipeptides. (a) Molecular structures of the screened β-lactam antibiotics. Specificity of (b) YePEPTWT, (c) YePEPTK314A and (d) YePEPTK314A-F311Y for selected penicillins and cephalosporins by competition assay (5 mM final concentration). e Specificity of YePEPTK314A-F311Y for selected dipeptides by competition assay (2.5 mM final concentration). Bars in (b)–(e) represent vector-subtracted uptake of [3H]Ala-Ala normalized to the uninhibited signal ± SEM of at least three independent experiments, each at least in triplicate. Figure 2. Specificity of YePEPT variants for β-lactam antibiotics and dipeptides. (a) Molecular structures of the screened β-lactam antibiotics. Specificity of (b) YePEPTWT, (c) YePEPTK314A and (d) YePEPTK314A-F311Y for selected penicillins and cephalosporins by competition assay (5 mM final concentration). e Specificity of YePEPTK314A-F311Y for selected dipeptides by competition assay (2.5 mM final concentration). Bars in (b)–(e) represent vector-subtracted uptake of [3H]Ala-Ala normalized to the uninhibited signal ± SEM of at least three independent experiments, each at least in triplicate. Results and discussion β l ibi i if The N- and C-terminal bundles of the YePEPTWT structure are coloured in green and blue, respectively. The Asp-Ala dipeptide was modelled into the binding pocket of YePEPTWT (PDB ID code: 4W6V) by superposition with the Ala-Phe-bound structure of PepTst in PyMol (The PyMol Molecular Graphics System, Schrödinger), followed by mutagenesis of the dipeptide as described32. Figure 3. Model of the substrate binding site of YePEPT. (a) YePEPTWT structure (PDB ID code: 4W6V32) with modelled bound Asp-Ala dipeptide. (b) YePEPTK314A-F311Y structure with modelled bound Asp-Ala dipeptide. Distances between the nitrogen atom of K314 and the carboxyl group of the aspartate side chain of Asp-Ala (a), between the hydroxyl group of Y311 and the α-amino group of Asp-Ala (b), and between the carboxyl group of E420 and the α-amino group of Asp-Ala (b) are indicated. Conserved residues involved in peptide backbone interactions in POTs are coloured in black34,35. Non-conserved residues in the substrate binding pocket are coloured in purple. The N- and C-terminal bundles of the YePEPTWT structure are coloured in green and blue, respectively. The Asp-Ala dipeptide was modelled into the binding pocket of YePEPTWT (PDB ID code: 4W6V) by superposition with the Ala-Phe-bound structure of PepTst in PyMol (The PyMol Molecular Graphics System, Schrödinger), followed by mutagenesis of the dipeptide as described32. a positively charged amino acid side chain at the N-terminal position32. As all tested β-lactam antibiotics contain an uncharged group (i.e., a phenyl- or hydroxyphenyl-group) at the position corresponding to the N-terminal amino acid side chain of dipeptides, we expected a potential beneficial effect by removing the positively charged K314 in YePEPTWT. In addition, the removal of this lysine side chain opens up space in the binding pocket, which was also considered to be beneficial, as β-lactam antibiotics are larger than dipeptides. Time course analy- sis of [3H]Ala-Ala uptake by the mutant YePEPTK314A demonstrated that the transport function is not impaired (Supplementary Fig. S1). pp y g Compared to wild-type (Fig. 2b), the inhibition pattern for YePEPTK314A (Fig. 2c) showed a strong increase in affinity for amoxicillin and cefadroxil, while ampicillin and cefalexin, as well as penicillin G and carbenicil- lin showed similar behaviours. Results and discussion β l ibi i if Structure‑based mutagenesis of YePEPTWT and β‑lactam antibiotic specificity determination of YePEPTK314A. Most residues involved in interactions with the backbone of di- and tripeptides are highly conserved in the POT family and were shown to be involved in substrate binding and transport (Fig. 3a, in black)34,35. Therefore, these residues cannot be mutated to tune and improve the specificity and affinity of POTs towards β-lactam antibiotics. K314 (Fig. 3a, in purple), one of the few non-conserved residues in the substrate binding pocket of YePEPTWT, was previously shown to be responsible for the recognition of dipeptides con- taining amino acids with negatively charged side chains at the N-terminal position through ionic interactions (Fig. 3a, dashed line). When mutated to glutamate (K314E), the specificity was inverted towards a dipeptide with https://doi.org/10.1038/s41598-021-96298-4 Scientific Reports \| (2021) 11:17205 \| www.nature.com/scientificreports/ Figure 3. Model of the substrate binding site of YePEPT. (a) YePEPTWT structure (PDB ID code: 4W6V32) with modelled bound Asp-Ala dipeptide. (b) YePEPTK314A-F311Y structure with modelled bound Asp-Ala dipeptide. Distances between the nitrogen atom of K314 and the carboxyl group of the aspartate side chain of Asp-Ala (a), between the hydroxyl group of Y311 and the α-amino group of Asp-Ala (b), and between the carboxyl group of E420 and the α-amino group of Asp-Ala (b) are indicated. Conserved residues involved in peptide backbone interactions in POTs are coloured in black34,35. Non-conserved residues in the substrate binding pocket are coloured in purple. The N- and C-terminal bundles of the YePEPTWT structure are coloured in green and blue, respectively. The Asp-Ala dipeptide was modelled into the binding pocket of YePEPTWT (PDB ID code: 4W6V) by superposition with the Ala-Phe-bound structure of PepTst in PyMol (The PyMol Molecular Graphics System, Schrödinger), followed by mutagenesis of the dipeptide as described32. Figure 3. Model of the substrate binding site of YePEPT. (a) YePEPTWT structure (PDB ID code: 4W6V32) with modelled bound Asp-Ala dipeptide. (b) YePEPTK314A-F311Y structure with modelled bound Asp-Ala dipeptide. Distances between the nitrogen atom of K314 and the carboxyl group of the aspartate side chain of Asp-Ala (a), between the hydroxyl group of Y311 and the α-amino group of Asp-Ala (b), and between the carboxyl group of E420 and the α-amino group of Asp-Ala (b) are indicated. Conserved residues involved in peptide backbone interactions in POTs are coloured in black34,35. Non-conserved residues in the substrate binding pocket are coloured in purple. Results and discussion β l ibi i if We concluded that the K314A mutation specifically increases the specificity for hydroxyphenyl containing β-lactam antibiotics, which might be due to the additional space to accommodate the hydroxyl group and/or a new interaction with the protein (e.g., a hydrogen bond (H-bond) of the hydroxyphenyl group with the protein). Structure‑based mutagenesis of YePEPTK314A and β‑lactam antibiotic specificity determina‑ tion of YePEPTK314A‑F311Y. To further increase the affinity of YePEPTK314A for β-lactam antibiotics, we looked for other residues in the binding pocket, which can be mutated without impairing the transport function of YePEPT. A prominent non-conserved residue pointing towards the substrate binding site in YePEPT is F311 (Fig. 3a, in purple). It was previously proposed that F311 forms a cation-π interaction with K314 in YePEPTWT, thus stabilizing this positive charge in the substrate binding pocket in absence of bound negatively charged dipeptides32. When mutated to a tyrosine (F311Y) in silico (Fig. 3b, in purple), the oxygen atom of the additional hydroxyl group is within H-bond distance of the α-amino group of the previously modelled32 Asp-Ala dipeptide (Fig. 3b, dashed line). The α-amino group probably already forms an ionic interaction with E420, as the cor- responding conserved residue in other POTs was shown to be involved in interactions with the α-amino group of bound dipeptides36–38. The distance of 3.2 Å from the carboxyl group of E420 in the structure of YePEPTWT to the α-amino group of the modelled Asp-Ala dipeptide (Fig. 3b, dashed line), supports this assumption. The predicted additional interaction of Y311 with the dipeptide backbone was hypothesized to increase the specific- ity and affinity for dipeptides as well as for peptidomimetics containing an α-amino group (e.g., aminopenicil- lins and aminocephalosporins). Time course analysis of [3H]Ala-Ala uptake using YePEPTK314A-F311Y expressing bacteria demonstrated that the transport function is retained in this mutant (Supplementary Fig. S1).h K314A F311Y The antibiotic inhibition pattern of YePEPTK314A-F311Y (Fig. 2d) showed an increased inhibition by all four β-lactam antibiotics containing an α-amino group (i.e., ampicillin, amoxicillin, cefalexin and cefadroxil) com- pared to YePEPTK314A, while there was still no inhibition by the non-aminopenicillins (i.e., penicillin G and carbenicillin). This behaviour fits with the hypothesis that the introduced tyrosine residue forms an additional H-bond with the substrate, most probably with the α-amino group of peptidomimetic substrates. Results and discussion β l ibi i if Data points in (b)–(d) epresent vector-subtracted uptake of [3H]Ala-Ala normalized to the uninhibited signal ± SEM of three ndependent experiments, each at least in triplicate. Numbers in brackets below Km- and IC50-values represent 5% confidence intervals in µM. acid), a basic- (i.e., lysine) or an aromatic residue (i.e., tyrosine) at the N- or C-terminal position was determined at a concentration of 2.5 mM (Fig. 2e). The same experiment with YePEPTWT showed previously, that only the tyrosine-containing dipeptides (i.e., Tyr-Ala and Ala-Tyr) as well Asp-Ala were able to inhibit [3H]Ala-Ala uptake32. In YePEPTK314E the specificity changed from Asp-Ala to Lys-Ala while there was no change on inhibition by the tyrosine-containing dipeptides. This led to the conclusion that a positively or negatively charged side chain at position 314 plays a role in substrate recognition of dipeptides containing a charged residue at the N-terminal position. This hypothesis was strengthened by the fact that in YePEPTK314A only the tyrosine-containing but none of the charged dipeptides were able to inhibit transport. In contrast, for YePEPTK314A-F311Y, all four charged dipep- tides (i.e., Asp-Ala, Ala-Asp, Lys-Ala and Ala-Lys) showed a reduction of [3H]Ala-Ala uptake (~ 40–70% residual uptake). The aromatic dipeptides (i.e., Tyr-Ala and Ala-Tyr) showed a slightly stronger inhibition compared to YePEPTK314A 32. These results support the hypothesis that the F311Y mutation leads to stronger interactions of YePEPT with ligands containing an α-amino group, independently of the side chains of the compound. Functional characterization of YePEPTK314A‑F311Y in E. coli cells. Based on the results from the β-lactam antibiotic inhibition assay, YePEPTK314A-F311Y represents a promising candidate for the desired anti- biotic translocation module and was therefore functionally characterized in more detail. First, to investigate the kinetics of the uptake of the radioligand, the Km for Ala-Ala was determined as 199 µM (Fig. 4a), which is the same as for YePEPTWT 32. Next, the inhibition of [3H]Ala-Ala uptake by different concentrations of amoxi- cillin and cefadroxil, the two antibiotics showing the strongest inhibition in the initial screen, was evaluated, resulting in IC50-values of 106 µM (Fig. 4b) and 229 µM (Fig. 4c), respectively. Results and discussion β l ibi i if To further test this hypothesis, the specificity of YePEPTK314A-F311Y for selected dipeptides containing an acidic- (i.e., aspartic Scientific Reports \| (2021) 11:17205 \| https://doi.org/10.1038/s41598-021-96298-4 www.nature.com/scientificreports/ Figure 4. Km and IC50 determinations for YePEPTK314A-F311Y. (a) Kinetics of [3H]Ala-Ala uptake in E. coli cells transformed with YePEPTK314A-F311Y. IC50 determination of YePEPTK314A-F311Y for (b) amoxicillin, (c) cefadroxil and (d) cefprozil by heterologous competition. Data points in (a) represent vector-subtracted uptake of [3H] Ala-Ala normalized to Vmax ± SEM of three experiments, each at least in triplicate. Data points in (b)–(d) represent vector-subtracted uptake of [3H]Ala-Ala normalized to the uninhibited signal ± SEM of three independent experiments, each at least in triplicate. Numbers in brackets below Km- and IC50-values represent 95% confidence intervals in µM. igure 4. Km and IC50 determinations for YePEPTK314A-F311Y. (a) Kinetics of [3H]Ala-Ala uptake in E. coli cells ransformed with YePEPTK314A-F311Y IC determination of YePEPTK314A-F311Y for (b) amoxicillin (c) cefadroxil Figure 4. Km and IC50 determinations for YePEPTK314A-F311Y. (a) Kinetics of [3H]Ala-Ala uptake in E. coli cells transformed with YePEPTK314A-F311Y. IC50 determination of YePEPTK314A-F311Y for (b) amoxicillin, (c) cefadroxil and (d) cefprozil by heterologous competition. Data points in (a) represent vector-subtracted uptake of [3H] Ala-Ala normalized to Vmax ± SEM of three experiments, each at least in triplicate. Data points in (b)–(d) represent vector-subtracted uptake of [3H]Ala-Ala normalized to the uninhibited signal ± SEM of three independent experiments, each at least in triplicate. Numbers in brackets below Km- and IC50-values represent 95% confidence intervals in µM. Figure 4. Km and IC50 determinations for YePEPTK314A-F311Y. (a) Kinetics of [3H]Ala-Ala uptake in E. coli cells transformed with YePEPTK314A-F311Y. IC50 determination of YePEPTK314A-F311Y for (b) amoxicillin, (c) cefadroxil and (d) cefprozil by heterologous competition. Data points in (a) represent vector-subtracted uptake of [3H] Ala-Ala normalized to Vmax ± SEM of three experiments, each at least in triplicate. Data points in (b)–(d) represent vector-subtracted uptake of [3H]Ala-Ala normalized to the uninhibited signal ± SEM of three independent experiments, each at least in triplicate. Numbers in brackets below Km- and IC50-values represent 95% confidence intervals in µM. 50 ( ) , ( ) nd (d) cefprozil by heterologous competition. Data points in (a) represent vector-subtracted uptake of [3H] Ala-Ala normalized to Vmax ± SEM of three experiments, each at least in triplicate. Results and discussion β l ibi i if While specificity for different β-lactam antibiotics differ between different members of the POT family, comparison of the obtained affini- ties for YePEPTK314A-F311Y with the values for PEPT1 (Ki(amoxicillin) ≥ 10 mM, Ki(cefadroxil) = 7.2 mM)22 and PEPT2 (Ki(amoxicillin) = 430 µM, Ki(cefadroxil) = 3 µM)23 allow for the classification of YePEPTK314A-F311Y as a high affinity trans- porter as is needed for an antibiotic transport module.fi p p To deepen our understanding of the structure-affinity relationship of antibiotics in YePEPTK314A-F311Y and to find antibiotics with higher affinities for YePEPTK314A-F311Y, we searched for commercially available analogues of amoxicillin and cefadroxil. Cefprozil (Fig. 2a 7), a cefadroxil analogue containing a longer aliphatic substituent https://doi.org/10.1038/s41598-021-96298-4 Scientific Reports \| (2021) 11:17205 \| www.nature.com/scientificreports/ e.com/scientificreports/ p Figure 5. SSM-based electrophysiology of YePEPTK314A-F311Y. (a) Electrophysiology data of antibiotic transport by YePEPTK314A-F311Y. (b) Representative electrophysiology traces of Ala-Ala transport by proteoliposomes containing YePEPTK314A-F311Y (blue) or empty control liposomes (red). Bars in (a) represent baseline-corrected transport normalized to the Ala-Ala signal ± SD from six sensors, each measured in quintuplicates. Individual measurements are depicted as white circles. Concentration of all compounds was 5 mM. Figure 5. SSM-based electrophysiology of YePEPTK314A-F311Y. (a) Electrophysiology data of antibiotic transport by YePEPTK314A-F311Y. (b) Representative electrophysiology traces of Ala-Ala transport by proteoliposomes containing YePEPTK314A-F311Y (blue) or empty control liposomes (red). Bars in (a) represent baseline-corrected transport normalized to the Ala-Ala signal ± SD from six sensors, each measured in quintuplicates. Individual measurements are depicted as white circles. Concentration of all compounds was 5 mM. n the six-membered ring of the cephem core structure, (i.e., a 1-propenyl- instead of a methyl group) showed stronger inhibition than cefadroxil and amoxicillin with an IC50 of 32 µM (Fig. 4d). Evaluation of antibiotic transport by YePEPTK314A‑F311Y. Considering that inhibition of Ala-Ala uptake does not provide proof that compounds are actually transported, we set up a solid-supported membrane (SSM)-based electrophysiology assay using proteoliposomes containing purified YePEPTK314A-F311Y to evaluate transport of the antibiotics, which showed inhibition in the cell-based uptake (Fig. 5a). In short, during the assay, the proteoliposomes adsorbed to a SSM-chip are alternatingly perfused with non-activating- (i.e., with- out potential substrates) and activating solutions (i.e., containing the potential substrate to be evaluated). If electrogenic transport occurs, transient currents after solution exchange can be detected (Fig. 5b, in blue). Materials and methodsh Cloning of YePEPT. The gene of the peptide transporter YePEPT from Y. enterocolitica (UniProt accession number: R9G739) was amplified and inserted into the pZUDF vector42 as previously described32. Mutants were prepared by site-directed mutagenesis using the QuikChange Lightning Multi Site-Directed Mutagenesis Kit (Agilent Technologies). Uptake assay with E. coli cells overexpressing YePEPT variants. 80 ml Luria Bertani (LB) medium supplemented with 100 µg/ml ampicillin was inoculated with 1 ml of an overnight pre-culture of E. coli BL21(DE3) pLysS transformed with YePEPTWT, YePEPTK314A, YePEPTK314A-F311Y or empty vector (control), and incubated at 37 °C and 180 rpm. Protein expression was induced at an OD600 of 0.7–0.8 with 300 µM isopropyl- β-D-thiogalactopyranoside (IPTG). After 3 h of induction, cells corresponding to 10 ml of an OD600 of 1.5 were harvested by centrifugation (5000 × g, 4 °C, 15 min) and resuspended in 1.5 ml uptake buffer (50 mM HEPES– NaOH, 150 mM NaCl, 5 mM glucose, pH 7.5), and kept on ice.hi g p ) p The final volume of the samples for the uptake assay was 50 µl and consisted of 20 µl of resuspended cells, 10 µl of 5 × substrate Master mix (250 µM Ala-Ala (50 µM final concentration) in uptake buffer spiked with [3H] Ala-Ala (Campro Scientific) to a specific activity of 0.1 Ci/mmol) and 20 µl of competitor in uptake buffer. For screening, dipeptide and antibiotic competitors were used at a final concentration of 2.5 mM and 5 mM, respec- tively. For Km-determination, various concentrations of Ala-Ala, spiked to a specific activity of 0.0125 Ci/mmol were used. For IC50-determinations, various concentrations of competitor (Fig. 4) were used. 50 p g To be able to measure the substrate transport in the linear regime, duration and temperature of the assay were optimized for YePEPT variants: 100 s at 18 °C for YePEPTWT, 60 s at 18 °C for YePEPTK314A and 100 s at 25 °C for YePEPTK314A-F311Y, see Supplementary Fig. S1. Transport was stopped after the indicated time by addi- tion of 450 µl of ice-cold stop buffer (50 mM HEPES–NaOH, 150 mM NaCl, 5 mM glucose, 2.5 mM Ala-Ala, pH 7.5) and the cells were pelleted by centrifugation (14,000 × g, room temperature, 2 min), washed once with 450 µl uptake buffer and pelleted again. Washed cells were resuspended in 50 µl 5% (w/v) SDS and transferred to white 96-well plates (OptiPlate, PerkinElmer, Waltham, MA, USA). Conclusion Vesicular molecular systems able to actively accumulate and degrade certain small organic molecules would be a promising approach to tackle the problem of environmental contamination by pollutants such as drugs and pesticides. One essential component needed for the development of such a system would be a transport module, which is able to accumulate the target molecule using energy provided by other components of the system (e.g., the proton gradient established by a light-driven proton pump). In this work, we engineered such a module specific for β-lactam antibiotics based on a bacterial peptide transporter. The specificity and affinity of the transporter YePEPT for certain β-lactam antibiotics was significantly improved using structure-based mutagenesis combined with functional characterization using an uptake assay in E. coli cells overexpressing the transporter. Two specific mutations were identified, which enhance the specificity and affinity of the transporter for β-lactam antibiotics containing certain structural features (i.e., hydroxyphenyl group or α-amino group). Screening of commercially available β-lactam antibiotics allowed for the identification of cefprozil as a ligand for YePEPTK314A-F311Y with relatively high affinity (IC50 = 32 µM; Fig. 4d). Cefprozil represents a promising candidate for future co-crystallization studies with YePEPTK314A-F311Y to elucidate the molecular binding mechanism for β-lactam antibiotics, which will facilitate a more targeted structure-based mutagenesis to further enhance the affinity for compounds of this group and to tailor the protein for compounds with specific structural features. Finally, SSM-based electrophysiology measurements provided evidence that β-lactam antibiotics with sufficient affinity not only inhibit transport (by binding competition), but are indeed translocated by YePEPTK314A-F311Y, with cefprozil reaching 46% transport relative to Ala-Ala. The extent to which certain β-lactam antibiotics are transported appears to be determined both, by the affinity and by structural features, as shown by the preference for hydroxyphenyl-containing cephalosporins over hydroxyphenyl-containing penicillins. Vesicular molecular systems able to actively accumulate and degrade certain small organic molecules would be a promising approach to tackle the problem of environmental contamination by pollutants such as drugs and pesticides. One essential component needed for the development of such a system would be a transport module, which is able to accumulate the target molecule using energy provided by other components of the system (e.g., the proton gradient established by a light-driven proton pump). In this work, we engineered such a module specific for β-lactam antibiotics based on a bacterial peptide transporter. Conclusion The specificity and affinity of the transporter YePEPT for certain β-lactam antibiotics was significantly improved using structure-based mutagenesis combined with functional characterization using an uptake assay in E. coli cells overexpressing the transporter. Two specific mutations were identified, which enhance the specificity and affinity of the transporter for β-lactam antibiotics containing certain structural features (i.e., hydroxyphenyl group or α-amino group). Screening of commercially available β-lactam antibiotics allowed for the identification of cefprozil as a ligand for YePEPTK314A-F311Y with relatively high affinity (IC50 = 32 µM; Fig. 4d). Cefprozil represents a promising candidate for future co-crystallization studies with YePEPTK314A-F311Y to elucidate the molecular binding mechanism for y g β-lactam antibiotics, which will facilitate a more targeted structure-based mutagenesis to further enhance the affinity for compounds of this group and to tailor the protein for compounds with specific structural features. Finally, SSM-based electrophysiology measurements provided evidence that β-lactam antibiotics with sufficient affinity not only inhibit transport (by binding competition), but are indeed translocated by YePEPTK314A-F311Y, with cefprozil reaching 46% transport relative to Ala-Ala. The extent to which certain β-lactam antibiotics are transported appears to be determined both, by the affinity and by structural features, as shown by the preference for hydroxyphenyl-containing cephalosporins over hydroxyphenyl-containing penicillins. www.nature.com/scientificreports/ were transported with transported charges relative to Ala-Ala of 9% and 23%, respectively. The preference of YePEPTK314A-F311Y for amoxicillin over ampicillin is in agreement with the higher specificity for antibiotics con- taining a hydroxyphenyl group observed in transport inhibition experiments (Fig. 2d). Of the (amino-)cepha- losporins (i.e., cefalexin, cefadroxil and cefprozil) only the latter two were transported with relative transported charges of 36% and 46%, respectively. Cefadroxil being more efficiently transported than amoxicillin, despite the lower affinity (i.e., higher IC50-value) determined in heterologous competition experiments (Fig. 4b,c) sug- gests that YePEPTK314A-F311Y preferentially transports β-lactam antibiotics containing a cephem ring structure (i.e., cephalosporins) over those containing a penam ring structure (i.e., penicillins). Nevertheless, in the case of cefalexin, the low specificity observed (Fig. 2d), due to the lack of a hydroxyl group at the phenyl ring, seems to outweigh this preference. The observed increased transport of cefprozil over cefadroxil can be attributed to the longer aliphatic side chain (Fig. 2a) either because of the higher affinity (Fig. 4c and d) or due to a structural preference of the transport mechanism of YePEPTK314A-F311Y. Results and discussion β l ibi i if As even small differences in solute concentrations (e.g., with or without the tested compound) can lead to artefact peaks, proteoliposome measurements were corrected by measuring the same solutions in liposomes devoid of transport protein (Fig. 5b, in red). This SSM-based electrophysiology technique for transporter research was described previously in detail39,40. p y In all cases where transport occurred, a positive peak (i.e., net positive charge transport) was detected, indicating proton-coupled symport as expected for a POT-family member. Studies on the bacterial POT PepTst showed proton:substrate stoichiometries of 3 for tripeptides and 4–5 for dipeptides12,41. Therefore and consider- ing a partial negative charge (-0.23) of the tested antibiotics at pH 6.7 (Supplementary Table S1), the observed positive transport peaks from antibiotics (Supplementary Figure S2) are in line with the co-transport of at least one proton per antibiotic molecule. Both tested (amino-)penicillin antibiotics (i.e., ampicillin and amoxicillin) https://doi.org/10.1038/s41598-021-96298-4 Scientific Reports \| (2021) 11:17205 \| www.nature.com/scientificreports/ Materials and methodsh The resin was transferred to a column (Promega Wizard Midicolumns) and washed with 15 ml of washing buffer and 3 ml of elution buffer (20 mM Tris–HCl, 150 mM NaCl, 0.2% (w/v) DM, pH 8). After addition of 400 µl of elution buffer supplemented with 400 mM imidazol and incubation for 30 min at 4 °C under gentle agitation, the protein was eluted by centrifugation (3,000 × g, 1 min, 4 °C). Imidazol was removed using a desalting column (Zeba spin desalting columns 7 k MWCO, Thermo Sci- entific) pre-equilibrated with elution buffer. Reconstitution. E. coli polar lipids (Avanti polar lipids, Inc.) dissolved in chloroform were evaporated under a gentle stream of nitrogen, dried under vacuum overnight and rehydrated with elution buffer to a final concentration of 5 mg/ml. The lipids were then solubilized in elution buffer supplemented with 2% DM at a final lipid concentration of 2 mg/ml and incubated for 1 h at room temperature under gentle agitation. Purified YePEPTK314A-F311Y was mixed with the solubilized lipids at an lipid-to-protein (LPR) ratio of 5 (1 mg/ml lipids, 0.2 mg/ml protein) and incubated for 10 min at room temperature under gentle agitation. The lipid-protein mixture was then transferred to 40 µl dialysis buttons (custom-made) with a 100,000 Da cut-off cellulose acetate membrane (Harvard apparatus) and dialysed against elution buffer for 6 days at 18 °C with two buffer exchanges (after 24 h and after 3 days). As a control sample, liposomes devoid of protein were prepared according to the same protocol. SSM‑based electrophysiology transport assay. SSM-based electrophysiology experiments were per- formed using a SURF2ER N1 instrument (Nanion Technologies) according to published protocols39,40. SSM were prepared as follows: 50 µl of thiol solution (0.5 mM 1-octadecanethiol in 100% isopropanol) was added into the sensor well and incubated for 3 h in a closed petri dish. After removal of the thiol solution, the sensors were washed five times with 100% isopropanol and five times with Milli-Q ultrapure water. The SSM was formed by application of 1.5 µl of lipid solution (7.5 µg/µl 1,2-diphytanoyl-sn-glycero-3-phosphocholine in 100% n-decane) directly onto the gold surface, followed immediately by addition of 50 µl of non-activating buffer (25 mM MES, 25 mM HEPES, 140 mM KCl, 2 mM MgCl2, pH 6.7). Materials and methodsh After addition of 150 µl scintillation cocktail (MicroScint 40, PerkinElmer), the plates were measured with a scintillation counter (2 min per well, Packard TopCount, PerkinElmer). p Raw data was processed by baseline-subtraction of the empty vector samples and normalized to the unin- hibited [3H]Ala-Ala signal. Km- and IC50-values were determined by nonlinear regression. All data processing and plotting was performed with the Prism GraphPad 6 software. Scientific Reports \| (2021) 11:17205 \| https://doi.org/10.1038/s41598-021-96298-4 www.nature.com/scientificreports/ Overexpression and membrane isolation. 24 l of LB medium supplemented with 100 µg/ml ampicillin were inoculated 1:100 with an overnight culture of E. coli BL21(DE3) pLysS transformed with pZUDF21-rbs- YePEPT-K314A-F311Y-3C-His10 and incubated at 37 °C and 180 rpm in an incubator shaker (Multitron, Infors HT). At an OD600 of 0.6–0.7, heterologous protein overexpression was induced by the addition of 300 µM IPTG and incubation was continued for 4 h. Cells were harvested by centrifugation (10,000 × g, 5 min, 4 °C). The pel- let was washed once with 2 l of membrane wash buffer (20 mM Tris–HCl, 500 mM NaCl, pH 8), centrifuged again (10,000 × g, 5 min, 4 °C), resuspended in 300 ml of lysis buffer (20 mM Tris–HCl, 50 mM NaCl, pH 8) and stored at −80 °C. Cells were thawed and lysed by sonication for 60 min (total ON time) in 5 s ON/3 s OFF pulses using a tip sonifier (Branson 450 Digital Sonifier) while cooled in an ice bath. Membranes were then harvested by ultracentrifugation (150,000 × g, 1 h, 4 °C), washed once with 240 ml of membrane wash buffer and homog- enized using a glass tissue homogenizer. The last ultracentrifugation was repeated once, the pellet resuspended in 36 ml of purification buffer (20 mM Tris–HCl, 300 mM NaCl, pH 8) and the membranes were flash-frozen in liquid nitrogen and stored at -80 °C. Purification of YePEPTK314A‑F311Y. YePEPTK314A-F311Y membranes from 2 l of cell culture were solubilized in 7 ml of purification buffer supplemented with 2% (w/v) of n-decyl-β-D-maltopyranoside (DM, Glycon Bio- chemicals GmbH) for 1 h at 4 °C under gentle agitation. After ultracentrifugation (150,000 × g, 1 h, 4 °C), the supernatant was diluted 1:1 with washing buffer (20 mM Tris–HCl, 300 mM NaCl, 5 mM L-histidine, 0.2% (w/v) DM, pH 8), supplemented with 500 µl (bed-volume) pre-equilibrated Ni–NTA superflow resin (Qiagen) and incubated for 4 h at 4 °C under gentle agitation. Materials and methodsh Proteoliposomes containing YePEPTK314A-F311Y, as well as empty control liposomes, were diluted 1:1 with non-activating buffer, sonicated for 30 s in a bath sonicator and 5 µl adsorbed to each sensor. Before measurements, sensors were centrifuged (3000 × g, 30 min, 20 °C) and tested for suitable capacitance and conductance values40. p All measurements were conducted at 20 °C. Each measurement consisted of alternating perfusions for 1 s with non-activating- and activating buffer (non-activating buffer supplemented with 5 mM of the tested com- pound). Ala-Ala was measured at the beginning and at the end of each measurement sequence to test for possible signal loss of the sensor during the experiment. The transported charge of each measurement was determined by integrating the transient current peak occurring after perfusion with activating buffer. As even small differ- ences in solute concentrations (e.g., with or without the tested compound) can lead to artefact peaks, the signals from sample sensors (sensors containing YePEPTK314A-F311Y-proteoliposomes) were corrected by subtracting the i p g p p y g averaged signals from five control sensors (sensors with adsorbed liposomes devoid of protein), each measured in quintuplicates. Baseline-corrected signals from six sample sensors, each measured in quintuplicates, were then normalized to the Ala-Ala signal and merged. Measurements (including peak integration) were performed using the SURF2ER N1 control software. Data processing and plotting was performed with the Prism GraphPad 6 software. References Interaction of anionic cephalosporins with the intestinal and renal peptide transporters PEPT 1 and PEPT 2. Biochim. Biophys. Acta - Biomembr. 1324, 296–308 (1997).fi p p p p y , 22. Bretschneider, B., Brandsch, M. & Neubert, R. Intestinal transport of β-lactam antibiotics: analysis of the affinity symporter (PEPT1), the uptake into Caco-2 cell monolayers and the transepithelial flux. Pharm. Res. 16, 55–61 p p p p y ( ) 22. Bretschneider, B., Brandsch, M. & Neubert, R. Intestinal transport of β-lactam antibiotics: analysis of the affinity at the H+/ , , , & , p β yfi y /p p orter (PEPT1), the uptake into Caco-2 cell monolayers and the transepithelial flux. Pharm. Res. 16, 55–61 (1999).fi fi symporter (PEPT1), the uptake into Caco-2 cell monolayers and the transepithelial flux. Pharm. Res. 16, 55–61 (1999). 23. Luckner, P. & Brandsch, M. Interaction of 31 β-lactam antibiotics with the H+/peptide symporter PEPT2: analysis of affinity constants and comparison with PEPT1. Eur. J. Pharm. Biopharm. 59, 17–24 (2005). p p 24. Prabhala, B. K. et al. Several hPepT1-transported drugs are substrates of the Escherichia coli proton-coupled oligopeptide trans- porter YdgR. Res. Microbiol. 168, 443–449 (2017).h p g 25. Prabhala, B. K. et al. The prototypical proton-coupled oligopeptide transporter YdgR from Escherichia coli facilitates chl phenicol uptake into bacterial cells. J. Biol. Chem. 293, 1007–1017 (2018). ug transport via the intestinal peptide transporter PepT1. Curr. Op p p 26. Brandsch, M. Drug transport via the intestinal peptide transporter PepT1. Curr. Opin. Pharmacol. 13, 881–887 (2013). p p . Brandsch, M. Drug transport via the intestinal peptide transpor 7. Minhas, G. S. & Newstead, S. Structural basis for prodrug recognition by the SLC15 family of proton-coupled peptide transporters Proc. Natl. Acad. Sci. USA 116, 804–809 (2019). 28. Ural-Blimke, Y. et al. Structure of prototypic peptide transporter DtpA from E. coli in comp into drug binding of human PepT1. J. Am. Chem. Soc. 141, 2404–2412 (2019). 8. Ural-Blimke, Y. et al. Structure of prototypic peptide transporter DtpA from E. coli in complex with valganciclovir provides insights into drug binding of human PepT1. J. Am. Chem. Soc. 141, 2404–2412 (2019). g g p 9. Brandsch, M., Knütter, I. & Bosse-Doenecke, E. Pharmaceutical and pharmacological importance of peptide transporters. J. Pharm Pharmacol. 60, 543–585 (2008).h 0. He, Y., Wang, K. & Yan, N. Data availabilityh The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. Received: 1 May 2021; Accepted: 23 July 2021 Received: 1 May 2021; Accepted: 23 July 2021 https://doi.org/10.1038/s41598-021-96298-4 Scientific Reports \| (2021) 11:17205 \| www.nature.com/scientificreports/ References 1. Benner, S. A. & Sismour, A. M. Synthetic biology. Nat. Rev. Genet. 6, 533–543 (2005). 1. Benner, S. A. & Sismour, A. M. Synthetic biology. Nat. Rev. Genet. 6, 533–543 (2005). ner, S. A. & Sismour, A. M. Synthetic biology. Nat. Rev. Genet. 6, 5 1. Benner, S. A. & Sismour, A. M. Synthetic biology. Nat. Rev. Genet. 6, 533–543 (2005). y gy 2. Miller, D. & Gulbis, J. Engineering protocells: prospects for self-assembly and nanoscale production-lines. Life 5, 1019–1053 (2015) 3 J i l S & Sh kl P Al i i f i bi l di i f ll i h i bi l F Mi bi l 11 2. Miller, D. & Gulbis, J. Engineering protocells: prospects for self-assembly and nanoscale production-lines. Life 5, 1019–1053 (2015). 3. Jaiswal, S. & Shukla, P. Alternative strategies for microbial remediation of pollutants via synthetic biology. Front. Microbiol. 11, 1–14 (2020). ( ) 4. Hirschi, S. et al. Engineering and assembly of protein modules into functional molecular systems. Chimia 70, 398–401 (2016). 5. Laohakunakorn, N. et al. Bottom-up construction of complex biomolecular systems with cell-free synthetic biology. Front. Bioeng Biotechnol. 8, 1–26 (2020). 6. Daughton, C. G. & Ternes, T. A. Pharmaceuticals and personal care products in the environment: agents of subtle change? Environ Toxicol. 28, 2663–2670 (2009). ( ) 7. Sta Ana, K. M., Madriaga, J. & Espino, M. P. β-Lactam antibiotics and antibiotic resistance in Asian lakes and rivers: An overview of contamination, sources and detection methods. Environ. Pollut. 275, 116624 (2021). 8. Czekalski, N., Berthold, T., Caucci, S., Egli, A. & Bürgmann, H. Increased levels of multiresistant bacteria and resistance genes after wastewater treatment and their dissemination into Lake Geneva, Switzerland. Front. Microbiol. 3, 1–18 (2012).h 8. Czekalski, N., Berthold, T., Caucci, S., Egli, A. & Bürgmann, H. Increased levels of multiresistant bacteria and res after wastewater treatment and their dissemination into Lake Geneva, Switzerland. Front. Microbiol. 3, 1–18 (2012) t 9. Paulsen, I. T. & Skurray, R. A. The POT family of transport proteins. Trends Biochem. Sci. 19, 404 (1994). rray, R. A. The POT family of transport proteins. Trends Biochem. h 0. Daniel, H., Spanier, B., Kottra, G. & Weitz, D. From bacteria to man: archaic proton-dependent peptide transporters at work Physiol. 21, 1 (2006). 1. Newstead, S. Recent advances in understanding proton coupled peptide transport via the POT family. Curr. Opin. Struct. Biol. References 45 17–24 (2017). ( ) 2. Prabhala, B. K., Rahman, M., Nour-eldin, H. H., Jørgensen, F. S. & Mirza, O. PTR2/POT/NPF transporters: what makes them tick? in Advances in Protein Chemistry and Structural Biology 123, 219–240 (Elsevier Ltd, 2021).h ein Chemistry and Structural Biology 123, 219–240 (Elsevier Ltd, 2 y gy . Steiner, H.-Y., Naider, F. & Becker, J. M. The PTR family: a new g 13. Steiner, H.-Y., Naider, F. & Becker, J. M. The PTR family: a new group of peptide transporters. Mol. Microbiol. 16, 825–834 (1995). 14. Reddy, V. S., Shlykov, M. A., Castillo, R., Sun, E. I. & Saier, M. H. Jr. The major facilitator superfamily (MFS) revisited. FEBS J. 279, 2022–2035 (2012). h y g p p p p 14. Reddy, V. S., Shlykov, M. A., Castillo, R., Sun, E. I. & Saier, M. H. Jr. The major facilitator superfamily (MFS) revisited. FEBS J. 279, 2022–2035 (2012). ( ) 5. Quistgaard, E. M., Löw, C., Guettou, F. & Nordlund, P. Understanding transport by the major facilitator superfamily (MFS) structures pave the way. Nat. Rev. Mol. Cell Biol. 17, 123–132 (2016). p y ( ) 16. Shi, Y. Common folds and transport mechanisms of secondary active transporters. Annu. Rev. Biophys. 42, 51–72 (2013). y 16. Shi, Y. Common folds and transport mechanisms of secondary active transporters. Annu. Rev. Bio & Fotiadis, D. Secondary active transporters. Subcell Biochem. 92, Bosshart, P. D. & Fotiadis, D. Secondary active transporters. Subce y 8. Daniel, H. & Kottra, G. The proton oligopeptide cotransporter family SLC15 in physiology and pharmacology. Pflugers Arch. Eur J. Physiol. 447, 610–618 (2004). y 9. Minhas, G. S. & Newstead, S. Recent advances in understanding prodrug transport through the SLC15 family of proton-coupled transporters. Biochem. Soc. Trans. 0, 1–10 (2020).f p ( ) 20. Ganapathy, M. E., Brandsch, M., Prasad, P. D., Ganapathy, V. & Leibach, F. H. Differential recognition of β-lactam antibiotics by intestinal and renal peptide transporters, PEPT 1 and PEPT 2. J. Biol. Chem. 270, 25672–25677 (1995). p p p 1. Ganapathy, M. E., Prasad, P. D., Mackenzie, B., Ganapathy, V. & Leibach, F. H. Interaction of anionic cephalosporins with the intestinal and renal peptide transporters PEPT 1 and PEPT 2 Biochim Biophys Acta Biomembr 1324 296 308 (1997) 21. Ganapathy, M. E., Prasad, P. D., Mackenzie, B., Ganapathy, V. & Leibach, F. H. Acknowledgements l f g Financial support from the University of Bern, the Swiss National Science Foundation (SNSF; Grant 310030_184980) and the NCCR Molecular Systems Engineering is kindly acknowledged. Author contributions M.S. and D.F. designed the experiments and analysed the data. Z.U. performed cloning and site-directed mutagenesis. M.S. and D.H. performed SSM-based electrophysiology experiments. M.S. performed all other experiments and prepared figures and graphs. M.S. and D.F wrote the manuscript. All authors read and approved the final manuscript. Competing interests h p g The authors declare no competing interests. The authors declare no competing interests. References The recombinant expression systems for structure determination of eukaryotic membrane proteins Protein Cell 5, 658–672 (2014).i 1. Opekarová, M. & Tanner, W. Specific lipid requirements of membrane proteins - A putative bottleneck in heterologous expression Biochim. Biophys. Acta - Biomembr. 1610, 11–22 (2003). p y 2. Boggavarapu, R., Jeckelmann, J.-M., Harder, D., Ucurum, Z. & Fotiadis, D. Role of electrostatic interactions for ligand recognition and specificity of peptide transporters. BMC Biol. 13, 58 (2015).hfi d specificity of peptide transporters. BMC Biol. 13, 58 (2015).h i y 3. Brandsch, M., Knütter, I. & Leibach, F. H. The intestinal H+/peptide symporter PEPT1: Structure–affinity relationships. Eur. J Pharm. Sci. 21, 53–60 (2004). 4. Doki, S. et al. Structural basis for dynamic mechanism of proton-coupled symport by the peptide transporter POT. Proc. Natl Acad. Sci. USA 110, 11343–11348 (2013). 5. Newstead, S. Molecular insights into proton coupled peptide transport in the PTR family of oligopeptide transporters. Biochim Biophys. Acta - Gen. Subj. 1850, 488–499 (2015). p y j 6. Guettou, F. et al. Selectivity mechanism of a bacterial homolog of the human drug-peptide transporters PepT1 and PepT2. Nat Struct. Mol. Biol. 21, 728–731 (2014).i 7. Lyons, J. A. et al. Structural basis for polyspecificity in the POT family of proton-coupled oligopeptide transporters. EMBO Rep 15, 886–893 (2014).i 8. Martinez Molledo, M., Quistgaard, E. M., Flayhan, A., Pieprzyk, J. & Löw, C. Multispecific substrate recognition in a proton dependent oligopeptide transporter. Structure 26, 467–476 (2018). p g p p p 39. Bazzone, A., Barthmes, M. & Fendler, K. SSM-based electrophysiology for transporter research. in Methods in enzymology (ed. Ziegler, C.) 594, 31–83 (Elsevier Inc., 2017). g 40. Bazzone, A. & Barthmes, M. Functional characterization of SLC transporters using solid supported membranes. in Biophysics of membrane proteins: methods and protocols (eds. Postis, V. L. G. & Goldman, A.) 73–103 (Springer US, 2020).h p p p g 1. Parker, J. L., Mindell, J. A. & Newstead, S. Thermodynamic evidence for a dual transport mechanism in a POT peptide transporter Elife 3, 1–13 (2014).i f 42. Ilgü, H. et al. Variation of the detergent-binding capacity and phospholipid content of membrane proteins when purified in dif- ferent detergents. Biophys. J. 106, 1660–1670 (2014). https://doi.org/10.1038/s41598-021-96298-4 Scientific Reports \| (2021) 11:17205 \| www.nature.com/scientificreports/ © The Author(s) 2021 Additional informationh Supplementary Information The online version contains supplementary material available at https://doi.org/ 10.1038/s41598-021-96298-4. Correspondence and requests for materials should be addressed to D.F. 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 nstitutional affiliations. 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) 2021 https://doi.org/10.1038/s41598-021-96298-4 Scientific Reports \| (2021) 11:17205 \|
https://openalex.org/W4362630185	https://figshare.com/articles/journal_contribution/Supplementary_Table_S1_from_A_Preclinical_and_Phase_Ib_Study_of_Palbociclib_plus_Nab-Paclitaxel_in_Patients_with_Metastatic_Adenocarcinoma_of_the_Pancreas/22545081/1/files/40008510.pdf	English	null	Supplementary Table S5 from A Preclinical and Phase Ib Study of Palbociclib plus Nab-Paclitaxel in Patients with Metastatic Adenocarcinoma of the Pancreas	null	2,023	cc-by	423	Supplementary Table S1. Representativeness of Study Participants Considerations related to: Pancreatic Ductal Adenocarcinoma (PDAC) Sex The incidence of PDAC is higher in males than females, 9.5/1,000 and 7.0/1,000, respectively. Male gender has been shown to be a risk factor for PDAC. Age The most frequent age of diagnosis for both genders is 60–69 years. Race/ethnicity Considerably higher incidence of PDAC in Black patients have been reported in studies from the United States and this is reflected by mortality rates. Black race has been found to be a risk factor for PDAC. Geography Europe has the highest age-standardized PDAC mortality rate, followed by North America. Africa and Central America have the lowest PDAC mortality rates. Other considerations The major risk factors for PDAC are family history, genetic disorders, complications and preferences. Overall representativeness of this study The age distribution and gender of our study population, median age 61 years and 55% male, reflect the demographics of PDAC in the general population. Our study population was relatively small (N=76) and patient recruitment was limited to the centers participating in the study, and thus no Black patients were recruited. Supplementary Table S1. Representativeness of Study Participants Considerations related to: Pancreatic Ductal Adenocarcinoma (PDAC) Sex The incidence of PDAC is higher in males than females, 9.5/1,000 and 7.0/1,000, respectively. Male gender has been shown to be a risk factor for PDAC. Age The most frequent age of diagnosis for both genders is 60–69 years. Race/ethnicity Considerably higher incidence of PDAC in Black patients have been reported in studies from the United States and this is reflected by mortality rates. Black race has been found to be a risk factor for PDAC. Geography Europe has the highest age-standardized PDAC mortality rate, followed by North America. Africa and Central America have the lowest PDAC mortality rates. Other considerations The major risk factors for PDAC are family history, genetic disorders, complications and preferences. Overall representativeness of this study The age distribution and gender of our study population, median age 61 years and 55% male, reflect the demographics of PDAC in the general population. Our study population was relatively small (N=76) and patient recruitment was limited to the centers participating in the study, and thus no Black patients were recruited. The age distribution and gender of our study population, median age 61 years and 55% male, reflect the demographics of PDAC in the general population. Our study population was relatively small (N=76) and patient recruitment was limited to the centers participating in the study, and thus no Black patients were recruited.
https://openalex.org/W2804610054	https://europepmc.org/articles/pmc5986524?pdf=render	English	null	Pro-Apoptotic and Anti-Cancer Properties of Diosgenin: A Comprehensive and Critical Review	Nutrients	2,018	cc-by	7,965	Received: 17 March 2018; Accepted: 16 May 2018; Published: 19 May 2018 Abstract: Novel and alternative options are being adopted to combat the initiation and progression of human cancers. One of the approaches is the use of molecules isolated from traditional medicinal herbs, edible dietary plants and seeds that play a pivotal role in the prevention/treatment of cancer, either alone or in combination with existing chemotherapeutic agents. Compounds that modulate these oncogenic processes are potential candidates for cancer therapy and may eventually make it to clinical applications. Diosgenin is a naturally occurring steroidal sapogenin and is one of the major bioactive compounds found in dietary fenugreek (Trigonella foenum-graecum) seeds. In addition to being a lactation aid, diosgenin has been shown to be hypocholesterolemic, gastro- and hepato-protective, anti-oxidant, anti-inﬂammatory, anti-diabetic, and anti-cancer. Diosgenin has a unique structural similarity to estrogen. Several preclinical studies have reported on the pro-apoptotic and anti-cancer properties of diosgenin against a variety of cancers, both in in vitro and in vivo. Diosgenin has also been reported to reverse multi-drug resistance in cancer cells and sensitize cancer cells to standard chemotherapy. Remarkably, diosgenin has also been reported to be used by pharmaceutical companies to synthesize steroidal drugs. Several novel diosgenin analogs and nano-formulations have been synthesized with improved anti-cancer efﬁcacy and pharmacokinetic proﬁle. In this review we discuss in detail the multifaceted anti-cancer properties of diosgenin that have found application in pharmaceutical, functional food, and cosmetic industries; and the various intracellular molecular targets modulated by diosgenin that abrogate the oncogenic process. Keywords: diosgenin; steroidal sapogenins; anti-cancer; apoptosis; oncogenic; metastasis Pro-Apoptotic and Anti-Cancer Properties of Diosgenin: A Comprehensive and Critical Review Gautam Sethi 1,2,3,,†, Muthu K. Shanmugam 3,†, Sudha Warrier 4, Myriam Merarchi 3, Frank Arfuso 5, Alan Prem Kumar 3 and Anupam Bishayee 6, 1 Department for Management of Science and Technology Development, Ton Duc Thang University, Ho Chi Minh City 700000, Vietnam 2 Faculty of Pharmacy, Ton Duc Thang University, Ho Chi Minh City 700000, Vietnam 3 Department of Pharmacology, Yong Loo Lin School of Medicine, National University of Singapore, Singapore 117600, Singapore; phcsmk@nus.edu.sg (M.K.S.); myriammerarchi@hotmail.fr (M.M.); phcapk@nus.edu.sg (A.P.K.) 4 Division of Cancer Stem Cells and Cardiovascular Regeneration, Manipal Institute of Regenerative Medicine, Manipal University, Bangalore 560065, India; sudha.warrier@manipal.edu 5 Stem Cell and Cancer Biology Laboratory, School of Biomedical Sciences, Curtin Health Innovation Rese Institute, Curtin University, Perth, WA 6102, Australia; frank.arfuso@curtin.edu.au y 6 Department of Pharmaceutical Sciences, College of Pharmacy, Larkin University, 18301 N. Miami Avenue Miami, FL 33169, USA * Correspondence: gautam.sethi@tdt.edu.vn or phcgs@nus.edu.sg (G.S.); * Correspondence: gautam.sethi@tdt.edu.vn or phcgs@nus.edu.sg (G.S.); abishayee@ULarkin org or abishayee@gmail com (A B ) * Correspondence: gautam.sethi@tdt.edu.vn or phcgs@nus.edu abishayee@ULarkin.org or abishayee@gmail.com (A.B.) p g p g abishayee@ULarkin.org or abishayee@gmail.com (A.B.) † These authors contributed equally to this work. nutrients nutrients www.mdpi.com/journal/nutrients 1. Introduction Cancer is a complex and heterogeneous disease that afﬂicts men and women worldwide; it is expected to increase due to human lifestyle changes and a rapidly aging population [1]. Hanahan Nutrients 2018, 10, 645; doi:10.3390/nu10050645 www.mdpi.com/journal/nutrients 2 of 12 Nutrients 2018, 10, 645 and Weinberg’s proposed hallmarks of cancer are widely accepted towards understanding the biology of cancer cells, through a multi-stage and progressive process [2–6]. These hallmarks include sustained proliferation of cells, constitutive activation of pro-survival transcription factors, deregulated cellular functions, evading cell death signals and growth suppressors, an increased pro-inﬂammatory tumor microenvironment, camouﬂaging against immune cell destruction, promoting angiogenesis, activating cell movement from the primary site and metastasis, enabling replicative immortality, and ﬁnally, severe genome instability [2,3,5–10]. In addition, several deregulated cellular signaling networks underlying the above hallmarks have been extensively investigated in pre-clinical and clinical drug development [3,4,11–18]. In spite of detailed information of these semi-synthetic and synthetic anti-cancer agents, they only provide limited therapeutic advantages to patients due to highly toxic unwanted side effects and the development of chemoresistance [19–21]. Emerging approaches should include the identiﬁcation of novel drug targets that are very effective in inhibiting the growth of cancer cells, while exhibiting fewer adverse effects. Natural products are a good source of compounds with novel chemical structures that are effective and less toxic [18,22–27]. High throughput technologies should be exploited for the screening of a large library of compounds for their anti-cancer activities [28–30]. The mainstream of United States Food and Drug Administration (US FDA)-approved compounds divulge that natural products and their derivatives occupy one-third of all novel drugs [22–24,31]. These natural compounds, in general, show multi-targeted effects and can modulate several oncogenic transcription factors that block the tumor microenvironment targets that usually sustain tumor growth [22,27,32]. These compounds can also be classiﬁed as cytotoxic or cytostatic compounds [30]. Therefore, many secondary metabolites and pure molecules isolated from herbs, spices, dietary fruits and vegetables, and even marine sources have been explored [33–35]. herbs, spices, dietary fruits and vegetables, and even marine sources have been explored [33–35]. Several novel bioactive molecules have been found in various edible cereals, pulses, roots, and in several parts of medicinal plants. Fenugreek (Trigonella foenum-graecum) belongs to the family Leguminosae and is considered a traditional medicinal herb that is commonly used in India, China, Thailand, and South-East Asian countries; it is also cultivated in the Mediterranean region and Northern Africa [36–41]. 1. Introduction Fenugreek seeds, shoots and leaves are used in Indian curry preparation as a condiment. Bibliometric data indicates that fenugreek extract has several pharmacological properties, such as being hypocholesterolemic, a lactation aid, antibacterial, a gastric stimulant, anti-anorexic, antidiabetic, galactogogic, hepatoprotective, and anti-cancer both in in vitro and in vivo studies [40–43]. Fenugreek seed extract contains several bioactive molecules in various classes of compounds, such as saponins, flavonoids, coumarins, and alkaloids that target several molecules involved in inflammation and cancer cell proliferation, invasion, migration, angiogenesis, and metastasis [40]. Sapogenins are a class of compounds that widely occur in natural products in their glycoside form and promote general healthy living. Among these compounds, steroidal sapogenins (otherwise known as spirostans) are the most potent bioactive compounds isolated from natural product sources [44,45]. Steroidal sapogenins exhibit ubiquitous pharmacological properties and the majority of them demonstrate anti-cancer activity in vitro and in pre-clinical animal models. Several clinical trials have been conducted with fenugreek seed extract, they are either completed or ongoing; however, diosgenin anti-cancer trials are yet to start [36,41]. Diosgenin is the most abundant steroidal sapogenin in fenugreek seeds. Fujii and Matsukawa isolated and identified diosgenin from Dioscorea tokoro Makino in 1935 [41,46] (Figure 1). Diosgenin is a phytosteroidal saponin and a major bioactive compound found in the seeds of T. foenum-graecum, commonly known as fenugreek, and in the roots of wild yam (Dioscorea villosa) [36,40, 41,47,48]. Interestingly, diosgenin is also found in high levels in numerous plant species including Costus speciosus, Smilax menispermoidea, species of Paris, Aletris and Trillum, and in species of Dioscorea [41,49,50]. The steroidal saponin, diosgenin, is biosynthesized from cholesterol. Cholesterol is formed from lanosterol and catalysed by the cytochrome P450 system. Several other routes of synthesis have been identified, such as from squalene-2,3-oxide in two ways: From cycloartenol through the formation of sitosterol [51] and from lanosterol via cholesterol [52]. Several studies have demonstrated the diverse biological activities of diosgenin, such as hypolipidemic, anti-inflammatory, anti-proliferative, 3 of 12 uch as 3 of 12 h Nutrients 2018, 10, 645 Several studies h Nutrients 2018, 10, x FO hypoglycemic activity, and as a potent anti-oxidant [44]. In addition, diosgenin inhibited cancer cell proliferation and induced apoptosis in a variety of cancer cell lines including colorectal, hepatocellular, breast, osteosarcoma, and leukemia [53–55]. 1. Introduction The primary mechanism of action of diosgenin is through the modulation of multiple cell signaling pathways that play prominent roles in cell-cycle regulation, differentiation, and apoptosis [56]. Remarkably, pharmaceutical companies use diosgenin as a principal precursor compound for the manufacturing of several steroidal drugs [48]. Diosgenin is also an attractive molecule with multifaceted properties that has found application in pharmaceutical, functional food, and cosmetic industries. In this review we provide an in-depth evaluation of literature on diosgenin and its pharmacodynamics and pharmacokinetics, and discuss several of its novel derivatives and nanoformulations that increase its bioavailability and therapeutic efficacy. Diosgenin, over the years, has provided abundant data on the prevention and treatment of various inflammation-driven diseases, including cancers [36] (Figure 2). yp p y p yp g y y p oxidant [44]. In addition, diosgenin inhibited cancer cell proliferation and induced apoptosis in a variety of cancer cell lines including colorectal, hepatocellular, breast, osteosarcoma, and leukemia [53–55]. The primary mechanism of action of diosgenin is through the modulation of multiple cell signaling pathways that play prominent roles in cell-cycle regulation, differentiation, and apoptosis [56]. Remarkably, pharmaceutical companies use diosgenin as a principal precursor compound for the manufacturing of several steroidal drugs [48]. Diosgenin is also an attractive molecule with multifaceted properties that has found application in pharmaceutical, functional food, and cosmetic industries. In this review we provide an in-depth evaluation of literature on diosgenin and its pharmacodynamics and pharmacokinetics, and discuss several of its novel derivatives and nanoformulations that increase its bioavailability and therapeutic efficacy. Diosgenin, over the years, has provided abundant data on the prevention and treatment of various inflammation-driven diseases, including cancers [36] (Figure 2). hypolipidemic, anti-inflammatory, anti-proliferative, hypoglycemic activity, and as a potent anti- oxidant [44]. In addition, diosgenin inhibited cancer cell proliferation and induced apoptosis in a variety of cancer cell lines including colorectal, hepatocellular, breast, osteosarcoma, and leukemia [53–55]. The primary mechanism of action of diosgenin is through the modulation of multiple cell signaling pathways that play prominent roles in cell-cycle regulation, differentiation, and apoptosis [56]. Remarkably, pharmaceutical companies use diosgenin as a principal precursor compound for the manufacturing of several steroidal drugs [48]. Diosgenin is also an attractive molecule with multifaceted properties that has found application in pharmaceutical, functional food, and cosmetic industries. 1. Introduction In this review we provide an in-depth evaluation of literature on diosgenin and its pharmacodynamics and pharmacokinetics, and discuss several of its novel derivatives and nanoformulations that increase its bioavailability and therapeutic efficacy. Diosgenin, over the years, has provided abundant data on the prevention and treatment of various inflammation-driven diseases, including cancers [36] (Figure 2). Figure 1. Chemical structure of diosgenin. Figure 2. Tumor stage-specific inhibition of molecular targets by diosgenin. ugreek Seed Bioactive Compounds Figure 1. Chemical structure of diosgenin. g g Figure 1. Chemical structure of diosgenin. Figure 2. Tumor stage-specific inhibition of molecular targets by diosgenin. Figure 2. Tumor stage-speciﬁc inhibition of molecular targets by diosgenin. Figure 1. Chemical structure of diosgenin. Figure 1 Chemical structure of diosgenin Figure 1 Chemical structure of diosgenin Figure 1. Chemical structure of diosgenin. Figure 1. Chemical structure of diosgenin. Figure 1. Chemical structure of diosgenin. g g p g y g Figure 2. Tumor stage-specific inhibition of molecular targets by diosgenin. Figure 2. Tumor stage-speciﬁc inhibition of molecular targets by diosgenin. 3. In Vitro Anti-Cancer Effects of Diosgenin 3. In Vitro Anti-Cancer Effects of Diosgenin Diosgenin, the major steroidal sapogenin in the fenugreek seed, has been shown to potently suppress constitutively-activated pro-inﬂammatory and pro-survival signaling pathways in a variety of cancer cells, and induced apoptosis [58]. Some of the earlier studies by Shishodia and Aggarwal [58] reported that diosgenin abrogated TNF-α-induced NF-κB activation and suppressed osteoclastogenesis in RAW 264.7 macrophage cells [58]. In Her-2 positive breast cancer cells, diosgenin inhibited the expression of AKT, mTOR, JNK and their associated pro-survival signaling pathways, and induced apoptosis in these cells [53]. In another study by Li et al., they reported that diosgenin could inactivate the STAT3 signaling pathway in hepatocellular carcinoma (HCC) cells, by inhibiting intracellular signaling molecules such as c-SRC, JAK1, and JAK2 (Figure 3). Diosgenin also suppressed STAT3 transcriptional activity and the expression of its downstream gene products involved in proliferation, invasion and metastasis. In addition, diosgenin sensitized HCC cells to doxorubicin and paclitaxel, and synergistically augmented apoptosis, thereby suggesting that diosgenin is a potential bioactive compound for the treatment of HCC and other cancers [54]. Diosgenin inhibited proliferation, AKT and JNK in a dose- and time-dependent manner and induced caspase-dependent apoptosis in A431 and Hep2 skin squamous cell carcinoma cells [59]. HT-29 colon cancer cells have been reported to be resistant to TRAIL-induced apoptosis. Diosgenin was shown to sensitize the HT-29 colon cancer cell to TRAIL. In addition, it potently suppressed cell proliferation and induced apoptosis by suppressing the p38/MAPK signaling pathway and the overexpression of DR5 [60]. Furthermore, Romero-Hernandez et al. [61] demonstrated that diosgenin-derived thio(seleno)ureas and glycomimetics, bearing a 1,2,3-triazolyl tether on C-3, showed more potent anti-cancer activity against MDA-MB-231 and MCF-7 breast cancer cells, HepG2 hepatocellular carcinoma cells, and induced apoptosis, compared to its parent compound diosgenin [61]. In another study, diosgenin conjugated to methotrexate was found to be more potent in inhibiting the growth of transport-resistant breast cancer cells and dihydrofolate reductase (enzyme involved in DNA synthesis), compared to the parent diosgenin [62]. In chronic myeloid leukemia cells, diosgenin-induced autophagy inhibited the mTOR signaling pathway and induced apoptotic cell death [63]. Diosgenin induced cytotoxicity and signiﬁcantly inhibited the growth and proliferation of MCF-7 breast cancer cells in a dose- and time-dependent manner. Diosgenin was also shown to inhibit N-nitroso-N-methylurea-induced breast cancer in rats [64]. g p Fenugreek contains several chemical 2. Fenugreek Seed Bioactive Compounds 2. Fenugreek Seed Bioactive Compounds g p Fenugreek contains several chemical 2. Fenugreek Seed Bioactive Compounds 2. Fenugreek Seed Bioactive Compounds Fenugreek contains several chemical constituents, such as alkaloids, steroidal sapogenins, saponins, flavonoids, lipids, amino acids, and carbohydrates. Diosgenin is a major bioactive steroidal sapogenin in the fenugreek seed, which is reported to have chemopreventive and therapeutic effects against inflammation and chronic inflammation-driven cancers in preclinical in vitro and in vivo models of cancer [36,57]. Table 1 illustrates the major constituents of fenugreek seeds and leaves. T bl M h h l f f k (T f ) Fenugreek contains several chemical constituents, such as alkaloids, steroidal sapogenins, saponins, flavonoids, lipids, amino acids, and carbohydrates. Diosgenin is a major bioactive steroidal sapogenin in the fenugreek seed, which is reported to have chemopreventive and therapeutic effects against inflammation and chronic inflammation-driven cancers in preclinical in vitro and in vivo models of cancer [36,57]. Table 1 illustrates the major constituents of fenugreek seeds and leaves. Fenugreek contains several chemical constituents, such as alkaloids, steroidal sapogenins, saponins, ﬂavonoids, lipids, amino acids, and carbohydrates. Diosgenin is a major bioactive steroidal sapogenin in the fenugreek seed, which is reported to have chemopreventive and therapeutic effects against inﬂammation and chronic inﬂammation-driven cancers in preclinical in vitro and in vivo models of cancer [36,57]. Table 1 illustrates the major constituents of fenugreek seeds and leaves. 4 of 12 Nutrients 2018, 10, 645 Table 1. Main phytochemical constituents of fenugreek (T. foenum-graecum). Table 1. Main phytochemical constituents of fenugreek (T. foenum-graecum). Class of Compounds Phytochemical Constituents Reference Steroidal sapogenins Diosgenin, Yamogenin, Smilagenin, Sarsasapogenin, Tigogenin, Neotigogenin, Gitogenin, Yuccagenin, Saponaretin [40] Flavonoids Quercetin, Rutin, Vitexin, Isovitexin [40] Saponins Graecunins, Fenugrin B, Fenugreekine, Trigofoenosides A–G [40] Alkaloids Trimethylamine, Neurin, Trigonelline, Choline, Gentianine, Carpaine, and Betain [40] Fibers Gum, Neutral detergent ﬁber [40] Lipids Lipids, Triacylglycerols, Diacylglycerols, Monoacylglycerols, Phosphatidylcholine, Phosphatidylethanolamine, Phosphatidylinositol, Free fatty acids [40] Others Coumarin, Amino acids, Vitamins, Minerals. 28% Mucilage; 22% Proteins; 5% of a stronger swelling, Bitter ﬁxed oil [40] Phytochemical Constituents 3. In Vitro Anti-Cancer Effects of Diosgenin Of several factors that contribute to the sustained growth and proliferation of tumors, one important factor is the abundant neovascularization or formation of new micro blood vessels at the tumor site, or within the tumor. This process is also known as tumor angiogenesis. Thus, the formation of new blood vessels in tumors actively supplies the essential nutrients and growth factors that allow tumors to 5 of 12 Nutrients 2018, 10, 645 acquire the ability to reject chemotherapeutic drugs and develop chemoresistance [27,65–68]. Diosgenin is also a potent inhibitor of cancer cell invasion, migration, and tumor-associated angiogenesis [27,69]. He et al., reported that diosgenin inhibited the invasion and migration of triple-negative breast cancer cells and was associated with the concomitant suppression of actin polymerization, phosphorylation of Vav2, and activation of Cdc42 oncoprotein expression. These proteins have been shown to be involved in the initiation of cancer cells’ invasive and migratory potential [70]. Similarly, diosgenin was found to inhibit PC3 androgen-independent prostate cancer cell invasion and migration. The inhibitory effect was mediated by the downregulation of matrix metalloproteinase (MMP)-2 and MMP-9, the key enzymes in matrix degradation and stroma invasion. Furthermore, diosgenin also downregulated the tissue inhibitors of metalloproteinase (TIMP)-2, vascular endothelial growth factor (VEGF), extracellular regulated kinase (ERK), Janus kinase (JNK), phosphotidyl-inositol-3 kinase/protein kinase B (PI3K/AKT), and NF-κB transcriptional activity [56]. In addition, diosgenin was reported to inhibit the expression of E-cadherin, integrin 5a and 6b, invasion, migration, and angiogenesis in hypoxia-sensitive BGC-823 gastric cancer cells [71]. Diosgenin was reported to inhibit proliferation of ER-positive MCF-7 breast cancer cells by the upregulation of the p53 tumor suppressor gene and activation of caspase 3, while it downregulated BCL2 in ER negative MDA-MB-231 triple-negative breast cancer cells [72]. Diosgenin, either alone, or in combination with thymoquinone, inhibited A431 and Hep2 squamous cell carcinoma cell proliferation, increased the Bax/Bcl2 ratio, and induced caspase 3-mediated apoptosis [59] (Table 2). The potential effect of diosgenin on NF-κB and STAT3 signaling pathways in tumor cells, is shown in Figure 3. Nutrients 2018, 10, x FOR PEER REVIEW 5 of 12 suppression of actin polymerization, phosphorylation of Vav2, and activation of Cdc42 oncoprotein expression. These proteins have been shown to be involved in the initiation of cancer cells’ invasive and migratory potential [70]. Similarly, diosgenin was found to inhibit PC3 androgen-independent prostate cancer cell invasion and migration. 3. In Vitro Anti-Cancer Effects of Diosgenin The inhibitory effect was mediated by the downregulation of matrix metalloproteinase (MMP)-2 and MMP-9, the key enzymes in matrix degradation and stroma invasion. Furthermore, diosgenin also downregulated the tissue inhibitors of metalloproteinase (TIMP)-2, vascular endothelial growth factor (VEGF), extracellular regulated kinase (ERK), Janus kinase (JNK), phosphotidyl-inositol-3 kinase/protein kinase B (PI3K/AKT), and NF-κB transcriptional activity [56]. In addition, diosgenin was reported to inhibit the expression of E-cadherin, integrin 5a and 6b, invasion, migration, and angiogenesis in hypoxia-sensitive BGC-823 gastric cancer cells [71]. Diosgenin was reported to inhibit proliferation of ER-positive MCF-7 breast cancer cells by the upregulation of the p53 tumor suppressor gene and activation of caspase 3, while it downregulated BCL2 in ER negative MDA-MB-231 triple-negative breast cancer cells [72]. Diosgenin, either alone, or in combination with thymoquinone, inhibited A431 and Hep2 squamous cell carcinoma cell proliferation, increased the Bax/Bcl2 ratio, and induced caspase 3-mediated apoptosis [59] (Table 2). The potential effect of diosgenin on NF-κB and STAT3 signaling pathways in tumor cells, is shown in Figure 3. Figure 3. Role of diosgenin in NF-κB and STAT3 signaling pathways. Diosgenin abrogates TNF-α - induced activation of NF-κB and IL6-induced STAT3 signaling pathways in tumor cells. Diosgenin can hence prevent proliferation, invasion and angiogenesis; and induce apoptosis, a characteristic vastly looked for in cancer therapy. Figure 3. Role of diosgenin in NF-κB and STAT3 signaling pathways. Diosgenin abrogates TNF-α -induced activation of NF-κB and IL6-induced STAT3 signaling pathways in tumor cells. Diosgenin can hence prevent proliferation, invasion and angiogenesis; and induce apoptosis, a characteristic vastly looked for in cancer therapy. Figure 3. Role of diosgenin in NF-κB and STAT3 signaling pathways. Diosgenin abrogates TNF-α - induced activation of NF-κB and IL6-induced STAT3 signaling pathways in tumor cells. Diosgenin can hence prevent proliferation, invasion and angiogenesis; and induce apoptosis, a characteristic vastly looked for in cancer therapy. Figure 3. Role of diosgenin in NF-κB and STAT3 signaling pathways. Diosgenin abrogates TNF-α -induced activation of NF-κB and IL6-induced STAT3 signaling pathways in tumor cells. Diosgenin can hence prevent proliferation, invasion and angiogenesis; and induce apoptosis, a characteristic vastly looked for in cancer therapy. 6 of 12 Nutrients 2018, 10, 645 Table 2. In vitro anti-cancer effects of diosgenin. Table 2. In vitro anti-cancer effects of diosgenin. 3. In Vitro Anti-Cancer Effects of Diosgenin Cancer model Cell Lines Diosgenin Dose Molecular Target References Breast carcinoma Estrogen receptor positive and estrogen receptor negative human breast cancer MCF-7 and MDA 231 cells 20 µM and 30 µM Inhibition of cell proliferation Induces apoptosis [72] MDA-MB-231 breast cancer cells 20 µM, 40 µM, and 60 µM Downregulation of Bcl2 [57] Her2 over-expressing breast cancer cells 5–20 µM Modulation of Akt, mTOR, and JNK phosphorylation [53] MCF-7 breast cancer cells 20 µM and 40 µM Upregulation of p53 tumor suppressor gene [73] Hepatocellular carcinoma C3A, HUH-7, and HepG2 cells 50 µM and 100 µM Downregulation of STAT3 signaling pathway Upregulates SH-PTP2 expression Induces apoptosis Potentiates the apoptotic effects of doxorubicin and paclitaxel [54] Prostate carcinoma PC3 cells 5 µM, 10 µM, and 20 µM Downregulates NF-κB signaling pathway Inhibits matrix metalloproteinases Inhibits invasion and migration of cells [56] Osteosarcoma 1547 cells 40 µM, 80 µM, and 100 µM Inhibits cell proliferation Induces apoptosis [55] 1547 cells 40 µM Inhibits cell proliferation Induces apoptosis Upregulation of p53 tumor suppressor gene [74] Human erythroleukemia HEL cells, K562 cells 40 µM Inhibits NF-κB signaling pathway [75] HEL cells 40 µM Inhibits proliferation Induces apoptosis Upregulation of p21 [76] Human Laryngocarcinoma Human Melanoma HEp-2 cells M4Beu cells 40 µM Inhibits cell proliferation Induces caspase-3 dependent apoptosis Upregulates p53 tumor suppressor gene [77] Human cancer cells Human epithelial carcinoma cell line (A431), human NSCLC cell line (A549), human ovarian cancer cell line (A2780), Human erythroleukemia (K562) and Dukes’ type C, colorectal adenocarcinoma (HCT-15) 10 mmol/L Induces apoptosis via mitochondrial dependent pathway [78] Multiple myeloma (U266), leukemia (U937), and breast cancer (MCF-7) 50 µM and 100 µM Inhibits NF-κB signaling pathway [58] Nutrients 2018, 10, 645 7 of 12 7 of 12 4. In Vivo Anti-Cancer Effects of Diosgenin In addition to in vitro inhibition of cancer cell proliferation by dietary fenugreek seeds and its bioactive constituent diosgenin, several studies have provided evidence that diosgenin is a potent inhibitor of tumor growth in vivo in rodent models of cancer. In a rat colorectal tumor model, administration of diosgenin, given during the promotional stage, reduced azoxymethane (AOM)-induced colonic aberrant crypt foci formation [79]. Similarly, Malisetty et al., showed that diosgenin, at a dose of 15 mg/kg, signiﬁcantly suppressed both the incidence and invasive potential of AOM-induced rat colon adenocarcinoma mass by 60% and colon tumor multiplicity (adenocarcinomas/rat) by 68% [80]. However, in the murine model of AOM/dextran sodium sulfate-induced colon aberrant crypt foci, diosgenin at doses of 20, 100 and 200 mg/kg b.w. in the diet did not reduce adenocarcinoma mass; nonetheless, a signiﬁcant reduction in tumor multiplicity was observed with all three doses tested [81]. In another study, diosgenin (at a dose of 10 mg/kg b.w. administered intra-tumorally) signiﬁcantly inhibited the growth of MCF-7 and MDA-MB-231 human breast cancer xenografts in mice [72]. In another study using inbred T739 mice, diosgenin was shown to signiﬁcantly inhibit the growth of mouse LA795 lung adenocarcinoma tumors by 33.94% [82]. Diosgenin, at a dose of 80 mg/kg administered by oral gavage, was reported to inhibit the growth of oral tumors in a DMBA-induced hamster buccal pouch model [64]. Diosgenin, in combination with thymoquinone, exhibited signiﬁcant tumor growth inhibition in a mice xenograft model [59]. Therefore, diosgenin modulates multiple targets and suppresses tumor growth in preclinical models of cancer. However, diosgenin’s poor solubility in organic solvents and its lack of bioavailability greatly hinder its translational process as a therapeutic compound. Further clinical trials are required to evaluate its potential either as a preventive or therapeutic anti-cancer agent. 5. Semisynthetic Derivatives of Diosgenin That Exhibit Anti-Cancer Activity Diosgenin is used in the pharmaceutical industry as the main precursor in the synthesis of steroids [83]. It has the ability to penetrate cell membranes and bind to specific receptors [84]. Steroidal sapogenins are bioactive molecules that have shown exceptional antiproliferative activity against several human cancer cells. By making specific changes in the steroidal structure of diosgenin, it can affect its biological activity. In a recent report, using diosgenin as the parent molecule, the authors synthesized two novel steroidal oxime compounds that showed significant antiproliferative activity on cervical cancer cells and human lymphocytes. These compounds induced apoptosis and activated caspase 3 [85]. In another study Mohammad et al., reported on the anti-proliferative activity of diosgenin and its semi-synthetic derivatives against breast (HBL-100), colon (HCT-116 and HT-19) and lung (A549) cancer cells. A structure-activity relationship study revealed that the potent anti-proliferative activity was mainly attributed to the analogs with the simple phenyl R moiety or electron-withdrawing ortho-substituted R moieties attached to the parent diosgenin [86]. In another study, diosgenin was used as a parent compound to synthesize 1α-hydroxysolasodine; it showed significant anti-cancer activity against prostate cancer (PC3), cervical carcinoma (HeLa), and hepatocellular carcinoma (HepG2) cells [87]. Twelve different analogs of diosgenin containing a long chain fatty acid/ester of diosgenin-7-ketoxime exhibited anti-cancer activity when tested against a panel of cancer cell lines. Compound 16 in this series exhibited potent anti-proliferative activity against DU145 prostate cancer cells, which was associated to the suppression of lipopolysaccharide-induced activation of TNF-α and IL6. The compound was also identified as safe, with a maximum tolerated dose of 300 mg/kg in Swiss albino mice [88]. In a recent article by Ghosh et al., they reported the synthesis of diosgenin functionalized iron oxide nanoparticles that exhibited anti-breast cancer activity by inhibiting proliferation and migration, and by inducing apoptosis [89]. Nutrients 2018, 10, 645 8 of 12 6. Conclusions and Future Perspectives Compounds derived either from medicinal or dietary plant sources embrace distinct advantages, such as novel bioactive structures, low toxicity, and being multi-targeted in abrogating oncogenic processes; thereby, they may form the source of improved therapeutic options. A vast body of pre- clinical experimental evidence suggests that diosgenin has great potential as an anti-cancer agent. In this review we have compiled and analyzed the role of diosgenin in modulating various oncogenic transcription factors and intracellular molecular targets that drive tumor initiation, progression and metastasis. It is well known that the majority of cancers are a consequence of chronic inﬂammation, infection, dysfunctional cell death mechanisms, and deregulation of cell cycle molecules. The ability of diosgenin to prevent carcinogenesis by acting as an anti-oxidant and anti-inﬂammatory agent, and its ability to induce apoptosis of cancer cells, suggests that it can be useful as an anti-carcinogenic agent. Due to the complexities in the cellular processes involved, several new studies need to be conducted to decipher the exact molecular targets that can be exploited to prevent cancer progression. Interestingly, there are 12 reported clinical trials on fenugreek seed extract on a variety of human ailments, as reported in www.clinicaltrials.gov. However, to date, there are no cancer-related clinical trials reported either on diosgenin or on fenugreek seed extract. Several novel synthetic diosgenin derivatives have been shown to improve its anti-cancer efﬁcacy. Several nano-formulations and delivery systems of diosgenin are also shown to improve its bioavailability. In conclusion, several challenges such as developing novel delivery systems, pharmaceutical formulations, and semi-synthetic derivatives that are water soluble, need to be overcome to uncover diosgenin’s beneﬁts either as a chemopreventive or therapeutic agent. Author Contributions: G.S. and M.K.S. wrote the paper, F.A., A.P.K., S.W., M.M. and A.B. critically analyzed and revised the manuscript. Acknowledgments: A.P.K. was supported by grants from National Medical Research Council of Singapore, Medical Science Cluster, Yong Loo Lin School of Medicine, National University of Singapore and by the National Research Foundation Singapore and the Singapore Ministry of Education under its Research Centers of Excellence initiative to Cancer Science Institute of Singapore, National University of Singapore. Conﬂicts of Interest: The authors declare no conﬂict of interest. Acknowledgments: A.P.K. was supported by grants from National Medical Research Council of Singapore, Medical Science Cluster, Yong Loo Lin School of Medicine, National University of Singapore and by the National Research Foundation Singapore and the Singapore Ministry of Education under its Research Centers of Excellence initiative to Cancer Science Institute of Singapore, National University of Singapore. Author Contributions: G.S. and M.K.S. wrote the paper, F.A., A.P.K., S.W., M.M. and A.B. critically analyzed and revised the manuscript. References 1. Siegel, R.L.; Miller, K.D.; Jemal, A. Cancer statistics, 2017. CA Cancer J. Clin. 2017, 67, 7–30. [CrossRef] [PubMed] 1. Siegel, R.L.; Miller, K.D.; Jemal, A. Cancer statistics, 2017. CA Cancer J. Clin. 2017, 67, 7–30. [CrossRef] [PubMed] . Hanahan, D.; Weinberg, R.A. The hallmarks of cancer. Cell 2000, 100, 57–70. [CrossRef] 3. Hanahan, D.; Weinberg, R.A. Hallmarks of cancer: The next generation. Cell 2011, 144, 646–674. [CrossRef] [PubMed] 3. Hanahan, D.; Weinberg, R.A. Hallmarks of cancer: The next generation. Cell 2011, 144, 646–674. [CrossRef] [PubMed] 4. Hanahan, D. Rethinking the war on cancer. Lancet 2014, 383, 558–563. [CrossRef] 4. Hanahan, D. Rethinking the war on cancer. Lancet 2014, 383, 558–563. [CrossRef] 5. Floor, S.L.; Dumont, J.E.; Maenhaut, C.; Raspe, E. Hallmarks of cancer: Of all cancer cells, all the time? Trends Mol. Med. 2012, 18, 509–515. [CrossRef] [PubMed] 5. Floor, S.L.; Dumont, J.E.; Maenhaut, C.; Raspe, E. Hallmarks of cancer: Of all cancer cells, all the time? Trends Mol. Med. 2012, 18, 509–515. [CrossRef] [PubMed] 6. Pietras, K.; Ostman, A. Hallmarks of cancer: Interactions with the tumor stroma. Exp. Cell Res. 2010, 316, 1324–1331. [CrossRef] [PubMed] Massague, J.; Obenauf, A.C. Metastatic colonization by circulating tumour cells. Nature 2016, 529, 298–306 [CrossRef] [PubMed] 7. Massague, J.; Obenauf, A.C. Metastatic colonization by circulating tumour cells. Nature 2016, 529, 298–306. [CrossRef] [PubMed] 8. Chaffer, C.L.; Weinberg, R.A. How does multistep tumorigenesis really proceed? Cancer Discov. 2015, 5, [CrossRef] [PubMed] 8. Chaffer, C.L.; Weinberg, R.A. How does multistep tumorigenesis really proceed? Cancer Discov. 2015, 5, 22 24 [C R f] [P bM d] Chaffer, C.L.; Weinberg, R.A. How does multistep tumorigenesis really proceed? Cancer Discov. 2015, 5 22–24. [CrossRef] [PubMed] 9. Iwatsuki, M.; Mimori, K.; Yokobori, T.; Ishi, H.; Beppu, T.; Nakamori, S.; Baba, H.; Mori, M. Epithelial- mesenchymal transition in cancer development and its clinical significance. Cancer Sci. 2010, 101, 293–299. [CrossRef] [PubMed] 10. Almendro, V.; Marusyk, A.; Polyak, K. Cellular heterogeneity and molecular evolution in cancer. Annu. Rev. Pathol. 2013, 8, 277–302. [CrossRef] [PubMed] 11. Steeg, P.S. Targeting metastasis. Nat. Rev. Cancer 2016, 16, 201–218. [CrossRef] [PubMed] 9 of 12 Nutrients 2018, 10, 645 12. Wang, Z.; Dabrosin, C.; Yin, X.; Fuster, M.M.; Arreola, A.; Rathmell, W.K.; Generali, D.; Nagaraju, G.P.; El-Rayes, B.; Ribatti, D.; et al. Broad targeting of angiogenesis for cancer prevention and therapy. Semin. Cancer Biol. 2015, 35, S224–S243. [CrossRef] [PubMed] 13. Sethi, G.; Ahn, K.S.; Aggarwal, B.B. References Targeting nuclear factor-kappa B activation pathway by thymoquinone: Role in suppression of antiapoptotic gene products and enhancement of apoptosis. Mol. Cancer Res. 2008, 6, 1059–1070. [CrossRef] [PubMed] 14. Tang, C.H.; Sethi, G.; Kuo, P.L. Novel medicines and strategies in cancer treatment and prevention. BioMed Res. Int. 2014, 2014, 474078. [CrossRef] [PubMed] 15. Ahn, K.S.; Sethi, G.; Aggarwal, B.B. Nuclear factor-kappa B: From clone to clinic. Curr. Mol. Med. 2007, 7, 619–637. [CrossRef] [PubMed] 16. Chai, E.Z.; Shanmugam, M.K.; Arfuso, F.; Dharmarajan, A.; Wang, C.; Kumar, A.P.; Samy, R.P.; Lim, L.H.; Wang, L.; Goh, B.C.; et al. Targeting transcription factor stat3 for cancer prevention and therapy. Pharmacol. Ther. 2016, 162, 86–97. [CrossRef] [PubMed] 17. Li, F.; Zhang, J.; Arfuso, F.; Chinnathambi, A.; Zayed, M.E.; Alharbi, S.A.; Kumar, A.P.; Ahn, K.S.; Sethi, G. NF-kappaB in cancer therapy. Arch. Toxicol. 2015, 89, 711–731. [CrossRef] [PubMed] 18. Shanmugam, M.K.; Kannaiyan, R.; Sethi, G. Targeting cell signaling and apoptotic pathways by dietary agents: Role in the prevention and treatment of cancer. Nutr. Cancer 2011, 63, 161–173. [CrossRef] [PubMed] 19. Deorukhkar, A.; Krishnan, S.; Sethi, G.; Aggarwal, B.B. Back to basics: How natural products can provide the basis for new therapeutics. Expert Opin. Investig. Drugs 2007, 16, 1753–1773. [CrossRef] [PubMed] 20. Kneller, R. The importance of new companies for drug discovery: Origins of a decade of new drugs. Nat. Rev. Drug Discov. 2010, 9, 867–882. [CrossRef] [PubMed] 21. Bishayee, A.; Sethi, G. Bioactive natural products in cancer prevention and therapy: Progress and promise. Semin. Cancer Biol. 2016, 40–41, 1–3. [CrossRef] [PubMed] 22. Newman, D.J.; Cragg, G.M. Natural products as sources of new drugs from 1981 to 2014. J. Nat. Prod. 2016, 79, 629–661. [CrossRef] [PubMed] 23. Newman, D.J. Developing natural product drugs: Supply problems and how they have been overcome. Pharmacol. Ther. 2016, 162, 1–9. [CrossRef] [PubMed] 4. Newman, D.J. Bioactive cyclic molecules and drug design. Expert Opin. Drug Discov. 2018, 13, 379– [CrossRef] [PubMed] 25. Shanmugam, M.K.; Lee, J.H.; Chai, E.Z.; Kanchi, M.M.; Kar, S.; Arfuso, F.; Dharmarajan, A.; Kumar, A.P.; Ramar, P.S.; Looi, C.Y.; et al. Cancer prevention and therapy through the modulation of transcription factors by bioactive natural compounds. Semin. Cancer Biol. 2016, 40–41, 35–47. [CrossRef] [PubMed] 26. Shanmugam, M.K.; Nguyen, A.H.; Kumar, A.P.; Tan, B.K.; Sethi, G. Targeted inhibition of tumor proliferation, survival, and metastasis by pentacyclic triterpenoids: Potential role in prevention and therapy of cancer. Cancer Lett. 2012, 320, 158–170. References [CrossRef] [PubMed] 27. Shanmugam, M.K.; Warrier, S.; Kumar, A.P.; Sethi, G.; Arfuso, F. Potential role of natural compounds as anti-angiogenic agents in cancer. Curr. Vasc. Pharmacol. 2017, 15, 503–519. [CrossRef] [PubMed] 28. Mishra, K.P.; Ganju, L.; Sairam, M.; Banerjee, P.K.; Sawhney, R.C. A review of high throughput technology for the screening of natural products. Biomed. Pharmacother. 2008, 62, 94–98. [CrossRef] [PubMed] 29. Cragg, G.M.; Grothaus, P.G.; Newman, D.J. Impact of natural products on developing new anti-cancer agents. Chem. Rev. 2009, 109, 3012–3043. [CrossRef] [PubMed] 29. Cragg, G.M.; Grothaus, P.G.; Newman, D.J. Impact of natural products on developing new anti-cancer agents. Chem. Rev. 2009, 109, 3012–3043. [CrossRef] [PubMed] 30. Harvey, A.L.; Edrada-Ebel, R.; Quinn, R.J. The re-emergence of natural products for drug discovery in the genomics era. Nat. Rev. Drug Discov. 2015, 14, 111–129. [CrossRef] [PubMed] 30. Harvey, A.L.; Edrada-Ebel, R.; Quinn, R.J. The re-emergence of natural products for drug discovery in the genomics era. Nat. Rev. Drug Discov. 2015, 14, 111–129. [CrossRef] [PubMed] 31. Newman, D.J. Natural products as leads to potential drugs: An old process or the new hope for drug discovery? J. Med. Chem. 2008, 51, 2589–2599. [CrossRef] [PubMed] 31. Newman, D.J. Natural products as leads to potential drugs: An old process or the new hope for drug discovery? J. Med. Chem. 2008, 51, 2589–2599. [CrossRef] [PubMed] 32. Chai, E.Z.; Siveen, K.S.; Shanmugam, M.K.; Arfuso, F.; Sethi, G. Analysis of the intricate relationship between chronic inﬂammation and cancer. Biochem. J. 2015, 468, 1–15. [CrossRef] [PubMed] 32. Chai, E.Z.; Siveen, K.S.; Shanmugam, M.K.; Arfuso, F.; Sethi, G. Analysis of the intricate relationship between chronic inﬂammation and cancer. Biochem. J. 2015, 468, 1–15. [CrossRef] [PubMed] 33. Sung, B.; Prasad, S.; Yadav, V.R.; Aggarwal, B.B. Cancer cell signaling pathways targeted by spice-derived nutraceuticals. Nutr. Cancer 2012, 64, 173–197. [CrossRef] [PubMed] 33. Sung, B.; Prasad, S.; Yadav, V.R.; Aggarwal, B.B. Cancer cell signaling pathways targeted by spice-derived nutraceuticals. Nutr. Cancer 2012, 64, 173–197. [CrossRef] [PubMed] 34. Newman, D.J.; Cragg, G.M. Current status of marine-derived compounds as warheads in anti-tumor drug candidates. Mar. Drugs 2017, 15, 99. [CrossRef] [PubMed] 34. Newman, D.J.; Cragg, G.M. Current status of marine-derived compounds as warheads in anti-tumor drug candidates. Mar. Drugs 2017, 15, 99. [CrossRef] [PubMed] 35. Newman, D.J.; Cragg, G.M. Drugs and drug candidates from marine sources: An assessment of the current “state of play”. Planta Med. 2016, 82, 775–789. [CrossRef] [PubMed] 35. References Newman, D.J.; Cragg, G.M. Drugs and drug candidates from marine sources: An assessment of the current “state of play”. Planta Med. 2016, 82, 775–789. [CrossRef] [PubMed] 10 of 12 10 of 12 Nutrients 2018, 10, 645 36. El Bairi, K.; Ouzir, M.; Agnieszka, N.; Khalki, L. Anticancer potential of trigonella foenum graecum: Cellular and molecular targets. Biomed. Pharmacother. 2017, 90, 479–491. [CrossRef] [PubMed] 37. Shabbeer, S.; Sobolewski, M.; Anchoori, R.K.; Kachhap, S.; Hidalgo, M.; Jimeno, A.; Davidson, N.; Carducci, M.A.; Khan, S.R. Fenugreek: A naturally occurring edible spice as an anticancer agent. Cancer Biol. Ther. 2009, 8, 272–278. [CrossRef] [PubMed] 38. Aggarwal, B.B.; Kunnumakkara, A.B.; Harikumar, K.B.; Tharakan, S.T.; Sung, B.; Anand, P. Potential of spice-derived phytochemicals for cancer prevention. Planta Med. 2008, 74, 1560–1569. [CrossRef] [PubMed] 39. Kenny, O.; Smyth, T.J.; Hewage, C.M.; Brunton, N.P. Antioxidant properties and quantitative UPLC-MS analysis of phenolic compounds from extracts of Fenugreek (Trigonella foenum-graecum) seeds and bitter melon (Momordica charantia) fruit. Food Chem. 2013, 141, 4295–4302. [CrossRef] [PubMed] 40. Wani, S.A.; Kumar, P. Fenugreek: A review on its nutraceutical properties and utilization in various food products. J. Saudi Soc. Agric. Sci. 2018, 17, 97–106. [CrossRef] 41. Kim, J.K.; Park, S.U. An update on the biological and pharmacological activities of diosgenin. EXCLI J. 2018, 17, 24–28. [PubMed] 42. Bahmani, M.; Shirzad, H.; Mirhosseini, M.; Mesripour, A.; Raﬁeian-Kopaei, M. A review on ethnobotanical and therapeutic uses of Fenugreek (Trigonella foenum-graceum L.). J. Evid. Based Complement. Altern. Med. 2016, 21, 53–62. [CrossRef] [PubMed] 43. Srinivasan, K. Fenugreek (Trigonella foenum-graecum): A review of health beneﬁcial physiological effects. Food Rev. Int. 2006, 22, 203–224. [CrossRef] 44. Jesus, M.; Martins, A.P.; Gallardo, E.; Silvestre, S. Diosgenin: Recent highlights on pharmacology and analytical methodology. J. Anal. Methods Chem. 2016, 2016, 4156293. [CrossRef] [PubMed] 45. Yang, S.F.; Weng, C.J.; Sethi, G.; Hu, D.N. Natural bioactives and phytochemicals serve in cancer treatment and prevention. Evid. Based Complement. Altern. Med. 2013, 2013, 698190. [CrossRef] [PubMed] 46. Djerassi, C.; Rosenkranz, G.; Pataki, J.; Kaufmann, S. Steroids, xxvii. Synthesis of allopregnane-3beta, 11beta, 17alpha-, 20beta, 21-pentol from cortisone and diosgenin. J. Biol. Chem. 1952, 194, 115–118. [PubMed] 47. Patel, K.; Gadewar, M.; Tahilyani, V.; Patel, D.K. A review on pharmacological and analytical aspects of diosmetin: A concise report. Chin. J. Integr. Med. 2013, 19, 792–800. [CrossRef] [PubMed] 47. Patel, K.; Gadewar, M.; Tahilyani, V.; Patel, D.K. References A review on pharmacological and analytical aspects of diosmetin: A concise report. Chin. J. Integr. Med. 2013, 19, 792–800. [CrossRef] [PubMed] 48. Chen, Y.; Tang, Y.M.; Yu, S.L.; Han, Y.W.; Kou, J.P.; Liu, B.L.; Yu, B.Y. Advances in the pharmacological ti iti d h i f di i Chi J N t M d 2015 13 578 587 [C R f] 48. Chen, Y.; Tang, Y.M.; Yu, S.L.; Han, Y.W.; Kou, J.P.; Liu, B.L.; Yu, B.Y. Advances in the pharmac activities and mechanisms of diosgenin. Chin. J. Nat. Med. 2015, 13, 578–587. [CrossRef] 49. Tang, Y.; Yi, T.; Chen, H.; Zhao, Z.; Liang, Z.; Chen, H. Quantitative comparison of multiple components in dioscorea nipponica and d. Panthaica by ultra-high performance liquid chromatography coupled with quadrupole time-of-ﬂight mass spectrometry. Phytochem. Anal. 2013, 24, 413–422. [CrossRef] [PubMed] 50. Tang, Y.N.; Pang, Y.X.; He, X.C.; Zhang, Y.Z.; Zhang, J.Y.; Zhao, Z.Z.; Yi, T.; Chen, H.B. UPLC-QTOF-MS identiﬁcation of metabolites in rat biosamples after oral administration of dioscorea saponins: A comparative study. J. Ethnopharmacol. 2015, 165, 127–140. [CrossRef] [PubMed] 51. Ciura, J.; Szeliga, M.; Grzesik, M.; Tyrka, M. Next-generation sequencing of representational difference analysis products for identiﬁcation of genes involved in diosgenin biosynthesis in Fenugreek (Trigonella foenum-graecum). Planta 2017, 245, 977–991. [CrossRef] [PubMed] 52. Vaidya, K.; Ghosh, A.; Kumar, V.; Chaudhary, S.; Srivastava, N.; Katudia, K.; Tiwari, T.; Chikara, S.K. De novo transcriptome sequencing in Trigonella foenum-graecum L. to identify genes involved in the biosynthesis of diosgenin. Plant. Genome 2013, 6, 1–11. [CrossRef] 53. Chiang, C.T.; Way, T.D.; Tsai, S.J.; Lin, J.K. Diosgenin, a naturally occurring steroid, suppresses fatty acid synthase expression in HER2-overexpressing breast cancer cells through modulating Akt, mTOR and JNK phosphorylation. FEBS Lett. 2007, 581, 5735–5742. [CrossRef] [PubMed] 54. Li, F.; Fernandez, P.P.; Rajendran, P.; Hui, K.M.; Sethi, G. Diosgenin, a steroidal saponin, inhibits STAT3 signaling pathway leading to suppression of proliferation and chemosensitization of human hepatocellular carcinoma cells. Cancer Lett. 2010, 292, 197–207. [CrossRef] [PubMed] 55. Moalic, S.; Liagre, B.; Corbiere, C.; Bianchi, A.; Dauca, M.; Bordji, K.; Beneytout, J.L. A plant steroid, diosgenin, induces apoptosis, cell cycle arrest and cox activity in osteosarcoma cells. FEBS Lett. 2001, 506, 225–230. [CrossRef] 56. Chen, P.S.; Shih, Y.W.; Huang, H.C.; Cheng, H.W. Diosgenin, a steroidal saponin, inhibits migration and invasion of human prostate cancer pc-3 cells by reducing matrix metalloproteinases expression. PLoS ONE 2011, 6, e20164. References [CrossRef] [PubMed] 11 of 12 Nutrients 2018, 10, 645 57. Raju, J.; Rao, C.V. Diosgenin, a steroid saponin constituent of yams and fenugreek: Emerging evidence for applications. In Bioactive Compounds Medicine in Phytomedicine; Rasooli, I., Ed.; InTech.: London, UK, 2012; pp. 125–142. 58. Shishodia, S.; Aggarwal, B.B. Diosgenin inhibits osteoclastogenesis, invasion, and proliferation through the downregulation of Akt, I kappa B kinase activation and NF-kappa B-regulated gene expression. Oncogene 2006, 25, 1463–1473. [CrossRef] [PubMed] 59. Das, S.; Dey, K.K.; Dey, G.; Pal, I.; Majumder, A.; MaitiChoudhury, S.; kundu, S.C.; Mandal, M. Antineoplastic and apoptotic potential of traditional medicines thymoquinone and diosgenin in squamous cell carcinoma. PLoS ONE 2012, 7, e46641. [CrossRef] [PubMed] 60. Lepage, C.; Leger, D.Y.; Bertrand, J.; Martin, F.; Beneytout, J.L.; Liagre, B. Diosgenin induces death receptor-5 through activation of p38 pathway and promotes trail-induced apoptosis in colon cancer cells. Cancer Lett. 2011, 301, 193–202. [CrossRef] [PubMed] 61. Romero-Hernandez, L.L.; Merino-Montiel, P.; Montiel-Smith, S.; Meza-Reyes, S.; Vega-Baez, J.L.; Abasolo, I.; Schwartz, S., Jr.; Lopez, O.; Fernandez-Bolanos, J.G. Diosgenin-based thio(seleno)ureas and triazolyl glycoconjugates as hybrid drugs. Antioxidant and antiproliferative proﬁle. Eur. J. Med. Chem. 2015, 99, 67–81. [CrossRef] [PubMed] 62. Cai, B.; Liao, A.; Lee, K.K.; Ban, J.S.; Yang, H.S.; Im, Y.J.; Chun, C. Design, synthesis of methotrexate-diosgenin conjugates and biological evaluation of their effect on methotrexate transport-resistant cells. Steroids 2016, 116, 45–51. [CrossRef] [PubMed] 63. Jiang, S.; Fan, J.; Wang, Q.; Ju, D.; Feng, M.; Li, J.; Guan, Z.B.; An, D.; Wang, X.; Ye, L. Diosgenin induces ros-dependent autophagy and cytotoxicity via mtor signaling pathway in chronic myeloid leukemia cells. Phytomedicine 2016, 23, 243–252. [CrossRef] [PubMed] 64. Jagadeesan, J.; Langeswaran, K.; Gowthamkumar, S.; Balasubramanian, M.P. Diosgenin exhibits beneﬁcial efﬁciency on human mammary carcinoma cell line MCF-7 and against N-nitroso-N-methylurea (NMU) induced experimental mammary carcinoma. Biomed. Prev. Nutr. 2013, 3, 381–388. [CrossRef] 65. Albini, A.; Tosetti, F.; Li, V.W.; Noonan, D.M.; Li, W.W. Cancer prevention by targeting angiogenesis. Nat. Rev. Clin. Oncol. 2012, 9, 498–509. [CrossRef] [PubMed] 6. Folkman, J. Tumor angiogenesis: Therapeutic implications. N. Engl. J. Med. 1971, 285, 1182–1186. [PubM 66. Folkman, J. Tumor angiogenesis: Therapeutic implications. N. Engl. J. Med. 1971, 285, 1182–1186. [PubMed] 67. Folkman, J. Role of angiogenesis in tumor growth and metastasis. Semin. Oncol. 2002, 29, 15–18. [CrossRef] [PubMed] 67. Folkman, J. Role of angiogenesis in tumor growth and metastasis. Semin. Oncol. 2002, 29, 15–18. [CrossRef] [PubMed] 68. Folkman, J. References Angiogenesis: An organizing principle for drug discovery? Nat. Rev. Drug Discov. 2007, 6, 273–286. [CrossRef] [PubMed] 69. Selim, S.; Al Jaouni, S. Anti-inﬂammatory, antioxidant and antiangiogenic activities of diosgenin isolated from traditional medicinal plant, costus speciosus (Koen ex.Retz.) Sm. Nat. Prod. Res. 2016, 30, 1830–1833. [CrossRef] [PubMed] 70. He, Z.; Chen, H.; Li, G.; Zhu, H.; Gao, Y.; Zhang, L.; Sun, J. Diosgenin inhibits the migration of human breast cancer MDA-MB-231 cells by suppressing VAV2 activity. Phytomedicine 2014, 21, 871–876. [CrossRef] [PubMed] 71. Mao, Z.J.; Tang, Q.J.; Zhang, C.A.; Qin, Z.F.; Pang, B.; Wei, P.K.; Liu, B.; Chou, Y.N. Anti-proliferation and anti-invasion effects of diosgenin on gastric cancer BGC-823 cells with HIF-1alpha shrnas. Int. J. Mol. Sci. 2012, 13, 6521–6533. [CrossRef] [PubMed] 72. Srinivasan, S.; Koduru, S.; Kumar, R.; Venguswamy, G.; Kyprianou, N.; Damodaran, C. Diosgenin targets AKT-mediated prosurvival signaling in human breast cancer cells. Int. J. Cancer 2009, 125, 961–967. [CrossRef] [PubMed] 73. Sowmyalakshmi, S.; Ranga, R.; Gairola, C.G.; Chendil, D. Effect of diosgenin (Fenugreek) on breast cancer cells. Proc. Am. Assoc. Cancer Res. 2005, 46, 1382. 74. Corbiere, C.; Liagre, B.; Bianchi, A.; Bordji, K.; Dauca, M.; Netter, P.; Beneytout, J.L. Different contribution of apoptosis to the antiproliferative effects of diosgenin and other plant steroids, hecogenin and tigogenin, on human 1547 osteosarcoma cells. Int. J. Oncol. 2003, 22, 899–905. [CrossRef] [PubMed] 75. Liagre, B.; Bertrand, J.; Leger, D.Y.; Beneytout, J.L. Diosgenin, a plant steroid, induces apoptosis in COX-2 deﬁcient K562 cells with activation of the p38 map kinase signalling and inhibition of NF-kappaB binding. Int. J. Mol. Med. 2005, 16, 1095–1101. [CrossRef] [PubMed] 12 of 12 Nutrients 2018, 10, 645 76. Leger, D.Y.; Liagre, B.; Corbiere, C.; Cook-Moreau, J.; Beneytout, J.L. Diosgenin induces cell cycle arrest and apoptosis in hel cells with increase in intracellular calcium level, activation of CPLA2 and COX-2 overexpression. Int. J. Oncol. 2004, 25, 555–562. [CrossRef] [PubMed] 76. Leger, D.Y.; Liagre, B.; Corbiere, C.; Cook-Moreau, J.; Beneytout, J.L. Diosgenin induces cell cycle arrest and apoptosis in hel cells with increase in intracellular calcium level, activation of CPLA2 and COX-2 overexpression. Int. J. Oncol. 2004, 25, 555–562. [CrossRef] [PubMed] 77. Corbiere, C.; Liagre, B.; Terro, F.; Beneytout, J.L. Induction of antiproliferative effect by diosgenin through activation of p53, release of apoptosis-inducing factor (AIF) and modulation of caspase-3 activity in different human cancer cells. Cell. Res. 2004, 14, 188–196. [CrossRef] [PubMed] 78. References Wang, S.L.; Cai, B.; Cui, C.B.; Liu, H.W.; Wu, C.F.; Yao, X.S. Diosgenin-3-O-alpha-L-rhamnopyranosyl- (1→4)-beta-d-glucopyranoside obtained as a new anticancer agent from dioscorea futschauensis induces apoptosis on human colon carcinoma HCT-15 cells via mitochondria-controlled apoptotic pathway. J. Asian Nat. Prod. Res. 2004, 6, 115–125. [CrossRef] [PubMed] 79. Raju, J.; Patlolla, J.M.; Swamy, M.V.; Rao, C.V. Diosgenin, a steroid saponin of Trigonella foenum graecum (Fenugreek), inhibits azoxymethane-induced aberrant crypt foci formation in F344 rats and induces apoptosis in HT-29 human colon cancer cells. Cancer Epidemiol. Biomarkers Prev. 2004, 13, 1392–1398. [PubMed] 80. Malisetty, V.S.; Patlolla, J.M.R.; Raju, J.; Marcus, L.A.; Choi, C.L.; Rao, C.V. Chemoprevention of colon cancer by diosgenin, a steroidal saponin constituent of fenugreek. Proc. Am. Assoc. Cancer Res. 2005, 46, 2473. 81. Miyoshi, N.; Nagasawa, T.; Mabuchi, R.; Yasui, Y.; Wakabayashi, K.; Tanaka, T.; Ohshima, H. Chemoprevention of azoxymethane/dextran sodium sulfate-induced mouse colon carcinogenesis by freeze-dried yam sanyaku and its constituent diosgenin. Cancer Prev. Res. (Phila.) 2011, 4, 924–934. [CrossRef] [PubMed] 82. Yan, L.L.; Zhang, Y.J.; Gao, W.Y.; Man, S.L.; Wang, Y. In vitro and in vivo anticancer activity of steroid saponins of paris polyphylla var. Yunnanensis. Exp. Oncol. 2009, 31, 27–32. [PubMed] 83. Marker, R.E.; Sterols, C.V. The preparation of testosterone and related compounds from sarsasapogenin and diosgenin. J. Am. Chem. Soc. 1940, 62, 2543–2547. [CrossRef] 84. Tietze, L.F.; Bell, H.P.; Chandrasekhar, S. Natural product hybrids as new leads for drug discovery. Angew. Chem. Int. Ed. 2003, 42, 3996–4028. [CrossRef] [PubMed] 85. Sanchez-Sanchez, L.; Hernandez-Linares, M.G.; Escobar, M.L.; Lopez-Munoz, H.; Zenteno, E.; Fernandez- Herrera, M.A.; Guerrero-Luna, G.; Carrasco-Carballo, A.; Sandoval-Ramirez, J. Antiproliferative, cytotoxic, and apoptotic activity of steroidal oximes in cervicouterine cell lines. Molecules 2016, 21, 1533. [CrossRef] [PubMed] 86. Masood Ur, R.; Mohammad, Y.; Fazili, K.M.; Bhat, K.A.; Ara, T. Synthesis and biological evaluation of novel 3-O-tethered triazoles of diosgenin as potent antiproliferative agents. Steroids 2017, 118, 1–8. [CrossRef] [PubMed] 87. Liu, C.; Xie, F.; Zhao, G.D.; Wang, D.F.; Lou, H.X.; Liu, Z.P. Synthetic studies towards 1alpha- hydroxysolasodine from diosgenin and the unexpected tetrahydrofuran ring opening in the birch reduction process. Steroids 2015, 104, 214–219. [CrossRef] [PubMed] 88. Hamid, A.A.; Kaushal, T.; Ashraf, R.; Singh, A.; Chand Gupta, A.; Prakash, O.; Sarkar, J.; Chanda, D.; Bawankule, D.U.; Khan, F.; et al. (22beta,25R)-3beta-hydroxy-spirost-5-en-7-iminoxy-heptanoic acid exhibits anti-prostate cancer activity through caspase pathway. Steroids 2017, 119, 43–52. [CrossRef] [PubMed] 89. © 2018 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/). References Ghosh, S.; More, P.; Derle, A.; Kitture, R.; Kale, T.; Gorain, M.; Avasthi, A.; Markad, P.; Kundu, G.C.; Kale, S.; et al. Diosgenin functionalized iron oxide nanoparticles as novel nanomaterial against breast cancer. J. Nanosci. Nanotechnol. 2015, 15, 9464–9472. [CrossRef] [PubMed] © 2018 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/). © 2018 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/W1639907664	https://europepmc.org/articles/pmc3470346?pdf=render	Latin	null	<i>N</i>-Ethyl-2-[1-(2-hydroxy-6-methoxyphenyl)ethylidene]hydrazinecarbothioamide	Acta crystallographica. Section E	2,012	cc-by	2,907	organic compounds b = 8.0393 (5) A˚ c = 10.3808 (8) A˚ = 103.510 (7) V = 695.26 (8) A˚ 3 Z = 2 Mo K radiation = 0.23 mm1 T = 173 K 0.46 0.32 0.24 mm Data collection Oxford Xcalibur (Eos, Gemini) diffractometer Absorption correction: multi-scan (CrysAlis RED; Oxford Diffraction, 2010) Tmin = 0.974, Tmax = 1.000 7338 measured reﬂections 3987 independent reﬂections 3287 reﬂections with I > 2(I) Rint = 0.030 Reﬁnement R[F 2 > 2(F 2)] = 0.047 wR(F 2) = 0.128 S = 1.05 3987 reﬂections 173 parameters 2 restraints H atoms treated by a mixture of independent and constrained reﬁnement max = 0.63 e A˚ 3 min = 0.21 e A˚ 3 Absolute structure: Flack (1983), with 1510 Friedel pairs Flack parameter: 0.00 (8) Acta Crystallographica Section E Structure Reports Online ISSN 1600-5368 Acta Crystallographica Section E Structure Reports Online ISSN 1600-5368 Related literature For thiosemicarbazone structures and their biological activity, see: Lobana et al. (2009). For thiosemicarbazones as ligands for metal-catalyzed reactions or hydrogenations, see: Xie et al. (2010); Pelagatti et al. (1998). For reference bond-length data, see: Allen et al. (1987). JPJ acknowledges the NSF–MRI program (grant No. CHE1039027) for funds to purchase the X-ray diffractometer. Supplementary data and ﬁgures for this paper are available from the IUCr electronic archives (Reference: CV5339). Experimental Crystal data C12H17N3O2S Mr = 267.35 Monoclinic, Pc a = 8.5681 (6) A˚ Table 1 In the title compound, C12H17N3O2S, the dihedral angle between the mean planes of the hydrazinecarbothioamide group and the benzene ring is 86.8 (4). In the crystal, intermolecular O—H S hydrogen bonds link the molecules into chains along [001]. The crystal studied was an inversion twin, the reﬁned ratio of the twin components being 0.98021 (3):0.01978 (7). Data collection: CrysAlis PRO (Oxford Diffraction, 2010); cell reﬁnement: CrysAlis PRO; data reduction: CrysAlis RED (Oxford Diffraction, 2010); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to reﬁne structure: SHELXL97 (Sheldrick, 2008); molecular graphics: SHELXTL (Sheldrick, 2008); software used to prepare material for publication: SHELXTL. o2982 Anderson et al. Brian J. Anderson, Christopher J. Kennedy and Jerry P. Jasinski* Department of Chemistry, Keene State College, 229 Main Street, Keene, NH 03435- 2001, USA Correspondence e-mail: jjasinski@keene.edu Received 8 September 2012; accepted 14 September 2012 Key indicators: single-crystal X-ray study; T = 173 K; mean (C–C) = 0.004 A˚; R factor = 0.047; wR factor = 0.128; data-to-parameter ratio = 23.0. Table 1 Hydrogen-bond geometry (A˚ , ). D—H A D—H H A D A D—H A O1—H1 S1i 0.82 2.35 3.1655 (19) 175 Symmetry code: (i) x; y; z þ 1 2. Refinement Atoms H1A and H2 were located on a difference map and refined isotropically. The remaining H atoms were placed in their calculated positions and then refined using the riding model, with C—H lengths of 0.93 Å (CH), 0.97 Å (CH2) or 0.96 Å (CH3) and the O—H length of 0.82 Å. The isotropic displacement parameters for these atoms were set to 1.2 (CH, CH2) or 1.5 (CH3, OH) times Ueq of the parent atom. The structure was refined as an inversion twin, with the twin law -1 0 0 0 -1 0 0 0 -1 2 and the refined ratio of twin components being 0.98021 (3):0.01978 (7). Comment Thiosemicarbazones are an important class of ligands whose metal complex structures and biological activity have been extensively investigated (Lobana et al., 2009). Recently, thiosemicarbazones have been studied as ligands for metal catalyzed reactions such as Mizoroki–Heck couplings (Xie et al., 2010) and hydrogenations (Pelagatti et al., 1998). The crystal structure of a novel thiosemicarbazone molecule is reported here. In the title compound, C12H17N3O2S (Fig. 1), the dihedral angle between the mean plane of the hydrazinecarbothioamide group (N1/S1/C3/N2/N3) and benzene ring is 86.8 (4)°. Bond lengths are in normal ranges (Allen et al., 1987). In the crystal, the intermolecular O—H···S hydrogen bonds (Table 1) link the molecules into chains in [001] (Fig. 2). Experimental A 50 ml round-bottomed flask was charged with 0.507 g (3.05 mmol) of 2′-hydroxy-6′-methoxyacetophenone and 0.363 g (3.05 mmol) of 4-ethyl-3-thiosemicarbazide followed by 35 ml of methanol, resulting in a clear yellow solution. The solution was refluxed for 5 h, and then the solvent was removed by rotary evaporation. The product was dissolved into 40°C acetonitrile and slowly allowed to cool to 0°C. Translucent crystals were observed after 48 h. (m.p. 458–460 K). Refinement References Allen, F. H., Kennard, O., Watson, D. G., Brammer, L., Orpen, A. G. & Taylor, R. (1987). J. Chem. Soc. Perkin Trans. 2, pp. S1–19. Flack, H. D. (1983). Acta Cryst. A39, 876–881. Lobana T S Sharma R Bawa G & Khanna S (2009) Coord Chem Rev Allen, F. H., Kennard, O., Watson, D. G., Brammer, L., Orpen, A. G. & Taylor, R. (1987). J. Chem. Soc. Perkin Trans. 2, pp. S1–19. , , , , , , , , p , y , R. (1987). J. Chem. Soc. Perkin Trans. 2, pp. S1–19. Flack, H. D. (1983). Acta Cryst. A39, 876–881. R. (1987). J. Chem. Soc. Perkin Trans. 2, pp. S1 19. Flack, H. D. (1983). Acta Cryst. A39, 876–881. Flack, H. D. (1983). Acta Cryst. A39, 876–881. Lobana, T. S., Sharma, R., Bawa, G. & Khanna, S. (2009). Coord. Chem. Rev. 253, 977–1055. Oxford Diffraction (2010). CrysAlis PRO and CrysAlis RED. Oxford Diffraction Ltd, Yarnton, Oxfordshire, England. Pelagatti, P., Venturini, A., Carcelli, M., Costa, M., Bacchi, A., Pelizzi, G. & Pelizza, C. (1998). J. Chem. Soc. Dalton Trans. pp. 2715–2721. Experimental Experimental Crystal data C12H17N3O2S Mr = 267.35 Sheldrick, G. M. (2008). Acta Cryst. A64, 112–122. Xie, G., Chellan, P., Mao, J., Chibale, K. & Smith, G. S. (2010). Adv. Synth. Catal. 352, 1641–1647. Monoclinic, Pc a = 8.5681 (6) A˚ Monoclinic, Pc a = 8.5681 (6) A˚ o2982 Anderson et al. Acta Cryst. (2012). E68, o2982 doi:10.1107/S1600536812039323 supplementary materials Acta Cryst. (2012). E68, o2982 [doi:10.1107/S1600536812039323] Acta Cryst. (2012). E68, o2982 [doi:10.1107/S1600536812039323] Acta Cryst. (2012). E68, o2982 [doi:10.1107/S1600536812039323] Acta Cryst. (2012). E68, o2982 [doi:10.1107/S1600536812039323] Computing details Data collection: CrysAlis PRO (Oxford Diffraction, 2010); cell refinement: CrysAlis PRO (Oxford Diffraction, 2010); data reduction: CrysAlis RED (Oxford Diffraction, 2010); program(s) used to solve structure: SHELXS97 (Sheldrick, 2008); program(s) used to refine structure: SHELXL97 (Sheldrick, 2008); molecular graphics: SHELXTL (Sheldrick, 2008); software used to prepare material for publication: SHELXTL (Sheldrick, 2008). sup-1 Acta Cryst. (2012). E68, o2982 supplementary materials Figure 1 Figure 2 g Packing diagram viewed along the c axis. Weak O—H···S intermolecular interactions are shown by dashed lines. C- bound H atoms were omitted for clarity. g Packing diagram viewed along the c axis. Weak O—H···S intermolecular interactions are shown by dashed lines. C- bound H atoms were omitted for clarity. Packing diagram viewed along the c axis. Weak O—H···S intermolecular interactions are shown by dashed lines. C- bound H atoms were omitted for clarity. Figure 1 Figure 1 The molecular structure of the title compound with the atom numbering scheme. Displacement ellipsoids are drawn at the 50% probability level. H atoms are presented as small spheres of arbitrary radius. The molecular structure of the title compound with the atom numbering scheme. Displacement ellipsoids are drawn at the 50% probability level. H atoms are presented as small spheres of arbitrary radius. Figure 2 Packing diagram viewed along the c axis. Weak O—H···S intermolecular interactions are shown by dashed lines. C- bound H atoms were omitted for clarity. N-Ethyl-2-[1-(2-hydroxy-6- methoxyphenyl)ethylidene]hydrazinecarbothioamide N-Ethyl-2-[1-(2-hydroxy-6- methoxyphenyl)ethylidene]hydrazinecarbothioamide N-Ethyl-2-[1-(2-hydroxy-6- methoxyphenyl)ethylidene]hydrazinecarbothioamide Crystal data C12H17N3O2S Mr = 267.35 Monoclinic, Pc Hall symbol: P -2yc a = 8.5681 (6) Å b = 8.0393 (5) Å c = 10.3808 (8) Å β = 103.510 (7)° V = 695.26 (8) Å3 Z = 2 F(000) = 284 Dx = 1.277 Mg m−3 Crystal data C12H17N3O2S Mr = 267.35 Monoclinic, Pc Hall symbol: P -2yc a = 8.5681 (6) Å b = 8.0393 (5) Å c = 10.3808 (8) Å β = 103.510 (7)° V = 695.26 (8) Å3 Z = 2 F(000) = 284 Dx = 1.277 Mg m−3 sup-2 Acta Cryst. (2012). E68, o2982 supplementary materials Mo Kα radiation, λ = 0.71070 Å Cell parameters from 2153 reflections θ = 3.2–32.3° µ = 0.23 mm−1 T = 173 K Chunk, colourless 0.46 × 0.32 × 0.24 mm Data collection Oxford Xcalibur (Eos, Gemini) diffractometer Radiation source: Enhance (Mo) X-ray Source Graphite monochromator Detector resolution: 16.1500 pixels mm-1 ω scans Absorption correction: multi-scan (CrysAlis RED; Oxford Diffraction, 2010) Tmin = 0.974, Tmax = 1.000 7338 measured reflections 3987 independent reflections 3287 reflections with I > 2σ(I) Rint = 0.030 θmax = 32.3°, θmin = 3.2° h = −12→11 k = −11→11 l = −15→15 Mo Kα radiation, λ = 0.71070 Å Cell parameters from 2153 reflections θ = 3.2–32.3° µ = 0.23 mm−1 T = 173 K Chunk, colourless 0.46 × 0.32 × 0.24 mm Data collection Oxford Xcalibur (Eos, Gemini) diffractometer Radiation source: Enhance (Mo) X-ray Source Graphite monochromator Detector resolution: 16.1500 pixels mm-1 ω scans Absorption correction: multi-scan (CrysAlis RED; Oxford Diffraction, 2010) Tmin = 0.974, Tmax = 1.000 7338 measured reflections 3987 independent reflections 3287 reflections with I > 2σ(I) Rint = 0.030 θmax = 32.3°, θmin = 3.2° h = −12→11 k = −11→11 l = −15→15 Refinement Refinement on F2 Least-squares matrix: full R[F2 > 2σ(F2)] = 0.047 wR(F2) = 0.128 S = 1.05 3987 reflections 173 parameters 2 restraints Primary atom site location: structure-invariant direct methods Secondary atom site location: difference Fourier map Hydrogen site location: inferred from neighbouring sites H atoms treated by a mixture of independent and constrained refinement w = 1/[σ2(Fo2) + (0.0636P)2 + 0.0627P] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max < 0.001 Δρmax = 0.63 e Å−3 Δρmin = −0.21 e Å−3 Absolute structure: Flack (1983), with 1510 Friedel pairs Flack parameter: 0.00 (8) Refinement Hydrogen site location: inferred from neighbouring sites H atoms treated by a mixture of independent and constrained refinement w = 1/[σ2(Fo2) + (0.0636P)2 + 0.0627P] where P = (Fo2 + 2Fc2)/3 (Δ/σ)max < 0.001 Δρmax = 0.63 e Å−3 Δρmin = −0.21 e Å−3 Absolute structure: Flack (1983), with 1510 Friedel pairs ρmin Absolute structure: Flack (1983), with 1510 Friedel pairs Flack parameter: 0.00 (8) Absolute structure: Flack (1983), with 1510 Acta Cryst. (2012). E68, o2982 Special details Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. Refinement of F2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F2, conventional R-factors R are based on F, with F set to zero for negative F2. The threshold expression of F2 > σ(F2) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F2 are statistically about twice as large as those based on F, and R-factors based on ALL data will be even larger. Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) x y z Uiso/Ueq S1 0.55727 (6) 0.29581 (7) 0.59523 (6) 0.04628 (16) O1 0.7690 (2) 0.0256 (2) 1.0893 (2) 0.0585 (5) H1 0.7155 −0.0595 1.0865 0.088 O2 1.1860 (2) 0.2488 (3) 0.92524 (19) 0.0549 (5) N1 0.6144 (3) 0.5442 (3) 0.7677 (2) 0.0457 (5) H1A 0.637 (4) 0.575 (4) 0.839 (3) 0.050 (9)* N2 0.7546 (2) 0.3070 (2) 0.83163 (19) 0.0379 (4) H2 0.782 (3) 0.198 (4) 0.823 (3) 0.047 (8)* N3 0.8239 (2) 0.3855 (2) 0.94956 (18) 0.0344 (4) C1 0.4947 (5) 0.8165 (4) 0.7392 (3) 0.0657 (9) H1B 0.4404 0.7948 0.8086 0.098* l atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) x y z Uiso/Ueq Acta Cryst. (2012). Special details E68, o2982 sup-3 supplementary materials supplementary materials H1C 0.4323 0.8925 0.6764 0.098 H1D 0.5980 0.8644 0.7765 0.098* C2 0.5155 (4) 0.6584 (4) 0.6716 (3) 0.0568 (7) H2A 0.4116 0.6086 0.6348 0.068* H2B 0.5674 0.6800 0.5996 0.068* C3 0.6451 (2) 0.3906 (2) 0.7390 (2) 0.0346 (4) C4 0.9297 (2) 0.3025 (2) 1.0323 (2) 0.0333 (4) C5 0.9822 (2) 0.1311 (3) 1.0085 (2) 0.0345 (4) C6 1.1173 (3) 0.1064 (3) 0.9568 (2) 0.0415 (5) C7 1.1719 (3) −0.0556 (4) 0.9435 (3) 0.0523 (6) H7 1.2623 −0.0731 0.9102 0.063* C8 1.0913 (4) −0.1879 (3) 0.9799 (3) 0.0532 (7) H8 1.1301 −0.2948 0.9731 0.064* C9 0.9556 (3) −0.1677 (3) 1.0258 (3) 0.0494 (6) H9 0.9002 −0.2595 1.0464 0.059* C10 0.9014 (3) −0.0067 (3) 1.0412 (2) 0.0411 (5) C11 1.0021 (3) 0.3811 (3) 1.1618 (3) 0.0510 (6) H11A 0.9585 0.4908 1.1644 0.077* H11B 1.1163 0.3883 1.1730 0.077* H11C 0.9782 0.3151 1.2319 0.077* C12 1.3067 (4) 0.2315 (5) 0.8506 (3) 0.0717 (9) H12A 1.3976 0.1733 0.9029 0.107* H12B 1.3397 0.3398 0.8281 0.107* H12C 1.2636 0.1700 0.7710 0.107* Atomic displacement parameters (Å2) U11 U22 U33 U12 U13 U23 S1 0.0464 (3) 0.0416 (3) 0.0444 (3) 0.0038 (3) −0.0023 (2) −0.0014 (3) O1 0.0607 (11) 0.0363 (9) 0.0872 (14) 0.0008 (8) 0.0352 (11) 0.0062 (10) O2 0.0481 (10) 0.0661 (12) 0.0562 (12) 0.0008 (9) 0.0238 (9) 0.0002 (9) N1 0.0463 (10) 0.0360 (10) 0.0482 (12) 0.0125 (8) −0.0027 (9) 0.0005 (9) N2 0.0434 (9) 0.0266 (8) 0.0390 (10) 0.0100 (7) 0.0001 (8) 0.0012 (7) N3 0.0367 (8) 0.0284 (8) 0.0371 (9) 0.0056 (7) 0.0066 (7) 0.0024 (7) C1 0.100 (3) 0.0467 (15) 0.0512 (15) 0.0374 (16) 0.0185 (16) 0.0129 (12) C2 0.0634 (16) 0.0465 (14) 0.0556 (16) 0.0212 (12) 0.0042 (13) 0.0089 (12) C3 0.0312 (9) 0.0284 (9) 0.0433 (11) 0.0026 (8) 0.0066 (8) 0.0047 (8) C4 0.0366 (10) 0.0291 (9) 0.0344 (10) 0.0049 (7) 0.0089 (8) 0.0048 (8) C5 0.0378 (10) 0.0314 (10) 0.0319 (9) 0.0105 (8) 0.0031 (8) 0.0030 (8) C6 0.0382 (10) 0.0493 (13) 0.0344 (10) 0.0081 (9) 0.0034 (9) −0.0021 (9) C7 0.0464 (12) 0.0651 (17) 0.0438 (12) 0.0242 (12) 0.0074 (10) −0.0085 (12) C8 0.0653 (16) 0.0414 (13) 0.0442 (13) 0.0226 (12) −0.0046 (12) −0.0054 (10) C9 0.0603 (15) 0.0317 (11) 0.0504 (14) 0.0123 (10) 0.0012 (12) 0.0045 (10) C10 0.0426 (11) 0.0355 (10) 0.0432 (12) 0.0084 (9) 0.0062 (9) 0.0030 (9) C11 0.0660 (15) 0.0382 (12) 0.0429 (13) 0.0100 (11) 0.0007 (11) −0.0026 (10) C12 0.0528 (16) 0.114 (3) 0.0560 (17) −0.0008 (17) 0.0279 (14) −0.0007 (18) Atomic displacement parameters (Å2) Acta Cryst. Special details (2012). E68, o2982 sup-4 supplementary materials su Acta Cryst. (2012). E68, o2982 Geometric parameters (Å, º) S1—C3 1.688 (2) C4—C11 1.484 (3) O1—C10 1.367 (3) C4—C5 1.487 (3) O1—H1 0.8200 C5—C10 1.389 (3) O2—C6 1.362 (3) C5—C6 1.399 (3) O2—C12 1.436 (3) C6—C7 1.402 (4) N1—C3 1.311 (3) C7—C8 1.368 (4) N1—C2 1.471 (3) C7—H7 0.9300 N1—H1A 0.76 (3) C8—C9 1.365 (4) N2—C3 1.355 (3) C8—H8 0.9300 N2—N3 1.382 (2) C9—C10 1.396 (3) N2—H2 0.92 (3) C9—H9 0.9300 N3—C4 1.281 (3) C11—H11A 0.9600 C1—C2 1.482 (4) C11—H11B 0.9600 C1—H1B 0.9600 C11—H11C 0.9600 C1—H1C 0.9600 C12—H12A 0.9600 C1—H1D 0.9600 C12—H12B 0.9600 C2—H2A 0.9700 C12—H12C 0.9600 C2—H2B 0.9700 C10—O1—H1 109.5 C6—C5—C4 120.3 (2) C6—O2—C12 117.1 (3) O2—C6—C5 114.6 (2) C3—N1—C2 123.2 (2) O2—C6—C7 125.7 (2) C3—N1—H1A 121 (2) C5—C6—C7 119.7 (2) C2—N1—H1A 116 (2) C8—C7—C6 119.6 (2) C3—N2—N3 119.04 (16) C8—C7—H7 120.2 C3—N2—H2 124.0 (18) C6—C7—H7 120.2 N3—N2—H2 116.8 (18) C9—C8—C7 122.0 (2) C4—N3—N2 116.48 (17) C9—C8—H8 119.0 C2—C1—H1B 109.5 C7—C8—H8 119.0 C2—C1—H1C 109.5 C8—C9—C10 118.9 (3) H1B—C1—H1C 109.5 C8—C9—H9 120.5 C2—C1—H1D 109.5 C10—C9—H9 120.5 H1B—C1—H1D 109.5 O1—C10—C5 116.17 (19) H1C—C1—H1D 109.5 O1—C10—C9 123.0 (2) N1—C2—C1 109.1 (2) C5—C10—C9 120.8 (2) N1—C2—H2A 109.9 C4—C11—H11A 109.5 C1—C2—H2A 109.9 C4—C11—H11B 109.5 N1—C2—H2B 109.9 H11A—C11—H11B 109.5 C1—C2—H2B 109.9 C4—C11—H11C 109.5 H2A—C2—H2B 108.3 H11A—C11—H11C 109.5 N1—C3—N2 116.6 (2) H11B—C11—H11C 109.5 N1—C3—S1 123.69 (17) O2—C12—H12A 109.5 N2—C3—S1 119.71 (15) O2—C12—H12B 109.5 N3—C4—C11 117.74 (18) H12A—C12—H12B 109.5 N3—C4—C5 124.46 (19) O2—C12—H12C 109.5 C11—C4—C5 117.79 (18) H12A—C12—H12C 109.5 C10—C5—C6 119.0 (2) H12B—C12—H12C 109.5 C10—C5—C4 120.74 (18) Geometric parameters (Å, º) sup-5 Acta Cryst. (2012). E68, o2982 supplementary materials C3—N2—N3—C4 −178.22 (19) C10—C5—C6—O2 178.6 (2) C3—N1—C2—C1 175.6 (3) C4—C5—C6—O2 −3.4 (3) C2—N1—C3—N2 171.6 (2) C10—C5—C6—C7 −2.3 (3) C2—N1—C3—S1 −8.0 (3) C4—C5—C6—C7 175.6 (2) N3—N2—C3—N1 1.3 (3) O2—C6—C7—C8 179.7 (2) N3—N2—C3—S1 −179.12 (15) C5—C6—C7—C8 0.7 (4) N2—N3—C4—C11 −177.9 (2) C6—C7—C8—C9 1.8 (4) N2—N3—C4—C5 0.6 (3) C7—C8—C9—C10 −2.7 (4) N3—C4—C5—C10 −87.9 (3) C6—C5—C10—O1 −179.0 (2) C11—C4—C5—C10 90.5 (3) C4—C5—C10—O1 3.0 (3) N3—C4—C5—C6 94.1 (3) C6—C5—C10—C9 1.5 (3) C11—C4—C5—C6 −87.4 (3) C4—C5—C10—C9 −176.5 (2) C12—O2—C6—C5 −169.3 (2) C8—C9—C10—O1 −178.5 (2) C12—O2—C6—C7 11.7 (4) C8—C9—C10—C5 1.0 (4) Hydrogen-bond geometry (Å, º) D—H···A D—H H···A D···A D—H···A O1—H1···S1i 0.82 2.35 3.1655 (19) 175 Symmetry code: (i) x, −y, z+1/2. Hydrogen-bond geometry (Å, º) D—H···A D—H H···A D···A D—H···A O1—H1···S1i 0.82 2.35 3.1655 (19) 175 S d (i) 1/2 Special details C3—N2—N3—C4 −178.22 (19) C10—C5—C6—O2 178.6 (2) C3—N1—C2—C1 175.6 (3) C4—C5—C6—O2 −3.4 (3) C2—N1—C3—N2 171.6 (2) C10—C5—C6—C7 −2.3 (3) C2—N1—C3—S1 −8.0 (3) C4—C5—C6—C7 175.6 (2) N3—N2—C3—N1 1.3 (3) O2—C6—C7—C8 179.7 (2) N3—N2—C3—S1 −179.12 (15) C5—C6—C7—C8 0.7 (4) N2—N3—C4—C11 −177.9 (2) C6—C7—C8—C9 1.8 (4) N2—N3—C4—C5 0.6 (3) C7—C8—C9—C10 −2.7 (4) N3—C4—C5—C10 −87.9 (3) C6—C5—C10—O1 −179.0 (2) C11—C4—C5—C10 90.5 (3) C4—C5—C10—O1 3.0 (3) N3—C4—C5—C6 94.1 (3) C6—C5—C10—C9 1.5 (3) C11—C4—C5—C6 −87.4 (3) C4—C5—C10—C9 −176.5 (2) C12—O2—C6—C5 −169.3 (2) C8—C9—C10—O1 −178.5 (2) C12—O2—C6—C7 11.7 (4) C8—C9—C10—C5 1.0 (4) sup-6 Acta Cryst. (2012). E68, o2982
https://openalex.org/W2790844793	https://europepmc.org/articles/pmc5812615?pdf=render	English	null	Reduced cognitive function, increased blood-brain-barrier transport and inflammatory responses, and altered brain metabolites in LDLr -/-and C57BL/6 mice fed a western diet	PloS one	2,018	cc-by	24,662	RESEARCH ARTICLE Reduced cognitive function, increased blood- brain-barrier transport and inflammatory responses, and altered brain metabolites in LDLr -/-and C57BL/6 mice fed a western diet Jennifer M. Rutkowsky1, Linda L. Lee2, Michelle Puchowicz3, Mari S. Golub4, Douglas E. Befroy5, Dennis W. Wilson6, Steven Anderson7, Gary Cline8, Jason Bini9, Kamil Borkowski10, Trina A. Knotts1, John C. Rutledge1, on behalf of the Mouse Metabolic Phenotyping Center Imaging Working Group¶ a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 1 Department of Molecular Biosciences, School of Veterinary Medicine, University of California, Davis, California, United States of America, 2 Division of Cardiovascular Medicine, Department of Internal Medicine, University of California, Davis, California, United States of America, 3 Department of Nutrition, School of Medicine, Case Western Reserve University, Cleveland, Ohio, United States of America, 4 Department of Environmental Toxicology, University of California, Davis, California, United States of America, 5 Magnetic Resonance Research Center, Department of Diagnostic Radiology, Yale University School of Medicine, New Haven, Connecticut, United States of America, 6 Department of Pathology, Microbiology, and Immunology, School of Veterinary Medicine, University of California, Davis, California, United States of America, 7 Department of Physiology and Membrane Biology, University of California, Davis, California, United States of America, 8 Department of Endocrinology, Yale University, New Haven, Connecticut, United States of America, 9 Yale PET Center, Department of Diagnostic Radiology, Yale University, New Haven, Connecticut, United States of America, 10 West Coast Metabolomics Center, Genome Center, University of California, Davis, California, United States of America OPEN ACCESS Citation: Rutkowsky JM, Lee LL, Puchowicz M, Golub MS, Befroy DE, Wilson DW, et al. (2018) Reduced cognitive function, increased blood-brain- barrier transport and inflammatory responses, and altered brain metabolites in LDLr -/-and C57BL/6 mice fed a western diet. PLoS ONE 13(2): e0191909. https://doi.org/10.1371/journal. pone.0191909 ¶ Membership of the Mouse Metabolic Phenotyping Center Imaging Working Group is provided in the acknowledgments. @ jrutkowsky@ucdavis.edu Introduction Previous studies have shown that elevated blood lipids and a diet high in saturated fats puts individuals at greater risk for dementia and cognitive impairment [1–5]. Moreover, ani- mal studies have shown that a high-fat/high-cholesterol diet not only induces cognitive impairment, but also increases neuroinflammation [6–8]. For instance, low-density lipopro- tein receptor null (LDLr -/-) mice are predisposed to elevated blood cholesterol levels and show evidence of cognitive impairment and increased brain inflammation when fed a high fat diet [9–12]. LDLr mediates the endocytosis of cholesterol rich low-density lipoproteins regu- lating plasma levels of cholesterol. It is prominently expressed in the liver, but also the gastro- intestinal tract, muscle (heart and skeletal) and brain [13]. Genetic knock out of LDLr leads to a twofold elevation in circulating cholesterol and 7-9-fold increase in LDL due to prolonged clearance rate [14]. Our previous work using brain microvascular endothelial cells and astro- cytes treated with lipids and lipoproteins showed a complex interaction of multiple cell stress response signaling mechanisms that was not adequately described by a single cell pathway [15–17]. In agreement, a western diet (WD) has been shown to decrease brain capillary expression of tight junction proteins and increase hippocampal blood-brain barrier (BBB) per- meability in the rat [18], potentially allowing for additional paracellular movement of blood components including lipids and lipoproteins. Diet has also been shown to activate microglia, resident brain inflammatory cells, and induce inflammation and cellular degeneration [8, 9, 19], each thought to contribute to the progression of cognitive impairment [20]. Competing interests: The authors have declared that no competing interests exist. Competing interests: The authors have declared that no competing interests exist. Other work has linked brain metabolic perturbations with cognitive impairment. For instance, studies using positron emission tomography (PET) to examine regional brain glucose metabolism show that Alzheimer’s disease (AD) and vascular dementia each exhibit a unique pattern of reduced brain glucose uptake [21, 22]. Further, metabolic stress, suggested by the elevation of lactate and glutamate, has been implicated in AD, ischemic stroke, epilepsy, and cognitive impairments [23] and a reduction of N-acetylaspartate accompanied by increases in glutamate & glutamine are correlated with brain injury and cognitive impairment [24–28]. However, the pathways by which major metabolic stressors such as a western diet or hyperlip- idemia influence brain metabolite levels and metabolic function are not fully understood. Diet & LDLr-/- alter brain function & metabolism National Science Foundation OSTI 97-24412 (KB). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Abstract Editor: Ma´ria A. Deli, Hungarian Academy of Sciences, HUNGARY Received: October 9, 2017 Accepted: January 12, 2018 Published: February 14, 2018 Recent work suggests that diet affects brain metabolism thereby impacting cognitive func- tion. Our objective was to determine if a western diet altered brain metabolism, increased blood-brain barrier (BBB) transport and inflammation, and induced cognitive impairment in C57BL/6 (WT) mice and low-density lipoprotein receptor null (LDLr -/-) mice, a model of hyperlipidemia and cognitive decline. We show that a western diet and LDLr -/- moderately influence cognitive processes as assessed by Y-maze and radial arm water maze. Also, western diet significantly increased BBB transport, as well as microvessel factor VIII in LDLr -/- and microglia IBA1 staining in WT, both indicators of activation and neuroinflammation. Interestingly, LDLr -/- mice had a significant increase in 18F- fluorodeoxyglucose uptake irre- spective of diet and brain 1H-magnetic resonance spectroscopy showed increased lactate and lipid moieties. Metabolic assessments of whole mouse brain by GC/MS and LC/MS/MS showed that a western diet altered brain TCA cycle and β-oxidation intermediates, levels of amino acids, and complex lipid levels and elevated proinflammatory lipid mediators. Our study reveals that the western diet has multiple impacts on brain metabolism, physiology, and altered cognitive function that likely manifest via multiple cellular pathways. Copyright: © 2018 Rutkowsky 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. All raw data files are available from the figshare database (https://doi.org/10.6084/m9. figshare.5821269.v1). Funding: This work was supported by the National Institute of Aging AG039094 and AG045541 and the Richard A. and Nora Eccles Harrison Endowed Chair in Diabetes Research Fund (JCR), and the 1 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Mice All animal studies were approved by the Animal Use and Care Committee at the University of California, Davis and Yale University School of Medicine and were maintained under similar environmental and identical experimental conditions (diet, time on diet, and age at start of diet) at each institution. Animals were euthanized by decapitation or exsanguination under anesthesia (isoflurane or ketamine). Male C57BL/6J (wild type, WT) and low-density lipopro- tein receptor knock out mice (LDLr -/-) were imported at ~6 weeks of age from JAX West Lab- oratory and maintained on a chow diet prior to administering the specialized diet. Mice were housed 2 to a cage that had a transparent divider with small holes to allow visual and olfactory interaction in a temperature controlled room with a 12 h light: 12 h dark cycle. They remained undisturbed except for weekly cage changes and monthly weighs. At eight weeks of age, half the mice in each genotype were fed ad lib either a western (WD; 42% kcal fat, 0.2% total choles- terol, and 34% sucrose by weight) or control diet (CD; 19.2% kcal fat, 0% added total choles- terol, and 12% sucrose by weight) (TD.88137 or TD.08485 respectively, Envigo, Indianapolis, IN) over the course of 12 weeks. Mice were subjected to behavioral analysis prior to sacrifice at 20 weeks of age when non-fasted tissues and plasma were collected. Separate cohorts were used for metabolic profile and immunohistochemistry, as well as cognitive/behavioral analyses (radial arm water maze, Y-maze, and Morris water maze) and MRI analysis of blood-brain barrier transfer coefficient. Plasma samples were analyzed for glucose, insulin, triglyceride, total cholesterol, HDL, and LDL by the UC Davis Mouse Metabolic Phenotyping Center Endo- crinology and Metabolism Core according to their standard protocols. For studies of pathol- ogy, anesthetized mice were perfusion fixed with 4% paraformaldehyde and sections of brain, liver, heart, kidney, pancreas, skeletal muscle, and lung were embedded in paraffin. Introduction Therefore, our goal for this project was to better understand the mechanisms of WD- induced cognitive impairment using molecular, cellular, biochemical, physiological, and imag- ing approaches. Here, we show that in mice, a WD or hyperlipidemia can alter brain glucose uptake and metabolite levels, activate resident inflammatory cells (microglia), increase brain factor VIII vascular expression and the BBB transfer coefficient, and induce moderate cogni- tive impairments. We first demonstrated that WD or genetically induced hyperlipidemia moderately impairs cognition as determined by Y and radial arm mazes. Using, Gd-DTPA contrast magnetic reso- nance imaging (MRI), we determined that a WD increases BBB transfer coefficient (Ki), potentially contributing to cognitive perturbation [18, 29]. Further, indicators of brain inflam- mation and activation, factor VIII and (ionized calcium binding adaptor molecule 1 (IBA1) protein and prostaglandin-endoperoxide synthase 2 RNA (previously correlated with cognitive disorders), were found to be elevated by WD. As members of the Mouse Metabolic Phenotyp- ing Center Imaging Working Group, we combined the collective and comprehensive expertise of our three universities, to assess how a WD in LDLr-/- and WT mice shifts brain metabolites. Our collaborators at Yale demonstrated an increase in glucose uptake by 18F-fluordeoxyglu- cose (18FDG) positron emission tomography (PET) and lactate concentration by 1H magnetic resonance spectroscopy (1H-MRS) in the brains of live LDLr-/- mice. Our collaborators at Case Western and UC Davis completed a more extensive metabolic analysis to establish that a PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 2 / 38 Diet & LDLr-/- alter brain function & metabolism WD modulates brain fatty acid, TCA cycle intermediates, acyl-CoA’s and complex lipid abun- dance and elevates proinflammatory lipid mediators. Finally, we review the basic physiological parameters and histopathological effects of diet and genotype revealed by this study that may contribute to the metabolic, inflammatory, and cognitive changes observed. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Cognitive assessment by Y, Morris water, and radial arm water mazes Y-maze. At age 18 weeks, 10 weeks after initiation of feeding, mice were subjected to test of spontaneous alternation in a Y-maze as previously described [30]. Mice, adapted to testing room for 30 min, were placed in the center of the Y-maze and were tracked with an overhead camera for the extent of an 8-min trial. An elevated white plastic Y-maze with three 40 cm arms at 120-degree angles illuminated by red light was used. Entry into each arm, total distance moved, and the amount of time spent resting and active were recorded, and an alternation score was computed as the number of times the three arms were sequentially entered. The % alternation score is the number of alternations divided by the total number of arms entered (n = 8/grp). Morris water maze. The Morris water maze (MWM) was administered at 19 weeks of age over a 5-day period as previously described [31]. The cohort (n = 8/grp) was relocated to a fixed position in the experimental room prior to testing. The first day consisted of a single visi- ble platform trial to ensure that the mice could swim to the platform and climb to the platform surface. For training trials, the platform was hidden below the water surface. Each mouse in the group completed a trial before a new trial was initiated for a total of 4 trials over 4 daily ses- sions. The water maze was 94 cm in diameter, with a 6 x 6 cm platform whose surface was located 2 cm below the surface of opacified water. Mice were released in one of three locations PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 3 / 38 Diet & LDLr-/- alter brain function & metabolism in non-platform quadrants using a random order that included each location once in the four daily trials. After the mouse reached the platform or 90 s elapsed, the mouse was placed on the platform for 30 s and then returned to a heated cage between trials. A probe trial was con- ducted on the fifth day, after completion of four daily sessions. The platform was removed from the maze and mice were given one 90 s trial to determine their search strategy for the spa- tial location of the platform. Radial arm water maze. Cognitive assessment by Y, Morris water, and radial arm water mazes A separate group of mice was subjected to a different spatial learning and memory test, the radial arm water maze (RAWM) [31]. Mice in the first cohort (n = 5/grp) were tested twice, prior to and after the 12-week experimental diet period to see if there was an early genotype effect and if there was a change with diet (genotype/diet interac- tion). Mice in the second cohort (n = 5/grp) were tested only after the experimental diet period to evaluate diet genotype interaction to eliminate pre-testing training effects. The test was con- ducted in a six-arm apparatus (77 cm diameter, 20 cm high, 25 cm long, 14cm wide) placed in a shallow pool of water. The center area was 28 cm in diameter. The escape 6 cm round plat- form was placed 1 cm below the opacified water surface. The platform was placed at the end of a different designated arm each day of the 9-day test. Each day consisted of 4 consecutive trials (at unique start location) followed by a fifth (retention) trial that was performed 30–40 min fol- lowing the fourth. Trials began when the mouse was placed in the start arm and ended when the mouse reached the platform or 60 s elapsed. The mouse then remained on the platform for a 30 s inter-trial interval before beginning, then began the next trial. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 MRI measurements of blood-brain barrier (BBB) transfer coefficient (Ki) experimental set-up BBB transport was assessed using MRI to map regional changes in the longitudinal relaxation time (T1) of the brain water signal following an infusion of contrast agent. A minimum of 6 animals from each group were studied and anesthetized with ketamine and xylazine. The left femoral vein was cannulated with PE-10 tubing for administration of saline and gadopentetate dimeglumine (Gd-DTPA—Magnevist; Bayer Healthcare Pharmaceuticals, Wayne, NJ, USA). MRI data were collected using a 7T Bruker Biospec MRS/MRI system (Bruker BioSpin MRI, Inc., Billerica, MA, USA) interfaced with ParaVision 4 (PV4) software (Bruker BioSpin GmbH, Rheinstetten, Germany) with a 32mm radiofrequency (RF) volume coil for transmis- sion and detection. Anesthetized cannulated mice were placed on a PVC animal stage in the prone position. Body temperature was maintained with circulating heated water bed (Gaymar Inc., Orchard Park, NY, USA) while the animal was in the magnet. The mice were placed near the center of the coil such that the isocenter was approximately 1 mm caudal to the bregma. Data acquisition. For each Gd-DTPA infusion, 10 T1 maps were obtained consecutively to follow the time course of contrast agent entry into the brain. Each T1 map was acquired using a rapid-acquisition refocused-echo (RARE) sequence with variable TR = 200ms, 531.8ms, 958.6ms, 1557.2ms, 2568.7ms, and 7500ms; effective TE = 30.80ms; RARE-factor = 16 A single slice was taken using 1mm in-plane thickness, 32x32 mm2 field of view, 128 x 128-pixel resolution (thus 250 μm x 250 μm x 1 mm voxel size), with a 1.8 min acquisition time per T1 map (18 min for 10 consecutive maps). A 0.5 mL/kg aliquot of the contrast agent Magnevist (0.5mmol/mL) was diluted 1:1 with saline and injected i.v. over approximately 10 secs followed by a 50uL saline flush immediately after the acquisition of data for the first T1 map in each 10 map series used for Patlak analysis. Data processing and analysis. T1 values were calculated using the T1 fit function in the Bruker PV4 image sequence analysis tools package. Post-processing image analysis was done using an in-house MATLAB (MATLAB 2014b, MathWorks, Natick, MA) script with Patlak PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 4 / 38 Diet & LDLr-/- alter brain function & metabolism linearized regression mathematical modeling for each pixel. MRI measurements of blood-brain barrier (BBB) transfer coefficient (Ki) experimental set-up The BBB transfer coefficient, Ki (permeability x surface area/volume - min-1), was calculated from the slope of the Patlak plot that best fit an impermeable, uni-directional influx, or bi-directional flux model selected using the F-test with p<0.05 as previously described [32]. Regions-of-interest (ROIs) were manually defined symmetrically about the midline, one from each hemisphere to exclude the median eminence and ventricles. Ki was calculated on a pixel-by-pixel basis and averaged to include both ROIs after pixels with CBF = 0 (see below) were excluded to minimize underestimates of Ki due to flow-limitation of contrast agent delivery [32]. Perfusion weighted imaging (PWI) analysis of cerebral blood flow (CBF) Data acquisition. CBF data were obtained before and after MRI measurements of BBB transfer coefficient using a Bruker PERF spin-echo sequence with TR = 1022.6ms and TE 12.8ms. Imaging parameters were the same as the RAREVTR but with a 4 min 22 s acquisition time. Data processing and analysis. Post-image acquisition analysis was done using Bruker PV4 image sequence analysis tools package with algebra as described by Williams et al. [33] and ROIs were manually defined in the CBF matrix to match the corresponding Ki matrix. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Assessment of brain factor VIII and IBA1 by immunohistochemistry and general pathology Analysis of vessel density and microglia activation in brain was determined using factor VIII and IBA1 staining, respectively. Paraffin embedded sections (4 μm) were deparaffinized through xylene to 100% reagent alcohol, and then treated with 0.3% hydrogen peroxide in 100% methanol for 30 min then rehydrated. Antigen retrieval was performed on sections for IBA1 with heat induced epitope retrieval using Target Retrieval Solution, pH 6 (Dako S1699) for 30 min at 95˚C, followed by a 20 min cool down and with Proteinase K (Dako S3020) at room temperature for 10 min for factor VIII. Slides were rinsed in deionized water and placed in 0.1M Phosphate Buffered Saline, pH 7.4 (PBS). The antibody diluent and blocking reagent were PBS-Tween 20 (0.02%) and 10% normal horse serum (NHS) in PBS-Tween 20, respec- tively. Sections were blocked for 20 min followed by the following primary antibodies: factor VIII (Dako A0082, AB_2315602, 1:2000) and IBA1 (Wako 19–19741, AB_839504, 1:600) for 1 h at room temperature. A single step, polymer based HRP (BioCare Medical, RC542H) was applied for 30 min to label rabbit anti-IBA1. A dual step, biotin-avidin based HRP (BioCare Medical, 4+ Detection System GR608) was applied for 10 min to link rabbit anti-factor VIII. Streptavidin-HRP (BioCare Medical HP604) was applied for 10 min to label the biotin link. All labels were visualized with NovaRed for peroxidase (Vector SK-4800), per manufacturer’s instructions. Sections were counterstained in Mayer’s Hematoxylin, air dried and cover slipped. A coronal section containing cortex, hippocampus, and thalamus at the level of the genicu- late bodies and amygdala was used for image analysis. Whole coronal sections were digitally scanned using an Olympus VS110 system. Images then were analyzed with Visiopharm soft- ware by manually outlining ROI (cortex, hippocampus, and thalamus) followed by automated detection of relative areas of immunopositivity in each area of interest. Histopathology was assessed in brain, heart, and liver by hematoxylin and eosin (H&E) staining of four-micron thick paraffin embedded sections (n = at least 5–6 grp). Paraffin was removed with xylene then then sections were rehydrated though a series of decreasing concentrations of ethanol. Sections were then stained with hematoxylin washed and then PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 5 / 38 Diet & LDLr-/- alter brain function & metabolism stained with eosin. Tissue was then dehydrated with ethanol and xylene and a coverslip mounted. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 mRNA expression by quantitative RT-PCR (qRT-PCR) Total RNA was extracted from powdered mouse brain hemisphere (n = at least 5/group) using TRI Reagent (Sigma-Aldrich, St. Louis, MO) and RNeasy Mini Kit (Qiagen, Valencia, CA) according to manufacturer’s protocol. Up to 5 μg of total RNA from each sample was reverse- transcribed to obtain cDNA in a final volume of 20 μL solution consisting of buffer, random hexamers, DTT, dNTPs and Superscript-III reverse transcriptase (Invitrogen, Carlsbad, CA). qRT-PCR with SYBR Green (Applied Biosystems, Foster City, CA) as fluorescent reporter was used to quantify the expression of selected genes identified by GeneChip analysis. All the gene specific primers (Table 1) were designed with Primer Express 1.0 software (Applied Biosys- tems) using the gene sequences obtained from Affymetrix Probeset IDs. The reaction was car- ried out in 384 well optical plates containing 25 ng RNA in each well. Quantitative RT-PCR using the ViiA™7 Real-Time PCR System (PE Applied Biosystems, Foster City, CA) measured transcript levels. The PCR amplification parameters were: initial denaturation step at 95˚C for 10 min followed by 40 cycles, each at 95˚C for 15 s (melting) and 60˚C for 1 min (annealing Table 1. qRT-PCR murine primers. The oligonucleotide sequences for each primer sequence were obtained from Affymetrix database using the probe set IDs and Primer3 software. The primers were custom prepared and used as described in the Methods. Gene Forward (5’-3’) Reverse (5’-3’) ARG GAACACGGCAGTGGCTTTAAC TGCTTAGCTCTGTCTGCTTTGC ATF3 AGAGCTGAGATTCGCCATCC GAGGACATCCGATGGCAGAG C1qA GCACCCAACGGGAAGGAT CTTTAAAACCTCGGATACCAGTC C1qB TCTGGGAATCCACTGCTGTC AGACCTCACCCCACTGTGTC C1qC CAGCGTCTTCTCTGGTTTCC TCCTGGAGGAAGAGGTCTGA CCL2 CTTCTGGGCCTGCTGTTCAC AGCCAACACGTGGATGCTC CD11b GGATCATAGGCGCCCACTT TCCTTACCCCCACTCAGAGACT CD16 TTTGGACACCCAGATGTTTCAG GTCTTCCTTGAGCACCTGGATC CD32 AATCCTGCCGTTCCTACTGATC GTGTCACCGTGTCTTCCTTGAG CD86 TTGTGTGTGTTCTGGAAACGGAG AACTTAGAGGCTGTGTTGCTGGG CD206 TCTTTGCCTTTCCCAGTCTCC TGACACCCAGCGGAATTTC CXCL2 (MIP2α) CCAACCACCAGGCTACAGG GCGTCACACTCAAGCTCTG DDIT3 AGGGCCAACAGAGGTCACAC GAATCTGGAGAGGGCT Factor VIII GAGGAACCACCGTCAAGCTTCATT CTGAAGGTGCATAGTCCCAGTCTT GAPDH GGATAGGGCCTCTCTTGCTCA GCAACAGGGTGGTGGACCT GDF15 GTGTCCCCACCTGTATCGCT CGTGCTTTGATCTGCGCAT IBA1 GGATCAACAAGCAATTCCTCGA CTGAGAAAGTCAGAGTAGCTGA IGF ACCCCACCCACAAAACAACA CGTCCCGGGTCGTTTACAC IkBα TCCTGCACTTGGCAATCATC AGCCAGCTCTCAGAAGTGCC IL-12a CCACCCTTGCCCTCCTAAAC GGCAGCTCCCTCTTGTTGTG IL-1B AAGGGCTGCTTCCAAACCTTT ATACTGCCTGCCTGAAGCTCT IL- 6 CTCGGCAAACCTAGTGCGTT GGAATGTCCACAAACTGATATGCT MAPK8 (JNK) GCTCCCAGAAAAGCAAGCAG CATCTTTTGGGGGAGTGCCT NR4A2 TGCAGGCAGAACCTGAAAGG CTAAATCCAGGATGCCCCG PTGS2 (COX2) CTTAGTTCCGTTTCTCGTGGTCA AACCCAATCAGCGTTTCTCG TNFα GGAACACGTCGTGGGATAATG GGCAGACTTTGGATGCTTCTT TREM2 GCCACCTATCCTGGGAACAG CCAACTCACCACAGATGTACACAC https://doi.org/10.1371/journal.pone.0191909.t001 Table 1. qRT-PCR murine primers. The oligonucleotide sequences for each primer sequence were obtained from Affymetrix database using the probe set IDs and Primer3 software. The primers were custom prepared and used as described in the Methods. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 6 / 38 Diet & LDLr-/- alter brain function & metabolism and extension). Relative changes in gene expression were determined from real-time quantita- tive PCR experiments by a ΔΔCT method [34] and was normalized with glyceraldehyde- 3-phosphate dehydrogenase (GAPDH). PET/CT imaging analysis of brain glucose uptake At Yale, a cohort of 24 mice were divided into four groups: WT CD (n = 8, mean weight 30.6 ± 3.7 g); WT WD (n = 6, mean weight 34.0 ± 2.9 g); LDLr -/- CD (n = 6, mean weight 29.9 ± 1.5 g); and LDLr -/- WD (n = 4), mean weight (32.9 ± 4.3 g) and were imaged using 18FDG PET to assess brain glucose uptake. All mice were injected via tail vein with mean activ- ity of 2.51 ± 3.86 MBq of 18FDG and listmode PET data were acquired for 30 min on a preclin- ical PET/CT scanner (Inveon PET/CT, Siemens Preclinical Systems, Knoxville, TN, USA). All animals were injected while in the scanner under isoflurane anesthesia. All animals had a low- dose CT acquisition after the PET acquisition to provide attenuation correction. All PET acquisitions were reconstructed using the vendor provided reconstruction algorithms (Inveon Acquisition Workplace; Siemens preclinical systems, Knoxville, TN, USA). Images for each acquisition were reconstructed with 3D-OSEM-MAP (OSEM: 2 iterations; 16 subsets) and 18 MAP iterations (β = 0.0023). Image analysis. ROI tracings and region-specific time activity curves (TACs) for the mice were performed using the vendor provided image software viewer (Inveon Research Work- place, Siemens preclinical systems, Knoxville, TN, USA). Due to the incidence in which the radiotracer did not exit the tail vein after injection, we calculated whole body activity at 30 min, using a full body contour ROI minus the tail activity, as our actual injected activity enter- ing the vasculature. Global brain uptake for all mice was determined by drawing a whole brain ROI. Both ROIs were drawn on the co-registered low-dose CT image provided for attenuation correction. To examine differences in 18FDG uptake between the four groups, brain ROI TACs were plotted for the entire 30 min acquisition. Based on the initial TACs, brain ROI quantification was also performed on a summed static image from 25–30 min post injection. All quantitative comparisons were performed using the body weight standardized uptake value (SUV) with correction for plasma glucose levels (SUVglu) between groups [35] mRNA expression by quantitative RT-PCR (qRT-PCR) The threshold cycle, Ct, which correlates inversely with the target mRNA levels, was measured as the cycle number at which the SYBR Green emission increases above a preset threshold level. The specific mRNA transcripts were expressed as fold difference from media control in the expression of the specific mRNAs. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 In vivo brain metabolomics profiling using 1H magnetic resonance spectroscopy (1H-MRS) Concentrations of cerebral metabolites in vivo were determined by localized 1H MRS on 20-week old male C57BL/6J and LDLr -/- mice after 12 weeks on either a CD or WD. Mice were anesthetized with ~1.5% isoflurane in an air (70%)/ O2 (30%) mixture delivered via a fit- ted nose-cone. The head was kept immobile using a stereotactic restraint with tooth bar and the skull was positioned under a 12mm diameter 1H-surface coil. Each mouse was continu- ously monitored using an MR compatible physiological monitoring system (SA instruments, Stony Brook, NY); body temperature was maintained using a heated water pad and the depth of anesthesia was adjusted as necessary to maintain stable physiological function. MR experi- ments were performed on a 9.4T/31cm Varian DirectDrive system (Agilent Technologies, Santa Clara, CA), using proprietary and custom-written pulse sequences. Multi-slice gradient echo scout images were obtained to ensure correct positioning and to establish the ROI for spectroscopy. To enhance spectral resolution and sensitivity, magnetic field (B0) homogeneity was optimized over a 3x3x3mm volume using a custom-written, adiabatic Fastmap sequence; PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 7 / 38 Diet & LDLr-/- alter brain function & metabolism typical 1H2O linewidths within the region of interest were 15–17 Hz. Localized, water-sup- pressed, 1H MR spectra were obtained from an 18.75 μL volume (2.5x2.5x3mm) using a LASER (Localization by Adiabatic Selective Refocusing) sequence [36] (1ms AHP excitation pulse, 2ms AFP refocusing pulses, sw = 6kHz, 4k points, TR = 3s, TE = 27.3ms) preceded by chemical shift selective (CHESS) water suppression (6x 10ms AFP pulses). Spectra were acquired from the cortex of each hemisphere over 256 transients; total acquisition time for each hemisphere was ~13min. Spectral processing was performed using MATLAB (Mathworks Inc, Natick, MA) software developed in house at the Yale Magnetic Resonance Research Center by Dr. Robin deGraaf and Dr. Douglas Befroy. The relative concentrations of cerebral metabolites were estimated using LCModel 6.3 [37]. Spectral fitting was performed using simulated spectra of a basis set of 17 metabolites (Alanine, Aspartate, β-Hydroxybutyrate, Choline, Creatine, γ-aminobutyric acid (GABA), α-Glucose, β-Glucose, Glutamate, Glutamine, Glutathione, Lactate, MyoInosi- tol, NAA, PhosphoCholine, PhosphoCreatine, Taurine). In vivo brain metabolomics profiling using 1H magnetic resonance spectroscopy (1H-MRS) The contribution of macromolecules to the in vivo spectrum was fitted using an in vivo macromolecule reference spectrum (composite of 4 regions) acquired with the metabolite peaks suppressed using a double-inver- sion module in the LASER sequence; lipid components were fitted using the standard reso- nances in the LCModel package. Metabolite content is expressed relative to the total creatine (Cr+PCr) content; estimates for individual metabolites with Craemer-Rao lower bounds (CRLB) >20% were discarded from the group analysis. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Brain metabolite analyses Targeted metabolic profiling was conducted at Case Western on whole homogenate brain tis- sues collected from 20-week-old mice (C57BL/6 and LDLr -/-) fed either CD or WD for 12 weeks. Tricarboxylic acid (TCA) cycle intermediates, total (free + bound) lipids and choles- terol, acyl-CoA’s and related amino acids, such as glutamine, glutamate, and aspartate, as well as ketone body (beta-hyroxybutyrate; BHB) and lactate were quantitatively assayed using sen- sitive gas chromatography (GC/MS) [38–41] and liquid chromatography (LC/MS/MS) mass spectrometry [42–44] analyses. Each of the selected intermediates was quantified against refer- ence standards to yield absolute concentrations (μmol/g wet weight tissue). Following decapi- tation under isoflurane twilight anesthesia, the brains from each mouse were rapidly collected and quickly frozen under liquid nitrogen [41, 43]. Metabolic profiles were conducted on a sin- gle mouse brain. Brain tissues (~200 mg, cortical-subcortical/mouse) were homogenized in a methanol-5% acetic acid (in milli-Q water) buffer solution (3 ml) following additions of refer- ence standards. Before centrifugation an aliquot (0.2 ml) of homogenate was reserved for fatty acid and cholesterol assays and the remaining homogenate then was centrifuged and the supernatant fraction (extract) was collected and divided into three additional aliquots. One ali- quot (2.2 ml) from the extract was used for acyl-CoA analysis (LC/MS/MS), and the other ali- quots were used for GC/MS analysis. Analysis of TCA and amino acid metabolite. For measurements of TCA and amino acid metabolites: an aliquot of extract (0.2 ml) was dried by nitrogen gas for 1–2 hr. and derivatized using MTBSTFA + 1% TBDMCS reagent (N-methyl-N-(tert-butyldimethylsilyl) trifluoroace- tamide + 1% tert-butyldimetheylchlorosilane, Regis Technologies, Inc. Morton Grove, IL, USA) and reacted at 70˚C for 30 min [39–41]. Each of the derivatized products were analyzed as trimethylsilyl derivatives on an Agilent 5973N-MSD equipped with an Agilent 6890 GC system (GC-MS) coupled to a DB-17MS capillary column (30m x 0.25mm x 0.25 μm) and operated in electron impact ionization (EI) sim mode during increasing oven temperature gradient. The corresponding ions monitored for each metabolite were as follows: BHB PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 8 / 38 Diet & LDLr-/- alter brain function & metabolism (m/z = 159), succinate (m/z = 289), fumarate (m/z = 287), (malate m/z = 419), citrate (m/ z = 459). Brain metabolite analyses Other intermediates and amino acids, including 3-hydroxyglutarate (m/z = 433), aspartate (m/z = 418), glutamate (m/z = 432), glutamine (m/z = 431) and GABA (m/z = 274) were also measured. The analysis of lactate involved using 0.05 ml of extract and the GC/MS system as describe above, but slightly modified methods. Lactate (m/z = 131) was analyzed as a PFBBR derivative and operated in chemical ionization mode (CI, ammonia gas) [42]. Fatty acid and cholesterol concentrations were analyzed from the aliquot reserved from the homogenate and then dried under nitrogen gas. The individual lipids were analyzed as tri- methylsilyl derivatives using the GC/MS system as described above (EI sim mode) [38, 45]. The corresponding ions monitored for each of the lipids were as follows: C:14 (m/z = 285), C16:0 (m/z = 313), C17:0 internal standard (m/z = 327), C:16:1 (m/z = 311), C:18 (m/z = 341), C18:1 (m/z = 339), C18:2 (m/z = 337), C:20 (m/z = 369), C:20:1 (m/z = 367), C:22 (m/z = 397) and cholesterol (m/z = 368). Acyl-CoA profiles analysis. Acyl-CoA profiles were assayed from the remaining extract (~2.2 ml) [41–44]; extract was loaded onto a Supelco solid-phase extraction cartridge ([2-(pyri- dyl)-ethyl functionalized silica gel]) pretreated with methanol. The cartridge was washed with a buffer containing methanol with 2% acetic acid, followed by elution of acyl-CoAs using buff- ers containing ammonium formate and/or methanol. The eluent was evaporated under nitro- gen gas, and applied to LC/MS/MS. The LC was coupled with an API4000 Qtrap MS (Applied Biosystems, Foster City, CA, USA) operated under positive ionization mode, and the Q1/Q3 components were monitored for each acyl-CoA species (C2:0 –C20:4 CoA). Complex lipids. Whole brain metabolic analysis of complex lipids was assessed by the West Coast Metabolomic Center at UC Davis by charged-surface hybrid column electrospray ionization quadrupole time of flight tandem mass spectrometry (CSH-ESI QTOF MS/MS) in both positive and negative modes using methods described previously [46–49]. Samples were extracted using the Matyash protocol using methyl tert-butyl ether (MTBE) [50]. 4 mg of pul- verized brain sample was mixed with 225 μL of ice-cold degassed MeOH and vortexed for 20s, 750 μL of ice-cold degassed MTBE was then added followed by vortexing for 20s and homoge- nize for 30s. MilliQ water was added (188 μL) followed by vortexing (20 s) and centrifugation (2 min; 14,000 g). Brain metabolite analyses Procedure shortly: ~25 mg mouse brain was homogenized using Geno rider 2000 sample homogenizer (SPEX Sample Prep; Metuchen, NJ) together with 5 μL BHT/EDTA (1:1 MeOH:water), 5 μL of 1 μM deuterated surrogates in methanol and PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 9 / 38 Diet & LDLr-/- alter brain function & metabolism 95 μL of methanol. Further, samples were mixed with 0.5 mL of deionized water and 1 mL of ethyl acetate and subsequently vortexed and centrifuged (10 min at 15,000 rcf). The organic phase was retreated and dried by applying a vacuum at 15 Hg for 10 min. Samples were recon- stituted in 100μl of 1-cyclohexyl ureido, 3-dodecanoic acid (CUDA) and 1-Phenyl 3-Hexade- canoic Acid Urea (PHAU) at 100nM in 1:1 methanol:acetonitrile. Residues within extracts were separated on a 2.1 x 150mm 0.17μm BEH column (Waters) and detected by electrospray ionization with multi reaction monitoring on a API 6500 QTRAP (Sciex; Redwood City, CA) and quantified against 7–9 point calibration curves of authentic standards using modifications of previously reported methods [52, 53]. Statistics Values are reported as mean ± SEM unless otherwise noted. Statistical calculations were done with the Sigma Stat software (Systat Software Inc., San Jose, CA), JMP (SAS institute Inc, Cary, NC, USA), and GraphPad Prism software (GraphPad Software Inc., La Jolla, CA). Two-way ANOVA tables were generated as previously described [54]. Grubb’s test was used to exclude outliers and samples not normally distributed were transformed using 1 x ; ﬃﬃﬃx p , or log x. Student t-tests, or one-way or two-way analysis of variance (ANOVA) followed by Tukey or Student- Newman-Keuls (SNK) post-hoc tests were used to test for differences between groups as indi- cated in the figure legends and statistical differences assigned at p0.05. Non-esterified oxyli- pins and endocannabinoid data were Log transformed and auto-scaled before multivariate analysis, as previously reported [55]. Statistical analyses including; partial least squares analy- sis, variable hierarchical clustering and repeated measures ANOVA were performed in JMP-Pro v 12.2. Repeated measures ANOVA was performed on variable clusters with a ran- dom subject effect, whereas variable cluster members, diet and genotype were used as fixed effects. The model tested for the diet, genotype, and the diet x genotype interaction effect. Brain metabolite analyses The resulting upper phase is then transferred (350 μL) to a separate tube, dried, and reconstituted with 65 μL MeOH:toluene+CUDA (9:1, v/v). Aliquots of 30 μL were transferred to two separate vials with micro-inserts for UHPLC-QTOF-MS analysis. Samples (3 μL) were injected at 65˚C and separated using a Waters Acquity UPLC CSH C18 column (100mm× 2.1 mm) with a particle size of 1.9 μm and a flow rate of 0.6 mL/min. Mass spec- trometry was conducted for positively charged ions (phosphatidylcholine (PC), lysoPC, PE, and PS) with an Agilent 6530 QTOF MS (resolution: 10,000) and for negatively charged ions (free fatty acids and phosphatidylinositols) with an Agilent 6550 QTOF MS (resolution: 20,000). Both mass spectrometers operated at full scan range m/z 65–1,700. Peak identification was processed in MassHunter Qual (Agilent) using the MS/MS information and Fiehn labora- tory LipidBlast spectral library [51] and then imported to Mass Profiler Professional for peak alignment. Results are provided as quantifier ion peak heights and normalized to the sum of all peak heights for all identified metabolites for each sample. In-depth details of the protocol can be found through the Metabolomics Workbench under protocol number 163 (http://www. metabolomicsworkbench.org/protocols/protocoldetails.php?file_id=163). Complex lipids. Whole brain metabolic analysis of complex lipids was assessed by the West Coast Metabolomic Center at UC Davis by charged-surface hybrid column electrospray ionization quadrupole time of flight tandem mass spectrometry (CSH-ESI QTOF MS/MS) in both positive and negative modes using methods described previously [46–49]. Samples were extracted using the Matyash protocol using methyl tert-butyl ether (MTBE) [50]. 4 mg of pul- verized brain sample was mixed with 225 μL of ice-cold degassed MeOH and vortexed for 20s, 750 μL of ice-cold degassed MTBE was then added followed by vortexing for 20s and homoge- nize for 30s. MilliQ water was added (188 μL) followed by vortexing (20 s) and centrifugation (2 min; 14,000 g). The resulting upper phase is then transferred (350 μL) to a separate tube, Non-esterified oxylipins and endocannabinoids analysis. Oxylipins, endocannabinoids, and fatty acids were isolated using methanol:ethyl acetate liquid-liquid extraction protocol from ~25 mg of the mouse brain. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Western diet leads to poorer Y-maze alternation and better Morris maze retention while LDLr -/- mice were more active and showed impaired learning in the radial arm maze To assess the cognitive effects of LDLr -/- and/or WD mice were tested with either both the Y- maze and Morris water maze (MWM), or the radial arm water maze (RAWM) at the age of 20 weeks (12 weeks on defined diet). In addition to cognitive performance as reflected in correct response, performance variables (like speed and thigmotaxis) were analyzed for each of the maze tests. In the Y-maze, LDLr -/- mice were consistently more active than C57BL/6 (WT) with an increased number of total arm entries (63.9±2 vs 54.3±2), total distance travelled (4103.7±104 vs 3538.2±102.5), and decreased % resting time (10.3±.06 vs 14.8±1.0). There was no effect of diet on performance endpoints; however, diet significantly affected the cognitive endpoint % alternation triplets regardless of genotype (p0.05). Poorer performance (fewer alternation triplets) was seen in the WD-fed group (Fig 1A). There was also a trend (p = 0.06) toward an interaction between genotype and diet, with the diet effect more prominent in the LDLr -/- subgroup. The MWM included visible, training, and probe trials. No effects of diet or genotype were seen for the ability of the mice to find a visible platform trial and this performance was not associated with body weight at the end of behavioral testing (20 weeks of age). With the train- ing trials, there were no effects of diet or genotype seen for performance endpoints (floating, thigmotaxis, and swimming speed). Similarly, no effects were seen on the cognitive measures PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 10 / 38 Diet & LDLr-/- alter brain function & metabolism Fig 1. Diet and genotype effects on cognitive endpoints from three different behavioral tests. The performanc C57BL/6 (WT) or LDLr-/- mice in the Y-maze alternation was determined after they were fed a control (CD) or western (WD) diet (A). The mean percentage of triplicates (number of alternations, number of times 3 adjacent ar divided by the total number of arms entered) is shown (n = 8/grp). These same animals were also exposed to the Morris Water Maze (MWM, B) where % time spent in the platform quadrant during the probe trial is shown. In Fig 1. Diet and genotype effects on cognitive endpoints from three different behavioral tests. The performance of C57BL/6 (WT) or LDLr-/- mice in the Y-maze alternation was determined after they were fed a control (CD) or western (WD) diet (A). PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Western diet leads to poorer Y-maze alternation and better Morris maze retention while LDLr -/- mice were more active and showed impaired learning in the radial arm maze The mean percentage of triplicates (number of alternations, number of times 3 adjacent arms, divided by the total number of arms entered) is shown (n = 8/grp). These same animals were also exposed to the Morris Water Maze (MWM, B) where % time spent in the platform quadrant during the probe trial is shown. In PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 11 / 38 Diet & LDLr-/- alter brain function & metabolism separate but identically treated groups of WT and LDLr -/- animals, mice were exposed to the Radial Arm Water Maze (RAWM, C) and the number of trial 4 errors in day’s 1–4 vs 5–9 of testing is shown (n = at least 9/grp). Values are mean ±SEM and analyzed by two-way ANOVA with Tukey’s multiple comparisons post hoc test. separate but identically treated groups of WT and LDLr -/- animals, mice were exposed to the Radial Arm Water Maze (RAWM, C) and the number of trial 4 errors in day’s 1–4 vs 5–9 of testing is shown (n = at least 9/grp). Values are mean ±SEM and analyzed by two-way ANOVA with Tukey’s multiple comparisons post hoc test. https://doi.org/10.1371/journal.pone.0191909.g001 separate but identically treated groups of WT and LDLr -/- animals, mice were exposed to the Radial Arm Water Maze (RAWM, C) and the number of trial 4 errors in day’s 1–4 vs 5–9 of testing is shown (n = at least 9/grp). Values are mean ±SEM and analyzed by two-way ANOVA with Tukey’s multiple comparisons post hoc test. https://doi.org/10.1371/journal.pone.0191909.g001 https://doi.org/10.1371/journal.pone.0191909.g001 of learning, the decrease in distance and time required to reach the platform between the first and last day of training. As a group (WT and LDLr-/-), the mice decreased distance to platform by 43±5% and time to platform by 30±7% during training. While there were no effects of diet or genotype on performance endpoints (swim speed, float time, or time in “wall zone”) in the probe trial, a diet effect (p0.05) was seen for the cognitive endpoint % time in the platform quadrant (Fig 1B). This finding was supported by a trend toward a diet effect (p = 0.09) for mean distance to target, or the ability of the mice to maintain a search pattern close to the plat- form. In both cases, the WD group showed greater time in platform quadrant and reduced dis- tance to quadrant compared to the CD fed group. Western diet leads to poorer Y-maze alternation and better Morris maze retention while LDLr -/- mice were more active and showed impaired learning in the radial arm maze Since the MWM and Y-maze were run on the same animals, we determined whether there was an association between the results of these tests by principal components analysis (PCA). We found no correlation between the main cognitive endpoints of the MWM and Y-maze, % platform quadrant time in the probe trial and % alternation triplets, (r = -0.03). This indicated that the two endpoints were measuring distinct cognitive processes with no interdependence in determining behavior for these endpoints. Body mass at the time of testing, cognitive, and performance endpoints were entered into PCA to determine the associations among these endpoints. Analysis indicated that performance variables were not strongly associated with cognitive variables in either task. To assess spatial learning and memory with a more complex test of learning, two additional cohorts of mice were tested by RAWM either pre- and post-12-week experimental diet inter- vention (twice), or only after the experimental diet intervention. We found no genotype effects on performance endpoints (float time or latency to platform) in mice tested pre-experimental diet. In addition, no genotype or diet effects were seen on performance endpoints in mice only post-tested after completion of the dietary exposure. We found that the initial pre-diet testing substantially improved the learning of the mice at the post-experimental diet test. Mice who had been tested previously averaged 1.8±0.2 errors during the first 4 RAWM sessions as com- pared to 3.4±0.2 errors for mice who had not been tested previously. Therefore only the first testing experience (cohort 1 = pre-diet, cohort 2 = post diet) was used to assess genotype effects. The LDLr -/- mice made the same number of errors as WT on the early problems (day 1–4), but made more on the later problems (days 5–9) demonstrating lack of improvement as seen in Fig 1C. Since both WD and LDLr-/- induce altered cognitive function, subsequent analysis of BBB transport, glucose uptake, neuroinflammation, and brain metabolites were assessed in both models as well as potential interaction. Western diet feeding increases blood-brain barrier transport WT and LDLr -/- mice fed a CD or WD were subjected to MRI analysis to determine if diet and/or the hyperlipidemic genotype altered BBB transport, indicated by a change in transfer coefficient (Ki = permeability x surface area/volume). Two-way ANOVA showed a significant effect of diet on the BBB transfer coefficient Ki where WD increased Ki when WT and LDLr-/- data were pooled (p0.05, Fig 2A). There was no significant difference between the genotypes nor was there a significant interaction between genotype and diet (p = 0.653 and 0.578, respec- tively). Increased Ki can be seen in the brain (circled in red) as a shift to yellow in the represen- tative images of Ki maps for each group (Fig 2B). These shifts in Ki were not associated with measurable differences in blood flow as perfusion weighted imaging of cerebral CBF to the PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 12 / 38 Diet & LDLr-/- alter brain function & metabolism Fig 2. Western diet increases BBB transfer coefficient (Ki). Wild type (WT) and LDLr -/- animals on either a control (CD) or western diet (WD) are plotted against their mean Ki (min-1) ± SEM (A). Representative images of the Ki maps of WT CD, LDLr -/- CD, WT WD and LDLr -/- WD are shown (B) where brain is circled in red. Two-way ANOVA showed a significant effect of diet on the BBB transfer coefficient Ki where WD increased Ki when WT and LDLr-/- data were pooled denoted by , p<0.05 (n = at least 6/grp). https://doi.org/10.1371/journal.pone.0191909.g002 Fig 2. Western diet increases BBB transfer coefficient (Ki). Wild type (WT) and LDLr -/- animals on either a control (CD) or western diet (WD) are plotted against their mean Ki (min-1) ± SEM (A). Representative images of the Ki maps of WT CD, LDLr -/- CD, WT WD and LDLr -/- WD are shown (B) where brain is circled in red. Two-way ANOVA showed a significant effect of diet on the BBB transfer coefficient Ki where WD increased Ki when WT and LDLr-/- data were pooled denoted by , p<0.05 (n = at least 6/grp). Fig 2. Western diet increases BBB transfer coefficient (Ki). Wild type (WT) and LDLr -/- animals on either a control (CD) or western diet (WD) are plotted against their mean Ki (min-1) ± SEM (A). PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 https://doi.org/10.1371/journal.pone.0191909.g002 Western diet feeding increases blood-brain barrier transport Representative images of the Ki maps of WT CD, LDLr -/- CD, WT WD and LDLr -/- WD are shown (B) where brain is circled in red. Two-way ANOVA showed a significant effect of diet on the BBB transfer coefficient Ki where WD increased Ki when WT and LDLr-/- data were pooled denoted by , p<0.05 (n = at least 6/grp). Fig 2. Western diet increases BBB transfer coefficient (Ki). Wild type (WT) and LDLr -/- animals on either a control (CD) or western diet (WD) are plotted against their mean Ki (min-1) ± SEM (A). Representative images of the Ki maps of WT CD, LDLr -/- CD, WT WD and LDLr -/- WD are shown (B) where brain is circled in red. Two-way ANOVA showed a significant effect of diet on the BBB transfer coefficient Ki where WD increased Ki when WT and LDLr-/- data were pooled denoted by , p<0.05 (n = at least 6/grp). https://doi.org/10.1371/journal.pone.0191909.g002 13 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Diet & LDLr-/- alter brain function & metabolism brain was essentially normal at 1.7±0.025 mL/100gm/sec and significant differences between mean CBF measured for phenotypes and genotypes were not found (data in supplement). Fur- ther, post-imaging venous blood samples were drawn from the vena cava and there were no significant differences in pH, pCO2, pO2 levels between the groups (data in supplement). Diet or genotype effects on inflammatory gene expression Since we found a significant increase in IBA1 staining in WT mice fed a WD, we attempted to determine the type of inflammatory response that was occurring. For this, whole brain expres- sion of genes known to be associated with an M1 or M2 microglia activation (CD16, CD32 & CD86 vs CD 206 & arginase 1) [56, 57], neurodegeneration (IL-1B and IL-6), neuroprotection (CX3CR1, TREM2, IGF1, GDF15, IL10) vascular inflammation (factor VIII and TNFα) or additional stress and inflammatory pathways (ATF3, CHOP, JNK, NFκβ) were assessed. While many of these showed no significant difference by genotype (WT vs LDLr -/-), diet (CD vs WD), or diet x genotype interaction, we found a significant decrease in ATF3, CD206, CD32, and IL-6 in LDLr -/- mice (when compared to WT) and a significant decrease in CD86 and a trend for a decrease in IBA1 with a WD (Table 2). In addition, prostaglandin-endoperoxide synthase 2 (PTGS2, also known as COX2) was significantly increased with a WD. Altered vessel density and microglia activation in western diet fed mice To better understand the influence of a WD on brain vascularization, an immunohistological analysis of vessel density was determined by factor VIII staining. We found a significant increase in the relative surface area stained with factor VIII in the hippocampus of LDLr -/- WD-fed mice relative to both control WT or LDLr -/- mice fed a CD, and in the thalamus compared to WT CD-fed mice (Fig 3A). Subjective evaluation suggests that this represents an increase in factor VIII expression in individual endothelial cells in larger vessels as well as expression in capillary endothelium that has a paucity of staining in other groups (Fig 3B and 3C). Subjective histopathologic analysis found no overt differences between genotype and diet groups in brain sections examined nor planimetry measurements of overall surface area for the mid temporal and occipital cortex. To determine the extent and location of microglia activation in the brains of WD-fed and/ or LDLr -/- mice, brain sections were stained for IBA1, which labels microglia and is increased in activated microglia. There was a significant increase in IBA1 staining in WT mice fed a WD compared with mice fed the CD in both the cortex and hippocampus (Fig 3D–3H). In addi- tion, microglial cells had markedly more prominent processes (Fig 3E and 3F). The density of IBA1 staining in both CD- and WD-fed LDLr -/- mice were no different from WT CD fed mice. Similar trends in the thalamus were not statistically significant, in part due to variability in the extent to which the cerebral peduncles were included in the section evaluated. Subjective observation clearly demonstrated that the cerebral peduncles contained a significantly greater density of IBA1 positive cells regardless of diet of genotype group. LDLr -/- mice show altered brain glucose metabolism To determine the effects on brain glucose uptake, C57BL/6 (WT) and LDLr -/- mice fed either a WD or CD underwent 18FDG PET/CT imaging. 18FDG SUVglu values for brain ROIs in each respective group were calculated from summed PET images from 25–30 min post-injec- tion (Fig 4) where LDLr -/- showed a significant increase in glucose uptake compared to WT mice (2.37 ± 0.10 and 2.00 ± 0.09, p0.05). PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 14 / 38 Diet & LDLr-/- alter brain function & metabolism Fig 3. Factor VIII immunostaining density increases in hippocampus and thalamus of LDLr -/- while IBA1 increases in the cortex and hippocampus of C57BL/6 (WT) fed a western diet. Area density of factor VIII and IBA1 immunostaining in coronal section of C57BL/6 (WT) or LDLr -/- mice fed control (CD) or western diet (WD) as determined by image analysis (A and D). Representative images comparing factor VIII immunostaining in thalamus of LDLr -/- mice fed CD (B) or WD (C) (scale bar = 50μm). WD fed animals had more granules of factor VIII positive material in capillaries (arrows). Microglia in control fed mice (E) have most IBA-1 Fig 3. Factor VIII immunostaining density increases in hippocampus and thalamus of LDLr -/- while IBA1 increases in the cortex and hippocampus of C57BL/6 (WT) fed a western diet. Area density of factor VIII and IBA1 immunostaining in coronal section of C57BL/6 (WT) or LDLr -/- mice fed control (CD) or western diet (WD) as determined by image analysis (A and D). Representative images comparing factor VIII immunostaining in thalamus of LDLr -/- mice fed CD (B) or WD (C) (scale bar = 50μm). WD fed animals had more granules of factor VIII positive material in capillaries (arrows). Microglia in control fed mice (E) have most IBA-1 Fig 3. Factor VIII immunostaining density increases in hippocampus and thalamus of LDLr -/- while IBA1 increases in the cortex and hippocampus of C57BL/6 (WT) fed a western diet. Area density of factor VIII and IBA1 immunostaining in coronal section of C57BL/6 (WT) or LDLr -/- mice fed control (CD) or western diet (WD) as determined by image analysis (A and D). Representative images comparing factor VIII immunostaining in thalamus of LDLr -/- mice fed CD (B) or WD (C) (scale bar = 50μm). WD fed animals had more granules of factor VIII positive material in capillaries (arrows). Increase of lactate in LDLr-/- mice and lipid moieties by proton magnetic resonance spectroscopy (1H-MRS) in LDLr -/- mice fed a WD To measure shifts in metabolites in vivo, 1H-MRS were analyzed using spectral fitting to esti- mate the content of 17 discrete metabolites plus lipid and macromolecules within the cortex (representative image and spectra shown in Fig 5). Lactate concentrations were significantly increased in the LDLr -/- mice compared to C57BL/6 (2.06±0.14 and 1.47±0.13, p0.05). In addition, LDLr -/- mice fed a WD had higher lipid concentrations (lipid 13a, lipid 9 and lipid 13a+13b) compared to LDLr -/- on CD or to WT on either diet (Table 3). While not signifi- cant, there was a trend (p = 0.084) for glutamine to decrease in the cortex of LDLr -/- when compared to control mice (2.92±0.14 and 3.70 ±0.41, respectively). Table 2. Shift in gene expression with western diet and/or Ldlr -/-. C57BL/6 (WT) or LDLr-/- mice were fed a control (CD) or western (WD) diet for 12 weeks and their brain gene expression was measured. Values are fold of WT on a CD ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two- way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = at least 5/grp). LDLr -/- mice show altered brain glucose metabolism Microglia in control fed mice (E) have most IBA-1 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 15 / 38 Diet & LDLr-/- alter brain function & metabolism positivity surrounding the nucleus while those from western diet (F) fed animals have denser staining that extends through elongate strands of cytoplasm (scale bar = 20μm). Representative images comparing IBA1 immunostaining in cortex (red), hippocampus (blue) and thalamus (green) of WT mice fed CD (G) or WD (H) (scale bar = ~0.5mm). Statistical differences from WT () or LDLr -/- (#) CD fed mice or LDLr-/-WD fed ($), as determined by ANOVA with Tukey’s multiple comparisons post hoc test p<0.05, n = at least 5/grp). positivity surrounding the nucleus while those from western diet (F) fed animals have denser staining that extends through elongate strands of cytoplasm (scale bar = 20μm). Representative images comparing IBA1 immunostaining in cortex (red), hippocampus (blue) and thalamus (green) of WT mice fed CD (G) or WD (H) (scale bar = ~0.5mm). Statistical differences from WT () or LDLr -/- (#) CD fed mice or LDLr-/-WD fed ($), as determined by ANOVA with Tukey’s multiple comparisons post hoc test p<0.05, n = at least 5/grp). htt //d i /10 1371/j l 0191909 003 https://doi.org/10.1371/journal.pone.0191909.g003 https://doi.org/10.1371/journal.pone.0191909.t002 Increase of lactate in LDLr-/- mice and lipid moieties by proton magnetic resonance spectroscopy (1H-MRS) in LDLr -/- mice fed a WD Variable WT LDLr -/- P-value CD (n = 6) WD (n = 5) CD (n = 6) WD (n = 6) Genotype Diet G×D1 ARG 1.06 ± 0.153 1.21 ± 0.22 0.933 ± 0.131 0.85 ± 0.0737 0.127 0.844 0.434 ATF3 1.02 ± 0.0988 0.956 ± 0.0453 0.863 ± 0.0286 0.833 ± 0.0205 0.023 0.426 0.763 C1qA 1.01 ± 0.0514 1.03 ± 0.104 0.833 ± 0.103 0.989 ± 0.0667 0.256 0.306 0.42 C1qB 1 ± 0.0109 0.95 ± 0.091 0.921 ± 0.0678 0.943 ± 0.065 0.494 0.828 0.584 C1qC 1 ± 0.0308 1.01 ± 0.088 1.04 ± 0.0332 1.08 ± 0.0522 0.335 0.688 0.836 CCL2 1.03 ± 0.0995 1.04 ± 0.0434 0.929 ± 0.0862 0.847 ± 0.0471 0.075 0.618 0.561 CD11b 1.01 ± 0.0616 1.06 ± 0.0537 1.14 ± 0.0801 1.07 ± 0.0515 0.287 0.839 0.367 CD16 1 ± 0.0441 0.957 ± 0.0346 1.07 ± 0.135 0.979 ± 0.0689 0.629 0.411 0.8 CD32 1.02 ± 0.0877a 1.01 ± 0.0465ab 0.782 ± 0.0369bc 0.753 ± 0.0406c <0.001 0.709 0.894 CD86 1.04 ± 0.118 0.957 ± 0.0637 1.14 ± 0.0623 0.869 ± 0.0715 0.973 0.042 0.273 CD206 1.07 ± 0.162 1.02 ± 0.0717 0.83 ± 0.0468 0.783 ± 0.0311 0.02 0.595 0.976 CXCL2 1.01 ± 0.0513 1.48 ± 0.41 1.18 ± 0.0795 0.995 ± 0.0866 0.473 0.473 0.114 DDIT3 1.01 ± 0.0507 1.01 ± 0.0381 1.12 ± 0.106 0.964 ± 0.0592 0.627 0.269 0.298 Factor VIII 1.04 ± 0.124 1.1 ± 0.0806 1.11 ± 0.0815 1.03 ± 0.0472 0.992 0.927 0.471 GDF15 1 ± 0.0387 1.07 ± 0.066 1.14 ± 0.126 1.14 ± 0.11 0.258 0.738 0.762 IBA1 1.01 ± 0.0487 0.912 ± 0.0291 1.14 ± 0.132 0.921 ± 0.0348 0.394 0.053 0.431 IGF 1.08 ± 0.186 1.36 ± 0.207 0.955 ± 0.0489 0.988 ± 0.179 0.169 0.366 0.457 IkBα 1.04 ± 0.136 1.15 ± 0.064 1.11 ± 0.112 1.32 ± 0.204 0.395 0.276 0.75 IL-12a 1.4 ± 0.616 0.851 ± 0.119 0.789 ± 0.068 1.23 ± 0.213 0.697 0.938 0.173 IL-1B 1.03 ± 0.103 0.971 ± 0.129 0.821 ± 0.123 0.711 ± 0.114 0.058 0.481 0.822 IL- 6 1.03 ± 0.112a 0.821 ± 0.191ab 0.536 ± 0.061b 0.572 ± 0.0925b 0.004 0.494 0.305 MAPK 8 1.02 ± 0.101 1.29 ± 0.19 1.04 ± 0.0821 0.899 ± 0.0603 0.125 0.628 0.081 NR4A2 1.01 ± 0.0784 0.994 ± 0.0758 0.902 ± 0.046 0.988 ± 0.0504 0.355 0.581 0.408 PTGS2 1.12 ± 0.231 1.3 ± 0.0868 0.942 ± 0.174 1.49 ± 0.172 0.952 0.048 0.322 TNFα 1.01 ± 0.0712 1.07 ± 0.0447 0.989 ± 0.0357 0.948 ± 0.0354 0.178 0.933 0.352 TREM2 1.02 ± 0.0839 1.04 ± 0.108 1.3 ± 0.141 1.05 ± 0.109 0.214 0.289 0.231 https://doi org/10 1371/journal pone 0191909 t002 Table 2. https://doi.org/10.1371/journal.pone.0191909.t002 Increase of lactate in LDLr-/- mice and lipid moieties by proton magnetic resonance spectroscopy (1H-MRS) in LDLr -/- mice fed a WD Shift in gene expression with western diet and/or Ldlr -/-. C57BL/6 (WT) or LDLr-/- mice were fed a control (CD) or western (WD) diet for 12 weeks and their brain gene expression was measured. Values are fold of WT on a CD ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two- way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = at least 5/grp). PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 16 / 38 Diet & LDLr-/- alter brain function & metabolism Fig 4. Genotype but not diet alters brain glucose uptake. 18FDG uptake measured by mean standardized uptake value (SUVglu) corrected for plasma glucose from a summed static image from 20–30 min post injection of C57BL/6 (WT) or LDLr -/- mice fed control (CD) or western diet (WD). Statistical differences by two-way ANOVA between WT and LDLr -/- mice () p<0.05, n = at least 4/grp. https://doi.org/10.1371/journal.pone.0191909.g004 Fig 4. Genotype but not diet alters brain glucose uptake. 18FDG uptake measured by mean standardized uptake value (SUVglu) corrected for plasma glucose from a summed static image from 20–30 min post injection of C57BL/6 (WT) or LDLr -/- mice fed control (CD) or western diet (WD). Statistical differences by two-way ANOVA between WT and LDLr -/- mice () p<0.05, n = at least 4/grp. https://doi.org/10.1371/journal.pone.0191909.g004 https://doi.org/10.1371/journal.pone.0191909.g004 Diet & LDLr-/- alter brain function & metabolism Table 3. Shift in brain metabolites assessed by 1H-MRS by western diet and/or LDLr -/-. C57BL/6 (WT) or LDLr-/ mice were fed a control (CD) or western (WD) diet for 12 weeks and their brain fatty acid and TCA cycle intermediates were measured. Values are mean of abundance relative to Cr+PCr ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (p 0.05, n = at least 7/grp). for 12 weeks and their brain fatty acid and TCA cycle intermediates were measured. Values are mean of abundance relative to Cr+PCr ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (p 0.05, n = at least 7/grp). Metabolite WT LDLr -/- P-value CD (n = 7) WD (n = 8) CD (n = 8) WD (n = 7) Genotype Diet G×D1 Alanine 1.2 ± 0.303 1.21 ± 0.143 1.12 ± 0.195 1.31 ± 0.319 0.951 0.691 0.727 Aspartate 3.2 ± 0.254 3.49 ± 0.375 3.6 ± 0.238 3.69 ± 0.327 0.387 0.566 0.765 Choline 0.598 ± 0.0718 0.726 ± 0.0676 0.663 ± 0.0767 0.566 ± 0.0826 0.517 0.836 0.146 Creatinine 5.19 ± 0.122 5.02 ± 0.107 5.18 ± 0.132 4.96 ± 0.164 0.872 0.152 0.88 GABA 1.42 ± 0.121 1.66 ± 0.0956 1.64 ± 0.179 1.5 ± 0.0827 0.798 0.692 0.155 αGlucose 1.65 ± 0.258 1.54 ± 0.148 1.39 ± 0.0925 1.46 ± 0.11 0.3 0.879 0.584 βGlucose 2.17 ± 0.316 2.1 ± 0.214 1.84 ± 0.147 1.94 ± 0.241 0.299 0.954 0.741 Glutamate 7.95 ± 0.282 8.17 ± 0.106 8.16 ± 0.315 8.36 ± 0.38 0.514 0.463 0.957 Glutamine 3.79 ± 0.639 3.63 ± 0.557 3.25 ± 0.174 2.56 ± 0.102 0.084 0.34 0.549 Glutathione 1.73 ± 0.121 1.65 ± 0.106 1.69 ± 0.0912 1.72 ± 0.0646 0.873 0.774 0.546 Lactate 1.79 ± 0.183ab 1.35 ± 0.163b 2.15 ± 0.191a 1.92 ± 0.227ab 0.018 0.186 0.666 MyoInositol 5.14 ± 0.295 5.21 ± 0.3 5.19 ± 0.0985 5.41 ± 0.218 0.633 0.55 0.756 NAA 6.06 ± 0.16 5.97 ± 0.129 5.94 ± 0.214 5.83 ± 0.113 0.47 0.525 0.963 PhosphoCholine 0.944 ± 0.0636 0.855 ± 0.0592 0.916 ± 0.0839 0.967 ± 0.0726 0.546 0.794 0.332 PhosphoCreatine 2.81 ± 0.122 2.98 ± 0.107 2.82 ± 0.132 3.04 ± 0.164 0.872 0.152 0.88 Taurine 6.73 ± 0.472 6.47 ± 0.221 7.01 ± 0.281 6.79 ± 0.405 0.367 0.496 0.957 MM C57 0.0667 ± 0.00303 0.0716 ± 0.00236 0.0695 ± 0.0041 0.0683 ± 0.00197 0.896 0.549 0.324 Cr+PCr 8 ± 0 8 ± 0 8 ± 0 8 ± 0 0.324 0.292 0.292 Glu+Gln 11.7 ± 0.643 11.8 ± 0.53 11.4 ± 0.22 10.9 ± 0.421 0.22 0.658 0.558 Lip13a 5.04 ± 0.353ab 4.72 ± 0.584ab 4.11 ± 0.389b 9.04 ± 2.31a 0.202 0.054 0.038 Lip 9 1.84 ± 0.2ab 1.6 ± 0.178ab 1.38 ± 0.208b 2.3 ± 0.337a 0.677 0.155 0.021 Lip 20 1.06 ± 0.129 1.05 ± 0.229 1.1 ± 0.175 1.09 ± 0.0886 0.818 0.965 0.998 Lip13a+Lip13b 5.05 ± 0.323 4.67 ± 0.583 4.31 ± 0.441 9.36 ± 2.73 0.186 0.093 0.053 https://doi org/10 1371/journal pone 0191909 t003 Many of the TCA intermediates and amino acids measured were affected by diet. Perturbation of fatty acid, TCA intermediate, and amino acid profiles in the brains of LDLr -/- and/or western diet fed mice Brains from a subset of LDLr -/- and WT, mice fed either a CD or WD were subjected to meta- bolic analysis by GC/MS and LC/MS/MS to determine fatty acid, cholesterol, TCA cycle inter- mediate, and amino acid profiles (Table 4). Genotype did not alter C14:0 (myristic acid) content in the brains but exposure to a WD increased C14:0 (p0.05) and there was a trend for a diet x genotype interaction (p = 0.054). In addition, there was a diet effect with respect to C16:1 content which was elevated in WD compared to CD group. A dietary effect on choles- terol profiles was observed in the WT group but not in the LDLr -/- group, as the WD-fed WT mice had significantly higher value than the CD group. Fig 5. In vivo brain metabolite content determined by 1H-MRS. Localized 1H spectra were obtained from an 18.75 μl volume (white box) positioned within the left or right hemisphere of the brain. Metabolite content was determined by spectral fitting with LC model using a basis set of 17 metabolites plus components due to macromolecules and intracellular lipids. https://doi.org/10.1371/journal.pone.0191909.g005 Fig 5. In vivo brain metabolite content determined by 1H-MRS. Localized 1H spectra were obtained from an 18.75 μl volume (white box) positioned within the left or right hemisphere of the brain. Metabolite content was determined by spectral fitting with LC model using a basis set of 17 metabolites plus components due to macromolecules and intracellular lipids. https://doi.org/10.1371/journal.pone.0191909.g005 17 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 https://doi.org/10.1371/journal.pone.0191909.t003 Diet & LDLr-/- alter brain function & metabolism Table 4. Shift in brain fatty acid and TCA cycle intermediate abundance with western diet and/or LDLr -/-. C57BL/6 (WT) or LDLr-/- mice were fed a control (CD) or western (WD) diet for 12 weeks and their brain fatty acid and TCA cycle intermediates were measured. Values are mean in μmol/g tissue ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = 4/grp). Table 4. Shift in brain fatty acid and TCA cycle intermediate abundance with western diet and/or LDLr -/-. C57BL/6 (WT) or LDLr-/- mice were fed a control (CD) or western (WD) diet for 12 weeks and their brain fatty acid and TCA cycle intermediates were measured. Values are mean in μmol/g tissue ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = 4/grp). Those with the most significant changes include; aspartate, GABA, 2-hydroxyglutarate (reduced α- ketoglutarate) and glutamine, as these intermediates were increased with WD irrespective of genotype. Additionally, succinate decreased with WD intervention in the WT but was not altered by diet in the LDLr -/- mice. β-hydroxybutyrate (BHB) was significantly increased with WD irrespective of genotype, and the LDLr -/- were significantly elevated over the WT group demonstrating a genotype effect. Lactate was significantly elevated in the WD groups; whereas the LDLr -/- genotype appeared to have no effect, as they were similar irrespective of diet. Diet or genotype had no effect on fumarate, malate, glutamate, or citrate. Acyl-CoA concentrations were significantly affected by WD and/or LDLr mutation. Ace- tyl-CoA, propionyl-CoA, butyryl-CoA, and BHB-CoA (intermediates of fatty acid oxidation) were significantly lower in the LDLr -/- group fed the WD compared to their diet matched WT groups. Although there were no significant differences in total-CoA concentrations with diet or genotype, the LDLr -/- group fed the WD trended lower compared to the WT WD group. C16:0-CoA, C18:0-CoA, C18:1-CoA, and C20:4 were not significantly different but C20:4 was significantly higher in the LDLr -/- group CD group. Although there were no signif- icant changes in malonyl-CoA (intermediate of fatty acid synthesis and regulator of carnitine acyl-transport system for fatty acid transport into mitochondria) with diet or genotype, there PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 18 / 38 Fatty Acids WT LDLr -/- P-value CD WD CD WD Genotype Diet G×D1 C14 1.62 ± 0.207ab 1.63 ± 0.123ab 1.35 ± 0.0777b 1.97 ± 0.134a 0.825 0.048 0.054 C16 264 ± 21.6 288 ± 5.08 277 ± 13.2 272 ± 10 0.921 0.527 0.305 C16:1 0.884 ± 0.0756b 1.13 ± 0.0512a 0.969 ± 0.0373ab 1.06 ± 0.0586ab 0.925 0.013 0.192 C18 288 ± 18.7 327 ± 10.2 317 ± 6.19 321 ± 10.1 0.36 0.103 0.175 C18:1 211 ± 27.2 212 ± 13.9 187 ± 16.9 174 ± 13 0.123 0.752 0.724 C18:2 2.2 ± 0.379 2.27 ± 0.388 2.07 ± 0.23 1.78 ± 0.0902 0.321 0.71 0.566 C 20 2.83 ± 0.307 3.7 ± 0.292 3.15 ± 0.429 3.09 ± 0.203 0.667 0.226 0.17 C20:1 11 ± 2.24 10.9 ± 0.735 8.74 ± 0.725 8.76 ± 0.462 0.102 0.976 0.962 C22 0.902 ± 0.152 0.992 ± 0.0792 0.864 ± 0.127 0.881 ± 0.0508 0.511 0.635 0.744 Cholesterol 55.8 ± 8.73b 93.7 ± 7.76a 88.9 ± 7.31a 81.5 ± 7.55ab 0.208 0.076 0.014 TCA Cycle intermediates Succinate 0.518 ± 0.0323a 0.355 ± 0.0274b 0.351 ± 0.0192b 0.386 ± 0.0325b 0.034 0.043 0.004 BHB 0.024 ± 0.00339b 0.0358 ± 0.00452ab 0.0345 ± 0.00259ab 0.0414 ± 0.00249a 0.033 0.016 0.478 GABA 4.66 ± 0.708 5.59 ± 0.743 3.84 ± 0.611 6.43 ± 0.504 0.982 0.019 0.223 Fumerate 0.591 ± 0.058 0.535 ± 0.0272 0.512 ± 0.0349 0.492 ± 0.0199 0.134 0.336 0.645 Malate 0.131 ± 0.00695 0.13 ± 0.0018 0.123 ± 0.00439 0.137 ± 0.00635 0.946 0.252 0.163 Aspartate 2.05 ± 0.0786bc 2.7 ± 0.134a 1.81 ± 0.137c 2.49 ± 0.201ab 0.143 0.001 0.91 2-Hydroxyglutarate 0.213 ± 0.00918b 0.272 ± 0.0143a 0.202 ± 0.0142b 0.271 ± 0.00987a 0.622 <0.001 0.677 Glutamate 16.6 ± 1.81 19.2 ± 1.34 18.2 ± 1.99 21.7 ± 0.989 0.224 0.081 0.779 Glutamine 6.46 ± 0.287bc 7.63 ± 0.363a 5.78 ± 0.205c 7.29 ± 0.159ab 0.077 <0.001 0.531 Citrate 0.131 ± 0.0309 0.136 ± 0.0131 0.119 ± 0.0172 0.125 ± 0.0126 0.585 0.803 0.986 Lactate 12.9 ± 0.828b 17.8 ± 1.34a 16.1 ± 0.825ab 17.5 ± 0.963a 0.185 0.009 0.112 Acyl-CoA’s Malonyl CoA 1.07 ± 0.144 1.23 ± 0.134 1.12 ± 0.119 1.25 ± 0.0971 0.751 0.293 0.914 Acetyl CoA 3.25 ± 0.217bc 4.01 ± 0.235ab 4.23 ± 0.181a 3.02 ± 0.169c 0.991 0.294 <0.001 Succinyl CoA 2.57 ± 0.287 2.93 ± 0.548 2.5 ± 0.404 3.36 ± 0.354 0.669 0.158 0.556 Butyryl CoA 1.01 ± 0.0306ab 1.11 ± 0.0198a 1.09 ± 0.0158a 0.986 ± 0.0239b 0.327 0.85 0.001 Propionyl CoA 0.171 ± 0.00543 0.191 ± 0.00179 0.188 ± 0.00511 0.172 ± 0.00682 0.86 0.725 0.004 HMG CoA 1.01 ± 0.0335 1.13 ± 0.0257 1.07 ± 0.0434 1.01 ± 0.039 0.415 0.437 0.022 BHB CoA 0.0788 ± 0.00618 0.105 ± 0.00713 0.0962 ± 0.0108 0.0727 ± 0.00538 0.495 0.945 0.01 C16 0 CoA 0.851 ± 0.107 0.857 ± 0.136 0.876 ± 0.069 0.868 ± 0.149 0.883 0.994 0.95 C18 1 CoA 0.324 ± 0.0709 0.363 ± 0.0704 0.371 ± 0.0505 0.343 ± 0.0717 0.843 0.927 0.623 C20 4 CoA 0.242 ± 0.0258 0.266 ± 0.0237 0.383 ± 0.0449 0.262 ± 0.0765 0.176 0.328 0.155 AcAc CoA 0.000871 ± 0.000157 0.000713 ± 3.95e-05 0.00084 ± 6.9e-05 0.00103 ± 0.000342 0.533 0.88 0.415 https://doi org/10 1371/journal pone 0191909 t004 p y g Complex lipid analysis by CSH-ESI QTOF MS/MS demonstrated that PC moieties 32:0–3, 36:3B, 38:4A, and 38:5A are all elevated in LDLr-/- while PC 30:0, 35:1, 38:1 in addition to diacylglycerol, and glucosylceramides 40:1 and 42:1 were elevated with a WD (negatively charged CSH, Table 5). was a decrease in the BHB-CoA to malonyl-CoA ratio in the LDLr -/- group fed the WD as compared to CD. This was likely due to the significant decrease in BHB-CoA. Complex lipid analysis by CSH-ESI QTOF MS/MS demonstrated that PC moieties 32:0–3, 36:3B, 38:4A, and 38:5A are all elevated in LDLr-/- while PC 30:0, 35:1, 38:1 in addition to diacylglycerol, and glucosylceramides 40:1 and 42:1 were elevated with a WD (negatively charged CSH, Table 5). PC moiety’s 31:0 and 33:1 were the only ones affected by both WD and LDLr-/- and PC 31:0 showed a significant additive effect for diet and genotype. Further, cere- mide 42:1(Cer18:1/24:0) and 42:2A (Cer18:1/24:1) in addition to fatty acid 18:1 and 20:1 were also found to be raised with a WD (Table 6, positively charged CSH). However, there were no Diet & LDLr-/- alter brain function & metabolism Table 5. Shift in negatively charged complex lipids with western diet and/or LDLr -/-. C57BL/6 (WT) or LDLr-/- mice were fed a control (CD) or western (WD) diet for 12 weeks and their brain complex lipids were measured by LC CSH-(+)ESI QTOF MS. Values are mean in μmol/g tissue ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = 4/grp). Table 5. Shift in negatively charged complex lipids with western diet and/or LDLr -/-. C57BL/6 (WT) or LDLr-/- mice were fed a control (CD) or western (WD) diet for 12 weeks and their brain complex lipids were measured by LC CSH-(+)ESI QTOF MS. Values are mean in μmol/g tissue ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = 4/grp). for 12 weeks and their brain complex lipids were measured by LC CSH (+)ESI QTOF MS. Values are mean in μmol/g tissue ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = 4/grp). PC moiety’s 31:0 and 33:1 were the only ones affected by both WD and LDLr-/- and PC 31:0 showed a significant additive effect for diet and genotype. Further, cere- mide 42:1(Cer18:1/24:0) and 42:2A (Cer18:1/24:1) in addition to fatty acid 18:1 and 20:1 were also found to be raised with a WD (Table 6, positively charged CSH). However, there were no PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 19 / 38 Oxylipins and endocannabinoids abundance segregates mice groups by diet or genotypes with a shift to a pro-inflammatory state Partial least square analysis, together with the mixed model ANOVA, were applied to deter- mine the impact of experimental groups on the level of free oxylipins, endocannabinoids and polyunsaturated fatty acids (PUFAS) (Fig 6A–6C). Forty seven percent of variables had VIP score greater than 1, suggesting that observed differences are most likely not false positives. Major diet and genotype effect were observed, however, without significant interaction between them. WD groups were characterized by decreased levels of 18 carbon PUFAs (alpha linolenic acid (aLA) and linoleic acid (LA)) as well as their corresponding ethanol amines (aLEA and LEA). Additionally, WD elevated the levels of soluble epoxide hydrolase metabo- lites of arachidonic acid (DiHETrEs) as well as 5- lipoxygenase (LOX) metabolites (LTB4 and 5- Hydroxyicosatetraenoic acid (HETE)) and COX metabolites (20-HETE and 6-keton PGF1a). LDLr -/- animals showed increased level of long chain (C 20 and 22) PUFAs as well as their 12-LOX metabolites (12-HETE, 12 HEPE, 14-HDoHE). Moreover, LDLr -/- decreased the levels of 18 carbon derived oxylipins (from both LOX and CYP pathways). Interestingly, the levels of their parent PUFAs were not impacted by the LDLr -/-. Diet & LDLr-/- alter brain function & metabolism Table 5. (Continued) Complex Lipids (neg) WT LDLr -/- P-value CD WD CD WD Genotype Diet G×D1 PE (36:2) 40900 ± 2570 54300 ± 8810 40100 ± 2390 43800 ± 3600 0.288 0.119 0.36 PE (38:4) 179000 ± 25800 152000 ± 16800 191000 ± 14300 206000 ± 14200 0.096 0.739 0.278 PE (38:5) 33100 ± 1200 38800 ± 4500 32600 ± 1510 35400 ± 1370 0.462 0.121 0.574 PE (38:6) 222000 ± 20700 224000 ± 24300 275000 ± 8730 269000 ± 14800 0.019 0.932 0.835 PE (40:4) 23800 ± 3770 21900 ± 3370 25900 ± 1240 28600 ± 2590 0.154 0.886 0.448 https://doi.org/10.1371/journal.pone.0191909.t005 significant changes in measured phosphatidylethanolamines, sphingomyelins or positively charged phosphatidylcholine moieties. Western diet exacerbates LDLr-/- elevation of plasma lipid levels without significantly changing body weight At the initiation of the diet, WT mice were heavier (23.5 ± 0.2g) than LDLr -/- (22.7 ± 0.2g; p 0.05) mice. Mice of both genotypes (WT and LDLr -/-) fed a WD had a significant increase in body weight compared to CD fed, and the C57BL/6 WD-fed were significantly heavier than western fed LDLr -/- (Table 7, p< 0.001). A subset of mice (n = 4/grp) had their blood collected at termination and non-fasting glucose, insulin, and lipid content was measured. There was no change in glucose but a significant difference between insulin levels was found in CD and WD fed mice (668.6 ± 148.2 and 3187.0 ± 2263.2 pg/mL, p 0.01). Total cholesterol, triglyceride levels, HDL and LDL levels were all found to have significantly effect with diet or genotype as well as having a diet x genotype interaction. Complex Lipids (neg) WT LDLr -/- P-value CD WD CD WD Genotype Diet G×D1 Acylcarnitine C16:0 2040 ± 344 2410 ± 298 2490 ± 56.9 2660 ± 227 0.197 0.305 0.709 Acylcarnitine C18:1 1840 ± 422 2450 ± 225 2320 ± 263 2620 ± 162 0.28 0.135 0.604 Ceramide (d42:2) 749 ± 142 1350 ± 244 1100 ± 110 1150 ± 222 0.682 0.108 0.17 Cholesterol 56000 ± 1300 57200 ± 1410 55700 ± 1230 54800 ± 1000 0.285 0.885 0.422 DG (34:1) 693 ± 125 1440 ± 244 835 ± 274 1100 ± 74.7 0.628 0.024 0.246 GlcCer (d40:1) 18300 ± 1600 35900 ± 8130 19900 ± 1450 22100 ± 2480 0.194 0.043 0.105 GlcCer (d42:1) 36900 ± 1960 69300 ± 14600 38700 ± 3850 44400 ± 6610 0.189 0.041 0.134 GlcCer (d42:2) 133000 ± 10700 279000 ± 72100 139000 ± 10800 155000 ± 24400 0.154 0.06 0.12 LPC (16:0) 4930 ± 487 5540 ± 616 5020 ± 203 5540 ± 390 0.917 0.232 0.92 LPC (18:0) 2460 ± 161 2990 ± 338 2620 ± 220 2900 ± 50.2 0.853 0.088 0.582 LPC (18:1) 1450 ± 130 1840 ± 252 1510 ± 61.1 1690 ± 37.2 0.751 0.073 0.476 PC (30:0) 102000 ± 11700b 135000 ± 20100ab 118000 ± 8530ab 162000 ± 8580a 0.136 0.012 0.694 PC (31:0) 3810 ± 336c 5800 ± 519b 4550 ± 274bc 8800 ± 628a 0.002 <0.001 0.031 PC (32:0) 1280000 ± 122000ab 1060000 ± 76100b 1570000 ± 65900a 1540000 ± 96100a 0.001 0.205 0.325 PC (32:1) 188000 ± 25200 175000 ± 25000 245000 ± 22800 229000 ± 15900 0.029 0.528 0.939 PC (32:3) 3730 ± 320b 4050 ± 317ab 4170 ± 221ab 5310 ± 528a 0.038 0.067 0.28 PC (33:1) 4400 ± 357b 6490 ± 666ab 5630 ± 277b 8670 ± 709a 0.008 <0.001 0.387 PC (34:0) 224000 ± 13900 243000 ± 38000 256000 ± 22400 281000 ± 14200 0.173 0.375 0.916 PC (34:1) 1820000 ± 23900 1800000 ± 36900 1850000 ± 11600 1860000 ± 7080 0.073 0.81 0.543 PC (35:1) 10900 ± 660 17400 ± 3480 11200 ± 835 17400 ± 612 0.928 0.005 0.937 PC (36:1) 483000 ± 47500 631000 ± 121000 502000 ± 33400 583000 ± 30200 0.838 0.123 0.632 PC (36:2) 171000 ± 10900 194000 ± 31300 181000 ± 10800 198000 ± 3510 0.691 0.284 0.882 PC (36:3) B 9830 ± 1380ab 7210 ± 1700b 13000 ± 1150ab 13800 ± 1330a 0.004 0.529 0.247 PC (36:4) B 232000 ± 38900 177000 ± 21600 277000 ± 29500 280000 ± 45400 0.056 0.472 0.426 PC (36:5) B 1280 ± 136 1000 ± 37.6 1420 ± 251 1300 ± 74.1 0.153 0.24 0.627 PC (38:1) 13500 ± 1330 28100 ± 7130 12300 ± 1110 14900 ± 1790 0.081 0.041 0.137 PC (38:2) 15000 ± 875 26000 ± 6360 15100 ± 1060 17900 ± 1680 0.26 0.062 0.245 PC (38:3) 6000 ± 448 7950 ± 1860 6800 ± 788 8150 ± 387 0.645 0.141 0.782 PC (38:4) A 309000 ± 39200 252000 ± 26400 343000 ± 27600 357000 ± 28500 0.044 0.498 0.274 PC (38:5) A 46800 ± 7670 37800 ± 3160 57600 ± 6080 57500 ± 6870 0.03 0.48 0.483 PC (38:5) B 4580 ± 1250 4380 ± 426 5890 ± 358 5810 ± 1750 0.24 0.902 0.958 PC (38:6) A 635000 ± 38400 682000 ± 65200 682000 ± 36600 691000 ± 55200 0.589 0.586 0.709 PC (38:6) C 961 ± 20.5 931 ± 72.4 955 ± 174 1100 ± 45.9 0.408 0.552 0.377 PC (39:6) 1040 ± 220 1660 ± 123 1060 ± 264 1510 ± 97.9 0.722 0.015 0.64 PC (40:4) 9480 ± 918 11500 ± 2750 10100 ± 1020 12400 ± 1030 0.658 0.212 0.924 PC (40:5) A 11700 ± 809 11400 ± 2710 12600 ± 809 13400 ± 492 0.35 0.865 0.703 PC (40:5) B 3100 ± 227 2820 ± 789 2720 ± 280 3910 ± 305 0.457 0.337 0.137 PC (40:6) A 1390 ± 165 1450 ± 145 1600 ± 134 1580 ± 162 0.271 0.897 0.807 PC (40:6) B 189000 ± 24100 218000 ± 41000 170000 ± 23900 194000 ± 29900 0.497 0.396 0.928 PC (40:7) 266000 ± 12700 331000 ± 46300 297000 ± 26300 312000 ± 29800 0.857 0.225 0.428 PC (40:8) 891 ± 222 1120 ± 68.6 1140 ± 282 1090 ± 325 0.663 0.735 0.566 PC (o-32:0) 6010 ± 981ab 4790 ± 627b 7120 ± 468ab 8260 ± 978a 0.014 0.961 0.162 PC (o-34:0) 2020 ± 111 1910 ± 179 2140 ± 114 2340 ± 205 0.106 0.746 0.339 PE (34:1) 28400 ± 2030 33600 ± 4730 29100 ± 1260 31600 ± 1240 0.811 0.178 0.632 PE (36:1) 42600 ± 3870 56700 ± 9250 42200 ± 1410 47100 ± 3660 0.369 0.105 0.411 (Continued) (Continued) PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 20 / 38 Diet & LDLr-/- alter brain function & metabolism Table 6. Shift in positively charged complex lipids with western diet and/or LDLr -/-. C57BL/6 (WT) or LDLr-/- mice were fed a control (CD) or western (WD) diet for 12 weeks and their complex lipids were measured by LC CSH- (-) ESI QTOF MS. Values are mean in μmol/g tissue ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = 4/grp). for 12 weeks and their complex lipids were measured by LC CSH- (-) ESI QTOF MS. Values are mean in μmol/g tissue ±SEM. Means in a row without a common subscript letter differ (p 0.05) as analyzed by two-way ANOVA and Tukey test. GxD = Genotype x Diet interaction effect (n = 4/grp). LDLr -/- and western diet increase aortic sinus plaque formation and hepatic lipidosis To determine the effects of diet and genotype on the general pathology of systemic organs, brain, liver, heart, kidney, pancreas, skeletal muscle, and lung were evaluated by routine histo- pathology. No gross changes in morphology and no significant lesions were seen in brain, PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 21 / 38 Diet & LDLr-/- alter brain function & metabolism Table 6. (Continued) Complex Lipids (pos) WT LDLr-/- P-value CD WD CD WD genotype diet G×D1 PC 38:4 A 45400 ± 5790 36500 ± 2050 47300 ± 2990 48000 ± 2230 0.088 0.276 0.203 PC 38:5 A 17700 ± 2350 14200 ± 784 19200 ± 1170 18600 ± 1240 0.074 0.2 0.359 PC 40:6 B 32500 ± 4090 40700 ± 3790 27800 ± 2610 30200 ± 3430 0.051 0.158 0.417 PC 40:7 42200 ± 4110 52300 ± 7920 45000 ± 2490 48900 ± 4150 0.954 0.195 0.551 PE 34:1 75300 ± 7710 80300 ± 9310 73200 ± 4210 77500 ± 3510 0.716 0.5 0.958 PE 36:2 92800 ± 5240 120000 ± 17200 86400 ± 4960 94700 ± 10200 0.165 0.123 0.399 PE 36:4 46700 ± 6390 39300 ± 1850 48900 ± 3720 48500 ± 2120 0.175 0.339 0.387 PE 38 2 838 ± 77.7 1130 ± 270 958 ± 96.7 1180 ± 115 0.697 0.147 0.839 PE 38:4 A 24600 ± 5950 25000 ± 2410 29600 ± 2020 24500 ± 6290 0.638 0.616 0.559 PE 38:4 B 444000 ± 64100 363000 ± 29200 460000 ± 32400 482000 ± 24100 0.121 0.475 0.227 PE 38:5 84400 ± 5320 94200 ± 7980 81900 ± 3580 88700 ± 3570 0.473 0.152 0.792 PE 38:6 819000 ± 124000 728000 ± 94300 964000 ± 48500 921000 ± 30300 0.064 0.437 0.777 PE 40:4 158000 ± 28100 139000 ± 24500 157000 ± 12200 171000 ± 18500 0.498 0.904 0.465 PE 40:6 1200000 ± 70300 1110000 ± 54200 1270000 ± 46900 1250000 ± 31100 0.064 0.32 0.482 PE 44:10 413000 ± 52500 468000 ± 74700 392000 ± 35200 459000 ± 36500 0.773 0.265 0.908 PG 32:0 12200 ± 1850 11800 ± 1430 10300 ± 552 11400 ± 559 0.357 0.81 0.549 PG 34:1 54800 ± 1980 57700 ± 5120 49800 ± 5130 54100 ± 4900 0.355 0.437 0.89 PI 36:4 3480 ± 1550 3200 ± 1400 6010 ± 68.3 6030 ± 222 0.026 0.904 0.886 PI 38:5 3620 ± 264 2960 ± 214 3510 ± 288 3670 ± 173 0.236 0.327 0.114 PS 36:2 43700 ± 1640 51100 ± 5040 39600 ± 2610 44300 ± 3130 0.128 0.096 0.691 SM d34:1 9590 ± 194 11200 ± 1290 8620 ± 477 10200 ± 555 0.212 0.056 0.992 SM d36:1 80000 ± 3370 71000 ± 7090 67000 ± 4770 75500 ± 3940 0.413 0.966 0.106 SM d36:2 16900 ± 3770 12200 ± 2600 15500 ± 303 18700 ± 2320 0.34 0.764 0.147 SM d38:1 8560 ± 291 10500 ± 872 7750 ± 614 8480 ± 807 0.059 0.073 0.387 SM d40:1 24500 ± 7000 31100 ± 9540 25600 ± 6590 17000 ± 7480 0.415 0.898 0.345 SM d41:1 11500 ± 3030 20200 ± 3580 13400 ± 780 13600 ± 3910 0.459 0.175 0.186 SM d42:2A 40300 ± 2120 58400 ± 10800 37100 ± 2320 43300 ± 6240 0.18 0.083 0.374 SM d42:3 856 ± 221 1440 ± 372 868 ± 254 1150 ± 260 0.746 0.164 0.613 SM d44:2 1420 ± 123 1510 ± 352 1130 ± 262 1410 ± 206 0.436 0.479 0.708 https://doi org/10 1371/journal pone 0191909 t006 kidney, pancreas, skeletal muscle, and lung. Complex Lipids (pos) WT LDLr-/- P-value CD WD CD WD genotype diet G×D1 Ceremide 34:1 14700 ± 1730 19300 ± 2980 15200 ± 2020 19000 ± 3340 0.967 0.136 0.885 Ceremide 34:2 1500 ± 398 677 ± 204 1330 ± 477 1740 ± 343 0.255 0.58 0.12 Ceremide 36:1 505000 ± 64300 535000 ± 60400 516000 ± 41700 568000 ± 55900 0.704 0.479 0.852 Ceremide 38:1 47600 ± 1080 46300 ± 3680 45400 ± 2830 46800 ± 2730 0.766 0.997 0.629 Ceremide 39:1 5210 ± 886 5090 ± 575 5530 ± 364 5910 ± 851 0.435 0.862 0.728 Ceremide 40:1 10500 ± 930 14000 ± 2110 10300 ± 774 12600 ± 1710 0.571 0.076 0.699 Ceremide 40:2 1340 ± 337 2340 ± 404 1810 ± 598 2320 ± 401 0.631 0.117 0.599 Ceremide 41:1 5740 ± 677 7530 ± 1260 5880 ± 381 6810 ± 686 0.725 0.12 0.605 Ceremide 42:1 5950 ± 623 9170 ± 1660 6130 ± 532 8140 ± 1540 0.732 0.05 0.626 Ceremide 42:2A 28000 ± 2870 46800 ± 8970 29100 ± 4290 36500 ± 5880 0.458 0.048 0.356 Ceremide 42:2B 986 ± 147 1790 ± 322 1350 ± 27.2 1810 ± 294 0.506 0.122 0.616 FA 18:1 165000 ± 27000 261000 ± 45800 178000 ± 20800 222000 ± 29400 0.684 0.049 0.429 FA 20:1 20400 ± 2580ab 37800 ± 7430a 18300 ± 2110b 23000 ± 3020ab 0.076 0.025 0.169 FA 20:2 1380 ± 209 1880 ± 385 1140 ± 134 1700 ± 244 0.435 0.064 0.889 FA 20:3 5620 ± 1210 7210 ± 1060 5410 ± 915 7920 ± 996 0.816 0.074 0.668 FA 20:3 5670 ± 880 8050 ± 1210 6410 ± 765 6990 ± 830 0.87 0.139 0.358 FA 20:4 464000 ± 51300 497000 ± 50600 524000 ± 45600 563000 ± 36100 0.2 0.45 0.942 FA 20:5 1070 ± 380 1470 ± 414 1920 ± 314 2060 ± 349 0.072 0.476 0.727 FA 22:0 4350 ± 1560 3280 ± 1630 6110 ± 937 5490 ± 1690 0.205 0.581 0.88 FA 22:6 85000 ± 10200 127000 ± 18200 95400 ± 8080 110000 ± 18900 0.844 0.076 0.378 FA 24:1 3760 ± 687ab 6760 ± 1530a 2910 ± 375b 3660 ± 572ab 0.05 0.061 0.239 GlcCer 38:1 22800 ± 2140 35400 ± 6160 23600 ± 1940 25600 ± 3160 0.251 0.073 0.182 GlcCer 40:1 107000 ± 9480 162000 ± 27200 110000 ± 6680 118000 ± 13100 0.227 0.076 0.176 GlcCer 41:1 106000 ± 7200 170000 ± 31500 104000 ± 7380 116000 ± 14500 0.146 0.055 0.18 GlcCer 42:1 307000 ± 18400 520000 ± 102000 306000 ± 28500 338000 ± 52100 0.15 0.061 0.153 GlcCer 42:2 840000 ± 56000 1180000 ± 189000 813000 ± 46700 862000 ± 116000 0.166 0.122 0.237 GlcCer d14:1/20:0(2OH 7510 ± 2470ab 3540 ± 1850b 7640 ± 2100ab 13900 ± 777a 0.018 0.565 0.02 LPC 16:0 10800 ± 1310 12000 ± 1520 10900 ± 508 11800 ± 870 0.938 0.358 0.908 LPC 18:0 6700 ± 585 7710 ± 903 6460 ± 216 7170 ± 492 0.529 0.175 0.805 LPC 18:1 2800 ± 203 3610 ± 472 2930 ± 82.2 3280 ± 377 0.765 0.096 0.5 LPE 16:0 1880 ± 131 2260 ± 249 2070 ± 123 2700 ± 526 0.326 0.125 0.686 LPE 18:0 6560 ± 718 7830 ± 996 7110 ± 487 9630 ± 2160 0.371 0.16 0.63 LPE 22:6 12500 ± 3620 8770 ± 552 9090 ± 470 7230 ± 1200 0.229 0.179 0.644 PC 32:0 271000 ± 25300ab 224000 ± 14200b 318000 ± 14900a 304000 ± 15800a 0.004 0.12 0.375 PC 32:1 18700 ± 3280 17600 ± 2670 21300 ± 1470 20600 ± 624 0.246 0.695 0.93 PC 33:1 1960 ± 851 4870 ± 309 3190 ± 1000 4110 ± 1110 0.795 0.049 0.276 PC 34 0 80200 ± 3220 75300 ± 5860 84400 ± 5240 85700 ± 2420 0.124 0.697 0.498 PC 34:1 342000 ± 28600 344000 ± 29300 356000 ± 15200 364000 ± 6760 0.454 0.822 0.895 PC 34:2 8480 ± 317 6830 ± 1630 9780 ± 500 9740 ± 490 0.038 0.367 0.39 PC 36:1 156000 ± 10800 178000 ± 19400 151000 ± 7220 160000 ± 9320 0.385 0.234 0.588 PC 36:2 28200 ± 1930 31900 ± 3310 28900 ± 1200 31000 ± 840 0.941 0.181 0.696 PC 36:3 B 4260 ± 463 3880 ± 233 4890 ± 315 3850 ± 1230 0.674 0.321 0.637 PC 36:4 B 67700 ± 8200 50400 ± 3050 71000 ± 3250 69900 ± 5910 0.062 0.119 0.169 PC 38:2 6350 ± 134 9180 ± 1630 5940 ± 384 6330 ± 747 0.102 0.106 0.209 PC 38:3 2260 ± 60.3 2420 ± 194 2230 ± 106 2240 ± 66.4 0.399 0.476 0.532 (Continued) PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 22 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Diet & LDLr-/- alter brain function & metabolism Fig 6. Shift in brain free oxylipins, endocannabinoids and polyunsaturated fatty acids (PUFA) to a proinflammatory profile by diet. Partial least squares analysis (A) discriminate between C57BL/6 (WT) or LDLr-/ mice, fed a control (CD) or western (WD) diet due to differences in brain metabolite pattern seen in associated loading plot (B). Hierarchical cluster analysis categorized measured variables into clusters and ANOVA analysis tested for effect of the diet, genotype, and the diet x genotype interaction. Clusters identified as having either a significant change are shown (C). Loading plot: variables with VIP scores >1 are shown. The variables are colored according to their clusters affiliations. https://doi.org/10.1371/journal.pone.0191909.g006 Fig 6. Shift in brain free oxylipins, endocannabinoids and polyunsaturated fatty acids (PUFA) to a proinflammatory profile by diet. Partial least squares analysis (A) discriminate between C57BL/6 (WT) or LDLr-/ mice, fed a control (CD) or western (WD) diet due to differences in brain metabolite pattern seen in associated loading plot (B). Hierarchical cluster analysis categorized measured variables into clusters and ANOVA analysis tested for effect of the diet, genotype, and the diet x genotype interaction. Clusters identified as having either a significant change are shown (C). Loading plot: variables with VIP scores >1 are shown. The variables are l d di t th i l t ffili ti Fig 6. Shift in brain free oxylipins, endocannabinoids and polyunsaturated fatty acids (PUFA) to a proinflammatory profile by diet. Partial least squares analysis (A) discriminate between C57BL/6 (WT) or LDLr-/ mice, fed a control (CD) or western (WD) diet due to differences in brain metabolite pattern seen in associated loading plot (B). Hierarchical cluster analysis categorized measured variables into clusters and ANOVA analysis tested for effect of the diet, genotype, and the diet x genotype interaction. Clusters identified as having either a significant change are shown (C). Loading plot: variables with VIP scores >1 are shown. The variables are colored according to their clusters affiliations. https://doi.org/10.1371/journal.pone.0191909.g006 https://doi.org/10.1371/journal.pone.0191909.g006 However, diet and genotype related lesions were present in heart and liver. LDLr -/- mice fed CD had foam cell plaques on the endocardial sur- face of the aortic sinus (Fig 7, heart). These focal accumulations were composed principally of aggregates of large mononuclear cells with cytoplasm filled with vacuoles. The aortic sinus of LDLr -/- mice fed WD also contained foam cell plaques but these were more extensive and foam cells were interspersed with extracellular accumulations of acicular clefts characteristic of cholesterol deposits, increased extracellular matrix, and scattered small mononuclear inflam- matory cells (complex atheroma). Both WT and LDLr -/- mice fed a WD had marked hepatic lipidosis (Fig 7, liver). In WT mice, a zonal accumulation of lipid in hepatocytes progressed from multiple lipid droplets in periacinar regions to larger single lipid accumulations in periportal regions. LDLr -/- mice fed WD had more prominent generalized hepatocyte lipid accumulations with randomly distributed individual necrotic hepatocytes and associated mononuclear inflammatory cell infiltration. In addition, Kuppfer cells were markedly enlarged due to microcystic intracytoplasmic vacuoles interpreted to represent lipid accumulation. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 23 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Physiological Parameter WT LDLr -/- P-value CD WD CD WD Genotype Diet G×D1 Body mass(g) 30.9 ± 0.45b 39.2 ± 0.809a 30.8 ± 0.538b 33 ± 0.605b <0.001 <0.001 <0.001 Glucose (pg/dl) 286 ± 34 300 ± 15 303 ± 10.2 310 ± 15 0.533 0.617 0.879 Insulin (pg/ml) 620 ± 77.3 5320 ± 2420 718 ± 176 1060 ± 291 0.115 0.062 0.1 TG (mg/dl) 164 ± 14.1b 56.7 ± 11.7b 393 ± 69.6b 1300 ± 161a <0.001 0.001 <0.001 TC (mg/dl) 153 ± 8.25c 273 ± 7.72c 567 ± 28.2b 2090 ± 75.8a <0.001 <0.001 <0.001 HDL (mg/dl) 122 ± 8.31b 206 ± 7b 211 ± 8.91b 476 ± 45.8a <0.001 <0.001 0.003 LDL (mg/dl) 31.6 ± 2.76c 64.3 ± 3.17c 489 ± 86.7b 1300 ± 46.4a <0.001 <0.001 <0.001 https://doi.org/10.1371/journal.pone.0191909.t007 activates microvessels and microglia, and increases BBB transport, all of which may be linked to the observed moderate altered cognitive function (Fig 8 and Tables 8 and 9). Previous studies have demonstrated that either a WD [29] and/or gene knockout-induced hyperlipidemia [9, 58] can result in cognitive impairment in mice while others have demon- strated a link of cognitive impairment to neuroinflammation or metabolic shifts [59, 60]. We extended these findings by comprehensively evaluating the consequences of WD and LDLr -/- genotype on behavior as an indicator of cognitive impairment. Both the MWM and the Y- maze alternation tests for cognition were sensitive to the effects of the WD with group sizes as small as 8 mice/group. While % of alternation triplets was decreased in WD-fed mice (an indi- cator of reduced cognitive function), these same animals showed an increase in % time in the platform quadrant on the probe trial for the MWM (indicating improved function). Both tests assess short-term spatial memory, but the MWM additionally evaluates day-to-day learning, which may contribute to the differences seen here. The Y-maze and MWM results were sensi- tive to WD, but they did not reflect an effect of LDLr-/- genotype on cognitive endpoints as seen by other groups [9]. Therefore, a second study was conducted where a different spatial learning and memory problem was presented on each of nine successive days of testing and number of errors on the last trial of each problem was compared across days. Diet & LDLr-/- alter brain function & metabolism Table 7. Non-fasting physiological parameters at 20 weeks. Data are represented as mean ± SEM for wild type (WT) LDLr -/- on either a control (CD) or western (WD) diet, n = 20 for body weight and 4 for all other parameters. a-c Means in a row without a common superscript letter differ (P < 0.05) as analyzed by two-way ANOVA and the TUKEY test. G × D1 = Genotype × Diet interaction effect. n = 20 for body weight and 4 for all other parameters. Table 7. Non-fasting physiological parameters at 20 weeks. Data are represented as mean ± SEM for wild type (WT) LDLr -/- on either a control (CD) or western (WD) diet, n = 20 for body weight and 4 for all other parameters. a-c Means in a row without a common superscript letter differ (P < 0.05) as analyzed by two-way ANOVA and the TUKEY test. G × D1 = Genotype × Diet interaction effect. n = 20 for body weight and 4 for all other parameters. Table 7. Non-fasting physiological parameters at 20 weeks. Data are represented as mean ± SEM for wild type (WT) LDLr -/- on either a control (CD) or western (WD) diet, n = 20 for body weight and 4 for all other parameters. a-c Means in a row without a common superscript letter differ (P < 0.05) as analyzed by two-way ANOVA and the TUKEY test. G × D1 = Genotype × Diet interaction effect. n = 20 for body weight and 4 for all other parameters. Discussion While there is significant epidemiologic evidence that a diet high in saturated fat and simple carbohydrates resulting in elevated plasma lipids and insulin resistance puts individuals at a greater risk for dementia and cognitive impairment, the cellular metabolic reasons for this are not fully understood. This study was conceived to better understand the mechanisms through which a western diet (WD—moderately high in saturated fat, sucrose, and cholesterol) induced cognitive impairment by assessing the brain molecular, cellular, biochemical, and physiological changes that occur. Our studies demonstrate that WD alters brain metabolism, PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 24 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 The RAWM detected an impairment in ability of LDLr -/- mice to solve a complex spatial problem with repeated exposure to different problems. Our studies thus indicate that Y-maze alternation can provide a rapid, inexpensive assessment of the cognitive effects of a WD. However, to avoid false negative findings in screening for cognitive deficits seen with LDLr-/- mutation more complex cognitive testing of the RAWM, are most sensitive and appropriate. These mod- est and potentially contradictory findings are not unexpected, as even genetic mouse models of Alzheimer’s (Tg2576) show cognitive impairments by T maze (modification of Y-maze) most consistently, followed by MWM and RAWM, but there is great variability between stud- ies [61]. Even though under normal conditions the blood-brain barrier (BBB) is highly regulated, metabolic stresses such as hyperlipidemia can cause neurovascular unit dysfunction and increase BBB permeability [62]. Studies by our lab have demonstrated that a bolus infusion of TGRL lipolysis products, similar to an increase that would be found in the postprandial state, can lead to a transient increase in the BBB Gd-DTPA transfer coefficient (Ki) [17, 63]. Here Ki was increased in animals fed a WD, while cerebral blood flow, determined by perfusion weighted imaging, was essentially normal. Interestingly, despite higher circulating lipids on a CD, LDLr -/- mice did not have elevated baseline Ki and no additive effect of a WD beyond that of WT. Our findings are consistent with a hypothesis that chronic consumption of a WD PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 25 / 38 Diet & LDLr-/- alter brain function & metabolism Fig 7. Cardiac and liver histopathology. In heart, CD (A) and WD-fed (B) wild type mice had normal aortic sinus structures, but the aortic sinus of LDLr -/- mice on CD had subendothelial foam cell plaques and medial accumulations of extracellular cholesterol (C), both were more extensive in LDLr -/- mice fed a WD (D). In liver, compared with CD fed (A), WT mice given a WD (B) had marked microvesicular and macrovesicular hepatic lipidosis that was most marked in the periacinar regions. CD fed LDLr -/-mice (C) had more hepatocellular glycogen than WT mice but were otherwise histologically normal. LDLr -/- mice given a WD (D) had generalized macrovesicular lipidosis, enlarged Kupffer cells with foamy cytoplasm (Arrowheads), periportal mononuclear inflammation, and individual hepatocyte necrosis accompanied by mixed inflammatory cell infiltrates (Arrow). Scale bar: heart = 100μM and liver = 50μM. Fig 7. Cardiac and liver histopathology. In heart, CD (A) and WD-fed (B) wild type mice had normal aortic sinus structures, but the aortic sinus of LDLr -/- mice on CD had subendothelial foam cell plaques and medial accumulations of extracellular cholesterol (C), both were more extensive in LDLr -/- mice fed a WD (D). In liver, compared with CD fed (A), WT mice given a WD (B) had marked microvesicular and macrovesicular hepatic lipidosis that was most marked in the periacinar regions. CD fed LDLr -/-mice (C) had more hepatocellular glycogen than WT mice but were otherwise histologically normal. LDLr -/- mice given a WD (D) had generalized macrovesicular lipidosis, enlarged Kupffer cells with foamy cytoplasm (Arrowheads), periportal mononuclear inflammation, and individual hepatocyte necrosis accompanied by mixed inflammatory cell infiltrates (Arrow). Scale bar: heart = 100μM and liver = 50μM. https://doi org/10 1371/journal pone 0191909 g007 Fig 7. Cardiac and liver histopathology. In heart, CD (A) and WD-fed (B) wild type mice had normal aortic sinus structures, but the aortic sinus of LDLr -/- mice on CD had subendothelial foam cell plaques and medial accumulations of extracellular cholesterol (C), both were more extensive in LDLr -/- mice fed a WD (D). In liver, compared with CD fed (A), WT mice given a WD (B) had marked microvesicular and macrovesicular hepatic lipidosis that was most marked in the periacinar regions. CD fed LDLr -/-mice (C) had more hepatocellular glycogen than WT mice but were otherwise histologically normal. LDLr -/- mice given a WD (D) had generalized macrovesicular lipidosis, enlarged Kupffer cells with foamy cytoplasm (Arrowheads), periportal mononuclear inflammation, and individual hepatocyte necrosis accompanied by mixed inflammatory cell infiltrates (Arrow). Scale bar: heart = 100μM and liver = 50μM. https://doi.org/10.1371/journal.pone.0191909.g007 increases microvascular leak, assuming no concurrent increase in capillary surface area, increases BBB permeability. Perhaps this is related to neuroinflammation, activation of endo- thelial apoptotic pathways [15, 64], or increased transcytosis. Regardless, an increase in BBB leak would likely increase movement of blood solutes including lipids and lipoproteins into the brain interstitial space which may alter neurovascular metabolism and inflammation. High-fat diet induced obesity in C57BL/6 mice not only impaired hippocampus-dependent memory and reduced long-term potentiation (as determined by Y-maze and novel object increases microvascular leak, assuming no concurrent increase in capillary surface area, increases BBB permeability. Perhaps this is related to neuroinflammation, activation of endo- thelial apoptotic pathways [15, 64], or increased transcytosis. Regardless, an increase in BBB leak would likely increase movement of blood solutes including lipids and lipoproteins into the brain interstitial space which may alter neurovascular metabolism and inflammation. p y High-fat diet induced obesity in C57BL/6 mice not only impaired hippocampus-dependent memory and reduced long-term potentiation (as determined by Y-maze and novel object PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 26 / 38 Diet & LDLr-/- alter brain function & metabolism Fig 8. Model of physiological pathway in which diet or genotype induced hyperlipidemia alters cognitive function. https://doi.org/10.1371/journal.pone.0191909.g008 https://doi.org/10.1371/journal.pone.0191909.g008 recognition) but has also shown increased microglia activation and loss of synapses [19, 65, 66]. Our immunohistochemical analysis demonstrates increased relative surface area of IBA1 in western diet fed WT mice where most immunopositive cells had multiple prominent radiat- ing elongate cytoplasmic processes typical of microglia. Since upregulation of IBA1is a marker of microglial activation [67–69], our results suggest neuroinflammation is increased by WD in WT but not LDLr -/- mice. However, it should be noted that our approach does not discrimi- nate between increased numbers of microglial cells vs. upregulation of IBA1 in greater propor- tions of microglial processes. recognition) but has also shown increased microglia activation and loss of synapses [19, 65, 66]. Our immunohistochemical analysis demonstrates increased relative surface area of IBA1 in western diet fed WT mice where most immunopositive cells had multiple prominent radiat- ing elongate cytoplasmic processes typical of microglia. Since upregulation of IBA1is a marker of microglial activation [67–69], our results suggest neuroinflammation is increased by WD in WT but not LDLr -/- mice. However, it should be noted that our approach does not discrimi- nate between increased numbers of microglial cells vs. upregulation of IBA1 in greater propor- tions of microglial processes. Upregulation of factor VIII immunostaining in WD-fed LDLr -/- mice was an unexpected finding as our initial hypothesis was that a WD would decrease the density of microvessels in brain as is seen in individuals with Alzheimer’s disease [70, 71]. However, as with IBA1 stain- ing, our approach does not distinguish between increased densities in pre-existing endothelial cells versus increased endothelial cell numbers. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Interestingly other studies associated elevated circulating plasma factor VIII as a risk factor for vascular dementia [72] and was found to be significantly elevated in plasma of patients with other brain injury models including ischemic stroke [73]. Further liver disease is associated with elevated plasma factor VIII levels which is thought to be due to production by sinusoidal endothelial cells [74]. It is possible that the increase in brain microvascular factor VIII may be due to an increase in endocytosis by BBB endothelial cells. Whether microvascular endothelial cell activation as indicated by an increased expression of factor VIII relate to hypoxia related angiogenesis or endocytosis of cir- culating factor VIII remains to be determined as does any correlation of these variables with changes in BBB transfer coefficient. CD16, CD32 & CD86 versus CD 206 & arginase expression have previously been used as markers of differentiation between M1 vs M2 microglia and microglia activation [56, 57]. The downregulation of CD86 with an increase in PTGS2 and IBA1 expression in WD fed mice may indicate a shift from M1- to M2- like phenotype as described by Abutbul et al. [75]. How- ever, LDLr -/- showed a decrease in CD32 and CD206 (M1 and M2, respectively) and a 27 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Diet & LDLr-/- alter brain function & metabolism Table 8. Functional changes by western diet (WD) or in a genetic (G) model of hyperlipidemia. https://doi.org/10.1371/journal.pone.0191909.t009 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 28 / 38 Diet & LDLr-/- alter brain function & metabolism decrease in IL-6 and ATF3, which indicate reduced inflammation and cell stress [64, 76, 77]. Whether these shifts in gene expression correlate to protein abundance or change in activity with a WD-fed mouse or LDLr -/- mice remains to be determined. With the observed cogni- tive impairment, and increase in BBB transport and neuroinflammation, we leveraged the col- lective and comprehensive expertise at our three universities to assess how a WD in LDLr-/- and WT mice shifts brain glucose uptake and metabolites. A previous study by Hu et al showed changes in behavior were correlated to a decrease in brain glucose uptake in the thalamus and striatum of high-fat fed rodents [78]. Yet, our study showed an increase in 18FDG uptake in LDLr -/- mice. These disparities may in part be due to differences in the diet (higher % cholesterol and different source of fat), the rodent species (rat vs mouse), length of time spent on the diet (9 vs 12 weeks), or a difference in fasting vs fed state. Although observing perturbations in neurological activity in defined portions of the brain would be beneficial, current imaging technology has limited resolution (~1mm) and measurements in small regions are difficult. While metabolic analysis of other cognitively impaired mouse models has shown elevation in lactate and glutamate [79, 80], we aimed to determine if hyperlipidemia alters metabolic pathways in vivo using 1H-MRS. Our 1H-MRS data indicated a significant increase in lactate levels with LDLr -/-, which has been observed previously in other cognitive decline models [79]. We also observed a trend for glutamine content to decrease. To further assess metabolite disruption, we utilized GC/MS and LC/MS/MS to examine whether a broader range of pathways including: fatty acid metabolism, TCA cycle, and β-oxi- dation may contribute to the progression of cognitive decline. As anticipated, we detected a significant increase in C14:0 and C16:1 free fatty acids with diet, but the LDLr -/- genotype appeared to blunt the effects of diet on brain fatty acid and cholesterol profiles. It is feasible that genetic abolition of the LDL receptor (known as a contributing transporter of cargo into the brain [81]) could reduce transport of fatty acids into the brain despite elevated plasma lipid levels. Parameter Overall Summary Cognitive function Y-maze # WD Decrease % spontaneous alterations in WD MWM " WD Increase % time in platform quadrant in WD in probe trial RAWM # G Increase in the number of Trial 4 errors in LDLr -/- BBB transport Ki " WD Increased in WD IHC Factor VIII " G x WD Elevated in LDLr -/- WD (hippocampus & thalamus) IBA1 " G x WD Elevated in WT WD (cortex & hippocampus) Gene Expression (whole brain RT-PCR) ATF3, IL-6, CD32 & CD206 #G Reduced in LDLr -/- CD86 # WD Reduced with WD Brain Metabolites (1H-MRS Spec) glutamine # G x WD Reduced in LDLr -/- WD lip13a+13b " G x WD Elevated in LDLr -/- WD Brain glucose utilization 18FDG-PET " G Increased Standardized Uptake Value in LDLr -/- Brain Metabolites (GC/MS & LC/MS) C14 " WD Elevated in WD C16:1 " WD Elevated in WD Cholesterol " G x WD Elevated in WT WD and LDLr -/- WD TCA cycle intermediates " WD Elevated BHB, GABA, Aspartate, 2-hydroxyguterate, Glutamine, Lactate in WD Acyl-CoA’s "# G x WD Acetyl-CoA elevated in LDLr -/- CD, Acetyl- and Butyryl-CoA reduced in LDLr -/- WD free 18 C PUFA’s # WD Reduced in WD HETEs & leukotriene " WD Elevated in WD Physiological Weight " WD, G, and G x WD Increased with WD, decrease with LDLr -/- Insulin " WD Increased with WD TG " WD, G, and G x WD Increased in LDLr -/- and WD fed LDLr -/-, but decreased in WD fed WT TC " WD, G, and G x WD Increased with WD or LDLr -/-, highest in LDLr -/- WD HDL " WD & G Increased with WD or LDLr -/-, highest in LDLr -/- WD LDL " G and G x WD Increased with WD or LDLr -/-, highest in LDLr -/- WD https://doi org/10 1371/journal pone 0191909 t008 Table 9. Cardiac and liver histopathology with western diet and/or LDLr -/-. Histology WT CD WT WD LDLr -/- CD LDLr-/- WD Summary Aortic sinus + ++ foam cell & plaque formation Liver lipidosis glycogen lipidosis enlarged, foamy Kupffer cells inflammation/necrosis https://doi.org/10.1371/journal.pone.0191909.t009 Table 9. Cardiac and liver histopathology with western diet and/or LDLr -/-. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 While LDLr-/- mice had more severe liver disease than WT, it seems unlikely to have resulted in hepatic failure and it remains uncertain whether this would be sufficient to alter brain metabolism. Furthermore, the WD-fed LDLr -/- mouse model has predominately been used as a model for atherosclerosis and only more recently cognitive impairment [9, 91]. Regardless, these perturbations in systemic metabolism may lead to a derangement in neurotransmission and result in cognitive defects seen in these mouse models. Interestingly, the altered cognitive function we see in the LDLr-/- are more modest than those seen in previous studies, which may be related in part to elevated phosphatidylcholine (PC) levels in the LDLr-/- mouse brain. The Cermenati et al. study showed that STZ diabetic mice had reduced myelin PC and may be linked to the neurodegenerative events in diabetes [92]. Further, we see an elevation in ceramide (d18:1/24:1), previously shown to have roles in arteriosclerosis, obesity, diabetes, and inflammation. In addition, long-chain sphingomyelins, derived from ceramides, activate macrophages inducing an inflammatory response [93]. These shifts in ceramides have been shown to have a role in astrocyte cell death and mediated cogni- tive impairment [94, 95]. However, the spatial distribution of lipid species is known to support the structural and metabolic functions of the central nervous system [96] and it remains to be determined if the spatial distribution of these metabolites are altered with WD. Our oxylipins and endocannabinoids analysis demonstrated an increase in the 5-lipoxygen- ase (5-LOX) related metabolites including 5-HETE and Leukotriene LTB4. Elevated arachi- donic acid (AA) and its 5-LOX products LTB4, LTD4, and 5-HETE have previously been shown to trigger apoptosis and suppress NFκB cell survival and are linked to hyperlipidemic inflammation [97, 98]. We also saw a decrease in anti-inflammatory 18:3n3 and 18:2n6 with a WD [99, 100]. However, this decrease may be due to a shift in dietary PUFA. The CD fat source is soybean oil, which provides 9.14g/kg of PUFAs while the WD, with milk fat provides 7.35g/kg (a decrease in n-3 PUFAS from 1.21 to 1.05 g/kg and n-6 PUFA’s from 7.93 to 6.3). Further evaluation of TCA cycle intermediates found an elevation in metabolic profile of aspartate, 2-hydroxygluterate, glutamine and lactate that appears to be predominately diet induced. This is interesting given that lactate, aspartate, and glutamate are increased in AD and db/db mouse models of cognitive decline [79, 80, 82] indicating that their elevation by WD may also be associated with the cognitive decline. While confounding, the difference in observed glutamate, lactate, and lipid moieties in vivo versus ex vivo may be due to methodol- ogy (GC/MS vs 1H-MRS), brain region (whole vs cortical focus), anesthesia interval effect (minutes vs hour), or genetic model (LDLr-/- vs db/db), yet, perturbations in lipid and meta- bolic intermediates are still present. Several studies have indicated that the type and depth of anesthesia can significantly modulate brain lactate and glutamine [83–85]. Additionally, the significant increase in GABA concentrations may suggest that GABA synthesis is induced by the WD in the LDLr-/- mutant mouse. Elevated cerebral spinal fluid levels of GABA have been linked to various neurological disorders including dementias, cerebellar cortical atrophy, and multiple sclerosis identifying this as a potential link between brain metabolic dysregulation and cognitive impairment [86]. WD in LDLr -/- mice showed decreased acyl-CoA species associated with fatty acid oxida- tion without altering the total acyl-CoA pool. These results suggest that there is a mismatch in fatty acid oxidation with diet or LDLr mutation suggestive of an inflammatory response due to oxidative stress-induced by altered lipid metabolism. These studies are consistent with previ- ous studies that demonstrate hypometabolism precedes the cognitive decline of AD, where a decline in brain glucose metabolism and mitochondrial function can appear decades prior to diagnosis of AD [87]. Our studies show that LDLr may play a role in oxidative metabolism of lipids, as revealed by the changes in the short and medium chain acyl-CoA’s. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 29 / 38 Diet & LDLr-/- alter brain function & metabolism While we hypothesize that WD directly alters astrocyte-neuron metabolic balance, resulting in impaired cognitive function, elevated brain lactate and glutamine has also been associated with acute liver disease [88]. Hepatic encephalopathy was accompanied by elevated aspartate, glutamine, glucose, and lactate throughout the brain with increasing neuronal injury [89, 90]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Interestingly, in LDLr -/- mice, we saw an increase in the long chain (20 and 22 carbon) neuro- protective PUFAs [101], and a decrease in 9-Hydroxy-10,12-octadecadienoic acid (9-HODE), a pro-inflammatory mediator [102] and 13-HODE, shown to regulate platelet vessel wall adhe- sion [103, 104], and commonly associated with a protective effect. This may be due to a shift in the types or quantities of lipids normally delivered across the BBB [81]. It is of interest that while the LDLr-/- animals had the highest circulating lipid levels (cho- lesterol, triglycerides, HDL, and LDL), yet they did not show increased BBB transport, neu- roinflammation, or cognitive impairment when compared to WT mice on WD. Having a constantly increased lipid level in the LDLr-/- mice may have a preconditioning/protective effect against the WD, or perhaps the temporal increase in circulating lipid levels associated with meals triggers metabolic and inflammatory pathways. Additionally, other undefined compensatory mechanism in the LDL-/- mice may protect the brain from elevated circulating lipids found in LDLr -/- mice. This indicates an increased serum lipid level alone may be insuf- ficient to elicit these changes and that other factors may play a role. In contrast, studies have shown that high carbohydrate diets are sufficient to induce obesity, metabolic inflexibility, and inflammation [105, 106]. Further, increased dietary sucrose is sufficient to alter cognition [107–109] and altering the source of dietary carbohydrates (sucrose or cornstarch) has been shown to impact life span in rodent models [110]. While not significant, WT animals on a WD in our study, have the highest nominal insulin level, suggestive of a shift toward insulin resis- tance, a contributor to inflammation and cognitive impairment [111]. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 30 / 38 Diet & LDLr-/- alter brain function & metabolism While we have focused on the metabolic, inflammatory and permeability changes associ- ated with diet-induced altered cognitive function, others have hinted at the influence of reac- tive oxygen species (ROS) generated by a deregulated metabolism on cognitive decline. For instance, NADPH oxidase-derived production of ROS was shown to be involved in learning and memory impairments in 16-month-old female rats [112], macromolecular ROS damages neurons from aged WT and 3xTg-AD mice [113], and late stage AD patients show significant oxidative DNA damage [114]. Furthermore, in an older population, a Mediterranean diet rich in antioxidants is associated with improved cognitive function [115]. Future studies evaluating any shift in ROS and the antioxidant potential due to a Mediterranean diet remain to be evalu- ated as well as their influence on cognitive impairment [114]. In summary, we found that a WD shifts brain metabolism to a more stressed state profile, activates the inflammatory and vascular system in the brain, and increases BBB transport; all of which likely play a role in the observed alteration in cognitive function seen in WT and LDLr-/- mice. By better understanding how hyperlipidemia and insulin resistance influences neurovascular dysregulation we can better understand neurovascular inflammation-induced cognitive impairments and identify novel targets for the treatment of these debilitating disorders. Acknowledgments We thank Dr. Jeffrey Walton from the UC Davis Nuclear Magnetic Resonance Facility for his assistance and expertise with Magnetic Resonance Imaging as well as Lynette Bower and Todd Tolentino who supervised the phenotyping at UC Davis Mouse Biology Program. We would also like to acknowledge the UC Davis MMPC Complications and Pathology Core and the Energy Balance, Exercise & Behavior Core, Yale MMPC Integrated Physiology Core-Mouse Imaging, and Case Western MMPC Analytical Core. Members of the Mouse Metabolic Pheno- typing Center Imaging. Finally, we would like to acknowledge Drs. K. C. Kent Lloyd (National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) U24-DK092993 and U24- DK092993-05S1), Gerald I. Shulman (NIDDK U24-DK059635), and Henri Brunengraber (NIDDK U24-DK076174) for their support of this work. Working Group include: Jennifer M. Rutkowsky, Michelle Puchowicz, Douglas E. Befroy, Gary Cline, and John C. Rutledge. Jenni- fer M. Rutkowsky (jrutkowsky@ucdavis.edu) serves as lead author for this group. References 1. Morris MC, Tangney CC. Dietary fat composition and dementia risk. Neurobiol Aging. 2014; 35 Suppl 2:S59–64. https://doi.org/10.1016/j.neurobiolaging.2014.03.038 PMID: 24970568 2. Dufouil C, Richard F, Fievet N, Dartigues JF, Ritchie K, Tzourio C, et al. APOE genotype, cholesterol level, lipid-lowering treatment, and dementia: the Three-City Study. Neurology. 2005; 64(9):1531–8. https://doi.org/10.1212/01.WNL.0000160114.42643.31 PMID: 15883313. 3. Haag MD, Hofman A, Koudstaal PJ, Stricker BH, Breteler MM. Statins are associated with a reduced risk of Alzheimer disease regardless of lipophilicity. The Rotterdam Study. J Neurol Neurosurg Psychi- atry. 2009; 80(1):13–7. https://doi.org/10.1136/jnnp.2008.150433 PMID: 18931004. 4. Kalmijn S, Launer LJ, Ott A, Witteman J, Hofman A, Breteler M. Dietary fat intake and the risk of inci- dent dementia in the Rotterdam Study. Annals of neurology. 1997; 42(5):776–82. https://doi.org/10. 1002/ana.410420514 PMID: 9392577 5. Whitmer RA, Sidney S, Selby J, Johnston SC, Yaffe K. Midlife cardiovascular risk factors and risk of dementia in late life. Neurology. 2005; 64(2):277–81. https://doi.org/10.1212/01.WNL.0000149519. 47454.F2 PMID: 15668425. 6. Maesako M, Uemura K, Iwata A, Kubota M, Watanabe K, Uemura M, et al. Continuation of exercise is necessary to inhibit high fat diet-induced beta-amyloid deposition and memory deficit in amyloid pre- cursor protein transgenic mice. PloS one. 2013; 8(9):e72796. https://doi.org/10.1371/journal.pone. 0072796 PMID: 24023774 7. Refolo LM, Malester B, LaFrancois J, Bryant-Thomas T, Wang R, Tint GS, et al. Hypercholesterolemia accelerates the Alzheimer’s amyloid pathology in a transgenic mouse model. Neurobiol Dis. 2000; 7 (4):321–31. Epub 2000/08/31. https://doi.org/10.1006/nbdi.2000.0304 PMID: 10964604. 8. Thirumangalakudi L, Prakasam A, Zhang R, Bimonte-Nelson H, Sambamurti K, Kindy MS, et al. High cholesterol-induced neuroinflammation and amyloid precursor protein processing correlate with loss of working memory in mice. J Neurochem. 2008; 106(1):475–85. Epub 2008/04/16. https://doi.org/10. 1111/j.1471-4159.2008.05415.x PMID: 18410513. 9. Buga GM, Frank JS, Mottino GA, Hendizadeh M, Hakhamian A, Tillisch JH, et al. D-4F decreases brain arteriole inflammation and improves cognitive performance in LDL receptor-null mice on a West- ern diet. Journal of lipid research. 2006; 47(10):2148–60. https://doi.org/10.1194/jlr.M600214-JLR200 PMID: 16835442. 10. de Oliveira J, Moreira EL, Mancini G, Hort MA, Latini A, Ribeiro-do-Valle RM, et al. Diphenyl Diselenide Prevents Cortico-cerebral Mitochondrial Dysfunction and Oxidative Stress Induced by Hypercholester- olemia in LDL Receptor Knockout Mice. Neurochem Res. 2013. https://doi.org/10.1007/s11064-013- 1110-4 PMID: 23881289. 11. Moreira EL, de Oliveira J, Nunes JC, Santos DB, Nunes FC, Vieira DS, et al. Age-related cognitive decline in hypercholesterolemic LDL receptor knockout mice (LDLr-/-): evidence of antioxidant imbal- ance and increased acetylcholinesterase activity in the prefrontal cortex. Project administration: Jennifer M. Rutkowsky. Project administration: Jennifer M. Rutkowsky. Supervision: John C. Rutledge. Writing – original draft: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Steven Anderson, Gary Cline, Jason Bini, Kamil Borkowski. Writing – original draft: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Steven Anderson, Gary Cline, Jason Bini, Kamil Borkowski. Writing – review & editing: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Steven Anderson, Gary Cline, Jason Bini, Kamil Borkowski, Trina A. Knotts, John C. Rutledge. Writing – review & editing: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Steven Anderson, Gary Cline, Jason Bini, Kamil Borkowski, Trina A. Knotts, John C. Rutledge. 13. Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science. 1986; 232(4746):34–47. PMID: 3513311. Author Contributions Conceptualization: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Steven Anderson, Gary Cline, Jason Bini, Trina A. Knotts, John C. Rutledge. Conceptualization: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Steven Anderson, Gary Cline, Jason Bini, Trina A. Knotts, John C. Rutledge. Data curation: Jennifer M. Rutkowsky, Steven Anderson. Data curation: Jennifer M. Rutkowsky, Steven Anderson. Formal analysis: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Steven Anderson, Gary Cline, Jason Bini, Kamil Borkowski. Funding acquisition: Michelle Puchowicz, Gary Cline, John C. Rutledge. Investigation: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Jason Bini, Kamil Borkowski, Trina A. Knotts. Methodology: Jennifer M. Rutkowsky, Linda L. Lee, Michelle Puchowicz, Mari S. Golub, Douglas E. Befroy, Dennis W. Wilson, Steven Anderson, Jason Bini, Kamil Borkowski, Trina A. Knotts. 31 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Diet & LDLr-/- alter brain function & metabolism References J Alzheimers Dis. 2012; 32 (2):495–511. Epub 2012/07/20. https://doi.org/10.3233/JAD-2012-120541 PMID: 22810096. 12. de Oliveira J, Hort MA, Moreira EL, Glaser V, Ribeiro-do-Valle RM, Prediger RD, et al. Positive correla- tion between elevated plasma cholesterol levels and cognitive impairments in LDL receptor knockout mice: relevance of cortico-cerebral mitochondrial dysfunction and oxidative stress. Neuroscience. 2011; 197:99–106. Epub 2011/09/29. https://doi.org/10.1016/j.neuroscience.2011.09.009 PMID: 21945034. 13. Brown MS, Goldstein JL. A receptor-mediated pathway for cholesterol homeostasis. Science. 1986; 232(4746):34–47. PMID: 3513311. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 32 / 38 Diet & LDLr-/- alter brain function & metabolism 14. Ishibashi S, Brown MS, Goldstein JL, Gerard RD, Hammer RE, Herz J. Hypercholesterolemia in low density lipoprotein receptor knockout mice and its reversal by adenovirus-mediated gene delivery. J Clin Invest. 1993; 92(2):883–93. https://doi.org/10.1172/JCI116663 PMID: 8349823 15. Aung HH, Tsoukalas A, Rutledge JC, Tagkopoulos I. A systems biology analysis of brain microvascu- lar endothelial cell lipotoxicity. BMC systems biology. 2014; 8:80. https://doi.org/10.1186/1752-0509- 8-80 PMID: 24993133 16. Yahiatene I, Aung HH, Wilson DW, Rutledge JC. Single-molecule quantification of lipotoxic expression of activating transcription factor 3. Physical chemistry chemical physics: PCCP. 2014; 16(39):21595– 601. https://doi.org/10.1039/c4cp03260h PMID: 25189785 17. Lee LL, Aung HH, Wilson DW, Anderson SE, Rutledge JC, Rutkowsky JM. Triglyceride-rich lipoprotein lipolysis products increase blood-brain barrier transfer coefficient and induce astrocyte lipid droplets and cell stress. Am J Physiol Cell Physiol. 2017: https://doi.org/10.1152/ajpcell.00120.2016 PMID: 28077357. 18. Kanoski SE, Zhang Y, Zheng W, Davidson TL. The effects of a high-energy diet on hippocampal func- tion and blood-brain barrier integrity in the rat. J Alzheimers Dis. 2010; 21(1):207–19. Epub 2010/04/ 24. https://doi.org/10.3233/JAD-2010-091414 PMID: 20413889. 19. Calvo-Ochoa E, Hernandez-Ortega K, Ferrera P, Morimoto S, Arias C. Short-term high-fat-and-fruc- tose feeding produces insulin signaling alterations accompanied by neurite and synaptic reduction and astroglial activation in the rat hippocampus. J Cereb Blood Flow Metab. 2014; 34(6):1001–8. https:// doi.org/10.1038/jcbfm.2014.48 PMID: 24667917 20. Gonzalez-Scarano F, Baltuch G. Microglia as mediators of inflammatory and degenerative diseases. Annu Rev Neurosci. 1999; 22:219–40. https://doi.org/10.1146/annurev.neuro.22.1.219 PMID: 10202538. 21. Kerrouche N, Herholz K, Mielke R, Holthoff V, Baron JC. 18FDG PET in vascular dementia: differentia- tion from Alzheimer’s disease using voxel-based multivariate analysis. J Cereb Blood Flow Metab. 2006; 26(9):1213–21. Epub 2006/03/10. https://doi.org/10.1038/sj.jcbfm.9600296 PMID: 16525414. 22. Kitagawa K. Cerebral blood flow measurement by PET in hypertensive subjects as a marker of cogni- tive decline. J Alzheimers Dis. 2010; 20(3):855–9. Epub 2010/02/26. https://doi.org/10.3233/JAD- 2010-091324 PMID: 20182039. 23. Stobart JL, Anderson CM. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 References J Cereb Blood Flow Metab. 2006; 26(3):310–20. https://doi.org/10.1038/sj.jcbfm.9600189 PMID: 16079791. 33. Williams DS, Detre JA, Leigh JS, Koretsky AP. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proceedings of the National Academy of Sciences of the United States of America. 1992; 89(1):212–6. Epub 1992/01/01. PMID: 1729691 34. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001; 25(4):402–8. https://doi.org/10.1006/meth.2001. 1262 PMID: 11846609. 35. Boellaard R, Oyen WJ, Hoekstra CJ, Hoekstra OS, Visser EP, Willemsen AT, et al. The Netherlands protocol for standardisation and quantification of FDG whole body PET studies in multi-centre trials. Eur J Nucl Med Mol Imaging. 2008; 35(12):2320–33. https://doi.org/10.1007/s00259-008-0874-2 PMID: 18704407. 36. Garwood M, DelaBarre L. The return of the frequency sweep: designing adiabatic pulses for contem- porary NMR. J Magn Reson. 2001; 153(2):155–77. https://doi.org/10.1006/jmre.2001.2340 PMID: 11740891. 37. Provencher SW. Automatic quantitation of localized in vivo 1H spectra with LCModel. NMR Biomed. 2001; 14(4):260–4. PMID: 11410943. 38. Bederman IR, Foy S, Chandramouli V, Alexander JC, Previs SF. Triglyceride synthesis in epididymal adipose tissue: contribution of glucose and non-glucose carbon sources. J Biol Chem. 2009; 284 (10):6101–8. Epub 2008/12/31. https://doi.org/10.1074/jbc.M808668200 PMID: 19114707 39. Kombu RS, Brunengraber H, Puchowicz MA. Analysis of the citric acid cycle intermediates using gas chromatography-mass spectrometry. Methods Mol Biol. 2011; 708:147–57. Epub 2011/01/06. https:// doi.org/10.1007/978-1-61737-985-7_8 PMID: 21207288. 40. Yang L, Kasumov T., Yu L., Jobbins K., David F., Previs S., Kelleher J., and Brunengraber B. Metabo- lomic assays of the concentration and mass isotopomer distribution of gluconeogenic and citric acid cycle intermediates. Metabolomics. 2006; 2:85–94. 41. Zhang Y, Zhang S, Marin-Valencia I, Puchowicz MA. Decreased carbon shunting from glucose toward oxidative metabolism in diet-induced ketotic rat brain. J Neurochem. 2015; 132(3):301–12. Epub 2014/10/15. https://doi.org/10.1111/jnc.12965 PMID: 25314677 42. Harris SR, Zhang GF, Sadhukhan S, Wang H, Shi C, Puchowicz MA, et al. Metabolomics and mass isotopomer analysis as a strategy for pathway discovery: pyrrolyl and cyclopentenyl derivatives of the pro-drug of abuse, levulinate. Chem Res Toxicol. 2013; 26(2):213–20. Epub 2012/11/23. https://doi. org/10.1021/tx3003643 PMID: 23171137 43. Li Q, Tomcik K, Zhang S, Puchowicz MA, Zhang GF. Dietary regulation of catabolic disposal of 4- hydroxynonenal analogs in rat liver. Free Radic Biol Med. 2012; 52(6):1043–53. Epub 2012/01/17. https://doi.org/10.1016/j.freeradbiomed.2011.12.022 PMID: 22245097 44. Zhang GF, Kombu RS, Kasumov T, Han Y, Sadhukhan S, Zhang J, et al. References Multifunctional role of astrocytes as gatekeepers of neuronal energy supply. Front Cell Neurosci. 2013; 7:38. Epub 2013/04/19. https://doi.org/10.3389/fncel.2013.00038 PMID: 23596393 24. Calderon-Garciduenas L, Mora-Tiscareno A, Franco-Lira M, Zhu H, Lu Z, Solorio E, et al. Decreases in Short Term Memory, IQ, and Altered Brain Metabolic Ratios in Urban Apolipoprotein epsilon4 Chil- dren Exposed to Air Pollution. J Alzheimers Dis. 2015; 45(3):757–70. Epub 2015/01/31. https://doi. org/10.3233/JAD-142685 PMID: 25633678. 25. Durazzo TC, Meyerhoff DJ, Mon A, Abe C, Gazdzinski S, Murray DE. Chronic Cigarette Smoking in Healthy Middle-Aged Individuals Is Associated With Decreased Regional Brain N-acetylaspartate and Glutamate Levels. Biological psychiatry. 2016; 79(6):481–8. Epub 2015/05/17. https://doi.org/10. 1016/j.biopsych.2015.03.029 PMID: 25979621 26. Fayed N, Andres E, Viguera L, Modrego PJ, Garcia-Campayo J. Higher glutamate+glutamine and reduction of N-acetylaspartate in posterior cingulate according to age range in patients with cognitive impairment and/or pain. Academic radiology. 2014; 21(9):1211–7. Epub 2014/07/02. https://doi.org/ 10.1016/j.acra.2014.04.009 PMID: 24981958. 27. Kaufman MJ, Janes AC, Hudson JI, Brennan BP, Kanayama G, Kerrigan AR, et al. Brain and cognition abnormalities in long-term anabolic-androgenic steroid users. Drug and alcohol dependence. 2015; 152:47–56. Epub 2015/05/20. https://doi.org/10.1016/j.drugalcdep.2015.04.023 PMID: 25986964 28. Pospisil P, Kazda T, Bulik M, Dobiaskova M, Burkon P, Hynkova L, et al. Hippocampal proton MR spectroscopy as a novel approach in the assessment of radiation injury and the correlation to neuro- cognitive function impairment: initial experiences. Radiation oncology (London, England). 2015; 10:211. Epub 2015/10/18. https://doi.org/10.1186/s13014-015-0518-1 PMID: 26474857 29. Davidson TL, Monnot A, Neal AU, Martin AA, Horton JJ, Zheng W. The effects of a high-energy diet on hippocampal-dependent discrimination performance and blood-brain barrier integrity differ for diet- induced obese and diet-resistant rats. Physiology & behavior. 2012; 107(1):26–33. https://doi.org/10. 1016/j.physbeh.2012.05.015 PMID: 22634281 30. Ohno M, Sametsky EA, Younkin LH, Oakley H, Younkin SG, Citron M, et al. BACE1 deficiency res- cues memory deficits and cholinergic dysfunction in a mouse model of Alzheimer’s disease. Neuron. 2004; 41(1):27–33. PMID: 14715132. 33 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Diet & LDLr-/- alter brain function & metabolism 31. Golub MS, Germann SL, Mercer M, Gordon MN, Morgan DG, Mayer LP, et al. Behavioral conse- quences of ovarian atrophy and estrogen replacement in the APPswe mouse. Neurobiol Aging. 2008 29(10):1512–23. https://doi.org/10.1016/j.neurobiolaging.2007.03.015 PMID: 17451844 32. Ewing JR, Brown SL, Lu M, Panda S, Ding G, Knight RA, et al. Model selection in magnetic resonance imaging measurements of vascular permeability: Gadomer in a 9L model of rat cerebral tumor. References Centering, scaling, and transformations: improving the biological information content of metabolomics data. BMC genomics. 2006; 7:142. Epub 2006/06/10. https://doi.org/10.1186/1471-2164-7-142 PMID: 16762068 56. Kigerl KA, Gensel JC, Ankeny DP, Alexander JK, Donnelly DJ, Popovich PG. Identification of two dis- tinct macrophage subsets with divergent effects causing either neurotoxicity or regeneration in the injured mouse spinal cord. J Neurosci. 2009; 29(43):13435–44. https://doi.org/10.1523/JNEUROSCI. 3257-09.2009 PMID: 19864556 57. Chhor V, Le Charpentier T, Lebon S, Ore MV, Celador IL, Josserand J, et al. Characterization of phe- notype markers and neuronotoxic potential of polarised primary microglia in vitro. Brain, behavior, and immunity. 2013; 32:70–85. https://doi.org/10.1016/j.bbi.2013.02.005 PMID: 23454862 58. Chico L, Simoncini C, Lo Gerfo A, Rocchi A, Petrozzi L, Carlesi C, et al. Oxidative stress and APO E polymorphisms in Alzheimer’s disease and in mild cognitive impairment. Free Radic Res. 2013; 47 (8):569–76. Epub 2013/05/15. https://doi.org/10.3109/10715762.2013.804622 PMID: 23668794. 59. Kim YS, Joh TH. Microglia, major player in the brain inflammation: their roles in the pathogenesis of Parkinson’s disease. Exp Mol Med. 2006; 38(4):333–47. https://doi.org/10.1038/emm.2006.40 PMID: 16953112. 60. Pascual B, Prieto E, Arbizu J, Marti-Climent J, Olier J, Masdeu JC. Brain glucose metabolism in vascu- lar white matter disease with dementia: differentiation from Alzheimer disease. Stroke. 2010; 41 (12):2889–93. Epub 2010/11/06. https://doi.org/10.1161/STROKEAHA.110.591552 PMID: 21051671. 61. Stewart S, Cacucci F, Lever C. Which memory task for my mouse? A systematic review of spatial memory performance in the Tg2576 Alzheimer’s mouse model. J Alzheimers Dis. 2011; 26(1):105– 26. https://doi.org/10.3233/JAD-2011-101827 PMID: 21558645. 62. Pallebage-Gamarallage M, Lam V, Takechi R, Galloway S, Clark K, Mamo J. Restoration of dietary-fat induced blood-brain barrier dysfunction by anti-inflammatory lipid-modulating agents. Lipids Health Dis. 2012; 11:117. https://doi.org/10.1186/1476-511X-11-117 PMID: 22978403 63. Ng KF, Anderson S, Mayo P, Aung HH, Walton JH, Rutledge JC. Characterizing blood-brain barrier perturbations after exposure to human triglyceride-rich lipoprotein lipolysis products using MRI in a rat model. Magnetic resonance in medicine. 2015. Epub 2015/10/21. https://doi.org/10.1002/mrm.25985 PMID: 26485349 64. Aung HH, Lame MW, Gohil K, An CI, Wilson DW, Rutledge JC. Induction of ATF3 Gene Network by Triglyceride-Rich Lipoprotein Lipolysis Products Increases Vascular Apoptosis and Inflammation. Arteriosclerosis, thrombosis, and vascular biology. 2013; 33(9):2088–96. https://doi.org/10.1161/ ATVBAHA.113.301375 PMID: 23868936. 65. Trevino S, Aguilar-Alonso P, Flores Hernandez JA, Brambila E, Guevara J, Flores G, et al. A high calo- rie diet causes memory loss, metabolic syndrome and oxidative stress into hippocampus and temporal cortex of rats. Synapse. 2015; 69(9):421–33. https://doi.org/10.1002/syn.21832 PMID: 26073877. 66. References Catabolism of 4-hydroxya- cids and 4-hydroxynonenal via 4-hydroxy-4-phosphoacyl-CoAs. J Biol Chem. 2009; 284(48):33521– 34. Epub 2009/09/18. https://doi.org/10.1074/jbc.M109.055665 PMID: 19759021 45. Brunengraber DZ, McCabe BJ, Kasumov T, Alexander JC, Chandramouli V, Previs SF. Influence of diet on the modeling of adipose tissue triglycerides during growth. Am J Physiol Endocrinol Metab. 2003; 285(4):E917–25. Epub 2003/06/12. https://doi.org/10.1152/ajpendo.00128.2003 PMID: 12799315. 46. Fiehn O, Kind T. Metabolite profiling in blood plasma. Methods Mol Biol. 2007; 358:3–17. Epub 2006/ 10/13. https://doi.org/10.1007/978-1-59745-244-1_1 PMID: 17035677. 47. Fiehn O. Metabolomics by Gas Chromatography-Mass Spectrometry: Combined Targeted and Untar- geted Profiling. Current protocols in molecular biology / edited by Ausubel Frederick M [et al]. 2016; 114:30.4.1–30.4.2. Epub 2016/04/03. https://doi.org/10.1002/0471142727.mb3004s114 PMID: 27038389. 48. Cajka T, Fiehn O. Comprehensive analysis of lipids in biological systems by liquid chromatography- mass spectrometry. Trends in analytical chemistry: TRAC. 2014; 61:192–206. Epub 2014/10/14. https://doi.org/10.1016/j.trac.2014.04.017 PMID: 25309011 49. McCoin CS, Piccolo BD, Knotts TA, Matern D, Vockley J, Gillingham MB, et al. Unique plasma meta- bolomic signatures of individuals with inherited disorders of long-chain fatty acid oxidation. Journal of PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 34 / 38 Diet & LDLr-/- alter brain function & metabolism inherited metabolic disease. 2016; 39(3):399–408. Epub 2016/02/26. https://doi.org/10.1007/s10545- 016-9915-3 PMID: 26907176 inherited metabolic disease. 2016; 39(3):399–408. Epub 2016/02/26. https://doi.org/10.1007/s10545- 016-9915-3 PMID: 26907176 50. 50. Matyash V, Liebisch G, Kurzchalia TV, Shevchenko A, Schwudke D. Lipid extraction by methyl-tert- butyl ether for high-throughput lipidomics. Journal of lipid research. 2008; 49(5):1137–46. Epub 2008/ 02/19. https://doi.org/10.1194/jlr.D700041-JLR200 PMID: 18281723 51. Kind T, Liu KH, Lee DY, DeFelice B, Meissen JK, Fiehn O. LipidBlast in silico tandem mass spectrom- etry database for lipid identification. Nature methods. 2013; 10(8):755–8. https://doi.org/10.1038/ nmeth.2551 PMID: 23817071 52. Agrawal K, Hassoun LA, Foolad N, Pedersen TL, Sivamani RK, Newman JW. Sweat lipid mediator profiling: a noninvasive approach for cutaneous research. Journal of lipid research. 2017; 58(1):188– 95. Epub 2016/11/23. https://doi.org/10.1194/jlr.M071738 PMID: 27875258 53. La Frano MR, Fahrmann JF, Grapov D, Fiehn O, Pedersen TL, Newman JW, et al. Metabolic perturba- tions of postnatal growth restriction and hyperoxia-induced pulmonary hypertension in a bronchopul- monary dysplasia model. Metabolomics. 2017; 13(4). ARTN 32 https://doi.org/10.1007/s11306-017- 1170-6 54. Assaad HI, Hou Y, Zhou L, Carroll RJ, Wu G. Rapid publication-ready MS-Word tables for two-way ANOVA. SpringerPlus. 2015; 4:33. https://doi.org/10.1186/s40064-015-0795-z PMID: 25635246 55. van den Berg RA, Hoefsloot HC, Westerhuis JA, Smilde AK, van der Werf MJ. 67. Ito D, Imai Y, Ohsawa K, Nakajima K, Fukuuchi Y, Kohsaka S. Microglia-specific localisation of a novel calcium binding protein, Iba1. Brain Res Mol Brain Res. 1998; 57(1):1–9. Epub 1998/06/19. PMID: 9630473. References Wu A, Ying Z, Gomez-Pinilla F. The interplay between oxidative stress and brain-derived neurotrophic factor modulates the outcome of a saturated fat diet on synaptic plasticity and cognition. Eur J Neu- rosci. 2004; 19(7):1699–707. Epub 2004/04/14. https://doi.org/10.1111/j.1460-9568.2004.03246.x PMID: 15078544. 67. Ito D, Imai Y, Ohsawa K, Nakajima K, Fukuuchi Y, Kohsaka S. Microglia-specific localisation of a novel calcium binding protein, Iba1. Brain Res Mol Brain Res. 1998; 57(1):1–9. Epub 1998/06/19. PMID: 9630473. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 35 / 38 Diet & LDLr-/- alter brain function & metabolism 68. Serrano-Pozo A, Gomez-Isla T, Growdon JH, Frosch MP, Hyman BT. A phenotypic change but not proliferation underlies glial responses in Alzheimer disease. The American journal of pathology. 2013; 182(6):2332–44. Epub 2013/04/23. https://doi.org/10.1016/j.ajpath.2013.02.031 PMID: 23602650 69. Wirths O, Breyhan H, Marcello A, Cotel MC, Bruck W, Bayer TA. Inflammatory changes are tightly associated with neurodegeneration in the brain and spinal cord of the APP/PS1KI mouse model of Alz- heimer’s disease. Neurobiol Aging. 2010; 31(5):747–57. Epub 2008/07/29. https://doi.org/10.1016/j. neurobiolaging.2008.06.011 PMID: 18657882. 70. Kitaguchi H, Ihara M, Saiki H, Takahashi R, Tomimoto H. Capillary beds are decreased in Alzheimer’s disease, but not in Binswanger’s disease. Neurosci Lett. 2007; 417(2):128–31. https://doi.org/10. 1016/j.neulet.2007.02.021 PMID: 17403574. 71. Buee L, Hof PR, Bouras C, Delacourte A, Perl DP, Morrison JH, et al. Pathological alterations of the cerebral microvasculature in Alzheimer’s disease and related dementing disorders. Acta neuropatho- logica. 1994; 87(5):469–80. PMID: 8059599. 72. Gallacher J, Bayer A, Lowe G, Fish M, Pickering J, Pedro S, et al. Is sticky blood bad for the brain?: Hemostatic and inflammatory systems and dementia in the Caerphilly Prospective Study. Arterioscle- rosis, thrombosis, and vascular biology. 2010; 30(3):599–604. Epub 2009/12/08. https://doi.org/10. 1161/ATVBAHA.109.197368 PMID: 19965782. 73. Stott DJ, Spilg E, Campbell AM, Rumley A, Mansoor MA, Lowe GD. Haemostasis in ischaemic stroke and vascular dementia. Blood Coagul Fibrinolysis. 2001; 12(8):651–7. Epub 2001/12/06. PMID: 11734665. 74. Hollestelle MJ, Geertzen HG, Straatsburg IH, van Gulik TM, van Mourik JA. Factor VIII expression in liver disease. Thromb Haemost. 2004; 91(2):267–75. Epub 2004/02/13. https://doi.org/10.1160/TH03- 05-0310 PMID: 14961153. 75. Abutbul S, Shapiro J, Szaingurten-Solodkin I, Levy N, Carmy Y, Baron R, et al. TGF-beta signaling through SMAD2/3 induces the quiescent microglial phenotype within the CNS environment. Glia. 2012; 60(7):1160–71. https://doi.org/10.1002/glia.22343 PMID: 22511296 76. Chen BP, Wolfgang CD, Hai T. Analysis of ATF3, a transcription factor induced by physiological stresses and modulated by gadd153/Chop10. Mol Cell Biol. References Estrogen regulation of glucose metabolism and mitochondrial function: therapeutic impli- cations for prevention of Alzheimer’s disease. Adv Drug Deliv Rev. 2008; 60(13–14):1504–11. Epub 2008/07/24. https://doi.org/10.1016/j.addr.2008.06.003 PMID: 18647624 88. Zwingmann C, Chatauret N, Leibfritz D, Butterworth RF. Selective increase of brain lactate synthesis in experimental acute liver failure: results of a [H-C] nuclear magnetic resonance study. Hepatology. 2003; 37(2):420–8. Epub 2003/01/24. https://doi.org/10.1053/jhep.2003.50052 PMID: 12540793. 89. Peeling J, Shoemaker L, Gauthier T, Benarroch A, Sutherland GR, Minuk GY. Cerebral metabolic and histological effects of thioacetamide-induced liver failure. Am J Physiol. 1993; 265(3 Pt 1):G572–8. Epub 1993/09/01. https://doi.org/10.1152/ajpgi.1993.265.3.G572 PMID: 8214078. 90. Rackayova V, Braissant O, McLin VA, Berset C, Lanz B, Cudalbu C. H and P magnetic resonance spectroscopy in a rat model of chronic hepatic encephalopathy: in vivo longitudinal measurements of brain energy metabolism. Metab Brain Dis. 2015. Epub 2015/08/09. https://doi.org/10.1007/s11011- 015-9715-8 PMID: 26253240. 91. Coenen KR, Hasty AH. Obesity potentiates development of fatty liver and insulin resistance, but not atherosclerosis, in high-fat diet-fed agouti LDLR-deficient mice. Am J Physiol Endocrinol Metab. 2007; 293(2):E492–9. Epub 2007/06/15. https://doi.org/10.1152/ajpendo.00171.2007 PMID: 17566116. 92. Cermenati G, Giatti S, Audano M, Pesaresi M, Spezzano R, Caruso D, et al. Diabetes alters myelin lipid profile in rat cerebral cortex: Protective effects of dihydroprogesterone. The Journal of steroid bio- chemistry and molecular biology. 2017; 168:60–70. Epub 2017/02/09. https://doi.org/10.1016/j.jsbmb. 2017.02.002 PMID: 28167298. 93. Sakamoto H, Yoshida T, Sanaki T, Shigaki S, Morita H, Oyama M, et al. Possible roles of long-chain sphingomyelines and sphingomyelin synthase 2 in mouse macrophage inflammatory response. Bio- chem Biophys Res Commun. 2017; 482(2):202–7. Epub 2016/11/12. https://doi.org/10.1016/j.bbrc. 2016.11.041 PMID: 27836537. 94. de la Monte SM, Tong M, Nguyen V, Setshedi M, Longato L, Wands JR. Ceramide-mediated insulin resistance and impairment of cognitive-motor functions. J Alzheimers Dis. 2010; 21(3):967–84. Epub 2010/08/10. https://doi.org/10.3233/JAD-2010-091726 PMID: 20693650 95. Pascual M, Valles SL, Renau-Piqueras J, Guerri C. Ceramide pathways modulate ethanol-induced cell death in astrocytes. J Neurochem. 2003; 87(6):1535–45. PMID: 14713309. 96. Martinez-Gardeazabal J, Gonzalez de San Roman E, Moreno-Rodriguez M, Llorente-Ovejero A, Manuel I, Rodriguez-Puertas R. Lipid mapping of the rat brain for models of disease. Biochim Biophys Acta. 2017. Epub 2017/02/27. https://doi.org/10.1016/j.bbamem.2017.02.011 PMID: 28235468. 97. Martinez-Clemente M, Ferre N, Gonzalez-Periz A, Lopez-Parra M, Horrillo R, Titos E, et al. 5-lipoxy- genase deficiency reduces hepatic inflammation and tumor necrosis factor alpha-induced hepatocyte damage in hyperlipidemia-prone ApoE-null mice. Hepatology. 2010; 51(3):817–27. Epub 2010/01/30. https://doi.org/10.1002/hep.23463 PMID: 20112424. 98. References 1996; 16(3):1157–68. Epub 1996/03/01. PMID: 8622660 77. Palomera-Avalos V, Grinan-Ferre C, Izquierdo V, Camins A, Sanfeliu C, Pallas M. Metabolic Stress Induces Cognitive Disturbances and Inflammation in Aged Mice: Protective Role of Resveratrol. Reju- venation research. 2017. Epub 2016/12/22. https://doi.org/10.1089/rej.2016.1885 PMID: 27998210. 78. Hu H, Xu Y, Liu C, Zhao H, Zhang H, Wang L. Changes in behavior and in brain glucose metabolism in rats after nine weeks on a high fat diet: a randomized controlled trial. Shanghai Arch Psychiatry. 2014; 26(3):129–37. Epub 2014/08/13. 79. Zheng H, Zheng Y, Zhao L, Chen M, Bai G, Hu Y, et al. Cognitive decline in type 2 diabetic db/db mice may be associated with brain region-specific metabolic disorders. Biochim Biophys Acta. 2017; 1863 (1):266–73. Epub 2016/11/07. https://doi.org/10.1016/j.bbadis.2016.11.003 PMID: 27816519. 80. Liguori C, Stefani A, Sancesario G, Sancesario GM, Marciani MG, Pierantozzi M. CSF lactate levels, tau proteins, cognitive decline: a dynamic relationship in Alzheimer’s disease. J Neurol Neurosurg Psychiatry. 2015; 86(6):655–9. https://doi.org/10.1136/jnnp-2014-308577 PMID: 25121572. 81. Molino Y, David M, Varini K, Jabes F, Gaudin N, Fortoul A, et al. Use of LDL receptor-targeting peptide vectors for in vitro and in vivo cargo transport across the blood-brain barrier. Faseb j. 2017. Epub 2017/01/22. https://doi.org/10.1096/fj.201600827R PMID: 28108572. 82. Algarzae N, Hebron M, Miessau M, Moussa CE. Parkin prevents cortical atrophy and Abeta-induced alterations of brain metabolism: (1)(3)C NMR and magnetic resonance imaging studies in AD models. Neuroscience. 2012; 225:22–34. Epub 2012/09/11. https://doi.org/10.1016/j.neuroscience.2012.08. 057 PMID: 22960314 83. Makaryus R, Lee H, Yu M, Zhang S, Smith SD, Rebecchi M, et al. The metabolomic profile during iso- flurane anesthesia differs from propofol anesthesia in the live rodent brain. J Cereb Blood Flow Metab. 2011; 31(6):1432–42. https://doi.org/10.1038/jcbfm.2011.1 PMID: 21266982 84. Shulman RG, Rothman DL, Hyder F. Stimulated changes in localized cerebral energy consumption under anesthesia. Proceedings of the National Academy of Sciences of the United States of America. 1999; 96(6):3245–50. PMID: 10077669 85. Shulman RG, Hyder F, Rothman DL. Cerebral metabolism and consciousness. C R Biol. 2003; 326 (3):253–73. PMID: 12806834. 86. Manyam NV, Katz L, Hare TA, Gerber JC 3rd, Grossman MH. Levels of gamma-aminobutyric acid in cerebrospinal fluid in various neurologic disorders. Arch Neurol. 1980; 37(6):352–5. Epub 1980/06/01 PMID: 6446278. 36 / 38 PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 Diet & LDLr-/- alter brain function & metabolism 87. Brinton RD. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 References Heemskerk MM, Giera M, Bouazzaoui FE, Lips MA, Pijl H, van Dijk KW, et al. Increased PUFA Con- tent and 5-Lipoxygenase Pathway Expression Are Associated with Subcutaneous Adipose Tissue Inflammation in Obese Women with Type 2 Diabetes. Nutrients. 2015; 7(9):7676–90. Epub 2015/09/ 18. https://doi.org/10.3390/nu7095362 PMID: 26378572 99. Monteiro J, Leslie M, Moghadasian MH, Arendt BM, Allard JP, Ma DW. The role of n—6 and n—3 poly- unsaturated fatty acids in the manifestation of the metabolic syndrome in cardiovascular disease and non-alcoholic fatty liver disease. Food & function. 2014; 5(3):426–35. Epub 2014/02/06. https://doi. org/10.1039/c3fo60551e PMID: 24496399. 100. Chen PY, Wang J, Lin YC, Li CC, Tsai CW, Liu TC, et al. 18-carbon polyunsaturated fatty acids ame- liorate palmitate-induced inflammation and insulin resistance in mouse C2C12 myotubes. J Nutr Bio- chem. 2015; 26(5):521–31. Epub 2015/02/18. https://doi.org/10.1016/j.jnutbio.2014.12.007 PMID: 25687616. 101. Hashimoto M, Hossain S. Neuroprotective and ameliorative actions of polyunsaturated fatty acids against neuronal diseases: beneficial effect of docosahexaenoic acid on cognitive decline in Alzhei- mer’s disease. J Pharmacol Sci. 116(2):150–62. Epub 2011/05/25. PMID: 21606627. 102. Moch D, Schewe T, Kuhn H, Schmidt D, Buntrock P. The linoleic acid metabolite 9DS-hydroxy-10,12 (E,Z)-octadecadienoic acid is a strong proinflammatory mediator in an experimental wound healing model of the rat. Biomedica biochimica acta. 1990; 49(4):201–7. Epub 1990/01/01. PMID: 2169722. 103. Buchanan MR, Bertomeu MC, Brister SJ, Haas TA. 13-Hydroxyoctadecadienoic acid (13-HODE) metabolism and endothelial cell adhesion molecule expression: effect on platelet vessel wall adhesion. Wiener klinische Wochenschrift. 1991; 103(14):416–21. Epub 1991/01/01. PMID: 1718092. 104. Weber E, Haas TA, Muller TH, Eisert WG, Hirsh J, Richardson M, et al. Relationship between vessel wall 13-HODE synthesis and vessel wall thrombogenicity following injury: influence of salicylate and PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 37 / 38 Diet & LDLr-/- alter brain function & metabolism dipyridamole treatment. Thrombosis research. 1990; 57(3):383–92. Epub 1990/02/01. PMID: 2315893. dipyridamole treatment. Thrombosis research. 1990; 57(3):383–92. Epub 1990/02/01. PMID: 2315893. 105. 105. Jorgensen W, Rud KA, Mortensen OH, Frandsen L, Grunnet N, Quistorff B. Your mitochondria are what you eat: a high-fat or a high-sucrose diet eliminates metabolic flexibility in isolated mitochondria from rat skeletal muscle. Physiol Rep. 2017; 5(6). ARTN e13207 https://doi.org/10.14814/phy2.13207 PMID: 28330953 106. Parvaresh Rizi E, Baig S, Shabeer M, Teo Y, Mok SF, Loh TP, et al. Meal rich in carbohydrate, but not protein or fat, reveals adverse immunometabolic responses associated with obesity. Nutr J. 2016; 15(1):100. Epub 2016/12/03. PLOS ONE \| https://doi.org/10.1371/journal.pone.0191909 February 14, 2018 References https://doi.org/10.1186/s12937-016-0219-0 PMID: 27903298 107. Beilharz JE, Maniam J, Morris MJ. Short exposure to a diet rich in both fat and sugar or sugar alone impairs place, but not object recognition memory in rats. Brain, behavior, and immunity. 2014; 37:134– 41. Epub 2013/12/07. https://doi.org/10.1016/j.bbi.2013.11.016 PMID: 24309633. 108. Best T, Howe P, Bryan J, Buckley J, Scholey A. Acute effects of a dietary non-starch polysaccharide supplement on cognitive performance in healthy middle-aged adults. Nutr Neurosci. 2015; 18(2):76– 86. Epub 2014/03/14. https://doi.org/10.1179/1476830513Y.0000000101 PMID: 24621069. 109. Kendig MD, Boakes RA, Rooney KB, Corbit LH. Chronic restricted access to 10% sucrose solution in adolescent and young adult rats impairs spatial memory and alters sensitivity to outcome devaluation. Physiology & behavior. 2013; 120:164–72. Epub 2013/08/21. https://doi.org/10.1016/j.physbeh.2013. 08.012 PMID: 23954407. 110. Murtagh-Mark CM, Reiser KM, Harris R Jr., McDonald RB. Source of dietary carbohydrate affects life span of Fischer 344 rats independent of caloric restriction. The journals of gerontology Series A, Bio- logical sciences and medical sciences. 1995; 50(3):B148–54. Epub 1995/05/01. PMID: 7743394. 111. Freeman LR, Haley-Zitlin V, Rosenberger DS, Granholm AC. Damaging effects of a high-fat diet to the brain and cognition: a review of proposed mechanisms. Nutr Neurosci. 2014; 17(6):241–51. https:// doi.org/10.1179/1476830513Y.0000000092 PMID: 24192577 112. Kan H, Hu W, Wang Y, Wu W, Yin Y, Liang Y, et al. NADPH oxidase-derived production of reactive oxygen species is involved in learning and memory impairments in 16-month-old female rats. Mol Med Rep. 2015. Epub 2015/06/11. https://doi.org/10.3892/mmr.2015.3894 PMID: 26058943. 113. Ghosh D, LeVault KR, Barnett AJ, Brewer GJ. A reversible early oxidized redox state that precedes macromolecular ROS damage in aging nontransgenic and 3xTg-AD mouse neurons. J Neurosci. 2012; 32(17):5821–32. Epub 2012/04/28. https://doi.org/10.1523/JNEUROSCI.6192-11.2012 PMID: 22539844 114. Lovell MA, Markesbery WR. Oxidative DNA damage in mild cognitive impairment and late-stage Alz- heimer’s disease. Nucleic acids research. 2007; 35(22):7497–504. Epub 2007/10/20. https://doi.org/ 10.1093/nar/gkm821 PMID: 17947327 115. Valls-Pedret C, Sala-Vila A, Serra-Mir M, Corella D, de la Torre R, Martinez-Gonzalez MA, et al. Medi- terranean Diet and Age-Related Cognitive Decline: A Randomized Clinical Trial. JAMA Intern Med. 2015; 175(7):1094–103. Epub 2015/05/12. https://doi.org/10.1001/jamainternmed.2015.1668 PMID: 25961184. 38 / 38
https://openalex.org/W4362416552	https://aacr.figshare.com/articles/journal_contribution/Supplementary_Tables_1-5_from_Identification_of_Novel_Gene_Amplifications_in_Breast_Cancer_and_Coexistence_of_Gene_Amplification_with_an_Activating_Mutation_of_i_PIK3CA_i_/22378355/1/files/39823706.pdf	English	null	Supplementary Tables 1-5 from Identification of Novel Gene Amplifications in Breast Cancer and Coexistence of Gene Amplification with an Activating Mutation of <i>PIK3CA</i>	null	2,023	cc-by	5,858	positive Residual mixed invasive ductal and lobular carcinoma BC1-BC42 samples were analyzed using Affymetrix Nsp I and Sty I SNP arrays while BC43-BC161 experiments were conducted using SNP5.0 arrays. BC_No histological _grade TNM ER PR HER2 invas BC1 3 pT2N1bivMX NA NA NA yes BC2 3 T3pN3 negative negative negative yes BC3 3 T2pN2 negative negative negative yes BC4 2 pT3N2Mx positive positive negative yes BC5 3 pN0 negative negative negative yes BC6 3 pN0 NA NA NA yes BC7 3 T3pN2 positive positive positive yes BC8 3 pT3N2aMx negative negative positive yes BC9 2 T2N1biMX positive negative negative yes BC10 3 T3pN3 negative negative positive yes BC11 3 pT2N3 positive negative negative yes BC12 2 T2N0Mx positive negative positive yes BC13 3 pN0 NA NA NA no BC14 3 pN0 NA NA NA yes BC15 3 T3pN2 NA NA NA yes BC16 3 T3pN0 NA NA NA yes BC17 2 T3pN0 negative negative negative no BC18 3 T2pN0 NA NA NA yes BC19 1 T3pN3 NA NA NA yes BC20 2 T2pN2 negative negative NA yes BC21 3 pN0 positive negative positive yes BC22 3 T3pN1 positive positive NA yes BC23 2 pN0 negative negative NA yes BC24 3 T3pN0 negative negative NA yes BC25 2 T3pN2 NA NA NA yes BC26 3 T2pN0 NA NA NA yes BC27 3 T2pN3 negative negative NA yes BC28 2 T3pN3 positive negative positive yes BC29 NA pN2 NA NA NA yes BC30 3 pN0 NA NA NA yes BC31 3 pN1 NA NA NA yes BC32 NA T3pN0 negative negative negative yes BC33 3 T3pN0 NA NA NA yes BC_No histological _grade TNM ER PR HER2 invas BC1 3 pT2N1bivMX NA NA NA yes BC2 3 T3pN3 negative negative negative yes BC3 3 T2pN2 negative negative negative yes BC4 2 pT3N2Mx positive positive negative yes BC5 3 pN0 negative negative negative yes BC6 3 pN0 NA NA NA yes BC7 3 T3pN2 positive positive positive yes BC8 3 pT3N2aMx negative negative positive yes BC9 2 T2N1biMX positive negative negative yes BC10 3 T3pN3 negative negative positive yes BC11 3 pT2N3 positive negative negative yes BC12 2 T2N0Mx positive negative positive yes BC13 3 pN0 NA NA NA no BC14 3 pN0 NA NA NA yes BC15 3 T3pN2 NA NA NA yes BC16 3 T3pN0 NA NA NA yes BC17 2 T3pN0 negative negative negative no BC18 3 T2pN0 NA NA NA yes BC19 1 T3pN3 NA NA NA yes BC20 2 T2pN2 negative negative NA yes BC21 3 pN0 positive negative positive yes BC22 3 T3pN1 positive positive NA yes BC23 2 pN0 negative negative NA yes BC24 3 T3pN0 negative negative NA yes BC25 2 T3pN2 NA NA NA yes BC26 3 T2pN0 NA NA NA yes BC27 3 T2pN3 negative negative NA yes BC28 2 T3pN3 positive negative positive yes BC29 NA pN2 NA NA NA yes BC30 3 pN0 NA NA NA yes BC31 3 pN1 NA NA NA yes BC32 NA T3pN0 negative negative negative yes BC33 3 T3pN0 NA NA NA yes BC_No histological _grade TNM ER PR HER2 invasion node histopathology_type BC_No histological _grade TNM ER PR HER2 invasion node histopathology_type positive invasive medullary carcinoma positive invasive ductal carcinoma positive invasive ductal carcinoma positive infiltrating ductal carcinoma negative invasive ductal carcinoma negative invasive lobular carcinoma positive infiltrating ductal carcinoma positive invasive ductal carcinoma positive invasive ductal carcinoma positive invasive poorly differentiated carcinom positive invasive carcinoma negative invasive ductal carcinoma negative ductal carcinoma in situ negative invasive ductal carcinoma positive invasive ductal carcinoma negative invasive ductal carcinoma negative ductal carcinoma in situ negative invasive ductal carcinoma positive invasive ductal carcinoma positive infiltrating ductal carcinoma negative invasive ductal carcinoma positive infiltrating ductal carcinoma negative infiltrating ductal carcinoma negative infiltrating ductal carcinoma positive infiltrating lobular carcinoma negative infiltrating ductal carcinoma positive infiltrating ductal carcinoma positive invasive breast carcinoma positive infiltrating ductal carcinoma negative infiltrating metaplastic carcinoma positive infiltrating ductal carcinoma negative invasive ductal carcinoma negative infiltrating ductal carcinoma positive invasive medullary carcinoma positive invasive ductal carcinoma positive invasive ductal carcinoma positive infiltrating ductal carcinoma negative invasive ductal carcinoma negative invasive lobular carcinoma positive infiltrating ductal carcinoma positive invasive ductal carcinoma positive invasive ductal carcinoma positive invasive poorly differentiated carcinoma positive invasive carcinoma negative invasive ductal carcinoma negative ductal carcinoma in situ negative invasive ductal carcinoma positive invasive ductal carcinoma negative invasive ductal carcinoma negative ductal carcinoma in situ negative invasive ductal carcinoma positive invasive ductal carcinoma positive infiltrating ductal carcinoma negative invasive ductal carcinoma positive infiltrating ductal carcinoma negative infiltrating ductal carcinoma negative infiltrating ductal carcinoma positive infiltrating lobular carcinoma negative infiltrating ductal carcinoma positive infiltrating ductal carcinoma positive invasive breast carcinoma positive infiltrating ductal carcinoma negative infiltrating metaplastic carcinoma positive infiltrating ductal carcinoma negative invasive ductal carcinoma negative infiltrating ductal carcinoma BC34 3 T2pN0 NA NA NA yes BC35 3 T1pN3 NA NA NA yes BC36 3 T3pN0 NA NA NA yes BC37 3 pN1 positive positive negative yes BC38 3 T2pN2 NA NA NA yes BC39 3 pN0 NA NA NA no BC40 2 pN0 NA NA NA yes BC41 3 pT2pN0pMX positive positive negative yes BC42 3 pT2pN0pMX positive positive NA yes BC43 2 T3pN1 negative negative positive yes BC44 2 pT3N2MX positive positive negative yes BC45 3 T3N1Mx negative negative NA yes BC46 3 T2 negative negative negative yes BC47 3 pT4dN2MX NA NA NA yes BC48 2 T2pN2 positive positive positive yes BC49 2 T3pN0 negative negative negative yes BC50 2 T3pN1 positive NA positive yes BC51 3 T1pN2 negative negative positive yes BC52 3 T3pN1 negative negative negative yes BC53 3 T3pN1 negative negative negative yes BC54 3 T1pN1 positive positive NA yes BC55 3 T2pN2 negative negative negative yes BC56 2 pT2N0Mx positive negative negative yes BC57 NA T3pN0 positive negative positive yes BC58 3 pT3N3Mx positive positive NA yes BC59 2 T2pN0 positive positive positive yes BC60 2 T2N0MX positive positive negative yes BC61 3 pT2pN2pMX NA NA NA yes BC62 3 T2N2aMX positive positive negative yes BC63 3 T3N1MX negative negative negative yes BC64 3 pT3N1Mi positive positive positive yes BC65 3 pT2N3c negative negative negative yes BC66 3 T3pN1 negative negative negative yes BC67 3 pN1 positive positive negative yes BC68 3 T2pN0 positive positive positive yes BC69 2 pT3N2aMx positive positive negative yes negative infiltrating ductal carcinoma positive infiltrating ductal carcinoma negative infiltrating ductal carcinoma positive infiltrating lobular carcinoma positive infiltrating ductal carcinoma negative ductal carcinoma in situ negative infiltrating ductal carcinoma negative invasive ductal carcinoma negative infiltrating ductal carcinoma positive Invasive ductal carcinoma positive infiltrating ductal carcinoma positive invasive ductal carcinoma NA poorly differentiated infiltrating ductal carcinoma positive poorly differentiated infiltrating ductal carcinoma positive infiltrating ductal carcinoma negative Invasive ductal carcinoma positive infiltrating ductal carcinoma positive Invasive, poorly differentiated ductal carcinoma positive invasive ductal carcinoma positive invasive ductal carcinoma positive Metastatic ductal carcinoma positive invasive ductal carcinoma positive moderately differentiated ductal carcinoma negative Invasive lobular carcinoma positive In situ and invasive ductal carcinoma negative invasive ductal carcinoma negative invasive lobular carcinoma positive infiltrating ductal carcinoma positive invasive ductal carcinoma positive invasive breast carcinoma positive Invasive carcinoma positive Invasive carcinoma positive Invasive ductal carcinoma positive Invasive ductal carcinoma negative invasive ductal carcinoma negative infiltrating ductal carcinoma positive infiltrating ductal carcinoma negative infiltrating ductal carcinoma positive infiltrating lobular carcinoma positive infiltrating ductal carcinoma negative ductal carcinoma in situ negative infiltrating ductal carcinoma negative invasive ductal carcinoma negative infiltrating ductal carcinoma positive Invasive ductal carcinoma positive infiltrating ductal carcinoma positive invasive ductal carcinoma NA poorly differentiated infiltrating ductal carcinoma positive poorly differentiated infiltrating ductal carcinoma positive infiltrating ductal carcinoma negative Invasive ductal carcinoma positive infiltrating ductal carcinoma positive Invasive, poorly differentiated ductal carcinoma positive invasive ductal carcinoma positive invasive ductal carcinoma positive Metastatic ductal carcinoma positive invasive ductal carcinoma positive moderately differentiated ductal carcinoma negative Invasive lobular carcinoma positive In situ and invasive ductal carcinoma negative invasive ductal carcinoma negative invasive lobular carcinoma positive infiltrating ductal carcinoma positive invasive ductal carcinoma positive invasive breast carcinoma positive Invasive carcinoma positive Invasive carcinoma positive Invasive ductal carcinoma positive Invasive ductal carcinoma negative invasive ductal carcinoma positive Residual mixed invasive ductal and lobular carcinoma BC70 1 pT1cN1aMx positive positive negative yes BC71 3 T2pN2 positive negative negative yes BC72 3 pT2N3aMX negative negative negative yes BC73 2 pN3 positive positive negative yes BC74 1 pT3N2aMX positive positive negative yes BC75 3 pT4cN1MX negative negative negative yes BC76 3 pT2N1aMX positive positive negative yes BC77 NA T3pN2 positive positive negative yes BC78 1 pN0 NA NA NA yes BC79 3 T3pN0 NA NA NA no BC80 3 T1pN3 negative positive negative yes BC81 2 T3pN0 NA NA NA no BC82 3 pN3 NA NA NA yes BC83 2 pN1 positive positive negative yes BC84 2 T3pN3 NA NA NA yes BC85 3 pN2 NA NA NA yes BC86 2 N1 negative negative negative yes BC87 3 T3pN3 negative negative positive yes BC88 3 pN0 negative negative NA yes BC89 3 pN0 positive positive NA yes BC90 3 T3pN0 NA NA NA no BC91 2 pN0 positive negative negative yes BC92 3 T3pN2 negative negative negative yes BC93 3 T2pN0 NA NA NA yes BC94 2 T3pN2 negative negative negative yes BC95 1 T2pN0 NA NA NA yes BC96 3 pN2 positive positive NA yes BC97 NA pN0 NA NA NA no BC98 3 T3pN0 NA NA NA yes BC99 2 T2pN0 positive negative negative yes BC100 2 T2pN1 positive positive NA yes BC101 2 T2pN0 NA NA NA yes positive Invasive ductal carcinoma positive Invasive ductal carcinoma positive Invasive ductal carcinoma positive invasive ductal carcinoma positive Invasive carcinoma with predominantly lobular features positive Infiltrating lobular carcinoma positive Invasive ductal carcinoma positive Significant residual invasive lobular carcinoma, involving previous biopsy site negative phyllodes tumor negative infiltrating ductal carcinoma positive invasive ductal carcinoma negative Ductal adenocarcinoma of colloid type positive inflammatory carcinoma positive invasive ductal carcinoma positive infiltrating ductal carcinoma positive invasive ductal carcinoma positive invasive ductal carcinoma positive invasive ductal carcinoma negative poorly differentiated infiltrating ductal carcinoma negative Infiltrating ductal carcinoma; carcinoma in situ negative Extensive ductal carcinoma in situ negative Moderately differentiated infiltrating ductal carcinoma positive invasive ductal carcinoma negative infiltrating ductal carcinoma positive infiltrating lobular carcinoma negative Well-differentiated infiltrating ductal carcinoma positive poorly differentiated infiltrating ductal carcinoma negative intraductal epithelial hyperplasia negative infiltrating ductal carcinoma negative Mixed infiltrating ductal and lobular carcinoma associated with DCIS positive Moderately differentiated infiltrating ductal carcinoma negative Infiltrating ductal carcinoma positive Moderately differentiated infiltrating ductal carcinoma negative Infiltrating ductal carcinoma BC102 2 pN0 NA NA NA yes BC103 3 T3 negative negative positive yes BC104 3 T2pN2 negative negative negative yes BC105 2 T2pN2 NA NA NA yes BC106 3 T2 negative negative positive yes BC107 2 T2pN0 NA NA NA yes BC108 2 T2pN1 positive positive NA yes BC109 2 T2pN0 NA NA NA yes BC110 3 pN3 positive negative NA yes BC111 NA T2pN0 NA NA NA no BC112 3 pN1 NA NA NA yes BC113 3 T3pN0 positive positive positive no BC114 NA positive positive positive yes BC115 2 pT2 NA NA NA yes BC116 3 T3pN2 NA NA NA yes BC117 2 pN1 NA NA NA yes BC118 3 T3N2 NA NA NA yes BC119 3 pN2 positive negative negative yes BC120 2 pN0 NA NA NA yes BC121 3 T2pN0 negative negative negative yes BC122 2 T2pN1 NA NA NA yes BC123 3 T2 negative negative negative yes BC124 3 negative negative NA yes BC125 3 T2pN0 NA NA NA yes BC126 3 T2pN3 NA NA NA yes BC127 3 T1pN0 NA NA NA yes BC128 3 T2pN0 NA NA NA yes BC129 1 T2pN0 NA NA NA yes BC130 3 T3pN0 NA NA NA yes BC131 NA T1pN1 NA NA NA yes BC132 2 T1pN1 positive positive positive yes BC133 3 T2pN1 negative negative NA yes BC134 3 T2pN0 NA NA NA yes BC135 3 T2pN0 positive positive NA yes BC136 1 T3pN1 NA NA NA yes BC137 3 T2pN2 NA NA NA yes BC138 3 T3N3Mx NA NA NA yes BC139 3 T2pN0 negative negative negative yes BC140 3 pT2N1MX negative negative NA yes BC141 3 T4bN2M1 NA NA NA yes BC142 3 pN0 NA NA NA no BC143 2 T3N1biMX positive positive negative yes BC144 3 pN0Mx NA NA NA yes BC145 3 T2N0MX NA NA NA yes BC146 1 N1biiiMX NA NA NA yes BC147 1 T1cN0M0,T2N0M0 NA NA NA yes BC148 2 pT3pN3apMX positive positive positive yes BC149 2 pT3pN1aM0 positive positive positive yes BC150 2 pN0 NA NA NA yes BC151 2 pT3pN1a NA NA NA yes BC152 2 pT2pN0pMX NA NA NA yes BC153 3 pT3pN2apMX NA NA NA yes BC154 3 pN1pM0 NA NA positive yes BC155 3 pT2pN0pMx positive positive negative yes BC156 3 pT2pNXpMX positive positive negative yes BC157 1 pT1cpNXpMX positive positive negative yes BC158 2 pT1cpN0pMX positive positive negative yes BC159 3 pT2pN2aM0 positive positive negative yes BC160 2 pT2pN3apMX positive positive negative yes BC161 2 pT2pN3apMX positive negative negative yes negative Infiltrating ductal carcinoma negative invasive breast carcinoma positive Infiltrating ductal carcinoma positive Mixed ductal and lobular invasive carcinoma positive Invasive ductal carcinoma negative Invasive ductal carcinoma positive infiltrating ductal carcinoma positive invasive ductal carcinoma negative ductal carcinoma in situ positive Invasive carcinoma negative High-grade DCIS; invasive ductal carcinoma negative Invasive ductal positive invasive ductal carcinoma negative Invasive ductal adenocarcinoma; Invasive lobular carcinoma positive infiltrating lobular carcinoma positive infiltrating ductal carcinoma negative Invasive ductal carcinoma positive invasive ductal carcinoma negative invasive ductal carcinoma positive infiltrating ductal carcinoma positive Mixed ductal-lobular carcinoma negative invasive ductal carcinoma NA invasive ductal carcinoma NA invasive ductal carcinoma with intraductal component negative Invasive ductal carcinoma positive Invasive ductal carcinoma with extensive intraductal component positive Invasive lobular positive Invasive lobular BC TBL1XR1 PIK3CA IRX2 PRDM1 EGFR FGFR1 ASPH NCALD MYC FGFR2 CCND1 POLD3 BC1 0.05 0.31 0.1 0.15 -0.06 0.21 0.03 0.06 0.06 -0.12 -0.28 -0.63 BC2 -0.09 0.25 0.16 0.14 0.28 -0.34 0.17 0 -0.28 -0.06 0.07 -0.12 BC3 -0.16 0.15 0.2 0.17 0.54 -0.44 0.36 0.95 0.7 -0.38 0.22 0.04 BC4 0.14 -0.33 0.17 -0.3 0.28 0.26 0.5 0.52 0.36 -0.14 0.55 0.74 BC5 -0.1 0.23 0.17 -0.09 0.22 0.31 0.09 0.13 0.67 -0.21 -0.08 -0.44 BC6 -0.18 -0.19 0.06 -0.13 -0.02 -0.12 0.06 0.32 -0.16 0.03 0.27 0.58 BC7 0.36 0.69 0.03 0.04 0.05 -0.11 -0.03 0.23 0.11 -0.09 0.06 0.2 BC8 0.04 0.72 -0.09 -0.15 -0.21 0.2 0.12 0.3 1.41 0.07 0.09 -0.56 BC9 -0.15 0.18 0.06 -0.05 0.22 0.22 0.08 0.05 -0.27 -0.2 -0.12 -0.1 BC10 0.14 0.56 0.27 0.01 0.17 -0.22 0.09 -0.16 0.29 0.12 -0.19 0.01 BC11 0.09 -0.35 0.18 0.16 -0.01 0.28 0.23 0.3 0.22 -0.05 0.16 0.08 BC12 -0.07 0.37 -0.14 -0.09 -0.03 -0.07 0.06 -0.1 -0.06 -0.08 -0.08 -0.42 BC13 -0.24 0.05 0.2 -0.09 -0.09 0.11 0.05 0.27 -0.1 -0.01 -0.23 -0.05 BC14 -0.08 0.26 0.12 -0.19 -0.39 0.48 0.13 0.49 0.01 -0.26 0.22 -0.14 BC15 -0.1 0.24 0.17 -0.14 0.19 -0.19 0.59 0.1 1.42 -0.16 0.09 -0.01 BC16 -0.08 -0.14 0.11 -0.12 0 -0.32 0.45 0.41 -0.01 -0.06 0.06 0.01 BC17 0.02 -0.1 -0.03 0.03 -0.05 0.3 -0.1 0.02 -0.03 -0.06 0.02 -0.34 BC18 0.1 -0.04 0 -0.02 -0.09 0.17 0.06 -0.11 -0.23 -0.01 0.04 -0.16 BC19 0.03 -0.06 0.27 -0.12 0.18 0.05 0.1 0.16 1.8 -0.12 -0.11 -0.62 BC20 0.09 0.28 -0.11 -0.08 -0.11 -0.44 0.11 0.3 0.22 -0.03 -0.09 -0.21 BC21 -0.19 0.09 0.04 -0.13 0.03 0.7 0.5 0.44 0.06 -0.04 0.56 -0.4 BC22 0.08 0.31 -0.18 -0.07 0.04 0.42 0.16 0.29 0.57 -0.13 -0.13 -0.25 BC23 -0.08 0.07 0.03 0.04 -0.1 0.05 -0.06 0.06 0.05 -0.03 0.07 -0.19 BC24 0.26 -0.2 0.31 0.1 0.14 0.04 0.18 0.2 0.23 0.14 0.11 0.06 BC25 -0.01 0.28 0.15 -0.01 0.04 -0.05 -0.07 -0.07 -0.36 -0.02 0.09 0.37 BC26 -0.02 0.27 0.04 -0.05 -0.01 -0.42 0.03 0.06 -0.23 0.07 -0.07 -0.05 BC27 0.07 0.41 0.24 -0.26 0.06 -0.18 -0.08 0.21 0.01 -0.05 -0.08 0.12 BC28 -0.16 0.41 0.01 0.02 0.04 0.73 0.1 0.14 0.19 -0.23 -0.17 -0.19 BC29 0.99 1.28 0.21 -0.08 0.2 0.61 0.15 0.2 0.03 -0.3 0.06 -0.66 BC30 -0.01 -0.17 1.2 -0.04 0.15 -0.03 0.17 0.12 -0.39 -0.05 -0.02 -0.38 BC31 0 0.18 0.15 0.12 0.11 0.51 0.29 0.16 -0.3 -0.08 0.13 0.05 BC32 -0.05 0.14 0.29 -0.05 0 0.07 0.01 0.02 0.13 -0.25 0 -0.43 BC33 -0.02 -0.08 0.06 0.04 0.16 -0.12 -0.1 -0.04 -0.14 -0.15 0.09 0.18 BC34 0.07 0.06 -0.06 -0.02 0.07 0.16 0.01 -0.13 -0.35 0 -0.07 0.23 BC35 -0.1 -0.21 0.1 -0.22 0.05 -0.15 0.05 -0.09 -0.11 -0.04 -0.06 0.04 BC36 -0.03 -0.33 0.5 -0.01 -0.15 0.04 -0.22 0.64 0.45 -0.14 0.08 0.06 BC37 -0.02 0.51 0.17 0.03 0.07 -0.04 -0.04 0.04 -0.05 -0.03 -0.03 -0.08 BC38 -0.03 0.28 0.15 -0.14 0.05 0.31 -0.03 0.06 -0.27 -0.2 0.4 -0.12 BC39 0.05 0.17 0.01 0.06 0.05 -0.31 -0.07 -0.03 -0.03 -0.05 -0.08 0.08 BC40 0.08 0.45 0.17 -0.22 0 -0.38 0.18 -0.03 -0.11 0.08 -0.05 0.18 BC41 -0.1 0.15 0.28 0 0.22 0.7 0.31 0.33 0.02 2.06 0.84 -0.33 BC42 0.07 0.27 0.09 -0.08 0.06 -0.15 -0.13 0.11 -0.23 0.1 0.08 -0.03 BC43 0.01 0.04 -0.02 -0.07 1.86 -0.16 0.05 0.23 0.04 -0.26 0.12 0.04 BC44 0.08 0.7 -0.03 -0.18 0 -0.28 0.01 0.06 -0.18 -0.04 -0.1 -0.09 BC45 -0.18 0.05 0.44 -0.03 0.05 0.16 0.12 0.01 -0.05 -0.22 0.14 -0.1 BC46 0.21 0.56 -0.14 -0.26 0 -0.16 0.21 0.5 0.23 0.02 -0.25 -0.14 BC47 0.18 0.49 0.2 -0.21 0.09 -0.21 0.02 0.74 0.92 0.05 0.04 -0.06 BC48 0.05 0.46 -0.06 -0.06 0.01 -0.43 0.03 0.18 0.02 -0.01 -0.07 -0.17 BC49 0.01 0.41 0.23 -0.09 0 -0.03 0.12 0.56 0.42 -0.4 0.08 -0.04 BC50 0.16 0.38 0.09 -0.27 0.03 0.12 0.12 0.18 -0.1 -0.02 0.1 0.13 BC51 -0.1 0.35 0 -0.24 0.04 -0.13 0.07 0.07 0.09 0.05 -0.08 -0.03 BC52 -0.13 0.28 0.11 -0.14 0.07 -0.04 0 0.28 -0.08 0.05 -0.06 0.11 BC53 -0.16 0.07 -0.04 -0.17 0.1 -0.12 0.08 0.19 0.11 0 -0.16 0.14 BC54 -0.15 0.05 0.07 -0.1 -0.03 -0.04 0.05 0.23 -0.11 -0.02 0.33 0.03 BC55 0.02 -0.06 -0.03 -0.07 0.17 -0.06 0.14 0.53 0.48 -0.09 -0.05 -0.05 BC56 0.1 -0.05 -0.05 -0.13 -0.07 0.15 0.25 0.42 0.59 -0.01 -0.27 -0.19 BC57 0.05 0.09 0.05 0.31 -0.05 0.12 0.02 0.16 0.03 -0.26 -0.04 -0.04 BC58 -0.11 0.52 0.09 -0.29 0.04 0.21 0.07 0.25 0.06 -0.07 0.17 0.75 BC59 0.08 0.4 -0.03 -0.13 -0.01 0.04 0.03 0.24 -0.01 -0.02 -0.07 -0.06 BC60 -0.05 -0.08 0.4 -0.17 -0.02 0.09 -0.03 -0.03 -0.13 0.05 0.52 -0.02 BC61 0.13 0.23 0.07 0.02 0.05 -0.03 0.06 0.31 0.34 -0.01 -0.02 0.08 BC62 0.25 -0.05 0.04 -0.05 -0.06 -0.06 0.06 0.46 0.04 -0.21 -0.03 0.12 BC63 -0.05 -0.14 0.08 -0.15 -0.04 0 0.12 0.15 -0.01 -0.04 0.05 0.07 BC64 0.64 0.63 -0.08 -0.19 -0.03 -0.13 0.05 0.18 -0.36 0.03 0.06 0.05 BC65 0.48 0.56 0.26 -0.29 -0.01 -0.04 0.1 0.66 0.12 -0.2 -0.1 0.32 BC66 -0.02 0.36 0.26 -0.11 -0.03 0.06 0.29 0.52 0.11 0.14 -0.22 0.09 BC67 -0.06 0 -0.04 -0.08 0.02 1.25 0.05 0.4 0.02 -0.14 1.15 -0.22 BC68 -0.09 0.25 0.17 -0.27 0.16 -0.5 0.23 0.26 0.14 0.28 0.33 0.72 BC69 0.09 0.34 0.16 -0.45 -0.01 0 0.03 0.29 0.14 0.04 0.64 0.44 BC70 -0.04 0.18 -0.04 -0.12 -0.01 0.06 0.08 0.02 -0.22 -0.08 -0.08 0 BC71 0.39 0.38 -0.03 -0.21 0.12 0.28 0.76 0.98 0.93 -0.04 -0.31 -0.23 BC72 -0.09 0.3 0.22 -0.08 0 -0.08 0.04 0.01 -0.08 -0.14 0.04 0.16 BC73 -0.11 0.09 0.19 -0.18 0.06 -0.04 0.01 0.02 -0.14 0.09 -0.08 -0.04 BC74 -0.05 0.18 0.09 -0.11 0.02 0 0.07 0.16 -0.41 -0.09 0.06 -0.04 BC75 0.02 0.1 0.16 0.25 0 0.09 0.12 0.23 0.19 0.24 -0.07 0.26 BC76 0.05 1.02 -0.03 0.1 0.05 0.14 0.1 0.68 0.11 -0.28 -0.16 -0.04 BC77 -0.01 0.42 -0.16 -0.24 -0.13 -0.31 -0.11 0.3 0.21 0.25 0.74 0.98 BC78 -0.02 0.43 1.67 -0.16 -0.11 -0.14 0.22 0.35 -0.11 -0.4 -0.44 -0.2 BC79 0.11 0.11 -0.05 -0.04 -0.08 -0.53 0.31 0.67 0.53 -0.31 -0.17 0.14 BC80 -0.09 -0.18 -0.01 -0.05 -0.06 0.12 0.3 1.75 0.44 -0.08 0.64 -0.14 BC81 -0.02 -0.03 -0.15 -0.34 -0.02 0.07 0.19 0.57 0.35 -0.08 -0.28 -0.21 BC82 -0.07 0.03 0.17 -0.15 -0.1 0.59 0.69 0.76 0.66 -0.14 0.15 0.06 BC83 -0.22 0.14 0.08 -0.34 0.23 0.18 0.18 0.19 0.05 -0.05 0.39 -0.15 BC84 0.14 0.44 -0.2 -0.11 -0.08 0 0.11 0.22 -0.35 -0.02 0.15 -0.17 BC85 0.27 0.49 0.14 -0.26 0.07 0 0.17 0.48 0.19 -0.18 -0.02 0.08 BC86 -0.22 -0.12 0.37 -0.11 0.83 0.15 0.29 0.31 0.19 -0.2 0.04 0.04 BC87 -0.08 0.19 0.09 -0.16 -0.04 0.23 0.9 0.41 0.72 -0.13 -0.01 -0.21 BC88 0.12 0.07 -0.11 -0.09 -0.06 -0.05 0.1 0.38 0.25 -0.09 -0.19 -0.1 BC89 0.12 0.24 0.04 0.05 0.1 0.28 0.04 0.31 0.11 0.11 0.29 0.42 BC90 0.17 -0.18 -0.08 -0.05 -0.04 -0.52 0.01 0.12 -0.03 -0.07 0.08 0.04 BC91 0.14 -0.18 0.06 -0.08 -0.07 -0.24 0.08 0.22 0.04 -0.1 -0.01 -0.01 BC92 -0.1 -0.1 0.27 -0.02 0.07 0.24 0.12 0.07 0.37 -0.07 0.14 0.31 BC93 0.12 0.08 -0.08 -0.4 0.17 0.15 0.27 0.62 0.53 -0.36 0 -0.26 BC94 -0.17 0.14 0.13 -0.13 0.08 -0.13 0.02 0.08 -0.22 -0.04 -0.21 -0.11 BC95 -0.02 -0.3 0.06 -0.03 0.18 -0.25 0.02 0.11 0.12 -0.15 -0.06 -0.1 BC96 0.04 -0.11 0.11 0 0.02 -0.5 0.23 0.46 0.29 -0.2 0.92 -0.28 BC97 0.07 -0.05 0 -0.01 -0.02 -0.28 0.1 0.36 0.41 -0.09 -0.21 0.05 BC98 0.15 0.07 -0.08 -0.01 0 0.71 0.13 0.21 0.15 -0.2 -0.02 -0.54 BC99 -0.08 0.22 0.06 -0.34 0.15 0.08 0.14 0.2 -0.04 0 0.42 -0.09 BC100 0.21 0.15 -0.04 -0.21 0.08 -0.18 0.05 0.23 0.31 -0.07 -0.23 -0.12 BC101 0.32 0.51 -0.16 -0.38 -0.04 0.02 0.03 0.42 0.14 -0.03 -0.04 0.19 BC102 -0.06 0.06 0.11 -0.13 -0.07 -0.17 0.01 0.09 -0.34 0.05 -0.01 0.13 BC103 0.26 0.68 0.12 -0.16 0.08 -0.22 0.37 0.3 0.21 -0.06 0.06 -0.18 BC104 -0.01 0.13 0.18 -0.28 -0.03 0.12 0.04 0.11 -0.13 0.05 -0.11 0.14 BC105 0.15 0.61 -0.05 0.02 -0.04 -0.16 0.06 0.43 0.31 -0.04 -0.12 -0.16 BC106 -0.14 0.08 0.18 -0.3 0.19 -0.22 0.15 0.15 -0.03 -0.05 0.08 0.25 BC107 -0.03 0.25 -0.02 -0.12 -0.05 -0.11 0.08 0.08 -0.17 -0.03 0.11 0.05 BC108 -0.09 0.1 0.14 -0.09 0.11 0.14 0.07 0.19 0.22 -0.08 -0.07 -0.01 BC109 0.11 0.08 0.06 -0.06 0.02 0.93 0.2 0.34 0.4 -0.14 0.77 0.89 BC110 0.07 0.04 0.08 -0.1 -0.02 0.13 0.06 0.22 0.23 -0.08 0.11 -0.17 BC111 0.11 0.27 -0.33 -0.21 -0.05 0.3 0.22 0.39 0.03 -0.05 -0.14 -0.27 BC112 0.11 0.13 0 0.07 1.65 -0.15 0.08 0.2 0.22 -0.15 -0.24 -0.18 BC113 0.02 -0.3 0 -0.01 -0.02 -0.18 -0.02 0.15 0.39 -0.04 -0.22 -0.06 BC114 -0.05 0.1 0.24 -0.25 -0.1 0.74 0.05 0.03 -0.1 -0.1 0.49 0.24 BC115 -0.1 0.09 0.12 -0.25 0.15 0.09 0.11 0.1 -0.16 0 0.02 0.05 BC116 0.06 0.33 -0.3 -0.14 -0.08 -0.28 0.06 0.42 0.12 -0.03 -0.15 0.01 BC117 0.06 0.12 0.08 -0.07 0.47 0.33 -0.29 0.63 0.57 -0.17 1.33 1.35 BC118 -0.11 -0.12 0.14 -0.16 0.06 -0.24 0.24 0.28 0.12 -0.14 -0.16 0.24 BC119 -0.04 0.42 -0.09 -0.26 0.11 -0.62 0.16 0.31 -0.03 -0.12 0.03 -0.03 BC120 0 0.23 -0.05 -0.03 -0.02 0.84 0.17 0.38 0.02 0.01 0.8 -0.14 BC121 0.39 0.08 -0.17 -0.03 0.04 0.36 0.12 0.34 0.17 0.87 0.07 0.33 BC122 0.46 0.61 -0.13 -0.34 0.07 -0.49 0.1 0.46 -0.04 0.27 -0.19 -0.02 BC123 0.1 0.23 0.06 0.1 0.17 -0.56 0.46 0.6 0.42 -0.07 -0.19 0 BC124 -0.33 -0.42 0.02 -0.45 0.05 -0.44 0.13 0.06 0.07 -0.45 -0.08 0.19 BC125 0.03 0.33 -0.06 -0.04 0.1 -0.16 0.54 0.62 0.33 0.09 -0.19 -0.21 BC126 0.12 0.34 -0.09 -0.19 0.13 -0.05 -0.07 0.16 0.01 -0.03 0.11 0.24 BC127 -0.02 -0.02 -0.05 -0.13 -0.06 -0.11 0 0.12 -0.21 -0.04 -0.1 0.04 BC128 0.01 0.03 0.11 0.02 0.04 -0.37 0.11 0.17 0.21 -0.08 0.19 0.42 BC129 -0.01 0.17 -0.21 -0.39 -0.03 0.72 0.03 0.18 0.39 -0.16 -0.29 -0.37 BC130 0.14 0.43 0.09 -0.07 0.06 -0.23 0.14 0.17 -0.29 0 -0.01 0.34 BC131 0.04 -0.2 0.04 -0.24 -0.02 -0.5 0.17 0.4 0.18 -0.07 0.05 -0.2 BC132 0.09 0.45 -0.04 -0.3 -0.01 1.41 0.22 0.19 0.05 -0.14 0.87 0.11 BC133 -0.01 -0.01 0.06 -0.11 0.1 -0.32 0.36 0.58 0.83 -0.07 -0.13 -0.06 BC134 0.03 0.06 0.62 -0.01 0.08 -0.4 0.27 0.58 0.59 -0.25 -0.12 0.01 BC135 -0.03 -0.15 0.16 0 -0.03 -0.6 0.03 0.18 0.31 -0.08 -0.12 -0.19 BC136 -0.05 0.46 0 -0.1 -0.07 -0.04 0.04 0.14 -0.16 -0.07 -0.14 -0.12 BC137 0.13 0.24 -0.01 -0.17 -0.16 0.23 0.11 0.64 0.18 -0.02 -0.06 -0.02 BC138 -0.01 -0.01 0.08 -0.5 0.05 -0.52 0.32 0.7 0.33 0.03 0.06 0.05 BC139 0.08 0.14 0.28 -0.09 -0.02 0.07 0.36 0.56 0.28 -0.06 0 0.26 BC140 -0.2 0.18 -0.05 -0.15 -0.02 -0.27 0.12 0.17 -0.19 -0.15 -0.23 0.12 BC141 -0.11 0.26 -0.02 -0.02 0.1 -0.13 0.26 0.27 -0.21 -0.1 -0.18 -0.2 BC142 -0.08 0.07 0 -0.11 -0.05 0.01 0.15 0.13 -0.01 -0.05 -0.08 0.09 BC143 -0.04 0.29 -0.05 -0.14 -0.08 0.06 0.04 0.1 0.09 0.03 0.01 -0.25 BC144 0.1 0.13 0.14 0.02 -0.01 -0.26 0.13 0.56 0.42 -0.18 0 -0.07 BC145 0.02 -0.2 0.07 -0.18 -0.05 -0.43 0.03 0.28 0 -0.18 0.08 -0.38 BC146 -0.03 -0.02 0.1 -0.19 -0.01 -0.25 0.01 0.28 0.1 -0.09 -0.07 -0.25 BC147 0.15 -0.28 -0.05 -0.19 -0.1 -0.23 0.04 0.27 0.18 -0.13 -0.09 0.03 BC148 0.09 0.19 -0.32 -0.11 -0.06 0.79 0.06 0.33 -0.13 -0.07 1.07 0.42 BC149 -0.07 0 -0.06 -0.11 0.04 0.09 0.15 0.19 0.13 -0.07 0.01 -0.09 BC150 1.52 0.36 0.09 -0.1 -0.05 0.35 0.02 0.1 -0.11 -0.01 0.09 -0.13 BC151 -0.14 0 0.02 -0.04 0.13 -0.46 0 0.13 -0.14 -0.13 0.06 -0.31 BC152 0.13 -0.06 0.11 -0.07 0.06 -0.49 0.3 0.52 0.63 -0.04 -0.05 -0.15 BC153 -0.03 -0.01 0.37 1.41 -0.02 -0.17 0.39 0.44 0.26 -0.2 -0.28 -0.29 BC154 -0.05 0.02 0.15 -0.17 0.04 -0.21 0.04 0.12 -0.29 -0.05 0.2 -0.14 BC155 0.56 1.06 0.25 0.91 0.11 0.19 0.81 0.63 0.35 -0.21 0.08 0.15 BC156 -0.15 -0.01 0.27 -0.1 0.13 -0.29 0.66 0.62 0.21 -0.16 -0.13 -0.24 BC157 -0.02 0.11 0.27 -0.15 -0.02 0 0.04 0.07 -0.12 -0.11 -0.16 0.11 BC158 -0.09 0.41 0.08 -0.16 0.06 -0.07 0.12 0.25 -0.08 -0.22 0.07 0.07 BC159 -0.1 0.11 0.28 -0.06 0.08 -0.03 0.09 0.16 -0.05 -0.2 -0.07 0.06 BC160 -0.03 0.39 0.01 -0.33 0.06 0.1 0.15 0.24 -0.16 -0.07 -0.1 -0.08 BC161 0.02 -0.15 0.03 -0.11 -0.02 0.02 0.07 0.14 -0.02 -0.06 -0.13 -0.01 CCND2 IRAK4 ERBB2 TLK2 NOTCH3 -0.01 0.08 -0.01 -0.09 -0.19 -0.1 0.23 -0.18 0.33 0.06 0.85 1.69 -0.16 0.04 0.03 0.05 -0.2 -0.04 0.22 0.17 -0.08 0.24 -0.13 0.04 0 -0.16 -0.18 -0.33 0.39 -0.13 -0.2 0.31 1.09 0.7 -0.09 0.32 -0.17 -0.18 0.03 -0.39 -0.1 -0.05 -0.18 -0.09 -0.25 0.32 0.33 -0.1 0.2 -0.11 0.01 0.27 1.43 0.1 -0.1 -0.34 -0.04 -0.43 0.32 -0.11 -0.17 -0.12 0.7 -0.09 0.04 -0.18 -0.08 0.2 0.51 0.6 -0.22 0.07 1.73 0.38 -0.32 -0.18 0.08 -0.04 0.29 -0.15 -0.18 0.01 0.15 0.16 -0.16 -0.01 -0.06 -0.24 0.33 0.04 -0.04 -0.12 1.33 0.23 -0.07 -0.04 0.18 1.06 0.28 -0.1 -0.18 0.09 -0.09 0.23 -0.02 -0.09 0.11 0.09 0.08 -0.17 -0.03 0.01 -0.26 0.21 0.08 -0.1 -0.43 0.14 -0.07 -0.18 -0.14 0.03 0.02 0.39 -0.07 -0.02 0.22 0.07 0.33 0.04 -0.25 0.33 1.02 0.4 -0.04 -0.14 -0.01 1.35 0.13 0 -0.29 0.15 -0.2 0.51 -0.18 -0.24 0.08 -0.28 0.4 -0.12 -0.22 1.19 1.14 0.87 0.09 -0.03 0.05 0 -0.03 0.12 -0.07 0.13 1.21 0.4 0.14 -0.13 -0.58 -0.04 0.31 -0.09 0 0.07 -0.1 0.31 -0.07 0.49 -0.11 0.02 0.29 0.19 0.01 -0.24 -0.03 0.28 0 -0.03 0.1 -0.39 0.15 -0.16 -0.03 0.15 0.04 0.24 -0.08 -0.11 0.15 1.16 -0.06 -0.01 -0.14 -0.08 -0.07 0.31 0.42 0.05 0.09 0.02 0.43 0.17 0.14 -0.08 -0.36 -0.18 0.19 -0.01 -0.17 -0.22 -0.21 -0.04 0.36 0.05 -0.25 -0.08 -0.05 0.35 0.31 -0.26 -0.29 -0.09 -0.03 -0.05 -0.17 -0.2 -0.04 0.03 -0.04 -0.21 -0.13 -0.13 0.04 0.05 -0.31 0.22 -0.01 0.09 -0.28 1.06 -0.17 0.08 0.05 -0.1 1.16 0.13 0.02 0.31 -0.16 -0.01 0.03 -0.15 0.3 -0.1 -0.04 -0.07 -0.13 -0.06 0.03 -0.27 -0.27 -0.28 0.32 0.72 -0.23 -0.11 0.02 0.07 0.06 -0.04 -0.32 -0.02 -0.21 0.2 -0.19 0.15 -0.02 -0.1 0.18 1.33 -0.02 -0.08 0.02 0.11 -0.06 -0.03 -0.02 0.01 0.21 -0.17 -0.03 -0.07 0.02 0.09 -0.15 -0.03 -0.16 -0.02 -0.1 0.08 0.04 0.02 0.8 -0.17 -0.46 -0.36 0.93 -0.1 0.12 1.46 -0.19 0.07 0.16 0.06 -0.43 -0.12 0.11 0.08 -0.09 -0.18 -0.25 -0.17 0.04 -0.01 -0.25 -0.09 0.38 0.13 0.13 1.19 -0.13 -0.23 -0.01 -0.07 -0.07 0.27 0.85 0.06 -0.02 -0.01 -0.17 -0.02 0.39 0.2 0.15 0.13 -0.02 0.14 0 -0.12 -0.04 -0.09 -0.09 -0.05 -0.37 -0.32 -0.23 -0.03 0.02 -0.03 -0.08 -0.01 0.24 -0.23 -0.65 -0.46 0.2 0.03 -0.36 -0.1 0.2 0.16 -0.06 0.05 -0.07 -0.01 0.09 0.02 0.22 -0.58 -0.5 -0.25 -0.01 0.08 -0.34 -0.09 -0.01 0.4 -0.26 -0.04 -0.15 -0.08 0.11 -0.04 -0.17 -0.21 -0.08 0.11 -0.14 -0.11 -0.03 -0.46 0.13 0.07 -0.25 -0.19 -0.08 0.03 0.18 0.01 -0.32 -0.2 0.37 0.14 -0.47 -0.45 0.04 0.13 0.24 -0.21 -0.11 0.04 -0.06 0.01 1.08 -0.01 -0.03 0.14 0.01 -0.2 0.03 0.13 0.03 0.15 -0.23 -0.32 -0.2 0.12 -0.08 0.31 -0.31 -0.09 0.13 -0.12 0.12 -0.04 0.1 0.1 -0.26 -0.44 -0.05 0.21 0.46 0.03 -0.37 0.19 0.15 -0.15 -0.09 -0.3 -0.31 0.29 0.16 -0.11 -0.21 -0.11 0.11 0.33 0.45 -0.17 -0.03 -0.03 0.09 0.09 -0.12 -0.15 0.02 0.08 -0.04 0.04 -0.11 0 -0.3 0.06 -0.28 -0.11 0.01 0.14 0.01 -0.24 -0.4 0.03 -0.03 0 0 -0.06 -0.07 -0.06 0.28 -0.11 -0.08 -0.14 0.26 -0.25 1.11 -0.07 -0.19 -0.02 0.1 -0.26 -0.11 0 0.07 -0.04 -0.11 -0.14 -0.12 -0.17 -0.05 0.36 -0.08 0.06 -0.11 0.05 -0.21 -0.3 -0.25 -0.03 -0.07 -0.25 -0.13 -0.08 0.22 0.06 0.01 0.22 -0.18 0.09 0.04 0.1 0.05 -0.02 0.11 -0.11 -0.02 0 0.13 -0.01 -0.18 -0.19 -0.11 0.01 0.09 -0.01 0.19 -0.05 0.02 -0.01 0.07 -0.04 0.02 -0.1 0.03 0.23 -0.17 -0.25 0.14 0.16 -0.03 1.31 -0.33 -0.14 0.14 -0.05 -0.06 -0.12 0.06 0.21 0.08 -0.34 -0.16 -0.04 -0.07 -0.09 -0.24 0.08 -0.23 -0.1 0.17 -0.18 -0.13 -0.15 0.4 0.01 -0.15 -0.05 0.04 -0.09 -0.14 -0.35 -0.23 -0.13 -0.03 -0.14 -0.19 -0.13 0.06 -0.11 0.03 -0.09 -0.1 0.49 -0.17 -0.29 -0.11 -0.03 -0.4 0.32 0.11 -0.05 -0.23 0.29 -0.01 -0.12 -0.16 -0.02 -0.07 0.44 -0.08 -0.25 -0.23 0.07 -0.04 0 -0.1 -0.27 -0.2 0.22 -0.06 -0.32 -0.21 -0.06 0.13 -0.11 -0.22 0.01 0 -0.14 -0.17 0.43 0.29 0.01 0.06 0.02 0.73 0.09 -0.18 0.18 -0.25 -0.28 -0.03 0.19 0.1 0 0.17 0.12 0.23 -0.08 -0.13 -0.31 -0.17 0.09 0 0 -0.29 -0.29 0.29 0.06 -0.13 -0.5 -0.48 0.58 0.51 0.28 -0.28 -0.18 0.05 0.08 0.11 -0.27 -0.11 -0.09 0.02 -0.09 -0.19 -0.04 -0.15 -0.16 0 0.72 -0.36 -0.06 -0.02 -0.01 -0.16 -0.22 -0.06 0.27 0.06 -0.03 -0.01 0.05 0.2 -0.07 -0.19 1.19 -0.03 0.1 -0.16 -0.09 -0.03 0 0.17 -0.16 -0.03 -0.04 0.11 0.11 -0.06 -0.05 -0.09 0 -0.01 -0.09 -0.3 -0.16 -0.07 -0.05 -0.22 -0.31 0.37 -0.04 0.12 -0.02 -0.37 -0.21 -0.09 0.14 0.05 1.16 -0.08 0.04 1.42 0.1 -0.09 -0.12 0.09 -0.01 -0.03 1.07 -0.12 0.27 -0.19 0.07 0.24 0.16 -0.43 0.05 0 -0.23 -0.08 0.13 0.01 -0.02 -0.32 0 0.08 0.05 -0.01 -0.11 -0.13 0 -0.01 -0.11 -0.28 -0.03 0.15 -0.05 -0.01 -0.07 -0.09 -0.01 -0.02 -0.03 -0.17 -0.1 0.04 BC TP53 PIK3CA AKT1 BC1 C275F BC2 BC3 P278R BC4 BC5 H1047R BC6 H1047R BC7 R248W BC8 BC9 BC10 R213STOP BC11 BC12 BC13 BC14 R248Q E545K BC15 BC16 BC17 BC18 BC19 H179Y BC20 K164E BC21 BC22 BC23 BC24 M246K P42T BC25 BC26 BC27 BC28 BC29 R248W E545K BC30 BC31 S241Y E545K BC32 R175H H1047R BC33 BC34 BC35 BC36 S215R BC37 BC38 BC39 BC40 Y163C H1047R BC41 R273C BC42 BC43 R248L BC44 H1047R BC45 G245V BC46 Y220C BC47 BC48 BC49 BC50 A159T BC51 BC52 H1047R BC53 H1047R BC54 BC55 BC56 BC57 BC58 BC59 BC60 H1047R BC61 BC62 BC63 frame shift BC64 BC65 Q144K BC66 R282W BC67 K291N BC68 BC69 BC70 BC71 splicing AG2 BC72 BC73 BC74 BC75 BC76 R273C K39N BC77 BC78 H1047L BC79 H193R BC80 BC81 BC82 BC83 E17K BC84 H1047L BC85 F134I BC86 H1047R BC87 H1047R BC88 BC89 BC90 BC91 BC92 frame shift BC93 H168R BC94 BC95 BC96 P278T H1047R E17K BC97 BC98 BC99 BC100 H1047R BC101 E1012K splicing AG2AT BC102 K39N BC103 frame shift G1049R BC104 BC105 E545K BC106 N239D BC107 E17K BC108 E17K BC109 BC110 BC111 BC112 BC113 BC114 BC115 BC116 BC117 BC118 BC119 BC120 H1047R BC121 L130I BC122 Y234C H1047R BC123 splicing BC124 R181C E17K BC125 E285V H1047R BC126 W146R BC127 BC128 R175H BC129 BC130 C238Y BC131 BC132 BC133 H1047R BC134 V272E BC135 BC102 K39N BC103 frame shift G1049R BC104 BC105 E545K BC106 N239D BC107 E17K BC108 E17K BC109 BC110 BC111 BC112 BC113 BC114 BC115 BC116 BC117 BC118 BC119 BC120 H1047R BC121 L130I BC122 Y234C H1047R BC123 splicing BC124 R181C E17K BC125 E285V H1047R BC126 W146R BC127 BC128 R175H BC129 BC130 C238Y BC131 BC132 BC133 H1047R BC134 V272E BC135 K39N BC136 BC137 R282W BC138 R248W BC139 BC140 BC141 H1047L BC142 BC143 BC144 BC145 BC146 frame shift BC147 BC148 BC149 BC150 D32Y BC151 BC152 BC153 complex BC154 BC155 P151S BC156 BC157 BC158 BC159 E17K BC160 BC161 clinical phenotype gene amplified total number of tumors p-value odds ratio CI left CI right or not staining neg staining pos staining neg staining pos no amplification no amplification with amplification with amplification HER2 ERBB2 51 14 1 9 75 5.26E-05 31 3.776 1451 HER2 FGFR1 50 18 2 5 75 0.02525 6.736 0.9967 76.54 PR CCND1 51 33 0 8 92 0.001027 infinite 2.409 infinite PR MYC 45 41 6 0 92 0.03156 0 0 1.004 PR POLD3 51 37 0 4 92 0.03624 infinite 0.8505 infinite ER FGFR1 39 47 0 7 93 0.03891 infinite 1.107 infinite size < 5 cm size >= 5 cm size < 5 cm size >= 5 cm no amplification no amplification with amplification with amplification tumor size ERBB2 81 57 8 15 161 0.04163 2.648 0.9767 7.726 mutation analysis without mutation with mutation without mutation with mutation no amplification no amplification with amplification with amplification TP53 NCALD 111 33 6 11 161 0.000727 6.079 1.893 21.64 TP53 CCND2 117 41 0 3 161 0.0194 infinite 1.125 infinite TP53 PIK3CA 113 38 4 6 161 0.02623 4.41 0.9863 22.43 PIK3CA PIK3CA 131 20 5 5 161 0.008968 6.43 1.352 30.76 number of tumors in each category odds ratio CI left CI right clinical phenotype gene amplified oligo LINE1_CN_qPCR_f LINE1_CN_qPCR_r ERBB2_CN_qPCR_f ERBB2_CN_qPCR_r MYC_CN_qPCR_f MYC_CN_qPCR_r BRD4_CN_qPCR_f BRD4_CN_qPCR_r IRX2_CN_qPCR_f IRX2_CN_qPCR_r NOTCH3_CN_cDNA_qPCR_f NOTCH3_CN_cDNA_qPCR_r TBL1XR1_CN_qPCR_f TBL1XR1_CN_qPCR_r BRD4_cDNA_qPCR_f BRD4_cDNA_qPCR_r IRX2_cDNA_qPCR_f IRX2_cDNA_qPCR_r TBL1XR1_cDNA_qPCR_f TBL1XR1_cDNA_qPCR_f AKT1.exon3.intron_f AKT1.exon3.intron_r BRAF.exon11.intron_f BRAF.exon11.intron_r BRAF.exon15.intron_f BRAF.exon15.intron_r HRAS_exon2.intron_f HRAS_exon2.intron_r HRAS_exon3.intron_f HRAS_exon3.intron_r PIK3CA.2.exon10.intron_f PIK3CA.2.exon10.intron_r PIK3CA.2.exon21.intron_f sequence AAAGCCGCTCAACTACATGG TGCTTTGAATGCGTCCCAGAG GGCTGGTGAGAAAGGTGGATT GGGTGTAGGCCACATTCATAGG AGCCGGCGAGAGAAAGAAG CGTCCCCACCCCCAGTAC CATCCCCACATGCGGG CAGCAACGATGTCCTGTGTATACTG AAACCCAGATTGCCACGAAA CGAAACACACCCAGGCTCAT GGGCTGCTGGTGGGAAA CCACATTTACAGGGACACAAAGG TGGCACGATCTCGGCTAAC GGGCTGAGGCAGGAGAATC GAGCGCTATGTCACCTCCTGTT GCCGGCAATCACATCAACTT TCGCCCTTCTACGGCAACT CCCTGGCCCTGCAGC TGGAATGCAGCAGGAGACAA TTCCGAAGGTCTAATACACAAACTGA GAGGGTCTGACGGGTAGAGTGTG AAAGAGGGCTCCAGCCAACC CATAAGGTTTTCTTTTTCTGTTTGGC CAAAATAAAAGTTGTTAAACATATCCT ATGCTTGCTCTGATAGGAAAATGAGA TCTAGTAACTCAGCAGCATCTCAGGG GTTTGCCCTTCAGATGGCCC AGCCCTATCCTGGCTGTGTCCT GGAGAGGCTGGCTGTGTGAACT CACCTGTGCGGCGTGGGCTC TCCAGAGGGGAAAAATATGACAAAGA ATGCTGAGATCAGCCAAATTCAGTTA TATCTAGCTATTCGACAGCATGCCAA oligo LINE1_CN_qPCR_f LINE1_CN_qPCR_r ERBB2_CN_qPCR_f ERBB2_CN_qPCR_r MYC_CN_qPCR_f MYC_CN_qPCR_r BRD4_CN_qPCR_f BRD4_CN_qPCR_r IRX2_CN_qPCR_f IRX2_CN_qPCR_r NOTCH3_CN_cDNA_qPCR_f NOTCH3_CN_cDNA_qPCR_r TBL1XR1_CN_qPCR_f TBL1XR1_CN_qPCR_r BRD4_cDNA_qPCR_f BRD4_cDNA_qPCR_r IRX2_cDNA_qPCR_f IRX2_cDNA_qPCR_r TBL1XR1_cDNA_qPCR_f TBL1XR1_cDNA_qPCR_f AKT1.exon3.intron_f AKT1.exon3.intron_r BRAF.exon11.intron_f BRAF.exon11.intron_r BRAF.exon15.intron_f BRAF.exon15.intron_r HRAS_exon2.intron_f HRAS_exon2.intron_r HRAS_exon3.intron_f HRAS_exon3.intron_r PIK3CA.2.exon10.intron_f PIK3CA.2.exon10.intron_r PIK3CA.2.exon21.intron_f PIK3CA.2.exon21.intron_r TP53.exon4.intron_f2 TP53.exon4.intron_r2 TP53.exon5.intron_f TP53.exon5.intron_r TP53.exon6.intron_f TP53.exon6.intron_r TP53.exon7.intron_f TP53.exon7.intron_r TP53.exon8.intron_f TP53.exon8.intron_r GTGGAATCCAGAGTGAGCTTTCATTT CCTGGTCCTCTGACTGCTCTTTTCACCCA GGCCAGGCATTGAAGTCTCAT CGTCTTCCAGTTGCTTTATCTGTTCA AAGAGCAATCAGTGAGGAATCAGAGG GGAGAGACGACAGGGCTGGTT CCACTGACAACCACCCTTAACCC CTTGCCACAGGTCTCCCCAAG GTCAGAGGCAAGCAGAGGCTG AGGTAGGACCTGATTTCCTTACTGCC AAGTGAATCTGAGGCATAACTGCACC This table has the primer sequences used in the experiments.
https://openalex.org/W2071295911	https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0014289&type=printable	English	null	Sublethal Doses of Anthrax Lethal Toxin on the Suppression of Macrophage Phagocytosis	PloS one	2,010	cc-by	8,985	Abstract Background: Lethal toxin (LT), the major virulence factor produced by Bacillus anthracis, has been shown to suppress the immune system, which is beneficial to the establishment of B. anthracis infections. It has been suggested that the suppression of MEK/MAPK signaling pathways of leukocytes contributes to LT-mediated immunosuppressive effects. However, the involvement of MAPK independent pathways has not been clearly elucidated; nor has the crucial role played by LT in the early stages of infection. Determining whether LT exerts any pathological effects before being enriched to an MEK inhibitory level is an important next step in the furtherance of this field. Methodology/Principal Findings: Using a cell culture model, we determined that low doses of LT inhibited phagocytosis of macrophages, without influencing MAPK pathways. Consistent low doses of LT significantly suppressed bacterial clearance and enhanced the mortality of mice with bacteremia, without suppressing the MEK1 of splenic and peripheral blood mononuclear cells. Conclusion/Significance: These results suggest that LT suppresses the phagocytes in a dose range lower than that required to suppress MEK1 in the early stages of infection. Citation: Kau J-H, Sun D-S, Huang H-S, Lien T-S, Huang H-H, et al. (2010) Sublethal Doses of Anthrax Lethal Toxin on the Suppression of Macrophage Phagocytosis. PLoS ONE 5(12): e14289. doi:10.1371/journal.pone.0014289 Editor: Adam J. Ratner, Columbia University, United States of America Received July 15, 2010; Accepted November 18, 2010; Published December 10, 2010 Received July 15, 2010; Accepted November 18, 2010; Published December 10, 2010 Copyright: 2010 Kau 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. supported by National Science Council of Taiwan ROC under grant no. 95-2311-B-320-006, 96-2311-B-320-005-MY3 and 99-2311-B-320- no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Funding: This work was supported by National Science Council of Taiwan ROC under grant no. 95-2311-B-320-006, 96-2311-B-320-00 003-MY3. The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscrip Competing Interests: The authors have declared that no competing interests exist. * E-mail: hhchang@mail.tcu.edu.tw * E-mail: hhchang@mail.tcu.edu.tw . These authors contributed equally to this work. . These authors contributed equally to this work. mediated generation of bactericidal superoxide and MEK- mediated survival of macrophage cells [12–13]. Abstract Whether or not LT suppresses macrophage function through a MEK/MAPK independent pathway, however, remains unclear. Jyh-Hwa Kau1., Der-Shan Sun2., Hsuan-Shun Huang2, Te-Sheng Lien2, Hsin-Hsien Huang1, Hung-Chi Lin1, Hsin-Hou Chang2* 1 Institute of Preventive Medicine, National Defense Medical Center, Taipei, Taiwan, Republic of China, 2 Department of Molecular Biology and Human Genetics, Tzu-Chi University, Hualien, Taiwan, Republic of China Sublethal Doses of Anthrax Lethal Toxin on the Suppression of Macrophage Phagocytosis Jyh-Hwa Kau1., Der-Shan Sun2., Hsuan-Shun Huang2, Te-Sheng Lien2, Hsin-Hsien Huang1, Hung-Chi Lin1, Hsin-Hou Chang2* Low plasma levels of LT detected in mice in early stages of infection To analyze the effects of low doses of LT on the suppression of phagocytosis in vivo, we examined the MEK1 level of mouse splenic mononuclear cells (SMCs) and peripheral blood mono- nuclear cells (PBMCs) after treatment with LT. Our data revealed that intravenous LT treatments with doses of 200 and 400 mg/kg, (but not 50 mg/kg) (LF:PA = 15:85) significantly suppressed the MEK1 level of SMCs and PBMCs of the mice (Fig. 4A). The corresponding mouse plasma levels of PA and LF within two and 24 hours following treatments with LT (50 mg/kg) are shown (Fig. 4B). The concentration of LF was much lower than the level required to suppress MEK1 and MAPK pathways in vitro (Fig. 4B LF levels: ,1 ng/mL in 2 hr groups, and not detectable [ND] in 24 hr groups vs. Fig. 1B, 1E 1.5 ng/mL groups). Theoretically, the effective dosage of LT in the early infectious stages is crucial to the survival of the bacterium. As a result, mouse plasma PA and LF concentrations in the early stages of infection were also examined. Mice in the experiment were treated with a lethal dose of B. anthracis spores (36107 CFU/kg; lethality began at 48– 72 hours after the treatment), and the plasma levels of PA and LF were measured at various points in time. Our data revealed that during the first 24 hours, PA and LF were not detectable in the plasma of mice; 48 hours following treatment, only a low level of PA was barely detectable. The low level of PA and LF detected in mice during the early stages of the infection was somewhat similar to a previous report using other experimental animals [27]. Our data implies that the plasma level of LT in the early stages of infection was much lower than the MEK1/MAPK suppressive levels under both in vitro and in vivo conditions (Fig. 1B and 1E 1.5 ng/mL LF groups vs. 4A and 4B 50 mg/kg LT groups vs. 4C 48 hrs groups). Introduction Anthrax refers to diseases caused by infection with the Gram- positive bacterium Bacillus anthracis. The bacteria replicates to very high numbers in the blood, leading to the death of the host. Because the bacteria spread extensively without an evident immune response, it has been suggested that the pathogen impairs the immune system of the host. Recent studies have shown that lethal toxin (LT), the major virulence factor of Bacillus anthracis [1– 3], strongly contributes to this intervention through pleiotropic action on various cells of the host immune system [4–5]. Lethal toxin isogenic mutant bacteria lacks single toxin components and is therefore greatly attenuated [6–7]. Lethal toxin is a bipartite protein complex comprising protective antigen (PA) and the lethal factor (LF), which functionally inactivates the mitogen activated protein kinase kinases (MAPKKs/MEKs) through metalloprotease activity of LF [8–10]. Mitogen activated protein kinases (MAPKs) are downstream effectors of MEKs, which were divided into the extracellular-regulated kinase (ERK, p42/44), p38, and c-Jun amino terminal protein kinase (JNK) subgroups. These have been implicated in multiple cellular events such as proliferation, survival, differentiation and inflammation [11]. Lethal toxin inhibits phagocytic function through the suppression of MEK- Some evidences have implied that LT might play an inhibitory role in phagocytic function of phagocytes at low doses without perturbing the MEK/MAPK pathways. Lethal toxin was shown to inhibit actin dynamics without influencing the MEK1 pathway; the inhibition of actin dynamics blocks the migration of neutrophil and Listeria-driven actin motility [14]. This implies that LT-mediated inhibition might also affect on phagocytosis, an actin dependent process. Phagocytic function of phagocytes is important for the clearance of bacteria. This was indicated by another major virulence factor of B. anthracis, the poly-c-D- glutamic acid capsule, inhibiting phagocytosis through nonim- munogenic surfaces, which is critical for full virulence [15]. It has also been shown that in vivo treatments for LT reduced fluorescent polystyrene beads engaged with mouse spleen macrophages [16], and may likely support the suppressive role of LT on phagocytosis. However, treatment under the same conditions also suppressed the MEK-C/EBP axis signaling pathways and induced apoptosis of spleen macrophages [16]. This leads to the conclusion that an MEK-dependent process is still involved, while PLoS ONE \| www.plosone.org December 2010 \| Volume 5 \| Issue 12 \| e14289 1 Anthrax LT on Phagocytosis lower doses of LT was not likely caused by cell death. In the above phagocytosis experiments (Fig. Various doses of LT affected cell survival, MEK/MAPK pathways and phagocytosis of macrophage cells Low plasma levels of LT detected in mice in early stages of infection Various doses of LT affected cell survival, MEK/MAPK pathways and phagocytosis of macrophage cells p y p g y p g The dosage effect of LT on the suppression of cell viability and intact MEK1 levels was analyzed using J774A.1 cells (Fig. 1A, B), a mouse macrophage cell line commonly applied for the study of LT function [13,18]. We found that the suppression on MEK1 is associated with the induction of cell death when treated with various doses of LT (Fig. 1A, 1B; significant suppression in 1.5 and 15 ng/mL LF groups vs. no significant suppression in the other groups). The phagocytosis of living bacteria Escherichia coli, Bacillus cereus and Bacillus subtilis, which were used as surrogates as B. anthracis, were then analyzed under treatments with LT in the same range of dosage (Fig. 1C, 1D). Our data revealed that LT significantly suppressed phagocytic activity, regardless of the bacterium used (Fig. 1C, D), suggesting that such suppression occurred without strain specificity. Intriguingly, the effective dose of LT on phagocytosis includes low doses that do not significantly suppress either cell survival or MEK1 (Fig. 1C and 1D vs. 1A and 1B, 1.5–150 pg/mL LF groups). In addition, MEK1, LT also blocks other MEKs [19]. To further investigate how LT treatments influence MAPKs, the downstream of MEKs, we performed Western blotting analysis (Fig. 1E). Respective gel densities of the phospho-p38 (pp38) MAPK and the phospho-p42/ 44 (pp42/44) MAPK under various treatments are shown (Fig. 1F). Our data indicated that LT primarily blocked the activation of p38 MAPK but not the p42/44 MAPK and JNK under treatment with lipopolysaccharide (LPS) (Fig. 1E, 1F; 1.5 ng/mL LF). Intriguingly, here we found that the inhibitory dose of LT on p38 MAPK (Fig. 1E, 1.5 ng/mL LF) was concurrent with the doses that affect both cell survival (Fig. 1A) and MEK1 (Fig. 1B). As LPS-induced cytokine production of macrophage was largely blocked when MAPKs were inactivated [20–21], we found that the inhibitory dose of LT for TNF-a production (Fig. 1G, 1.5 ng/ mL LF groups) was also consistent with that of the cell viability, MEK1 and p38MAPK data (Fig. 1A, 1B, 1E, 1.5 ng/mL LF groups). These results suggest that LT influenced MEK1 and MEK3/MEK6, the upstream of p38 MAPK, in a similar dose range in murine macrophages. In addition, such suppression of phagocytosis could be induced by low dose LT without influencing the MAPK pathways or cell survival. Introduction 1C, 1D), we measured phagocy- tosis using a gentamicin protection assay [22–26]. After plating the macrophage lysate on agar plates, the LT treated groups with low E. coli counts were considered to exhibit low phagocytosis activity. Low E. coli counts, however, might be explained by an enhanced killing of the engulfed E. coli in LT-activated macrophages rather than a lower uptake of these bacteria, or alternatively, increased permeability of LT-treated cells to the antibiotic gentamicin. To confirm the suppressive effect of LT, fluorescent beads were used as phagocytosis targets and analyzed using flow cytometry (Fig. 3). Macrophage J774A.1 cells were allowed to engulf immunoglobulin-opsonized fluorescent beads. We were able to distinguish the beads from the macrophages using particle/cell-size differences (Fig. 3A vs. 3C, 3E, 3G; 3I; R1 regions: J774A.1 cells). By differentiating the fluorescent signal of cells, we were able to measure the beads-phagocytosed phago- cytes of PA and LT treated groups (Fig. 3F, 3H, 3J, right panels). Relative phagocytosis was determined by calculating the beads- phagocytosed/non-phagocytosed ratio of the cell population (Fig. 3K). We found that treatments using doses higher than 15 pg/mL of LF (with 85 ng/mL PA) significantly reduced the engulfment on beads (Fig. 3K, * P,0.05, ** P,0.01, 15 pg/mL LF groups vs. LF untreated groups), which is partly consistent with our phagocytosis data using living bacteria (Fig. 1C, 1D, 15 and 150 pg/mL LF groups). When the kinase data was also taken into consideration (Fig. 1B and 1E, no inhibition in 15 and 150 pg/mL LF groups), these results suggested that low dose LT- mediated suppression on phagocytosis involving a MAPK independent pathway. the involvement of MEK-independent pathway remains unclear. As the role of LT in the early phases of infection is an important issue deserving of further investigations [17], it would be interesting to address the potential pathological role of low dose LT, by which it would not influence the MEK pathway. As a result, in this study we sought to investigate the inhibitory role of low dose LT on phagocytosis. Related issues regarding whether or not this suppression on phagocytes is beneficial for the survival of infected bacteria are also discussed. Low dose of LT suppressed the bacterial clearance and enhanced mortality in septic mice result, a septic shock mouse model was employed. First, we measured the sublethal and lethal doses of E. coli and LT, respectively. We found that treatment using E. coli at doses higher than 3.56109 CFU/kg, or using LT doses higher than 400 mg/kg could induce mortality in mice (Fig. 5A, 5B). The LT dose 50 mg/ kg is not only a sublethal dose for mice (Fig. 5B), but also a sub- MEK1 inhibitory dose of mouse splenic and peripheral blood mononuclear cells (Fig. 4A, SMCs and PBMCs groups). Notably, even though sublethal treatments using either E. coli (Fig. 5A, 26109 CFU/kg) or LT (Fig. 5B, 50 mg/kg) alone showed no mortality, and a combination of both treatments resulted in 100% mortality among the mice (Fig. 5C). In contrast, combined treatments using vehicle or control protein bovine serum albumin (BSA) plus E. coli (26109 CFU/kg) did not lead to mortality (Fig. 5C). To investigate the inhibitory role of sub-MEK1 dose of LT (Fig. 4A, 50 mg/kg) on the bacterial clearance in vivo, viable bacteria (CFU) in mouse blood specimens were analyzed after combined treatment with LT and E. coli for 2, 16 and 24 hours (Fig. 5D). The control mice treated with BSA significantly eliminated bacteria from their circulation within 24 hours, while LT-treated mice showed no obvious clearance during the same period, as evidenced by the 4–6 fold higher levels of residual surviving bacteria in their circulation (Fig. 5D, LT vs. BSA groups; * P,0.05, ** P,0.01). These findings suggested that LT-mediated suppression with a sub-MEK1 inhibitory dose of splenic and peripheral blood mononuclear cells is sufficiently beneficial for bacteria to survive host clearance. According to our results, sub-MEK1 inhibitory doses of LF (15 pg/mL to 150 pg/mL) plus adequate PA (85 ng/mL) are sufficient to suppress the macrophage phagocytosis in vitro (Fig. 1C, 1D, 3K). Theoretically, such low doses of LT might be sufficient to enhance the mortality and bacterial survival in septic mice. As a Figure 2. Flow cytometry analysis for LT-induced cell death of macrophage cells. Flow cytometry analysis was employed to determine the cell death of macrophage cells (A–D). After 3-hour treatments of various concentrations of LT, mouse macrophage J774A.1 cells were stained with annexin V-APC and PI. The upper right quadrants (Rl) and the lower right quadrants (R2) were positive for PI uptake representing the non-viable cells (A–D). Flow cytometry analyses of cell death and phagocytosis of fluorescent beads We employed flow cytometry to determine whether LT-treated cells were actually killed by the toxin, and, therefore, unable to phagocytose the bacteria. Through annexin V and propidium iodide (PI) staining, we found that the LT treatments could only induce cell death in the 1.5 ng/mL LF groups but not the other groups treated with lower doses of LT (Fig. 2; 2B vs. 2C, 2D; 2E, 1.5 ng/mL LF groups vs. 1.5–150 pg/mL groups). This is consistent with our WST-1 data (Fig. 1A, 1.5 ng/mL LF groups), and further indicates that the suppression of phagocytosis by PLoS ONE \| www.plosone.org PLoS ONE \| www.plosone.org December 2010 \| Volume 5 \| Issue 12 \| e14289 2 Anthrax LT on Phagocytosis PLoS ONE \| www.plosone.org 3 December 2010 \| Volume 5 \| Issue December 2010 \| Volume 5 \| Issue 12 \| e14289 December 2010 \| Volume 5 \| Issue 12 \| e14289 December 2010 \| Volume 5 \| Issue 12 \| e14289 PLoS ONE \| www.plosone.org 3 Anthrax LT on Phagocytosis Anthrax LT on Phagocytosis Figure 1. LT-mediated suppression of macrophage cells. Inhibitory effects of various concentrations of LT treatments (3 hours) on cell viability (A), intact MEK1 levels (B), phagocytosis of E. coli (C), B. cereus and B. subtilis (D), MAPK pathways (E, F) and TNF-a production (G) of mouse macrophage J774A.1 cells are shown. The numbers of surviving cells (A) and surviving bacteria after phagocytosis (C, D) in the PA groups (LF untreated) were adjusted to 100%. Levels of pp38, p38, pp42/44, p42/44 MAPKs in J774A.1 cells treated with LPS and LT were analyzed by Western blotting (E). Phosphorylated c-Jun (p-c-Jun) levels revealed JNK activity in J774A.1 cells (E). Quantitative results of pp42/44 and pp38 levels were obtained by measuring gel intensity normalized with internal controls p42/44 and p38, respectively (F); the levels in the PA groups (LF untreated) were adjusted to 100%. After various PA, LT and LPS treatments, the level of TNF-a secreted by J774A.1 cells was determined by ELISA (G); PA groups (without LF and LPS treatment) were adjusted to 100%. Error bars indicate standard deviations. * P,0.05, P,0.01, * P,0.001 compared to the PA controls (PA: 85 ng/mL). n = 6 (3 experiments with 2 replicates) (A, C, D, G). doi:10.1371/journal.pone.0014289.g001 Low dose of LT suppressed the bacterial clearance and enhanced mortality in septic mice One representative diagram out of three experiments is showed, in which the percentage of cells in each quadrant was indicated (A–D). The quantitative results of relative dead cells (% of cell subpopulations in R1 + R2) are presented as mean 6 standard deviation (E). P,0.01, compared to the PA control groups (LF untreated). doi:10.1371/journal.pone.0014289.g002 Discussion The time course of LT and PA treatments was 3 hours. In pa cell size (forward scatter) and the Y-axis represents the granularity of cells (side scatter). A and B: fluo only, E and F: beads + PA (85 ng/mL) treated cells, G and H: beads + LT (LF 150 pg/mL + PA 85 ng/mL + PA 85 ng/mL) treated cells. One representative diagram out of three experiments is shown (A–J). The population of the macrophage, which was specifically gated to obtain the graphs B, D, F, H, J, respectiv fluorescent intensity, which indicates the levels of macrophage-engaged beads, and the Y-axis repre PLoS ONE \| www.plosone.org 5 Dec Figure 3. Flow cytometry analysis for phagocytosis. Flow cytometry analysis was employed to determine the phagocytosed level of fluorescent beads by macrophages (A–J). The time course of LT and PA treatments was 3 hours. In panels A, C, E, G, I, the X-axis represents particle/ cell size (forward scatter) and the Y-axis represents the granularity of cells (side scatter). A and B: fluorescent beads only, C and D: macrophage cells only, E and F: beads + PA (85 ng/mL) treated cells, G and H: beads + LT (LF 150 pg/mL + PA 85 ng/mL) treated cells, I and J: beads + LT (LF 15 pg/mL + PA 85 ng/mL) treated cells. One representative diagram out of three experiments is shown (A–J). The R1 region in panels A, C, E, G and I reveals the population of the macrophage, which was specifically gated to obtain the graphs B, D, F, H, J, respectively. In panels B, D, F, H, J, the X-axis represents fluorescent intensity, which indicates the levels of macrophage-engaged beads, and the Y-axis represents the granularity of the cells (side scatter). Figure 3. Flow cytometry analysis for phagocytosis. Flow cytometry analysis was employed to determine the phagocytosed level of fluorescent beads by macrophages (A–J). The time course of LT and PA treatments was 3 hours. In panels A, C, E, G, I, the X-axis represents particle/ cell size (forward scatter) and the Y-axis represents the granularity of cells (side scatter). Discussion Phagocytosis is a process relying highly on actin cytoskeletal dynamics [28]. Specific mechanisms that block actin dynamics, in so doing suppress phagocytic uptake into macrophages. Such mechanisms are evolved by various pathogens to avoid robust bactericidal effector functions [29]. Given that the MEK/MAPK pathways play an important role in the regulation of cytoskeletal dynamics [30], and that LT, an MEK inhibitor, could block actin dynamics [14,31], it is not surprising to observe the inhibitory role of LT on phagocytosis. However, whether or not there involves a MAPK-independent pathway in LT-mediated suppression on phagocytosis, is not clearly demonstrated. Intriguingly, our data suggested that such inhibitory pathways are likely to exist. Using a cell culture model, here we find that LT could suppress the phagocytosis of J774A.1 macrophage cells at low doses without influencing the MAPK pathways (Fig. 1C, 1D and 3K; 15 pg/mL groups). A septic shock mouse model revealed that the anti- phagocytic properties of LT could exacerbate the bacterial clearance and enhance mortality in mice (Fig. 5C). Enhanced mortality in septic animal after LT treatment is to a certain degree consistent with a previous study [32]. These authors shed light on the anti-inflammatory role of LT on the amelioration of LPS treatment in mice, while the exacerbation of LT on E. coli treated mice, however, was not specifically linked to the suppression of macrophage phagocytosis [32]. Our data provides an alternative Figure 2. Flow cytometry analysis for LT-induced cell death of macrophage cells. Flow cytometry analysis was employed to determine the cell death of macrophage cells (A–D). After 3-hour treatments of various concentrations of LT, mouse macrophage J774A.1 cells were stained with annexin V-APC and PI. The upper right quadrants (Rl) and the lower right quadrants (R2) were positive for PI uptake representing the non-viable cells (A–D). One representative diagram out of three experiments is showed, in which the percentage of cells in each quadrant was indicated (A–D). The quantitative results of relative dead cells (% of cell subpopulations in R1 + R2) are presented as mean 6 standard deviation (E). P,0.01, compared to the PA control groups (LF untreated). d l December 2010 \| Volume 5 \| Issue 12 \| e14289 PLoS ONE \| www.plosone.org 4 Anthrax LT on Phagocytosis Figure 3. Flow cytometry analysis for phagocytosis. Flow cytometry analysis was employ fluorescent beads by macrophages (A–J). Discussion Cellular actin levels were used as internal controls of respective groups (A). The relative gel densities after being normalized with respective actin levels are shown; the level of untreated groups was adjusted to 100% (A). The plasma PA and LF levels were determined using ELISA from specimens of mice treated with LT (50 mg/kg, a sublethal dose) (B), or spores of B. anthracis (36107 CFU/kg, a lethal dose) (C). ND: not detectable. n = 6 (3 experiments with 2 replicates) (B, 2 hr groups; C). n = 10 (3 experiments with 3 or 4 replicates) (B, 24 hr groups). doi:10.1371/journal.pone.0014289.g004 q g One of the candidate pathways is through NALP1. It has been shown that the susceptibility of different mouse strains is linked to the polymorphic locus Nalp1b [35]. The product of this locus, NALP1, is a component of inflammasone, a protein complex consisting of NALP1 and caspases 1, leading to the processing and release of pro-inflammatory cytokines IL-1b and IL-18 [36]. Activation of caspase-1 by NALP1 mediates murine macrophage cytolysis through the mitochondrial protein Bnip3 and Bnip3L [35,37]. Because MEK cleavage occurs in both resistant and susceptible cells [13,38], but NALP1 activation occurs in susceptible macrophages only [35], these evidences suggest that LT-mediated NALP1 suppression might be independent of MEK pathways. However, NALP1 is linked to LT-mediated suppres- sion of other cellular functions, while the phagocytosis has not been characterized in these studies. As a result, it may be suggested that NALP1 or a yet unidentified pathway is responsible for the suppression of phagocytosis by low dose LT. Further investigations are required to characterize the specific pathways involved. Figure 4. Leukocyte MEK1 levels and plasma LT levels in LT treated mice. After mice were treated with various doses (50, 200 and 400 mg/kg) of LT intravenously for 3 hours, intact MEK1 levels of mouse splenic mononuclear cells (SMCs) and peripheral blood mononuclear cells (PBMCs) were analyzed using Western blotting (A). Cellular actin levels were used as internal controls of respective groups (A). The relative gel densities after being normalized with respective actin levels are shown; the level of untreated groups was adjusted to 100% (A). The plasma PA and LF levels were determined using ELISA from specimens of mice treated with LT (50 mg/kg, a sublethal dose) (B), or spores of B. anthracis (36107 CFU/kg, a lethal dose) (C). ND: not detectable. Discussion 3K indicate the respective conditions as showed in Fig. 3F, 3H and 3J, respectively (K). Error bars indicate standard deviations of three experiments. * P,0.05, ** P,0.01 compared to the PA (LF untreated) control groups. n = 6 (3 experiments with 2 replicates). doi:10.1371/journal.pone.0014289.g003 explanation in which such phenomena could be attributed to the suppression of phagocytosis by LT, because LT greatly reduced bacterial clearance (Fig. 5D). Because we used low doses of LT without influencing the MEK1 of splenic and peripheral blood mononuclear cells (Fig. 4A, 50 mg/kg LT groups), the MEK- dependent suppression on bactericidal superoxide [12] is not likely to be involved. Because phagocytic cells play an important roles in the cross road of innate immunity and adaptive immunity [33], our findings may broaden our current understanding of the immunosuppressive properties of LT. It is well known that LT paralyzes the immune system primarily mediated through the inhibition of MEK/MAPK pathways, and in part mediated through the initiation of leukocyte cell death [5,17,34]. Intrigu- ingly, here we found that the phagocytic inhibitory doses of LT could be far lower than that of eliciting cell death of phagocytes (Fig. 1A, 2E vs. 1C, 1D, 3K), and that such inhibition did not significantly influence the MEK/MAPK pathways (Fig. 1B, 1E, vs. 1C, 1D, 3K). A reasonable speculation is that the immunosup- pressive effect of LT might be initiated at a relatively earlier infectious stage, in which LT is produced at low doses. Accordingly, here we proposed a hypothetical model (Fig. 6). During the early infections, B. anthracis expressed low levels of LT that are insufficient to suppress the MEK pathway of phagocytes (Fig. 4A, 4C), but critical for the inhibition of phagocytic clearance of bacterial pathogens (Fig. 5D; 6A–C). After B. anthracis is amplified to a greater population in the host (Fig. 6D), the LT gradually reached to an MEK/MAPK inhibitory dose (Fig. 6E) and finally led to the suppression of MEK/MAPK and cell death of macrophages (Fig. 6F). The molecular target of LT under such low doses, however, requires further investigation. Figure 4. Leukocyte MEK1 levels and plasma LT levels in LT treated mice. After mice were treated with various doses (50, 200 and 400 mg/kg) of LT intravenously for 3 hours, intact MEK1 levels of mouse splenic mononuclear cells (SMCs) and peripheral blood mononuclear cells (PBMCs) were analyzed using Western blotting (A). Discussion A and B: fluorescent beads only, C and D: macrophage cells only, E and F: beads + PA (85 ng/mL) treated cells, G and H: beads + LT (LF 150 pg/mL + PA 85 ng/mL) treated cells, I and J: beads + LT (LF 15 pg/mL + PA 85 ng/mL) treated cells. One representative diagram out of three experiments is shown (A–J). The R1 region in panels A, C, E, G and I reveals the population of the macrophage, which was specifically gated to obtain the graphs B, D, F, H, J, respectively. In panels B, D, F, H, J, the X-axis represents fluorescent intensity, which indicates the levels of macrophage-engaged beads, and the Y-axis represents the granularity of the cells (side scatter). December 2010 \| Volume 5 \| Issue 12 \| e14289 PLoS ONE \| www.plosone.org 5 The percentage of the cells with relatively high fluorescence intensity on the right panels of B, D, F, H, J, indicated the phagocytosed levels of beads by macrophages. In contrast, the percentage of cells on the left panels with relatively low fluorescence intensity indicates the number of cells without phagocytosed beads (B, D, F, H, J). The relative phagocytosis of macrophages was quantified by calculating the beads-phagocytosed (right panel: R)/ non-phagocytosed (left panel: L) ratio of macrophages (K), the level of LF untreated (PA only) groups were normalized to 100%. The F, H, J labels in the columns of Fig. 3K indicate the respective conditions as showed in Fig. 3F, 3H and 3J, respectively (K). Error bars indicate standard deviations of three experiments. * P,0.05, ** P,0.01 compared to the PA (LF untreated) control groups. n = 6 (3 experiments with 2 replicates). doi:10.1371/journal.pone.0014289.g003 Anthrax LT on Phagocytosis Anthrax LT on Phagocytosis Anthrax LT on Phagocytosis The percentage of the cells with relatively high fluorescence intensity on the right panels of B, D, F, H, J, indicated the phagocytosed levels of beads by macrophages. In contrast, the percentage of cells on the left panels with relatively low fluorescence intensity indicates the number of cells without phagocytosed beads (B, D, F, H, J). The relative phagocytosis of macrophages was quantified by calculating the beads-phagocytosed (right panel: R)/ non-phagocytosed (left panel: L) ratio of macrophages (K), the level of LF untreated (PA only) groups were normalized to 100%. The F, H, J labels in the columns of Fig. Discussion n = 6 (3 experiments with 2 replicates) (B, 2 hr groups; C). n = 10 (3 experiments with 3 or 4 replicates) (B, 24 hr groups). doi:10.1371/journal.pone.0014289.g004 In summary, here we demonstrated that LT suppresses macrophage phagocytosis through both MEK/MAPK dependent and independent pathways. Our results suggest that a two-step inhibition of macrophage phagocytosis by LT. The first stage involved a MEK-independent inhibition when the LT concentra- tion was lower than the MEK/MAPK suppressive dosage (Fig. 6B). In this stage, LT suppresses macrophage phagocytosis and December 2010 \| Volume 5 \| Issue 12 \| e14289 PLoS ONE \| www.plosone.org 6 Anthrax LT on Phagocytosis Figure 5. Treatment with sublethal doses of LT enhanced mortality and impaired bacterial clearance in mice with bacteremia. To test the sublethal dose of LT, mortality in C57BL/6J mice was measured after treatment with different doses of E. coli (A, n = 8) or LT (B, n = 8) (4 experiments with 2 replicates). The mortality of mice with E. coli bacteremia could be enhanced by sub-MAPK LT treatments (50 mg/kg), but not by equivalent quantities of control protein BSA or vehicle treatments (C, n = 8) (4 experiments with 2 replicates). The surviving E. coli in the plasma of the septic mice was enhanced by the treatment of sublethal doses of LT (50 mg/kg) (D, n = 6) (3 experiments with 2 replicates). The { marks shown in A and B indicates the dosages of E. coli and LT that used in C and D. * P,0.05, ** P,0.01. doi:10.1371/journal.pone.0014289.g005 Figure 5. Treatment with sublethal doses of LT enhanced mortality and impaired bacterial clearance in mice with bacteremia. To test the sublethal dose of LT, mortality in C57BL/6J mice was measured after treatment with different doses of E. coli (A, n = 8) or LT (B, n = 8) (4 experiments with 2 replicates). The mortality of mice with E. coli bacteremia could be enhanced by sub-MAPK LT treatments (50 mg/kg), but not by equivalent quantities of control protein BSA or vehicle treatments (C, n = 8) (4 experiments with 2 replicates). The surviving E. coli in the plasma of the septic mice was enhanced by the treatment of sublethal doses of LT (50 mg/kg) (D, n = 6) (3 experiments with 2 replicates). The { marks shown in A and B indicates the dosages of E. Ethics Statement The research methods applied regarding the experimental mice were approved by the Animal Care and Use Committee of Tzu Chi University (approval ID: 95017 and 98104). Discussion coli and LT that used in C and D. * P,0.05, ** P,0.01. doi:10.1371/journal.pone.0014289.g005 bacterial clearance in vivo to facilitate bacterial infection (Fig. 6C). Because anthrax can cause massive bacteremia [34], in the second stage, LT might progressively increase and eventually reach an MEK/MAPK suppressive dose, thereby leading to a p38 MAPK dependent inhibition (Fig. 6D-F). Our results suggest that LT is able to transmit a phagocytic suppressive signal through an MEK/ MAPK independent pathway, which could be beneficial for the survival of B. anthracis during early infectious stages. Figure 6. A hypothetical model. In the early infectious stage of B. anthracis (A), LT likely suppresses phagocytosis at a low level resulting in an increase in bacterial survival and decrease in host clearance without influencing the MEK/MAPK pathways (B–C). As the surviving bacteria accumulated to a high level (D), the LT level may gradually reach to a MAPK inhibitory dose (E) leading to further disruption of host homeostasis and immune responses (F). Q Activation processes. H Inhibition processes. doi:10.1371/journal.pone.0014289.g006 ELISA analysis for circulating LT in mouse plasma y g p A sandwich ELISA modified from previously described methods [27] was performed. To investigate the turnover of LT in the blood circulation, mice were treated with 50 mg/kg body weight LT (LF:PA = 15:85). Blood specimens were collected from the mice 2 hours and 24 hours after the treatment, and mixed with the anticoagulant acid citrate dextrose (5:1) (ACD: 20 mmol/L citric acid, 110 mol/L sodium citrate, and 5 mmol/L glucose; pH 7.3) to avoid blood clotting. Anti-PA and anti-LF antibodies were purified with serum from rabbits and mice immunized with high-performance liquid chromatography (HPLC)-purified PA or LF (6 cycles at 3-week intervals; 100 mg each cycle) as previously described [39]. Protein A column-purified polyclonal rabbit anti- PA or anti-LF antibodies (50 ml, 40 mg/ml) were incubated Cell viability analysis well dishes. The cell culture medium was then collected from each well. The TNF-a levels produced by macrophage cells was determined using an enzyme-linked immunosorbent assay (ELISA) kit from eBioscience, (San Diego, CA, USA) according to the manufacturer’s instructions. The mouse macrophage-like cell line J774A.1 (ATCC TIB67) was maintained in Dulbeco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS) used in cytotoxicity analysis [18,26]. After being changed with serum free medium, cells (16105/well) were treated with different concentrations (0.15 fg/mL to 15 ng/mL) of LF plus 85 ng/mL PA in a 96- well cell culture dish. Three hours after treatments, the level of viable cells was analyzed using the WST-1 kit (Roche Diagnostics, Taipei, Taiwan) according to the manufacturer’s instructions [26,42–43]. The assay principle was based on a reduction of the tetrazolium salt WST-1 to formazan by cellular dehydrogenases in viable cells. The generation of yellow formazan was measurable at 450 nm and was correlated to the number of cells. A standard curve plotted between WST-1 values with serious diluted J774A.1 cells (0 to 56105) was obtained first, and the relative number of viable cells was then calculated, accordingly. Phagocytosis analysis by gentamicin protection assay Phagocytosis analysis by gentamicin protection assay Gentamicin protection assay were performed as described in previous literatures [24–25,44–45]. Live bacteria E. coli, B. cereus and B. subtilis (16108 CFU/mL in serum free cell culture medium; 1 mL/reaction) were incubated with J774A.1 cells (16108/well, with or without one hour LT pretreatment) in an antibiotic free condition. After 90-minutes of incubation, a cell membrane impermeable antibiotic gentamicin (10 mg/mL) was added to the medium for 30 minutes to eliminate the extracellular bacteria. Macrophage cells containing engulfed bacteria were washed with 16phosphate-buffered saline (PBS), and lysed in 1% Triton-X 100, 16PBS solution. The cell lysate was then plated on LB agar plates to determine the number of living bacteria (CFU) engulfed by macrophages. Cell death analysis by flow cytometry y y y y Following the previously described method [50], the dead cell subpopulations was determined by staining the cells with allophycocyanin (APC) conjugated annexin V (annexin V-APC) and propidium iodide (PI). The macrophage J774A.1 cells were treated with PA or various doses of LT for 3 hours. After the treatments, the total number of cells (56105 per sample) was washed twice with PBS for staining. The cells were stained by annexin V-APC (0.5 mg/ml) and PI (2 mg/ml) in the staining buffer (10 mM HEPES [4-(2-hydroxyethyl)-1-piperazineethane- sulfonic acid], pH 7.4, 150 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2) at room temperature for 20 min in the dark. After labeling, the cells were diluted in 10 mM HEPES buffer to a final volume of 600 ml, and analyzed by flow cytometry. Analyses of MEK/MAPK pathways For in vivo analysis, C57BL/6J mice were intravenously treated with various dosages of LT (50, 200, 400 mg/kg body weight, LF:PA = 15:85). Three hours after treatment, the splenic and peripheral mononuclear cells were isolated and purified using a Ficoll-Paque PLUS kit (GE healthcare, Singapore, Singapore) according to the manufacturers’ instructions. The cell lysate was then subjected to MEK1 analysis, as indicated in the following methods. For in vitro analysis, J774A.1 cells (16106/well) were cultured on six-well dishes treated with various concentrations of LT for three hours with or without lipopolysaccharide (LPS, 0.1 mg/mL, Sigma-Aldrich, St. Louis, MO, USA) [39]. PA treatments (100 ng/mL) were used as negative controls. Cell lysate was separated by SDS-PAGE and then transferred to a nitrocellulose paper, whereupon Western blotting analysis was performed, as previously described [39]. The levels of MEK1 were measured by specific antibody (anti-NH2-terminal MEK1, Upstate Biotechnology, Lake Placid, NY, USA). The cellular MAPKs were measured by probing the antibodies against MAPK pathway kinases (p42/44, phospho-p42/44 [pp42/44], p38, and phospho- Jun kinase (JNK), Cell Signaling Technology, Danvers, MA; anti- pp38 Promega, Madison WI, USA). Jun kinase activity was measured by phosphorylation of c-Jun using a kit (Cell Signaling Technology, Danvers, MA, USA). The intensity of the blots was measured using Image J (version 1.32; National Institutes of Health, USA) software. Phagocytosis analysis by flow cytometry Phagocytosis analysis using fluorescent beads by flow cytometry was modified from previously described methods [16,46–47]. Silica beads (mean diameter 1.1 mm, Bangs Laboratories, Fishers, IN, USA) [48] were opsonized with fluorescein-labeled rat anti- mouse immunoglobulin (Ig, 1 mg/mL, Jackson ImmunoResearch Laboratories) for 30 min at room temperature (25uC) in 16PBS. After blocking with 5% bovine serum albumin (BSA) in 16PBS for one hour, the beads (16107) were mixed with J774A.1 cells (16106, with or without one hour LT pretreatment) in 200 mL cell-culture medium. After two washes (PBS-albumin, 300 g, 10 min, 25uC), and fixation (2% paraformaldehyde in PBS, 30 min) the levels of phagocytosis were revealed by the florescent intensity of cell-engulfed beads, and could be identified using fluorescence flow cytometry (FACScalibur, BD Biosciences). Beads were distinguished from the J774A.1 cells according to their forward and side scatter characteristics (Fig. 3A vs. 3C; R1 region: J774A.1 cells). We then analyzed the fluorescent signals (beads) from the R1 regions (J774A.1 cells) to identify the relative level of phagocytosis (Fig. 3B, 3D, 3F, 3H, 3J, right panels). The J774A.1 cells were gated and 10,000 events acquired for each sample. The background fluorescence (,5%) was used to set the [left (L)/right (R)] border lines in Fig. 3B, 3D, 3F, 3H and 3J by beads-untreated J774A.1 cells. The level of phagocytosis/non-phagocytosis ratio in the beads was normalized to 100% by PA treated (LF untreated) J774A.1 cells. To avoid inefficient binding to cellular receptors [49], all LT treatments were supplied at the same doses levels as PA (85 ng/mL). Mice, bacteria and toxin The C57BL/6J mice (males, 8-11 weeks of age) were purchased from National Laboratory Animal Center (NLAC, Taipei, Taiwan). Mice were housed in the Laboratory Animal Center of Tzu Chi University (Hualien, Taiwan). B. anthracis (ATCC 14186), which contains both pXO1 and pXO2 plasmids that express functional lethal toxin (LT) and edema toxin (ET), was grown and maintained as previously described [18,39]. Native B. anthracis lethal toxins, PA and LF, were purified from the culture supernatants of B. anthracis (ATCC 14186). The bacteria E. coli (OP50, a laboratory strain [40–41]), B. cereus (ATCC 7004) and B. subtilis (ATCC 21336) were used in the phagocytosis experiments. Figure 6. A hypothetical model. In the early infectious stage of B. anthracis (A), LT likely suppresses phagocytosis at a low level resulting in an increase in bacterial survival and decrease in host clearance without influencing the MEK/MAPK pathways (B–C). As the surviving bacteria accumulated to a high level (D), the LT level may gradually reach to a MAPK inhibitory dose (E) leading to further disruption of host homeostasis and immune responses (F). Q Activation processes. H Inhibition processes. doi:10.1371/journal.pone.0014289.g006 December 2010 \| Volume 5 \| Issue 12 \| e14289 PLoS ONE \| www.plosone.org 7 Anthrax LT on Phagocytosis References 18. Kau JH, Lin CG, Huang HH, Hsu HL, Chen KC, et al. (2002) Calyculin A Sensitive Protein Phosphatase Is Required for Bacillus anthracis Lethal Toxin Induced Cytotoxicity. Curr Microbiol 44: 106–111. 1. Dixon TC, Meselson M, Guillemin J, Hanna PC (1999) Anthrax. N Engl J Med 341: 815–826. 1. Dixon TC, Meselson M, Guillemin J, Hanna PC (1999) Anthrax. N Engl J Med 341: 815–826. 2. Collier RJ, Young JA (2003) Anthrax toxin. Annu Rev Cell Dev Biol 19: 45– 70. y y 19. Tonello F, Montecucco C (2009) The anthrax lethal factor and its MAPK kinase-specific metalloprotease activity. Mol Aspects Med 30: 431–438. 3. Moayeri M, Leppla SH (2004) The roles of anthrax toxin in pathogenesis. Curr Opin Microbiol 7: 19–24. 20. Scherle PA, Jones EA, Favata MF, Daulerio AJ, Covington MB, et al. (1998) Inhibition of MAP kinase kinase prevents cytokine and prostaglandin E2 production in lipopolysaccharide-stimulated monocytes. J Immunol 161: 5681–5686. 4. Abrami L, Reig N, van der Goot FG (2005) Anthrax toxin: the long and winding road that leads to the kill. Trends Microbiol 13: 72–78. 5. Fukao T (2004) Immune system paralysis by anthrax lethal toxin: the roles of innate and adaptive immunity. Lancet Infect Dis 4: 166–170. 5. Fukao T (2004) Immune system paralysis by anthrax lethal t 21. Chen G, Li J, Ochani M, Rendon-Mitchell B, Qiang X, et al. (2004) Bacterial endotoxin stimulates macrophages to release HMGB1 partly through CD14- and TNF-dependent mechanisms. J Leukoc Biol 76: 994–1001. innate and adaptive immunity. Lancet Infect Dis 4: 166–170. 6. Pezard C, Berche P, Mock M (1991) Contribution of individual toxin components to virulence of Bacillus anthracis. Infect Immun 59: 3472–3477. 22. Mandell GL (1973) Interaction of intraleukocytic bacteria and antibiotics. J Clin Invest 52: 1673–1679. 7. Pezard C, Duflot E, Mock M (1993) Construction of Bacillus anthracis mutant strains producing a single toxin component. J Gen Microbiol 139(Pt 10): 2459–2463. 23. Vaudaux P, Waldvogel FA (1979) Gentamicin antibacterial activity in the presence of human polymorphonuclear leukocytes. Antimicrob Agents Che- mother 16: 743–749. 8. Duesbery NS, Webb CP, Leppla SH, Gordon VM, Klimpel KR, et al. (1998) Proteolytic inactivation of MAP-kinase-kinase by anthrax lethal factor. Science 280: 734–737. 24. Isberg RR, Voorhis DL, Falkow S (1987) Identification of invasin: a protein that allows enteric bacteria to penetrate cultured mammalian cells. Cell 50: 769–778. 9. References Vitale G, Pellizzari R, Recchi C, Napolitani G, Mock M, et al. (1998) Anthrax lethal factor cleaves the N-terminus of MAPKKs and induces tyrosine/threonine phosphorylation of MAPKs in cultured macrophages. Biochem Biophys Res Commun 248: 706–711. 25. Chang HH, Shih KN, Lo SJ (2000) Receptor-mediated endocytosis as a selection force to enrich bacteria expressing rhodostomin on their surface. J Biomed Sci 7: 42–50. 26. Kau JH, Sun DS, Huang HH, Wong MS, Lin HC, et al. (2009) Role of visible light-activated photocatalyst on the reduction of anthrax spore-induced mortality in mice. PLoS ONE 4: e4167. 10. Turk BE (2007) Manipulation of host signalling pathways by anthrax toxins. Biochem J 402: 405–417. 11. Chang L, Karin M (2001) Mammalian MAP kinase signalling cascades. Nature 410: 37–40. 27. Mabry R, Brasky K, Geiger R, Carrion R, Jr., Hubbard GB, et al. (2006) Detection of anthrax toxin in the serum of animals infected with Bacillus anthracis by using engineered immunoassays. Clin Vaccine Immunol 13: 671–677. 12. Crawford MA, Aylott CV, Bourdeau RW, Bokoch GM (2006) Bacillus anthracis toxins inhibit human neutrophil NADPH oxidase activity. J Immunol 176: 7557–7565. 13. Park JM, Greten FR, Li ZW, Karin M (2002) Macrophage apoptosis by anthrax lethal factor through p38 MAP kinase inhibition. Science 297: 2048–2051. 28. Groves E, Dart AE, Covarelli V, Caron E (2008) Molecular mechanisms of phagocytic uptake in mammalian cells. Cell Mol Life Sci 65: 1957–1976. 29. Coombes BK, Valdez Y, Finlay BB (2004) Evasive maneuvers by secreted bacterial proteins to avoid innate immune responses. Curr Biol 14: R856–867. 14. During RL, Li W, Hao B, Koenig JM, Stephens DS, et al. (2005) Anthrax lethal toxin paralyzes neutrophil actin-based motility. J Infect Dis 192: 837–845. xin paralyzes neutrophil actin-based motility. J Infect Dis 192: 837–84 30. Pullikuth AK, Catling AD (2007) Scaffold mediated regulation of MAPK signaling and cytoskeletal dynamics: a perspective. Cell Signal 19: 1621–1632. 15. Drysdale M, Heninger S, Hutt J, Chen Y, Lyons CR, et al. (2005) Capsule synthesis by Bacillus anthracis is required for dissemination in murine inhalation anthrax. Embo J 24: 221–227. 31. During RL, Gibson BG, Li W, Bishai EA, Sidhu GS, et al. (2007) Anthrax lethal toxin paralyzes actin-based motility by blocking Hsp27 phosphorylation. Embo J 26: 2240–2250. J 16. Buck M, Chojkier M (2007) C/EBPbeta phosphorylation rescues macrophage dysfunction and apoptosis induced by anthrax lethal toxin. Am J Physiol Cell Physiol 293: C1788–1796. 32. Statistical analysis All results were calculated from data of no fewer than three independent experiments. Statistical differences between groups were calculated using Student t test and presented as mean 6 standard deviation (SD). A P value of less than 0.05 (P,0.05) was considered statistically significant. The statistical tests were carried out and output to graphs using Microsoft Excel (Microsoft Taiwan, Taipei, Taiwan) and SigmaPlot (Systat Software, Point Richmond, CA, USA) software. Analysis of TNF-a production The plate was then washed again, followed by the addition of goat anti-mouse immunoglobulin G (IgG) (diluted in 5% BSA-PBS). After one hour of incubation and two washes (PBS), donkey anti-goat horseradish peroxidase (HRP) conjugate (minimal cross-reactivity to mouse and rabbit IgGs, Jackson ImmunoResearch Laboratories, West Grove, PA, USA) was diluted in 5% BSA-PBS and incubated for additional one hour. ELISA reactions were developed with the substrate o-phenylene- diamine (OPD) (Sigma). Experiments were run in triplicate, averaging each data point with standard deviation. received sequential intravenous treatments of E. coli (26109 to 56109 CFU/kg) and LT (50 mg/kg to 400 mg/kg). The mortality of the mice was recorded, and blood specimens were collected from the mice subjected to combined treatments of E. coli (26109 CFU/kg) and LT (50 mg/kg) at 2, 16, and 24 hours. Because mortality occurred within 24 hours (Fig. 5C, day1 groups), the data of the LT-treated 24 hour groups (Fig. 5D, LT + E. coli, 24 hr groups), was obtained from the survivors. Blood samples were mixed with anticoagulant ACD (2:1), and the blood/ ACD mixture (15 ml) was applied directly to LB agar plates. The number of cells formed in the colony after an overnight culture was recorded. Septic shock mouse model Conceived and designed the experiments: JHK DSS HHC. Performed the experiments: JHK HSH TSL HHC. Analyzed the data: JHK DSS HHC. Contributed reagents/materials/analysis tools: HHH HCL. Wrote the paper: HHC. Conceived and designed the experiments: JHK DSS HHC. Performed the experiments: JHK HSH TSL HHC. Analyzed the data: JHK DSS HHC. Contributed reagents/materials/analysis tools: HHH HCL. Wrote the paper: HHC. Following the previously described method [51], fresh E. coli culture was collected when the OD reached 0.4–0.5. Using a predetermined standard curve [51], we adjusted the E. coli suspension to 26107 CFU/mL in normal saline. C57BL/6J mice Analysis of TNF-a production Macrophage J774A.1 cells (26105) were treated with 0.1 mg/ mL LPS and various concentrations of LT for three hours, in 96- PLoS ONE \| www.plosone.org December 2010 \| Volume 5 \| Issue 12 \| e14289 8 Anthrax LT on Phagocytosis overnight in a 96-well ELISA plate (Corning Inc., Corning, NY, U.S.A.) at 4uC. The plates were then blocked with 5% bovine serum albumin (BSA; Sigma)/PBS for 2 h at room temperature. The plates were washed three times with 0.5% Tween 20-16PBS and once with 16PBS. To establish the standard cures, quantified (HPLC-purified B. anthracis-derived) native PA and LF were diluted with mouse serum at 1 mg/ml and serially diluted onto the ELISA plate. To detect PA and LF in blood of infected animals, serum was added to the plate initially diluted 1:1 in 5% BSA-PBS, and serially diluted across the plate. After one-hr incubation, the plates were washed as described above. Protein A column-purified mouse anti-PA or anti-LF polyclonal antibodies were diluted 1:1000 in 5% BSA-PBS and added to the plate for one-hr of incubation. The plate was then washed again, followed by the addition of goat anti-mouse immunoglobulin G (IgG) (diluted in 5% BSA-PBS). After one hour of incubation and two washes (PBS), donkey anti-goat horseradish peroxidase (HRP) conjugate (minimal cross-reactivity to mouse and rabbit IgGs, Jackson ImmunoResearch Laboratories, West Grove, PA, USA) was diluted in 5% BSA-PBS and incubated for additional one hour. ELISA reactions were developed with the substrate o-phenylene- diamine (OPD) (Sigma). Experiments were run in triplicate, averaging each data point with standard deviation. overnight in a 96-well ELISA plate (Corning Inc., Corning, NY, U.S.A.) at 4uC. The plates were then blocked with 5% bovine serum albumin (BSA; Sigma)/PBS for 2 h at room temperature. The plates were washed three times with 0.5% Tween 20-16PBS and once with 16PBS. To establish the standard cures, quantified (HPLC-purified B. anthracis-derived) native PA and LF were diluted with mouse serum at 1 mg/ml and serially diluted onto the ELISA plate. To detect PA and LF in blood of infected animals, serum was added to the plate initially diluted 1:1 in 5% BSA-PBS, and serially diluted across the plate. After one-hr incubation, the plates were washed as described above. Protein A column-purified mouse anti-PA or anti-LF polyclonal antibodies were diluted 1:1000 in 5% BSA-PBS and added to the plate for one-hr of incubation. Anthrax LT on Phagocytosis 33. Silva MT (2010) When two is better than one: macrophages and neutrophils work in concert in innate immunity as complementary and cooperative partners of a myeloid phagocyte system. J Leukoc Biol 87: 93–106. 43. Chang HH, Shyu HF, Wang YM, Sun DS, Shyu RH, et al. (2002) Facilitation of cell adhesion by immobilized dengue viral nonstructural protein 1 (NS1): arginine-glycine-aspartic acid structural mimicry within the dengue viral NS1 antigen. J Infect Dis 186: 743–751. y p g y y J 34. Baldari CT, Tonello F, Paccani SR, Montecucco C (2006) Anthrax toxins: A paradigm of bacterial immune suppression. Trends Immunol 27: 434–440. g 44. Ortega-Barria E, Pereira ME (1991) A novel T. cruzi heparin-binding protein promotes fibroblast adhesion and penetration of engineered bacteria and trypanosomes into mammalian cells. Cell 67: 411–421. 35. Boyden ED, Dietrich WF (2006) Nalp1b controls mouse macrophage susceptibility to anthrax lethal toxin. Nat Genet 38: 240–244. y , ( ) p susceptibility to anthrax lethal toxin. Nat Genet 38: 240–244. yp 45. Chang HH, Lo SJ (2000) Modification with a phosphoryla 36. Tschopp J, Martinon F, Burns K (2003) NALPs: a novel protein family involved in inflammation. Nat Rev Mol Cell Biol 4: 95–104. 45. Chang HH, Lo SJ (2000) Modification with a phosphorylation tag of PKA in the TraT-based display vector of Escherichia coli. J Biotechnol 78: 115–122. TraT-based display vector of Escherichia coli. J Biotechnol 78 37. Ha SD, Ng D, Lamothe J, Valvano MA, Han J, et al. (2007) Mitochondrial proteins Bnip3 and Bnip3L are involved in anthrax lethal toxin-induced macrophage cell death. J Biol Chem 282: 26275–26283. 46. Liao G, Simone J, Simon SR (1994) Paracrine downregulation of Fc gamma RIII in human monocyte-derived macrophages induced by phagocytosis of nonopsonized particles. Blood 83: 2294–2304. 38. Alileche A, Serfass ER, Muehlbauer SM, Porcelli SA, Brojatsch J (2005) Anthrax lethal toxin-mediated killing of human and murine dendritic cells impairs the adaptive immune response. PLoS Pathog 1: e19. 47. Huang HS, Sun DS, Lien TS, Chang HH (2010) Dendritic cells modulate platelet activity in IVIg-mediated amelioration of ITP in mice. Blood: In press. 48. Sun DS, King CC, Huang HS, Shih YL, Lee CC, et al. (2007) Antiplatelet autoantibodies elicited by dengue virus non-structural protein 1 cause thrombocytopenia and mortality in mice. J Thromb Haemost 5: 2291–2299. 39. Kau JH, Sun DS, Tsai WJ, Shyu HF, Huang HH, et al. References Cui X, Li Y, Li X, Haley M, Moayeri M, et al. (2006) Sublethal doses of Bacillus anthracis lethal toxin inhibit inflammation with lipopolysaccharide and Escherichia coli challenge but have opposite effects on survival. J Infect Dis 193: 829–840. 17. Tournier JN, Rossi Paccani S, Quesnel-Hellmann A, Baldari CT (2009) Anthrax toxins: a weapon to systematically dismantle the host immune defenses. Mol Aspects Med 30: 456–466. PLoS ONE \| www.plosone.org December 2010 \| Volume 5 \| Issue 12 \| e14289 December 2010 \| Volume 5 \| Issue 12 \| e14289 9 Anthrax LT on Phagocytosis Anthrax LT on Phagocytosis (2005) Antiplatelet activities of anthrax lethal toxin are associated with suppressed p42/44 and p38 mitogen-activated protein kinase pathways in the platelets. J Infect Dis 192: 1465–1474. 49. Wigelsworth DJ, Krantz BA, Christensen KA, Lacy DB, Juris SJ, et al. (2004) Binding stoichiometry and kinetics of the interaction of a human anthrax toxin receptor, CMG2, with protective antigen. J Biol Chem 279: 23349–23356. 40. Cheng CL, Sun DS, Chu WC, Tseng YH, Ho HC, et al. (2009) The effects of the bacterial interaction with visible-light responsive titania photocatalyst on the bactericidal performance. J Biomed Sci 16: 7. 50. Vermes I, Haanen C, Steffens-Nakken H, Reutelingsperger C (1995) A novel assay for apoptosis. Flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein labelled Annexin V. J Immunol Methods 184: 39–51. p J 41. Wong MS, Sun DS, Chang HH (2010) Bactericidal performance of visible-light responsive titania photocatalyst with silver nanostructures. PLoS One 5: e10394. 51. Wong MS, Chu WC, Sun DS, Huang HS, Chen JH, et al. (2006) Visible-light- induced bactericidal activity of a nitrogen-doped titanium photocatalyst against human pathogens. Appl Environ Microbiol 72: 6111–6116. p p y 42. Chang HH, Kau JH, Lo SJ, Sun DS (2003) Cell-adhesion and morphological changes are not sufficient to support anchorage-dependent cell growth via non- integrin-mediated attachment. Cell Biol Int 27: 123–133. PLoS ONE \| www.plosone.org December 2010 \| Volume 5 \| Issue 12 \| e14289 10
https://openalex.org/W3008189172	http://plymsea.ac.uk/id/eprint/8893/1/Dixonetalmicroorganisms%20Published.pdf	Latin	null	Seasonal Changes in Microbial Dissolved Organic Sulfur Transformations in Coastal Waters	Microorganisms	2,020	cc-by	11,857	Article Seasonal Changes in Microbial Dissolved Organic Sulfur Transformations in Coastal Waters Joanna L Dixon 1,, Frances E Hopkins 1, John A Stephens 1 and Hendrik Schäfer 2 1 Plymouth Marine Laboratory, Prospect Place, West Hoe, Plymouth, Devon PL1 3DH, UK; fhop@pml.ac.uk (F.E.H.), JAS@pml.ac.uk (J.A.S.) 2 School of Life Sciences, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, UK; H.Schaefer@warwick.ac.uk Correspondence: jod@pml.ac.uk; Tel.: +44‐(0)1752‐633100 R i d 13 D b 2019 A t d 24 F b 2020 P bli h d 27 F b 2020 Abstract: The marine trace gas dimethylsulfide (DMS) is the single most important biogenic source of atmospheric sulfur, accounting for up to 80% of global biogenic sulfur emissions. Approximately 300 million tons of DMS are produced annually, but the majority is degraded by microbes in seawater. The DMS precursor dimethylsulfoniopropionate (DMSP) and oxidation product dimethylsulphoxide (DMSO) are also important organic sulfur reservoirs. However, the marine sinks of dissolved DMSO remain unknown. We used a novel combination of stable and radiotracers to determine seasonal changes in multiple dissolved organic sulfur transformation rates to ascertain whether microbial uptake of dissolved DMSO was a significant loss pathway. Surface concentrations of DMS ranged from 0.5 to 17.0 nM with biological consumption rates between 2.4 and 40.8 nM∙d−1. DMS produced from the reduction of DMSO was not a significant process. Surface concentrations of total DMSO ranged from 2.3 to 102 nM with biological consumption of dissolved DMSO between 2.9 and 111 nM∙d−1. Comparisons between 14C2‐DMSO assimilation and dissimilation rates suggest that the majority of dissolved DMSO was respired (>94%). Radiotracer microbial consumption rates suggest that dissimilation of dissolved DMSO to CO2 can be a significant loss pathway in coastal waters, illustrating the significance of bacteria in controlling organic sulfur seawater concentrations. Keywords: dimethylsulfide; dimethylsulfoxide; bacteria; dissimilation to CO2; radiotracers; stable tracers; coastal variability Microorganisms 2020, 8, 337; doi:10.3390/microorganisms8030337 Article Seasonal Changes in Microbial Dissolved Organic Sulfur Transformations in Coastal Waters Joanna L Dixon 1,, Frances E Hopkins 1, John A Stephens 1 and Hendrik Schäfer 2 1 Plymouth Marine Laboratory, Prospect Place, West Hoe, Plymouth, Devon PL1 3DH, UK; fhop@pml.ac.uk (F.E.H.), JAS@pml.ac.uk (J.A.S.) 2 School of Life Sciences, University of Warwick, Gibbet Hill Road, Coventry CV4 7AL, UK; H.Schaefer@warwick.ac.uk Correspondence: jod@pml.ac.uk; Tel.: +44‐(0)1752‐633100 Received: 13 December 2019; Accepted: 24 February 2020; Published: 27 February 2020 1. Introduction The marine trace gas dimethylsulfide (DMS) is the single most important biogenic source of atmospheric sulfur [1]. It accounts for up to 80% of global biogenic sulfur emissions, and plays a key role in transporting sulfur to the terrestrial environment [2,3]. Approximately 300 million tons of DMS are produced annually in the marine environment [4]. However, only around 16% is transferred into the atmosphere [5], because the majority (~84%) is degraded by microbes in seawater [5–7]. The emission of DMS provides important precursors for the formation of secondary organic aerosols, and thus plays a vital role in atmospheric chemistry and climate processes [8,9]. In seawater, DMS along with its precursors (particulate and dissolved dimethylsulfoniopropionate; DMSPp and DMSPd, respectively) provides important sources of carbon and sulfur for marine micro‐organisms [10–13]. Microbial oxidation of DMS to dimethylsulfoxide (DMSO) in the mixed surface layer of the ocean is often the major sink for DMS [13,14]. However, the dominant processes affecting DMSO concentrations in marine waters remain largely unquantified, but microbes are likely to be key players determining not only marine DMS (and thus DMS flux to the atmosphere) and DMSPd, but also DMSOd concentrations. www.mdpi.com/journal/microorganisms www.mdpi.com/journal/microorganisms Microorganisms 2020, 8, 337; doi:10.3390/microorganisms8030337 Microorganisms 2020, 8, 337 2 of 20 Microorganisms 2020, 8, 337 The dominant source of DMS is thought to be via the microbial (bacteria and/or phytoplankton) enzymatic cleavage of algal‐derived DMSP [15–17], although photochemically‐derived production mechanisms [18] and DMSO reduction [15,19,20] could also contribute. The majority of DMSP in marine waters is particulate bound in intracellular pools (DMSPp), which gets released into the dissolved phase through algal cell lysis caused by grazing, viral attack or autolysis, or exudation [21,22]. Rapid utilization and turnover of DMSPd by bacteria, and some phytoplankton containing extracellular DMSP‐lyases, typically maintains relatively low nano‐molar concentrations [12]. y yp y y Dimethylsulfoxide is also an important ubiquitous reservoir of organic sulfur in the ocean, where the total pool of DMSO is often greater than that of DMS [23], and equal to or greater than the total pool of DMSP [20,24]. Conventionally, the main sources of DMSOd are attributed to photochemical and microbial oxidation of DMS [14,25]. The microbial DMS oxidation process is thought to occur via a methylamine‐dependent co‐oxidation pathway, with bacteria like the marine Roseobacter clade using an enzyme called trimethylamine monooxygenase [26,27]. 1. Introduction However, direct biosynthesis within cells (DMSOp) coupled with transformation to the dissolved phase via a number of pathways, including permeative diffusion, cell lysis, and as a byproduct of cell activity, have also been suggested [25,28]. The function of DMSO in cells is still under debate, but hypotheses revolve around cryoprotection, osmotic pressure regulation, modification of intracellular electrolytes, and oxidative stress defense [25,29]. The marine sinks of DMSOd remain essentially uncharacterised, although biological DMSOd consumption in seawater was previously observed [30,31]. Possible DMSOd loss pathways include bacterial consumption, reduction to DMS, oxidation to dimethylsulphone, and export via sinking particles [32]. Several cultured phytoplankton species have been shown to reduce DMSOd [33]. DMSO reductases are widespread in bacteria [34], and a variety of aerobic and anaerobic bacteria have been shown to reduce DMSOd to DMS during anaerobic respiration [19,35]. Growth on DMSO as a carbon source has also been reported for isolates of Hyphomicrobium [36], Arthrobacter [36,37], and Methylophaga [38]. However, our understanding of the production and consumption pathways of DMSO in the surface oceans and their controls are poorly understood [24]. Our objective was to employ a novel combination of stable and radiotracers in order to simultaneously determine seasonal changes in multiple dissolved organic matter sulfur transformation rates, and to ascertain whether microbial carbon DMSOd uptake and dissimilation to CO2 were significant DMSOd loss pathways. Our results suggest that dissimilation of dissolved DMSO to CO2 can be a significant loss pathway (for DMSOd) in coastal waters. 2.1. Organic Sulfur Concentrations Concentrations of DMS, DMSP, and DMSO in seawater were analysed sequentially using a purge and cryotrapping system coupled with sulfur specific gas chromatography using a Varian 3800 gas chromatograph with a pulsed flame photometric detector (GC‐PFPD) using methodology described by Simó et al. [46], Simó and Vila‐Costa [47], and Archer et al. [48], as modified by Vila‐Costa et al. [12] for DMSO analysis. Briefly, for DMS, 5 mL samples were gently filtered through a 25 mm GF/F (glass fiber) filter directly into a purge tower, avoiding any contact with air, and immediately analysed via GC‐PFPD (purged for 5 min at 60 mL min−1 and cryogenically trapped in a PTFE sample loop submerged in liquid nitrogen before desorption using boiling water to GC). For dissolved DMSP (DMSPd), the purged DMS sample was transferred to a glass vial with 10 M NaOH and hydrolysed for 6–24 h (to convert DMSPd to DMS). Following hydrolysis, samples were analysed as above. For dissolved DMSO (DMSOd), ~10 mg cobalt‐doped NaBH4 was added to the purged DMSPd sample (which reduces DMSOd to DMS) and purged for a further 10 min, with DMS analysis as previous. For total DMSP (DMSPt), which includes particulate DMSP (DMSPp) and a minor fraction from dissolved DMSP (DMSPd), 7 mL of whole seawater was pipetted into a glass vial with 1 mL 10 M NaOH, and left for 12–24 h for hydrolysis to convert DMSP to DMS. Then, 1 mL was carefully pipetted to a glass purge tower for extraction of DMS as above. For DMSPp, 7 mL of whole seawater was gravity filtered through a 25 mm GF/F filter. The filter was placed in a glass vial with 7 mL MQ and 1 mL 10 M NaOH, and left for 12–24 h for hydrolysis. Then, 1 mL was pipetted to the purge tower and analysed as previously. For particulate DMSO (DMSOp), ~10 mg cobalt‐doped NaBH4 was added to the purged DMSPp sample to reduce DMSOp to DMS, and subsequently analysed by GC as above. The detection limit of the system was approximately 2.9 pmol S. The standard deviation of at least duplicate experimental samples was on average 6%, 10%, 9%, 9%, and 7% of the mean for DMS, DMSPt, DMSPp, DMSOp, and DMSOd, respectively. DMS standards for calibration were prepared from DMSP (>98% purity; Dr Sinan Battah, University of Essex, Colchester, UK) in a 10 M NaOH solution in Milli‐Q water. 2. Materials and Methods Surface samples (≤10 m) were collected from the Western Channel Observatory long term monitoring station L4, situated ~13 km south west of Plymouth (50.3 N, 04.22 W, water depth ~55 m). Water samples were collected by the RV Plymouth Quest using 10 L Niskin bottles mounted on a rosette sampler, which also housed a Seabird 19 + CTD. Sub‐samples were decanted into acid‐washed brown glass bottles, sealed with ground glass stoppers with no headspace, and at in situ temperature for the ~2 h transit back to the laboratory. Temperature variability during transit was +1 °C, which is within the in situ diurnal variability at station L4. Seawater temperature was determined from the Seabird CTD, which has an accuracy of ±0.001 °C [39]. The concentration of chlorophyll a, nutrients, bacteria, and phytoplankton community composition were determined weekly at station L4 as part of the western English Channel Observatory (https://www.westernchannelobservatory.org.uk/). Chlorophyll a concentrations were determined through fluorometric analysis of acetone extracted pigments [40]. Nutrient analysis was conducted using recognized analytical techniques for nitrate [41,42] and phosphate [43]. Numbers of bacterial cells were determined by flow cytometry (Accuri C6 instrument) using SYBR Green I DNA‐stained cells to determine high nucleic acid (HNA) and low nucleic acid (LNA) containing cells from 1.8 mL seawater samples fixed in paraformaldehyde (5% final concentration). Synechococcus sp. numbers were determined by flow cytometry (Accuri C6 instrument) on unstained 3 of 20 Microorganisms 2020, 8, 337 samples based on their light scattering and autofluorescence properties [44]. Phytoplankton enumeration and composition were conducted using established microscopy [45]. samples based on their light scattering and autofluorescence properties [44]. Phytoplankton enumeration and composition were conducted using established microscopy [45]. 2.2. Stable Isotope Tracer Rate Experiments Isotope tracer incubation experiments were also conducted using surface waters collected from station L4 between May and October 2014. On each date, approximately 350 mL seawater was siphoned directly into four acid washed and rinsed Tedlar (1L, Supelco, from Merck, Gillingham, Dorset, UK)bags without exposure to ambient air. We simultaneously added DMS (d3‐DMS, 99 atom % d, Merck, Gillingham, Dorset, UK), DMSPd (d6‐DMSP 99 atom %,, Australian Government, National Measurement Institute, Sydney, Australia) and DMSOd (13C2‐DMSO 99 atom % 13C, Merck, Gillingham, Dorset, UK) at ~10% of in situ concentrations into triplicate experimental bags. These and a control experimental bag (without any stable tracer additions) were incubated for 3–4 h in the dark at in situ temperature. During experiments, sub‐samples were collected over 3–4 time points from each bag using a 50 mL glass syringe via the inlet on the Tedlar bag. Approximately 30 mL was withdrawn (ensuring no headspace or bubbles) at each time point and immediately gently filtered through a Millipore filtration unit containing a 25 mm GF/F filter directly into a 20 mL glass receiving syringe. This was immediately injected into a purge tower and analysed by PTR‐MS as above. This technique allows the simultaneous quantification of DMS derived from DMSP cleavage and DMSO reduction, gross DMS loss (we assume this equates to biological consumption because photochemical reactions and sea to air flux were eliminated in our closed dark experimental bags), and net change in DMS concentrations (gross production–biological consumption). DMS production from DMS cleavage was determined as the rate of accumulation of d6‐DMS from d6‐DMSP, while DMS production from DMSO reduction was measured as the rate of accumulation of 13C2‐DMS from 13C2‐DMSO. Biological consumption of DMS was calculated from the rate of decrease in d3‐DMS. Net change in DMS results from the rate of change of DMS, and thus gross DMS production, is calculated as net change in DMS plus biological consumption. On two of the sampling dates (21 July and 26 August 2014), additional samples were also taken during the time course incubations for the determination of DMSOd derived from the microbial oxidation of DMS (rate of appearance of d3‐DMSO from d3‐DMS), the biological consumption of DMSOd (from loss of 13C2‐DMSO corrected for any conversion to 13C2‐DMS), and the biological consumption of DMSPd (from loss of d6‐DMSPd corrected for any conversion to d6‐DMS and d6‐DMSPd). 2.1. Organic Sulfur Concentrations Typically, 4–5‐point calibration curves were carried out twice per month during the sampling period, with an r2 for the resulting linear regression of ng sulfur versus square root of the peak area of typically ≥0.996. We report DMS, DMSPt, DMSPp, DMSOp, and DMSOd data. We additionally determined the change in concentration of DMS from Tedlar bag incubation experiments (see stable isotope tracer rate experiments below) by withdrawing ~30 mL and immediately gently filtering through a Millipore filtration unit containing a 25 mm GF/F filter directly into a 20 mL glass receiving syringe (ensuring no headspace, bubbles, or exposure to the atmosphere). This was immediately injected into a purge tower, and purged with high purity nitrogen at a flow rate of ~100 mL min‐1 for 15 min directly into the proton transfer reaction mass spectrometer (m/z 66, PTR‐MS, Ionicon,Innsbruck, Austria). This results in an exponentially decaying peak, allowing the total amount of DMS in a sample to be calculated by integration of the total peak area. Baseline levels were attained after 15 min of purging. Calibration curves were prepared using pure DMS (Merck, Gillingham, Dorset, UK). A primary DMS standard was prepared gravimetrically followed by dilution to produce a secondary standard, using gas tight vials. Five working (tertiary) standards were made up by dilution of the secondary standard in ultra‐pure water in 100 mL glass syringes, to produce a 5‐point calibration. For analysis, sub‐samples of each standard were taken using 20 mL glass syringes without exposing the sample to the air, and purged and analyzed as above. Calibrations were performed on each sampling date. DMSPd and DMSOd were sequentially reduced to DMS after adding NaOH (DMSPd) and cobalt‐doped NaBH4 (DMSOd) [12,47] into the purge tower and direct analysis by PTR‐MS as above for the GC method. On 11 dates, DMS concentrations analysed via GC‐PFPD and PTR‐MS were compared and show good agreement: y (DMS PTR‐MS) = 0.957 x (DMS Microorganisms 2020, 8, 337 4 of 20 GC‐PFPD), where r = 0.962 (n = 11, p < 0.001), suggesting that DMS concentrations derived from PTR‐MS analysis were on average 4% lower than those from GC measurements. GC‐PFPD), where r = 0.962 (n = 11, p < 0.001), suggesting that DMS concentrations derived from PTR‐MS analysis were on average 4% lower than those from GC measurements. 2.2. Stable Isotope Tracer Rate Experiments For subsequent isotope DMSPd and DMSOd analysis at each time point, 20 mL was withdrawn from the Tedlar incubation bags using a glass syringe and placed immediately into a 20 mL serum vial containing two pellets of sodium hydroxide, which were immediately crimp sealed. These samples were stored in the dark at in situ temperature for between 4 and 8 weeks [49–51]. Before analysis by PTR‐MS, 10 mL sub‐samples were taken with a glass syringe and filtered as previously described. The filtered sub‐sample was immediately injected in a purge tower. Stable isotopes of DMSPd and DMSOd were sequentially reduced to DMS after adding NaOH (DMSPd) and cobalt‐doped NaBH4 (DMSOd) into the purge tower and direct analysis by PTR‐MS as previously. Concentrations of stable isotopes were determined via PTR‐MS at m/z of 63, 65, 66, and 69 for unlabeled DMS, 13C2‐DMS, d3‐DMS, and d6‐DMS, respectively (as this method of soft ionization within the PTR‐MS adds a proton to each compound with no fragmentation of compounds). Final concentrations were calculated using standard curves. To scale the rate of tracer consumption or production to in situ values, the calculated rates were divided by the concentration of added tracer (yielding the apparent rate constant, h−1) and multiplied by the concentration of natural DMS, DMSPd, or DMSOd as appropriate [15]. Biological turnover times for DMS, DMSPd, and DMSOd were calculated from the inverse of the rate constants for the loss of d3‐DMS, d6‐DMSPd, and 13C2‐DMSOd, respectively. 2.3. Radiotracer Rate Experiments Seawater samples from the coastal station L4 were collected from Niskin bottles via acid washed Teflon tubing directly into the gas tight dark glass bottles (305 mL volume, acid washed, and 5 of 20 Microorganisms 2020, 8, 337 rinsed with hot water). Labelled 14C2‐DMSO was added to each bottle and incubated in the dark at in situ temperature (no headspace). Tracer nano‐molar (≤1.6 nM, representing ≤4% in situ DMSOd concentrations) additions of 14C2‐DMSOd (14CH3SO14CH3) were added to samples to determine microbial assimilation into biomass and dissimilation to 14CO2. Labelled 14C2‐DMSO was purchased from American Radiolabeled Chemicals, Inc (St.Louis, Missouri, USA) with a specific activity of 30 mCi mmol−1 and a radiochemical purity of >99% (based on high performance liquid chromatography). A primary stock was made by diluting 52 μCi into 25 mL of 18 MΩ milli‐Q water (2.1 μCi mL−1), and was stored in gas tight amber vials in the dark at 4 °C. Storage trials suggest <6% loss in activity over 12 months. Addition volumes of 14C2‐DMSO to seawater samples were <1% of the sample volume incubation experiments. 2.3.1. Carbon Assimilation from DMSOd For DMSOd carbon assimilation, a volume of 100 mL of the seawater sample was withdrawn from the bottom of the gas tight sampling bottles with a Teflon tube attached to a gas tight glass syringe. The tube was detached and the glass syringe attached to a Swinnex filter holder containing a 47 mm Supor 0.2 μm filter [47]. Supor filters (0.2 μm) were used because of their superior retention of particulate material [52]. Procedural blanks were routinely assessed by incubating 0.2 μm filtered seawater (with added Mercuric Chloride, 0.01% final concentration) and filtration, resulting in average counts of <60 ± 3 DPM per filter (n = 10, for ~34,000 DPM added per incubation). Samples and procedural blanks were filtered in a downward position with application of a very gentle pressure (as in Simό & Vila‐Costa [47]). It took about 6–8 min to filter each sample. Filters were rinsed (using a three‐way luer lock and pre‐loaded 2 mL syringe) with approximately 2 mL of 0.2 μm filtered seawater (but were not allowed to dry out). Filters were placed into scintillation vials, covered with 4 mL liquid scintillation fluid (Optiphase HiSafe 3; Perkin‐Elmer, High Wycombe, UK), and counted on a Tri‐carb 3100 (Perkin Elmer, High Wycombe, UK) liquid scintillation counter. Typically, for seawater samples, the coefficient of variation based on 3–6 replicates is <3%. It is possible that filtration artefacts caused release of DMSOd from particulate material (cf. DMSP, Kiene & Slezak [53]), so DMSO assimilation rates should be considered as minimal estimates. Exposure of filters to air at the end of filtration was avoided in our approach, which has previously been reported to cause severe DMSPd release [53]. 2.3.2. Carbon Dissimilation to CO2 from DMSOd DMSOd carbon microbial oxidation to 14CO2 (dissimilation) was determined in triplicate by pipetting 1 mL samples (each from replicate 305 mL gas tight incubation bottles) into 2 mL micro centrifuge tubes (o ring sealed), and adding 0.5 mL of SrCl2.6H2O (1 M to precipitate the 14CO2 as Sr14CO3), 20 μL of NaOH (1 M, to neutralise the HCl produced), and 100 μL of Na2CO3 (1 M, to ensure adequate pellet formation) (as in Goodwin et al. [54] for 14C labelled methyl halides). The efficiency of the process assessed by mass balance of added 14C label was 96% ± 3% (n = 6). After centrifugation, the supernatant was aspirated and the pellet washed twice with ethanol (80%), resuspended in 1 mL of NaOH solution (~10 nM) that had been adjusted to a pH of 11.7, before addition of Optiphase HiSafe III scintillant to create a slurry. The samples were vortex mixed and stored in the dark for >24 h before being analysed on the scintillation counter. This period ensures that any chemiluminescence arising from interactions between the added NaOH and the Optiphase scintillant subsides [52]. Procedural blanks were routinely assessed as previously described, and resulted in average counts of <22 ± 7 DPM mL−1 (n = 10, for ~34 000 DPM added per incubation). g p Microbial assimilation and dissimilation rates of DMSOd were determined from linear time course experiments (refer to Figure S1), where the apparent rate constants k (h−1) was initially calculated from a ratio of the 14C counts collected on either the filter (assimilation) or as precipitated 14CO2 (dissimilation, DPM mL−1∙h−1) divided by the 14C2‐DMSO spike (DPM mL−1). The apparent rate constant was multiplied by the in situ concentration of DMSOd to calculate DMSOd assimilation or dissimilation rates (nM d−1). All rate constants were corrected by subtracting killed sample counts. In 6 of 20 Microorganisms 2020, 8, 337 addition, dissimilation rates were corrected to account for a minor (typically <1%) contribution from assimilation into particles entrained within the Sr14CO3 precipitate. addition, dissimilation rates were corrected to account for a minor (typically <1%) contribution from assimilation into particles entrained within the Sr14CO3 precipitate. 3.1. Environmental and Biological Variables at Station L4 Station L4 is situated in northern temperate waters (salinity ~ 35.0 PSU [39]) and, typically, surface water temperature does not increase above 10 °C until mid‐April (Figure 1a). This is coincident with decreasing nutrient concentrations (Figure 1b), increasing concentrations of chlorophyll a (Figure 1c), and the start of water column stratification [39]. Average winter (Jan–Mar) nitrate and phosphate concentrations were 8.6 ± 0.6 and 0.6 ± 0.1 μM, respectively. Concentrations of nitrate and phosphate rapidly declined to <0.1 μM by the beginning of June and generally remain limited until early October, when the water column becomes fully mixed and nutrients begin to increase to typical winter values coincident with the decreasing sea surface temperature (Figure 1). The concentration of chlorophyll a showed a maxima mid‐April of 2.4 μg L−1 (Figure 1c), which is average compared with the long term trends (1992–2008 Smyth et al. [39]). This peak was associated with a typical spring diatom bloom (3.04 × 103 cells mL−1), mainly comprising of Pseudo‐nitzschia (0.8 × 103 cells mL−1) and large (≥4 μm) Thalassiosira phytoplankton cells (2.2 × 103 cells mL−1 Figure 2a). This was followed in May by a slightly smaller chlorophyll a peak (1.8–2.0 μg L−1, Figure 1c), but longer lasting phytoplankton bloom consisting mainly (23–55% of total phytoplankton) of Phaeocystis (1.5–4.4 × 103 cells mL−1 Figure 2a). Thereafter, chlorophyll a concentrations generally showed a decreasing pattern for the rest of 2014 (Figure 1c). From mid‐July to mid‐September, the phytoplankton was dominated by phytoflagellates (~2–5 μm) and did not show the more typical late August/September dinoflagellate bloom [39]. There was a relatively small bloom of Emiliania huxleyi during late August (up to 1.1 × 103 cells mL−1 Figure 2a). The two relatively small peaks in dinoflagellate abundance that occurred during June and July (Figure 2b) were dominated by Heterocapsa sp. (118 cells mL−1, 90% of total Dinoflagellate species) and Neoceratium lineatum (118 cells mL−1, 97% of total dinoflagellate species), respectively. From flow cytometry analysis, the numbers of nanophytoplankton (2–20 μm), picophytoplankton (<2.0 μm), and Synechococcus ranged between 0.12 and 15.9, 1.46 and 41.5, and 0.15 and 62.0 × 103 cells mL−1, respectively, and showed peaks in abundance during September (Figure 2c). Total bacteria ranged between 2.87 and 22.3 x105 cells mL−1 and were generally dominated by the high nucleic acid fraction. Bacterial numbers were generally highest during June–September months (Figure 2d). 3.2. DMS, DMSP, and DMSO Concentrations Near surface concentrations of DMS ranged from 0.5 nM during October to a maximum of 17.0 nM in mid‐June (Figure 3a). While DMS concentrations close to the bottom at 50 m showed less pronounced variability, ranging between 0.4 and 5.0 nM. Total DMSP near surface concentrations did not show any distinct maxima like DMS, but were on average 68.1 ± 18.5 nM during spring and summer months before generally decreasing to 10.6 ± 0.4 nM in October (Figure 3b). In close to bottom waters, DMSPt concentrations averaged 13.5 ± 8.9 nM (June–October). However, there were noticeably higher concentrations during May (average 108.0 ± 36.0 nM, Figure 3b), which could have been because of decaying and/or settling Phaeocystis cells, which were relatively abundant during this month (Figure 2a). The DMS/DMSPt ratio (Figure 3a) clearly followed the same pattern as the DMS, suggesting that elevated DMS concentrations were not just a product of higher concentrations of DMSPt. For near surface waters, 66–100% of the DMSPt was particulate (57–100% for near bottom samples). The total DMSO concentration in surface waters ranged between 2.3 and 102 nM, with on average ~56% in the dissolved phase (Figure 3c). The maxima in DMSOt occurred during June, coincident with a relatively high concentration of DMS at 11.4 nM. Minima in DMSOt concentrations were observed during autumn months, and were on average 7.0 ± 1.0 nM. 7 of 20 Microorganisms 2020, 8, 337 8 10 12 14 16 18 20 01 02 03 04 05 06 07 08 09 10 11 12 Temperature (oC) (a) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 1 2 3 4 5 6 7 8 9 10 01 02 03 04 05 06 07 08 09 10 11 12 Phosphate (µM) Nitrate (µM) Nitrate Phosphate (b) 1.5 2.0 2.5 g L-1) (c) 8 10 12 14 16 18 20 01 02 03 04 05 06 07 08 09 10 11 12 Temperature (oC) (a) (a) Figure 1. Change in (a) sea surface temperature, (b) inorganic nutrient concentrations of nitrate and phosphate, and (c) chlorophyll a in surface waters of station L4 in the western English Channel during 2014. 3.2. DMS, DMSP, and DMSO Concentrations 8 10 12 14 16 18 01 02 03 04 05 06 07 08 09 10 11 12 Temperature (oC) (a) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 1 2 3 4 5 6 7 8 9 10 01 02 03 04 05 06 07 08 09 10 11 12 Phosphate (µM) Nitrate (µM) Nitrate Phosphate (b) 0.0 0.5 1.0 1.5 2.0 2.5 01 02 03 04 05 06 07 08 09 10 11 12 Chl a (µg L-1) Month (2014) (c) 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 1 2 3 4 5 6 7 8 9 10 01 02 03 04 05 06 07 08 09 10 11 12 Phosphate (µM) Nitrate (µM) Nitrate Phosphate (b) 0.0 0.5 1.0 1.5 2.0 2.5 01 02 03 04 05 06 07 08 09 10 11 12 Chl a (µg L-1) Month (2014) (c) (c) Figure 1. Change in (a) sea surface temperature, (b) inorganic nutrient concentrations of nitrate and phosphate, and (c) chlorophyll a in surface waters of station L4 in the western English Channel during 2014. of 2 tal phytoplankton and dinoflagellate cells; (c) Synechococcus, pico, and nanophytoplankton; and (d) total h Channel during 2014. HNA, high nucleic acid; LNA, low nucleic acid. 8 tal phytoplankton and dinoflagellate cells; (c) Synechococcus, pico, and nanophytoplankton; and (d) total h Channel during 2014. HNA, high nucleic acid; LNA, low nucleic acid. s, pico, and nanophytoplankton; and (d) to nucleic acid. ton and dinoflagellate cells; (c) Synechococcu ing 2014. 3.2. DMS, DMSP, and DMSO Concentrations HNA, high nucleic acid; LNA, low l phytopla Channel d Microorganisms 2020, 8, 337 9 of 20 0.00 0.05 0.10 0.15 0.20 0.25 0 2 4 6 8 10 12 14 16 18 01 02 03 04 05 06 07 08 09 10 11 12 DMS (nM) surface 10 m 50 m average (sur & 10 m) DMS:DMSPt (a) 0 20 40 60 80 100 120 140 01 02 03 04 05 06 07 08 09 10 11 12 DMSPt (nM) 2 m 10 m average (2 & 10 m) 50 m (b) 80 100 120 140 SO (nM) DMSOd DMSOp DMSOt (2m) DMSOt (2 & 10m) (c) Microorganisms 2020, 8, 337 0.00 0.05 0.10 0.15 0.20 0.25 0 2 4 6 8 10 12 14 16 18 01 02 03 04 05 06 07 08 09 10 11 12 DMS (nM) surface 10 m 50 m average (sur & 10 m) DMS:DMSPt (a) Figure 3. Changes in concentrations of (a) dimethylsulfide (DMS) and average molar ratio of DMS/total dimethylsulfoniopropionate (DMSPt), (b) DMSPt, and (c) dimethylsulphoxide (DMSO) measured in the water column at station L4. 3.3. Stable‐Isotope Tracer Experiments 0.00 0.05 0.10 0.15 0.20 0.25 0 2 4 6 8 10 12 14 16 18 01 02 03 04 05 06 07 08 09 10 11 12 DMS (nM) surface 10 m 50 m average (sur & 10 m) DMS:DMSPt (a) 0 20 40 60 80 100 120 140 01 02 03 04 05 06 07 08 09 10 11 12 DMSPt (nM) 2 m 10 m average (2 & 10 m) 50 m (b) 0 20 40 60 80 100 120 140 01 02 03 04 05 06 07 08 09 10 11 12 DMSO (nM) Month (2014) DMSOd DMSOp DMSOt (2m) DMSOt (2 & 10m) (c) 0 20 40 60 80 100 120 140 01 02 03 04 05 06 07 08 09 10 11 12 DMSPt (nM) 2 m 10 m average (2 & 10 m) 50 m (b) (b) (b) 0 20 40 60 80 100 120 140 01 02 03 04 05 06 07 08 09 10 11 12 DMSO (nM) Month (2014) DMSOd DMSOp DMSOt (2m) DMSOt (2 & 10m) (c) (c) Figure 3. Changes in concentrations of (a) dimethylsulfide (DMS) and average molar ratio of DMS/total dimethylsulfoniopropionate (DMSPt), (b) DMSPt, and (c) dimethylsulphoxide (DMSO) measured in the water column at station L4. 3.3. 3.2. DMS, DMSP, and DMSO Concentrations Stable‐Isotope Tracer Experiments 3.3. Stable‐Isotope Tracer Experiments 10 of 20 10 of 20 Microorganisms 2020, 8, 337 Biological consumption of DMS (DMS BC) ranged between 2.4 and 40.8 nM d−1 (Figure 4a). The average DMS BC was 5.5 ± 2.3 nM d−1 (n = 8), excluding the three maxima that occurred during June and September. Net DMS production (change in 12C‐DMS with time) ranged between 0.0 and 10.5 nM d−1 (average 3.7 ± 3.4 nM d−1 n = 12). Summing net DMS production and DMS BC yields gross DMS production rates of 2.7–42.9 nM d−1 (average 14.4 ± 11.2 nM d−1, n = 11 Figure 4a). We only detected DMS production from DMSP cleavage on half of the sampling dates ranging between 0.2 and 21.5 nM d−1 (Figure 4b). DMS produced from the reduction of DMSO was only detected at the beginning of September (1.2 ± 0.0 nM d−1, Figure 4b), and was thus not a significant process during May–September 2014. We determined changes in concentrations of d6‐DMSPd and 13C2‐DMSOd on two dates in July and August (organic S transformations are summarized in Figure 5). There was a net loss of DMSPd (loss of 12C‐DMSPd) of 20.7 ± 8.8 and 29.2 ± 13.5 nM d−1 on 21 July and 26 August, respectively. We calculated biological consumption of DMSP as 75.7 ± 14.3 and 48.4 ± 15.6 nM d−1 for July and August, respectively (Figure 5). During these two experiments, we did not detect any DMSPd cleavage or any direct oxidation of DMSPd to DMSOd, and thus we calculated gross DMSPd production (biological DMSPd consumption – net loss of DMSPd) of at least 55.0 and 19.2 nM d−1 for July and August, respectively (Figure 5). Biological consumption of DMSOd was 23.2 ± 5.9 and 25.6 ± 11.6 nM d−1 for July and August, respectively (Figure 5). We observed a net loss of DMSOd (loss of 12C‐DMSOd) of 8.1 ± 3.3 and 18.9 ± 11.3 nM d−1, and thus calculated that gross DMSOd production must be at least 15.1 ± 6.4 and 6.7 ± 16.2 nM d‐1 for July and August, respectively (Figure 5). 3.3. Radiotracer Experiments When nano‐molar concentrations of 14C2‐DMSOd were added to seawater samples, 14C‐carbon was incorporated into cellular biomass and respired to 14CO2 linearly for ~3.5 h (for examples, see Figure S1). Thus, marine microbes assimilated and dissimilated DMSOd carbon, using it for growth and energy. Microbial uptake of DMSOd into biomass (assimilation) ranged between <0.01 and 0.49 nM d‐1 during June–December July 2014, and was at a maximum during summer months (Figure 4c). Microbial conversion of carbon from DMSOd to CO2 (dissimilation) was significantly higher and ranged between 2.7 and 111.0 nM d−1, with maximum rates during June (Figure 3c). The combination of DMSO assimilation and dissimilation thus ranged between 2.9 and 111.0 nM d−1, with between <0.1% and 5.3% of DMSOd being used for microbial growth, although this may represent a lower limit if filtration artefacts led to significant cell lysis and subsequent loss of assimilated DMSO. 4. Discussion The average surface DMS concentration found at L4 between May and October 2014 was 5.1 ± 4.0 nM, which compares well with the average found in U.K. shelf waters over the same monthly span of 5.4 ± 8.6 nM (data retrieved from the Global Surface Seawater DMS database: http://saga.pmel.noaa.gov/dms, number of records 2637). The ranges of DMS, DMSPt, and DMS/DMSPt presented here also agree with those found in a previously published seasonal cycle at L4 [55]. Variations in DMS and DMSP are partly a consequence of taxonomic succession, particularly of dinoflagellate species [55,56]. Our maximum DMS was ~6 nM lower than Archer et al. [55], possibly reflecting the absence of Karenia mikimotoi (cf. ~100 cells mL‐1 in Archer et al. [55]) and lower numbers of Scrippsiella trochoidea (8.2 compared with 25 cells mL‐1 in Archer et al. [55]). Surface DMS concentrations showed a statistically significant positive correlation with the total abundance of dinoflagellate cells (r = 0.529, n = 19 p = 0.02), perhaps reflecting the ability of several species, for example, Scrippsiella trochoidea and Heterocapsa triquetra, to directly produce DMS by cleaving dissolved DMSP [57–60]. Interestingly, there was a bloom of Heterocapsa sp. on 23 June 2014 (118 cells mL−1), which 11 of 20 Microorganisms 2020, 8, 337 11 of 20 Microorganisms 2020, 8, 337 11 of 20 0.0 1.0 2.0 0 5 10 15 20 25 30 35 40 45 01 02 03 04 05 06 07 08 09 10 11 12 DMS turn over (d) DMS rate (nM d-1) Net Production Biological Consumption Gross Production Turn over (a) 0 5 10 15 20 25 30 01 02 03 04 05 06 07 08 09 10 11 12 DMS rate (nM d-1) DMS produced from DMSP cleavage (b) 60 80 100 120 0.3 0.4 0.5 ation/consumption M d-1) miliation (nM d-1) 14C-DMSO assimilation 14C-DMSO dissimilation Biological consumption 13C-DMSOd (c) 0.0 1.0 2.0 0 5 10 15 20 25 30 35 40 45 01 02 03 04 05 06 07 08 09 10 11 12 DMS turn over (d) DMS rate (nM d-1) Net Production Biological Consumption Gross Production Turn over (a) Figure 4. Changes in rates of (a) production and consumption of DMS and turn over time, (b) DMS produced from the cleavage of DMSP, and (c) microbial utilization of DMSO in surface waters at station L4. Error bars represent ±1 standard deviation based on three replicates. 4. Discussion 0.0 1.0 2.0 0 5 10 15 20 25 30 35 40 45 01 02 03 04 05 06 07 08 09 10 11 12 DMS turn over (d) DMS rate (nM d-1) Net Production Biological Consumption Gross Production Turn over (a) 0 5 10 15 20 25 30 01 02 03 04 05 06 07 08 09 10 11 12 DMS rate (nM d-1) DMS produced from DMSP cleavage (b) 0 20 40 60 80 100 120 0.0 0.1 0.2 0.3 0.4 0.5 01 02 03 04 05 06 07 08 09 10 11 12 DMSO dissimilation/consumption (nM d-1) DMSO assimiliation (nM d-1) Month (2014) 14C-DMSO assimilation 14C-DMSO dissimilation Biological consumption 13C-DMSOd (c) 0 5 10 15 20 25 30 01 02 03 04 05 06 07 08 09 10 11 12 DMS rate (nM d-1) DMS produced from DMSP cleavage (b) (b) 0 20 40 60 80 100 120 0.0 0.1 0.2 0.3 0.4 0.5 01 02 03 04 05 06 07 08 09 10 11 12 DMSO dissimilation/consumption (nM d-1) DMSO assimiliation (nM d-1) Month (2014) 14C-DMSO assimilation 14C-DMSO dissimilation Biological consumption 13C-DMSOd (c) Figure 4. Changes in rates of (a) production and consumption of DMS and turn over time, (b) DMS produced from the cleavage of DMSP, and (c) microbial utilization of DMSO in surface waters at station L4. Error bars represent ±1 standard deviation based on three replicates. 12 of 20 Microorganisms 2020, 8, 337 Figure 5. Box model summarising organic sulfur transformation rates (nM d−1) under (a) mid‐high concentrations of DMS (14.0 nM, sampled on 21 July 2014) and (b) low concentrations of DMS (1.0 nM sampled on 26 August 2014). These were the only two dates when all organic sulfur transformation rates were simultaneously determined. NP refers to net production and GP to gross production. Added stable tracers (in grey boxes) and their transformation products are indicated in grey text. 4. Discussion In comparison with DMS and DMSP, concentrations of DMSO are relatively less well documented [32]. During the majority of our sampling dates, surface DMSOt concentrations remained lower than DMSPt concentrations, with maximum values not exceeding ~50 nM, as was similarly reported for coastal Antarctic waters [20]. Our DMSOt values are within ranges reported globally [24,30,32,47]. One notable exception was a large peak of DMSOt (DMSOd 62 nM, DMSOp 46 nM), which corresponded to an over 140‐fold increase in the number of dinoflagellate cells from a pre‐bloom average of 0.91 to 131 cell mL−1. During this time, the dinoflagellate abundance was dominated (90%) by Heterocapsa cells, which were previously absent. We thus hypothesise that either there is direct production of DMSOd from the high‐DMSP producing Heterocapsa [58], or the bacterial community associated with these dinoflagellates oxidises DMS to DMSOd (as Sagittula stellata have been shown to do using DMS as an energy source [63]) or metabolises DMSP straight to DMSO. Dominant pelagic bacteria such as the marine Roseobacter clade are thought to use methylamine‐dependent monoxygenases to oxidise DMS to DMSO [27]. This trimethylamine monoxygenase is also present in the SAR11 clade, which, together with the Roseobacter group, could account for 20% of bacterial cells in surface seawater [27]. The SAR11 subgroups Ia and Ib have also been suggested to be the main potential DMSP consumers [64], although Roseobacter sp. also metabolise DMSP [65]. Growth experiments with a Roseobacter isolate also revealed its potential plasticity in metabolism, by demonstrating a shift from DMS oxidation and DMSP degradation under aerobic conditions to DMSO (and nitrate) reduction under anaerobic conditions [66]. Alternatively, Thume et al. [67] have recently reported that a new sulfur metabolite dimethylsulfoxonium propionate (DMSOP) is synthesized by several DMSP‐producing phytoplankton and marine bacteria, which is further metabolized (by marine bacteria) to DMSO. On average, DMSOd accounted for 55% of DMSOt and showed no obvious trends over the sampling period. The concentration of DMSOt in surface waters also showed a significant relationship with dinoflagellate abundance (r = 0.498, n = 19, p < 0.05). The biological consumption rates of DMS determined during this temperate coastal study were generally in the same range as those reported for coastal Antarctic [20] and subarctic Pacific waters [65]. In these polar waters, Asher et al. [20,68] suggest that their measured gross DMS loss/consumption rates should largely reflect biological consumption owing to low calculated DMS photo‐oxidation rates. 4. Discussion Transformations labelled 1–9 are all microbial processes: (1) oxidation of DMS to DMSOd (appearance of d3‐DMSO), (2) consumption of DMS (corrected loss of d3‐DMS), (3) reduction of DMSOd to DMS (appearance of 13C2 DMS), (4) enzymatic cleavage of DMSPd to DMS (appearance of d6‐DMS), (5) conversion of DMSPd to DMSOd (appearance of d6‐DMSO), (6) consumption of DMSPd (corrected loss of d6‐DMSP), (7) consumption of DMSOd (corrected loss of 13C2‐DMSOd), (8) DMSOd i il ti f th (i ti f 14C DMSO i t th ti l t h ) d (9) DMSO Figure 5. Box model summarising organic sulfur transformation rates (nM d−1) under (a) mid‐high concentrations of DMS (14.0 nM, sampled on 21 July 2014) and (b) low concentrations of DMS (1.0 nM sampled on 26 August 2014). These were the only two dates when all organic sulfur transformation rates were simultaneously determined. NP refers to net production and GP to gross production. Added stable tracers (in grey boxes) and their transformation products are indicated in grey text. Transformations labelled 1–9 are all microbial processes: (1) oxidation of DMS to DMSOd (appearance of d3‐DMSO), (2) consumption of DMS (corrected loss of d3‐DMS), (3) reduction of DMSOd to DMS (appearance of 13C2 DMS), (4) enzymatic cleavage of DMSPd to DMS (appearance of d6‐DMS), (5) conversion of DMSPd to DMSOd (appearance of d6‐DMSO), (6) consumption of DMSPd (corrected loss of d6‐DMSP), (7) consumption of DMSOd (corrected loss of 13C2‐DMSOd), (8) DMSOd assimilation for growth (incorporation of 14C‐DMSOd into the particulate phase), and (9) DMSOd dissimilation to CO2 (14CO2 precipitated as 14CO3). An “X” denotes no detectable rate was determined during the incubation experiment. Rates are shown as ±1 standard deviation based on three replicates. may have also contributed to the DMS maxima observed at this time. The range in DMSPt concentrations found at the coastal L4 station is within the range often reported for a variety of other marine environments, including the North Sea, North Atlantic, Mediterranean, and subarctic Pacific 13 of 20 Microorganisms 2020, 8, 337 [20,30], but did not reach the elevated concentrations associated with intense dinoflagellate or Phaeocystis sp. blooms (>200 nM) [30,61,62]. The majority of the DMSP was in the particulate phase, as usually reported [24,30,56,62]. The average DMSPp/DMSOp for our study was 5.4 ± 4.8, in agreement with the average reported in Simó & Vila‐Costa [47]. 4. Discussion Biological DMS consumption rates were reportedly lower in the Ross Sea, Antarctica (0.02–8.8 nM d−1), possibly because of lower temperatures minimizing bacterial activity [13]. Biological processes are generally reported to dominate DMS removal compared with sea‐air flux, photo‐oxidation, and mixing at the base of the mixed layer [20,68,69]. Our biological consumption rates of DMS also showed a statistically significant positive linear correlation with both the numbers of the dinoflagellate Scrippsiella trochoidea (n = 11, r = 0.9858 p < 0.001) and the diatoms Pseudo‐nitzschia pungens (n = 11, r = 0.9908, p < 0.001) and Leptocylindrus danicus (n = 11, r = 0.8543, p < 0.001), where the numbers of phytoplankton cells varied between <0.04 and 7.68, <0.16 and 2.6, and <0.10 and 196.0 cells mL−1, respectively. The most noticeable was the peak in DMS BC during June, which coincided with an increase in the numbers of Scrippsiella trochoidea (from average background of 0.16 to 7.92 cells mL−1), which accounted for ~72% of all the dinoflagellates. Despite generally occurring in relatively low numbers (cf. ~25 cells mL−1 at L4 reported in Archer et al. [55]), Scrippsiella trochoidea is a prolific DMSP producer, with cellular concentrations reported as high as millimolar [70] or 174–380 pg DMSPt cell−1 [58,71]. The bacterial species associated with Scrippsiella trochoidea in culture have been shown to consume DMS, mostly oxidising it to DMSO [72]. We cannot confirm DMS oxidation to DMSO during these June experiments as we did not undertake complimentary stable tracer DMSO analysis on these dates. The biological turnover time for DMS at station L4 ranged between 0.2 and 1.8 days, in agreement with previous marine estimates [30,68], and showed the quickest turnover coincident with maximum rates of DMS BC. DMS produced from DMSP cleavage was highly variable at our coastal station, ranging from non‐detectable to 21.5 nM d−1 14 of 20 14 of 20 Microorganisms 2020, 8, 337 (average 9.1 ± 9.3 nM d−1, n = 6). Generally, these rates are in the range previously reported for a variety of marine environments [15,20,30,68]. 4. Discussion Our rates of DMSPd cleavage correlated with the abundance of the grazing ciliate Tontonia ovalis (n = 6, r = 0.9790, p < 0.001) and large flagellates (≥15 μm, where n = 6, r = 0.8346, p < 0.05), possibly suggesting enhanced DMSP lyase activity owing to grazing pressures, perhaps as some chemical “don’t eat me” cue [73], or because of physical disruption of cells during grazing [74]. DMS produced from DMSO reduction was not detectable at the coastal station L4, in sharp contrast to Antarctic environments [15,20]. Following a simple DMS mass balance approach, which assumes that the observed net change in the DMS pool must equal the DMS produced by DMSP cleavage and DMSO reduction minus biological DMS consumption [15], allows an assessment of the significance of other DMS sources. Our data suggest that up to 42.9 nM d−1 of DMS could be excreted from biological particles (and/or the conversion of unlabeled DMSP or DMSO that has leaked from cells into the dissolved pool [15]). However, a correlation between estimated DMS release from particles and biological consumption (n = 11, r = 0.7903, p < 0.001) suggests a tight coupling in coastal waters. gg g p g The biological loss of DMSOd due to assimilation and dissimilation (determined using radiotracers) ranged between 2.9 and 111 nM d−1. However, the maximum loss rate was driven by the relatively high DMSOd concentration of 62 nM compared with the otherwise seasonal average of 9.2 ± 6.4 nM. Excluding the observed maxima, radiotracer‐derived DMSOd biological consumption ranged between 2.9 and 24.3 nM d−1. Independent stable tracer experiments using 13C2‐DMSO during July–August also suggest gross biological DMSO consumption rates of 23.3 ± 5.9 and 25.6 ± 11.6 nM d‐1 (Figure 4c). A comparison between stable tracer‐derived microbial DMSOd biological consumption rates and radiotracer‐derived DMSOd assimilation plus dissimilation loss rates during July suggests that the majority of microbial DMSOd loss (23.2 ± 5.9 nM d−1 Figure 5a) was because of microbial dissimilation of DMSOd to CO2 (24.2 ± 0.3 nM d‐1, Figure 5a), presumably in order to provide reducing power. 4. Discussion Our seasonally resolved data suggest that DMSO assimilation is highest during summer (up to 5%) when nutrients are depleted, 15 of 20 Microorganisms 2020, 8, 337 and lowest during nutrient replete winter months, in agreement with an annual study of methanol metabolism at station L4 [80]. The rates of biological consumption of DMSPd (75.7 and 48.4 nM d−1 for 21 July and 26 August respectively, Figure 5) were between 9–19 and 2–3 times higher than that of DMS and DMSOd, respectively, under both medium (14.0 nM) and low DMS (1.0 nM) conditions (Figure 5). During these two sampling dates, we did not detect any bacterial enzymatic conversion of DMSPd to DMS, which perhaps indicates sustained bacterial sulfur demand and consumption of DMSPd via demethylation or demethiolation pathways [1,83]. Although it has been demonstrated that Synechococcus cells (Figure 2c) assimilate DMSP sulfur [84], our dark incubations would minimise their contribution to DMSP uptake. Our biological consumption rates of DMSPd are within the range of marine DMSPd turnover rates summarized in Kiene et al. [83], where up to 100% of DMSPd was reportedly metabolized via demethylation. The turnover times of DMSPd were estimated at 0.3 ± 0.1 d, in agreement with other marine waters [30]. DMS has been previously demonstrated to be metabolized by bacteria much more slowly than DMSPd [10], with our bacterial DMS consumption data suggesting that it is utilised up to 19 times slower than DMSPd. Stable tracer derived biological consumption rates of DMSOd are intermediate between DMS and DMSPd. The combined biological consumption of dissolved DMSP, DMS, and DMSO was ~77–108 nM d−1 (Figure 5). We estimate the contribution made by these dissolved organic species to both bacterial carbon and sulfur demand by assuming that bacterial heterotrophic production (BP) was 0.18 ± 0.04 (n = 3) and 0.08 ± 0.04 (n = 3) μg C L−1∙h−1 during July and August, respectively (data derived from Sergeant et al. [80] using a theoretical leucine‐to‐carbon conversion factor of 1.55 kg C mol leu−1 [85]). Bacterial respiration (BR) was calculated from production, where BR = 3.69BP0.58 [86]. Bacterial growth efficiency (BGE) was calculated in two different ways; firstly using production (BP) and respiration (BR) estimates (BGE = BP/(BP + BR)), and secondly using chlorophyll a concentrations (Figure 1c). 4. Discussion Bacterial carbon demand was calculated by dividing bacterial production (BP) by the average of the two estimates of BGE as 3.35 ± 0.83 and 1.74 ± 0.79 μmoles C L−1 d−1 for July and August, respectively. The rates of biological consumption of dissolved DMSP, DMS, and DMSO were converted to carbon units (by multiplying by 5, 2, and 2, respectively) to yield rates of 0.44 and 0.30 μmol C L−1 d−1 for July and August, respectively. This suggests that these three organic sulfur species could support 13%–17% of the estimated bacterial carbon demand during summer months. Bacterial carbon demand was converted to sulfur demand, assuming a carbon to sulfur ratio of 86 [11,87], resulting in 0.04 and 0.02 μmoles S L−1∙d−1 for July and August, respectively. Calculations suggest that DMSPd alone could supply all of the sulfur to meet microbial demand (195% and 234% for July and August, respectively, DMSP S microbial consumption/bacterial S demand). Even if only ~50% of DMSPd sulfur used by bacteria was incorporated into biomass, that is, used for assimilatory rather than dissimilatory purposes, then the combined uptake of DMS, DMSPd, and DMSOd (277% and 378% for July and August, respectively) would still meet bacterial sulfur demands. DMSPd is known to be a widespread substrate for heterotrophic bacteria, with literature suggesting that it can provide up to 15% and 100% of their carbon and sulfur requirements, respectively [11,83]. Similar calculations for DMSOd suggest that this compound alone could supply 60–126% and 1.4–2.9% of bacterial sulfur and carbon demand, respectively. Bacterial utilization of such reduced organic sulfur species over dissolved sulfate is thought to be energetically preferable for the synthesis of methionine [88]. In summary these data demonstrate that throughout the productive months of the year at a In summary, these data demonstrate that, throughout the productive months of the year at a temperate coastal location, the biological consumption of DMS is highly variable, and largely decoupled from the amount of DMS produced from cleavage of DMSPd, and its oxidation to DMSOd. Stable tracer experiments suggest that DMS produced from the reduction of DMSOd is not a common pathway in temperate coastal waters, which contrasts to Antarctic regions. Microbial consumption rates of organic sulfur species follow the order DMSPd > DMSOd > DMS, where the microbial dissimilation of DMSOd to CO2 can be a significant loss pathway for DMSOd in coastal waters. 4. Discussion However, during August, stable tracer‐derived gross consumption rates of 25.6 ± 11.6 nM d−1 were higher than the radiochemical‐derived DMSOd assimilation plus dissimilation combined rate of 10.5 ± 0.3 nM d−1 (Figure 5b), suggesting that other DMSOd loss reactions were dominant, for example, perhaps further oxidation to dimethylsulphone [32], as we did not detect any reduction to DMS. Our experimental design precluded DMSO losses owing to photochemical reactions [75]. Excluding the maxima, the biological DMSO loss rates (assimilation plus dissimilation) in coastal waters determined in our study averaged 11.9 ± 2.8 nM d−1, which is comparable to DMSOd loss rates of 4–10 nM d−1 determined via changes in concentrations of DMSO during dark incubations [30,76]. By comparison, biological DMSOd uptake rates determined at coastal stations in the Gulf of Mexico were lower than ours, ranging between 1.7 and 3.9 nM d−1 [31], possibly because of comparatively lower chlorophyll a levels and the influence of riverine outflows during the latter study. Turnover times of DMSOd were estimated from the reciprocal of the total apparent rate constant (14C‐derived assimilation plus dissimilation) at 0.6–1.9 d, which is comparable to 0.5–0.7 d derived from the 13C2‐DMSO stable tracer experiments, and not dissimilar to previous literature estimates of 2–5 d [30,76]. However, Tyssebotyn et al. [31] report a much slower median DMSOd turnover time of 7.4 d in coastal river plume stations, presumably because of their lower DMSOd oxidation rates. The majority of DMSOd utilized by the heterotrophic community was respired (>94%), like a variety of other low nano‐molar organic methylated substrates in seawater such as methanol, methylamines, glycine betaine, and trimethylamine N‐oxide [77–81], although up to 30% of acetaldehyde was assimilated into biomass by SAR11 bacterioplankton in culture [82]. The microbial oxidation of C1 units, in these cases, methyl groups, has been previously hypothesized to be a significant conduit by which dissolved organic carbon is recycled to CO2 in the upper ocean [81], and our DMSOd respiration data lend support to this idea. Tyssebotyn et al. [31] similarly concluded that DMSO was mostly metabolized for energy, although they reported a lower proportion of dissimilation (62–75%) compared with our data, and suggested that the proportion metabolized does not change with dissolved carbon or nutrient status. 4. Discussion However, what controls the loss of DMSOd and the identification of bacteria responsible (and their biochemical pathways/genes) in seawater largely remains elusive. 16 of 20 Microorganisms 2020, 8, 337 16 of 20 Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Figure S1: Typical time course experiments at the coastal station L4 showing the amount of radioactive carbon from the added 14C2‐DMSO that was used during (a) assimilation into particulate material and (b) oxidation to 14CO2 after radiotracer addition of ≤1.6 nM. Where DPM is disintegrations per minute (1 DPM = 4.51 × 10−13 Ci). Error bars represent ±1 standard deviation of three replicate samples. Supplementary Materials: The following are available online at www.mdpi.com/xxx/s1, Figure S1: Typical time course experiments at the coastal station L4 showing the amount of radioactive carbon from the added 14C2‐DMSO that was used during (a) assimilation into particulate material and (b) oxidation to 14CO2 after radiotracer addition of ≤1.6 nM. Where DPM is disintegrations per minute (1 DPM = 4.51 × 10−13 Ci). Error bars represent ±1 standard deviation of three replicate samples. Author Contributions: Conceptualization, J.L.D., F.E.H., and H.S.; methodology, J.L.D., F.E.H., and J.A.S.; formal analysis, J.L.D., F.E.H., and J.A.S.; writing—original draft preparation, J.L.D.; writing—review and editing, J.L.D. and F.E.H.; funding acquisition, J.L.D., F.E.H., and H.S. All authors have read and agreed to the published version of the manuscript. Author Contributions: Conceptualization, J.L.D., F.E.H., and H.S.; methodology, J.L.D., F.E.H., and J.A.S.; formal analysis, J.L.D., F.E.H., and J.A.S.; writing—original draft preparation, J.L.D.; writing—review and editing, J.L.D. and F.E.H.; funding acquisition, J.L.D., F.E.H., and H.S. All authors have read and agreed to the published version of the manuscript. Funding: This work was funded by a NERC grant NE/L004151/1 and the Western Channel Observatory, which is is funded by the UK Natural Environment Research Council through its National Capability Long‐term Single Centre Science Programme, Climate Linked Atlantic Sector Science, grant number NE/R015953/1. Funding: This work was funded by a NERC grant NE/L004151/1 and the Western Channel Observatory, which is is funded by the UK Natural Environment Research Council through its National Capability Long‐term Single Centre Science Programme, Climate Linked Atlantic Sector Science, grant number NE/R015953/1. Acknowledgments: We thank Denise Cummings and the boat crew of the RV Quest for sampling at station L4, which is provided by the Plymouth Marine Laboratory, Western Channel Observatory (www.westernchannelobservatory.org.uk). 4. Discussion We also thank Glen Tarran, Claire Widdicombe, and Malcolm Woodward for provision of flow cytometry, microscope phytoplankton identification and enumeration, and nutrients, respectively. Acknowledgments: We thank Denise Cummings and the boat crew of the RV Quest for sampling at station L4, which is provided by the Plymouth Marine Laboratory, Western Channel Observatory (www.westernchannelobservatory.org.uk). We also thank Glen Tarran, Claire Widdicombe, and Malcolm Woodward for provision of flow cytometry, microscope phytoplankton identification and enumeration, and nutrients, respectively. Conflicts of Interest: The authors declare no conflict of interest References 1. Levasseur, M. Impact of Arctic meltdown on the microbial cycling of sulfur. Nat. Geosci. 2013, 6, 691–700. 2. Lovelock, J.E.; Maggs, R.J.; Rasmussen, R.A. Atmospheric dimethyl sulphide and the natural sulphur cycle. Nature 1972, 237, 452–453. 3. Kettle, A.J.; Andreae, M.O. Flux of dimethylsulfide from the oceans: A comparison of updated datasets and flux models. J. Geophys. Res. Atmos. 2000, 105, 26793–26808. 4. Curson, A.R.; Todd, J.D.; Sullivan, M.; Johnston, A.W. Catabolism of dimethylsulphoniopropionate: Microorganisms, enzymes and genes. Nat. Rev. Microbiol. 2011, 9, 849–859. 5. Galí, M.; Simó, R. A meta‐analysis of oceanic DMS and DMSP cycling processes: Disentangling the summer paradox. Glob. Biogeochem. Cycles 2015, 29, 496–515. 6. Kiene, R.P.; Bates, T.S. Biological removal of dimethyl sulphide from sea water. Nature 1990, 345, 702– 7. Herrmann, M.; Najjar, R.G.; Neeley, A.R.; Vila‐Costa, M.; Dacey, J.W.; DiTullio, G.R.; Kieber, D.J.; Kiene, R.P.; Matrai, P.A.; Simo, R.; et al. Diagnostic modeling of dimethylsulfide production in coastal water west of the Antarctic Peninsula. Cont. Shelf Res. 2012, 32, 96–109. 8. Andreae, M.O. Ocean‐atmosphere interactions in the global biogeochemical sulfur cycle. Mar. Chem. 1990, 30, 1–29. 9. Von Glasow, R.V.; Crutzen, P.J. Model study of multiphase DMS oxidation with a focus on halogens. Atmos. Chem. Phys. 2004, 4, 589–608. 10. Kiene, R.P.; Linn, L.J. The fate of dissolved dimethylsulfoniopropionate (DMSP) in seawater: Tracer studies using 35S‐DMSP. Geochim. Cosmochim. Acta 2000, 64, 2797–2810. 11. Simó, R.; Archer, S.D.; Pedrós‐Alió, C.; Gilpin, L.; Stelfox‐Widdicombe, C.E. Coupled dynamics of dimethylsulfoniopropionate and dimethylsulfide cycling and the microbial food web in surface waters of the North Atlantic. Limnol. Oceanogr. 2002, 47, 3–61. 12. Vila‐Costa, M.; Kiene, R.P.; Simó, R. Seasonal variability of the dynamics of dimethylated sulfur compounds in a coastal northwest Mediterranean site. Limnol. Oceanogr. 2008, 53, 198–211. 13. del Valle, D.A.; Kieber, D.J.; Toole, D.A.; Brinkley, J.; Kienea, R.P. Biological consumption of dimethylsulfide (DMS) and its importance in DMS dynamics in the Ross Sea, Antarctica. Limnol. Oceanogr. 2009, 54, 785–798. 14. Del Valle, D.A.; Kieber, D.J.; Kiene, R.P. Depth‐dependent fate of biologically consumed dimethylsulfide in the Sargasso Sea. Mar. Chem. 2007, 103, 197–208. 15. Asher, E.C.; Dacey, J.W.H.; Mills, M.M.; Arrigo, K.R.; Tortell, P.D. High concentrations and turnover rates of DMS, DMSP and DMSO in Antarctic sea ice. Geophys. Res. Letts. 2011, 38, L23609 doi:10.1029/2011GL049712. 17 of 20 Microorganisms 2020, 8, 337 16. Steinke, M.; Daniel, C.; Kirst, G.O. References DMSP Lyase in Marine Macro‐ and Microalgae. In Biological and Environmental Chemistry of DMSP and Related Sulfonium Compounds; Kiene, R.P., Visscher, P.T., Keller, M.D., Kirst, G.O., Eds.; Springer: Boston, MA, USA, 1996; pp. 317–324. 17. Stefels, J.; van Boekel, W.H.M. Production of DMS from dissolved DMSP in axenic cultures of the marine phytoplankton species Phaeocystis sp. Mar. Ecol. Prog. Ser. 1993, 97, 11–18. 18. Spiese, C.E. Cellular Production and Losses of Dimethylsulfide in Marine Phytoplankton. Ph.D. Thesis, State University of New York College of Environmental Science and Forestry, Syracuse, NY, USA, 2010. 19. González, J.M.; Kiene, R.P.; Moran, M.A. Transformation of sulphur compounds by an abundant lineage of the marine bacteria in the α‐subclass of the class Proteobacteria. Appl. Environ. Microbiol. 1999, 65, 3810– 3819. 20. Asher, E.C.; Dacey, J.W.H.; Stukel, M.; Long, M.C.; Tortell, P.D. Processes driving seasonal variability in DMS, DMSP, and DMSO concentrations and turnover in coastal Antarctic waters. Limnol. Oceanogr. 2017, 62, 104–124. 21. Stefels, J. Physiological aspects of the production and conversion of DMSP in marine algae and higher plants. J. Sea Res. 2000, 43, 183–197. 22. Simó, R. Production of atmospheric sulfur by oceanic plankton: Biogeochemical, ecological and evolutionary links. Trends Ecol. Evol. 2001, 16, 287–294. 23. Simό, R.; Grimalt, J.O.; Albaigés, J. Dissolved dimethylsulphide, dimethylsulphoniopropionate dimethylsulphoxide in western Mediterranean waters. Deep Sea Res. II 1997, 44, 929–950. 24. Zindler, C.; Bracher, A.; Marandino, C.A.; Taylor, B.; Torrecilla, E.; Kock, A.; Bange, H.W. Sulphur compounds, methane, and phytoplankton: Interactions along a north‐south transit in the western Pacific Ocean. Biogeosciences 2013, 10, 3297–3311, doi:10.5194/bg‐10‐3297‐2013. 25. Lee, P.A.; de Mora, S.J. Intracellular dimethylsulphoxide (DMSO) in unicellular marine algae: Speculations on its origin and possible biological role. J. Phycol. 1999, 358–18. 26. Chen, Y.; Patel, N.A.; Crombie, A.; Scrivens, J.H.; Murrell, J.C. Bacterial flavin‐containing monooxygenase is trimethylamine monooxygenase. Proc. Natl. Acad. Sci. USA 2011, 108, 17791–17796. 27. Lidbury, I.; Kröber, E.; Zhang, Z.; Zhu, Y.; Murrell, J.C.; Chen, Y.; Schäfer, H. A mechanism for bacterial transformation of dimethylsulfide to dimethylsulfoxide: A missing link in the marine organic sulfur cycle. Environ. Microb. 2016, 18, 2754–2766. 28. Andreae, M.O. Dimethylsulfoxide in marine and freshwaters. Limnol. Oceanogr. 1980, 25, 1054–1063. 29. Sunda, W.; Kieber, D.J.; Kiene, R.P.; Huntsman, S. An antioxidant function for DMSP and DMS in algae. Nature 2002, 418, 317–320. 30. Simό, R.; Pedrόs‐Aliό, C.; Malin, G.; Grimalt, J.O. References Long‐term phytoplankton community dynamics in the Western English Channel. J. Plankton Res. 2010, 32, 643–655. 46. Simό, R.; Grimalt, J.O.; Albaigés, J. Sequential method for the determination of nanomolar concentrations of dimethyl sulfoxide in natural waters. Anal. Chem. 1996, 68, 1493–1498. Simό, R.; Vila‐Costa, M. Ubiquity of algal dimethylsulfoxide in the surface ocean: Geographic and temporal distribution patterns. Mar. Chem. 2006, 100, 136–146. 48. Archer, S.D.; Kimmance, S.A.; Stephens, J.A.; Hopkins, F.E.; Bellerby, R.G.J.; Schulz, K.G.; Piontek, J.; Engel, A. Contrasting responses of DMS and DMSP to ocean acidification in Arctic waters. Biogeosciences 2013, 10, 1893–1908. 49. Townsend, D.W.; Keller, M.D. Dimethylsulfide (DMS) and dimethylsulfoniopropionate (DMSP) in relation to phytoplankton in the Gulf of Maine. Mar. Ecol. Prog. Ser. 1996, 137, 229–241. 50. Simό, R.; Malin, G.; Liss, P. Refinement of the borohydride reduction method for trace analysis of dissolved and particulate dimethyl sulfoxide in marine water samples. Anal. Chem. 1998, 70, 4864–4867. 51. Kinsey, J.D.; Kieber, D.J. Microwave preservation method for DMSP, DMSO and acrylate in unfiltered seawater and phytoplankton culture samples. Limnol. Oceanogr. Methods 2016, 14, 196–209. 52. Kiene, R.P.; Hoffmann Williams, L.P. Glycine betaine uptake, retention, and degradation by microorganisms in seawater. Limnol. Oceanogr. 1998, 43, 1592–1603. 53. Kiene, R.P.; Slezak, D. Low dissolved DMSP concentrations in seawater revealed by small‐volume gra filtration and dialysis sampling. Limnol. Oceanogr: Methods 2006, 4, 80–95. 54. Goodwin, K.D.; Schaefer, J.K.; Oremland, R.S. Bacterial oxidation of dibromomethane and methyl brom in natural waters and enrichement cultures. Appl. Environ. Microbiol. 1998, 64, 4629–4636. 55. Archer, S.,D.; Cummings, D.G.; Llewellyn, C.A.; Fishwick, J.R. Phytoplankton taxa, irradiance and nutrient availability determine the seasonal cycle of DMSP in temperate shelf seas. Mar. Ecol. Prog. Ser. 2009, 394, 111–124. 56. Zindler, C.; Peeken, I.; Marandino, C.A.; Bange, H.W. Environmental control on the variability of DMS and DMSP in the Mauritanian upwelling region. Biogosciences 2012, 9, 1041–1051, doi:10.5194/bg‐9‐1041‐2012. 57. Niki, T.; Kunugi, M.; Otsuki, A. DMSP‐lyase activity in five marine phytoplankton species: Its potential importance in DMS production. Mar. Biol. 2000, 136, 759–764. 58. Caruana, A.M.N.; Steinke, M.; Turner, S.M.; Malin, G. Concentrations of dimethylsulphoniopropionate and activities of dimethylsulphide‐producing enzymes in batch cultures of nine dinoflagellate species. Biogeochemistry 2012, 110, 87–107. g y 59. Steinke, M.; Malin, G.; Archer, S.D.; Burkill, P.H.; Liss, P.S. DMS production in a coccolithophorid bloom: Evidence for the importance of dinoflagellate DMSP lyases. Aquat. Microb. Ecol. 2002, 26, 259–270. 60. References Biological turnover of DMS, DMSP and DMSO in contrasting open‐sea waters. Mar. Ecol. Prog. Ser. 2000, 203, 1–11. 31. Tyssebotn, I.M.B.; Kinsey, J.D.; Kieber, D.J.; Kiene, R.P.; Rellinger, A.N.; Motard‐Côté, J. Concentrations, biological uptake, and respiration of dissolved acrylate and dimethylsulfoxide in the northern Gulf of Mexico. Limnol. Oceanogr. 2017, 62, 1198–1218. 32. Hatton, A.D.; Darroch, L.; Malin, G. The role of dimethylsulphoxide in the marine biogeochemical cycle of dimethylsuphide. Oceanogr. Mar. Biol. Annu. Rev. 2004, 42, 29–56. 33. Spiese, C.E.; Kieber, D.J.; Nomura, C.T.; Kiene, R.P. Reduction of dimethylsulfoxide to dimethylsulfide by marine phytoplankton. Limnol. Oceanogr. 2009, 54, 560–570. 34. McCrindle, S.L.; Kappler, U.; McEwan, A.G. Microbial dimethylsulfoxide and trimethylamine‐N‐oxide respiration. Adv. Microb. Physiol. 2005, 50, 147–198. 35. Vogt, C.; Rabenstein, A.; Rethmeier, J.; Fischer, U. Dimethyl sulphoxide reduction with reduced sulp compounds as electron donors by anoxygenic phototrophic bacteria. Microbiology 1997, 143, 767–773. 36. deBont, J.A.M.; van Dijken, J.P.; Harder, W. Dimethyl sulphoxide and dimethyl sulfide as a car sulphur and energy source for growth of Hyphomicrobium. S. J. Gen. Microbiol. 1981, 127, 315–323. 37. Borodina, E.; Kelly, D.P.; Rainey, F.A.; Ward‐Rainey, N.L.; Wood, A.P. Dimethylsulfone as a growth substrate for novel methylotrophic species of Hyphomicrobium and Arthrobacter. Arch. Microbiol. 2000, 173, 425–437. 38. Kim, H.G.; Doronina, N.V.; Trostsenko, Y.A.; Kim, S.W. Methylophaga aminisulfidivorans sp. nov., a restricted facultatively methylotrophic marine bacterium. Int. J. Evol. Syst. Microbiol. 2007, 57, 2096–2101. 18 of 20 Microorganisms 2020, 8, 337 39. Smyth, T.J.; Fishwick, J.R.; Al‐Moosawi, L.; Cummings, D.G.; Harris, C.; Kitidis, V.; Rees, A.; Martinez‐Vicente, V.; Woodward, E.M.S.W. A broad spatio‐temporal view of the Western English Channel observatory. J. Plankton Res. 2010, 32, 585–601. 40. Tilstone, G.; Smyth, T.A.P.; Hutson, R. Measured and remotely sensed estimates of primary production in the Atlantic Ocean from 1998 to 2005. Deep Sea Res. II 2009, 56, 918−930. 41. Brewer, P.G.; Riley, J.P. The automatic determination of nitrate in sea water. Deep Sea Res. 1965, 12, 765– 772. 42. Grasshoff, H. Methods of Seawater Analysis; Verlag Chemie: Weinheim, Germany; New York, NY, USA, 1976; p. 317. 43. Zhang, J.Z.; Chi. J. Automated analysis of nanomolar concentrations of phosphate in natural waters with liquid waveguides. Environ. Sci. Technol. 2002, 36, 1048−105. 44. Tarran, G.A.; Bruun, J.T. Nanoplankton and picoplankton in the Western English Channel: Abundance and seasonality from 2007–2013. Prog. Oceanogr. 2015, 137, 446–455. 45. Widdicombe, C.E.; Eloire, D.; Harbour, D.; Harris, R.P.; Somerfield, P.J. References Franklin, D.J.; Poulton, A.J.; Steinke, M.; Young, J.; Peeken. I.; Malin, G. Dimethylsulphide, DMSP‐lyase activity and microplankton community structure inside and outside of the Mauritanian upwelling. Prog. Oceanogr. 2009, 83, 134–142. 61. Michaud, S.; Levasseur, M.; Cantin, G. Seasonal variations in dimethylsulfoniopropionate and dimethylsulfide concentrations in relation to the plankton community in the St. Lawrence Estuary. Estuar. Coast. Shelf Sci. 2007, 71, 741–750. 19 of 20 Microorganisms 2020, 8, 337 62. Speeckaert, G.; Borges, A.V.; Champenois, W.; Royer, C.; Gypens, N. Annual cycle of dimethylsulfoniopropionate (DMSP) and dimethylsulfoxide (DMSO) related to phytoplankton succession in the Southern North Sea. Sci. Total Environ. 2018, 622, 362–372. 63. Boden, R.; Murrell, J.C.; Schäfer, H. Dimethylsulfide is an energy source for the heterotrophic marine bacterium Sagittula stellata. FEMS Microbiol. Lett. 2011, 322, 188–193. 64. Cui, Y.; Suzuki, S.; Omori, Y.; Wong, A.‐K.; Ijichi, M.; Kaneko, R.; Kameyama, S.; Tanimoto, H.; Hamasaki, K. Abundance and distribution of dimethylsulfoniopropionate degradation genes and the corresponding bacterial community structure are dimethyl sulfide hot spots in the tropical and subtropical Pacific Ocean. Appl. Environ. Microbiol. 2015, 81, 4184–4194. 65. Newton, R.J.; Griffin, L.E.; Bowles, K.M.; Meile, C.; Gifford, S.; Givens, C.E.; Howard, E.C., King, E.; Oakley, C.A.; Reisch, C.Y.; et.al. Genome characteristics of a generalist marine bacterial lineage. ISME J. 2010, 4, 784–798, doi:10.1038/ismej.2009.150. 66. Kanukollu, S.; Wemheuer, B.; Herber, J.; Billerbeck, S.; Lucas, J.; Daniel, R.; Simon, M.; Cypionka, H.; Engelen, B. Distinct compositions of free‐living, particle‐associated and benthic communities of the Roseobacter group in the North Sea. FEMS Microbiol. Ecol. 2016, 92, fiv145 doi:10.1093/femsec/fiv145. 67. Thume, K.; Gebser, B.; Chen, L.; Meyer, N.; Kieber, D.J.; Pohnert, G. The metabolite dimethylsulfoxonium propionate extends the marine organosulfur cycle. Nature 2018, 563, 412–415, doi:10.1038/s41586‐018‐0675‐0. 68. Asher, E.; Dacey, J.W.; Ianson, D.; Peña, A.; Tortell, P.D. Concentrations and cycling of DMS, DMSP, and DMSO in coastal and offshore waters of the Subarctic Pacific during summer, 2010–2011. J. Geophys. Res. Oceans 2017, 122, 3269–3286. 69. Yang, M.; Archer, S.D.; Blomquist, B.W.; Ho, D.T.; Lance, V.P.; Torres, R.J. Lagrangian evolution of DMS during the Southern Ocean gas exchange experiment: The effects of vertical mixing and biological community shift. J. Geophys. Res. Oceans 2013, 118, 6774–6790. 70. Hatton, A.D.; Wilson, S.T. Particulate dimethylsulphoxide and dimethylsulphoniopropionate in phytoplankton cultures and Scottish coastal waters. Aquat. Sci. 2007, 69, 330‐340. 71. Keller, M.D.; Bellows, W.K.; Guillard, R.R.L. Dimethylsulfide production in marine phytoplankton. ACS Symp. Ser. References 1989, 393, 167–182. 72. Hatton, A.D.; Shenoy, D.M.; Hart, M.C.; Mogg, A.; Green, D.H. Metabolism of DMSP, DMS and DMSO by the cultivable bacterial community associated with the DMSP‐producing dinoflagellate Scrippsiella trochoidea. Biogeochemistry 2012, 110, 131–146. 73. Strom, S.; Wolfe, G.; Holmes, J.; Stecher, H.; Shimeneck, C.; Lambert, S. Chemical defense in the microplankton I: Feeding and growth rates of heterotrophic protists on the DMS‐producing phytoplankter Emiliania huxleyi. Limnol. Oceanogr. 2003, 48, 217–229. 74. Evans, C.; Kadner, S.V.; Darroch, L.J.; Wilson, W.H.; Liss, P.S.; Malin, G. The relative significance of viral lysis and microzooplankton grazing as pathways of dimethylsulfoniopropionate (DMSP) cleavage: An Emiliania huxleyi culture study. Limnol. Oceanogr. 2007, 52, 1036–1045. 75. Lee, P.A.; de Mora, S.J.; Levasseur, M. A review of dimethylsulfoxide in aquatic environments. Atmos. Ocean 1999, 37, 439–456. 76. Kiene, R.P.; Gerard, G. Determination of trace levels of dimethylsulphoxide (DMSO) in seawater and rainwater. Mar. Chem. 1994, 47, 1–12. 77. Dixon, J.L.; Beale, R.; Nightingale, P.D. Microbial methanol uptake in northeast Atlantic waters. ISME J. 2011, 5, 704–716. 78. Dixon, J.L.; Beale, R.; Nightingale, P.D. Rapid biological oxidation of methanol in the tropical Atla Significance as a microbial carbon source. Biogeosciences 2011, 8, 2707–2716. 79. Dixon, J.L.; Sargeant, S.; Nightingale, P.D.; Murrell, J.C. Gradients in microbial methanol uptake: Productive coastal upwelling waters to oligotrophic gyres in the Atlantic Ocean. ISME J. 2013, 7, 568–680 doi:10.1038/ismej.2012.130. 80. Sargeant, S.L.; Murrell, J.C.; Nightingale, P.D.; Dixon, J.L. Seasonal variability in microbial methanol utilisation in coastal waters of the western English Channel. Mar. Ecol. Prog. Ser. 2016, 550, 53–64, doi:10.3354/meps11705. 81. Sun, J.; Steindler, L.; Thrash, J.C.; Halsey, K.H.; Smith, D.P.; Carter, A.E.; Landry, Z.C.; Giovannoni, S.J. One carbon metabolism in SAR11 pelagic marine bacteria. PLoS ONE 2011, 6, e23973. 20 of 20 20 of 20 Microorganisms 2020, 8, 337 82. Halsey, K.H.; Giovannoni, S.J.; Graus, M.; Zhao, Y.; Landry, Z.; Thrash, J.C.; Vergin, K.L.; de Gouw, J. Biological cycling of volatile organic carbon by phytoplankton and bacterioplankton. Limnol. Oceanogr. 2017, 62, 2650–2661. 83. Kiene, R.P.; Linn, L.J.; Bruton, J.A. New and important roles for DMSP in marine microbial communities. J. Sea Res. 2000, 43, 209–224. 84. Vila‐Costa, M.; Simó, R.; Harada, H.; Gasol, J.M.; Slezak, D.; Kiene, R.P. Dimethylsulfoniopropionate uptake by marine phytoplankton. Science 2006, 314, 652–654. 85. Simon, M.; Azam, F. Protein content and protein synthesis rates of planktonic marine bacteria. Mar. Ecol. Prog. Ser. 1989, 51, 201–213. 86. © 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/). References Robinson, C. Heterotrophic bacterial respiration. In Microbial Ecology of the Oceans; Kirchman, D.L., Ed.; Wiley: Hoboken, NJ, USA, 2008; pp. 299–334. 87. Fagerbakke, K.M.; Heldal, M.; Norland, S. Content of carbon, nitrogen, oxygen, sulfur and phosphorus in native aquatic and cultured bacteria. Aquat. Microb. Ecol. 1996, 10, 15–27. 88. Kiene, R.P.; Linn, L.J.; González, J.; Moran, M.A.; Bruton, J.A. Dimethylsulfoniopropionate and methanethiol are important precursors of methionine and protein‐sulfur in marine bacterioplankton. Appl. Environ. Microbiol. 1999, 65, 4549–4558. © 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/W4378904843	https://www.mdpi.com/2313-0105/9/6/301/pdf?version=1685501542	English	null	Hybrid Modeling of Lithium-Ion Battery: Physics-Informed Neural Network for Battery State Estimation	Batteries	2,023	cc-by	15,091	Citation: Singh, S.; Ebongue, Y.E.; Rezaei, S.; Birke, K.P. Hybrid Modeling of Lithium-Ion Battery: Physics-Informed Neural Network for Battery State Estimation. Batteries 2023, 9, 301. https://doi.org/ 10.3390/batteries9060301 Keywords: Li-ion battery; battery modeling; state estimation; state of health (SOH); state of charge (SOC); hybrid modeling; physics-informed neural network (PINN); single-particle model (SPM) batteries batteries Soumya Singh 1,* , Yvonne Eboumbou Ebongue 1, Shahed Rezaei 2 and Kai Peter Birke 1, 1 Fraunhofer Institute for Manufacturing Engineering and Automation IPA, Nobelstr. 12, 70569 Stuttgart, Germany 1 Fraunhofer Institute for Manufacturing Engineering and Automation IPA, Nobelstr. 12, 70569 Stuttgart, Germany 2 Mechanics of Functional Materials Division, Institute of Materials Science, Technical University of Darmstadt, Otto-Berndt-Str. 3, 64287 Darmstadt, Germany 3 Institute for Photovoltaics, Electrical Energy Storage Systems, University of Stuttgart, Pfaffenwaldring 47, 70569 Stuttgart, Germany * C d i h@i f h f d 2 Mechanics of Functional Materials Division, Institute of Materials Science, Technical University of Darmstadt, Otto-Berndt-Str. 3, 64287 Darmstadt, Germany 3 Institute for Photovoltaics, Electrical Energy Storage Systems, University of Stuttgart, Pfaffenwaldring 47, 70569 Stuttgart, Germany * Correspondence: soumya.singh@ipa.fraunhofer.de Abstract: Accurate forecasting of the lifetime and degradation mechanisms of lithium-ion batteries is crucial for their optimization, management, and safety while preventing latent failures. However, the typical state estimations are challenging due to complex and dynamic cell parameters and wide variations in usage conditions. Physics-based models need a tradeoff between accuracy and complexity due to vast parameter requirements, while machine-learning models require large training datasets and may fail when generalized to unseen scenarios. To address this issue, this paper aims to integrate the physics-based battery model and the machine learning model to leverage their respective strengths. This is achieved by applying the deep learning framework called physics-informed neural networks (PINN) to electrochemical battery modeling. The state of charge and state of health of lithium-ion cells are predicted by integrating the partial differential equation of Fick’s law of diffusion from a single particle model into the neural network training process. The results indicate that PINN can estimate the state of charge with a root mean square error in the range of 0.014% to 0.2%, while the state of health has a range of 1.1% to 2.3%, even with limited training data. Compared to conventional approaches, PINN is less complex while still incorporating the laws of physics into the training process, resulting in adequate predictions, even for unseen situations. Article Hybrid Modeling of Lithium-Ion Battery: Physics-Informed Neural Network for Battery State Estimation Soumya Singh 1,* , Yvonne Eboumbou Ebongue 1, Shahed Rezaei 2 and Kai Peter Birke 1,3 1. Introduction In automotive applications, portable devices, or energy storage applications, lithium- ion batteries (LIB) need to operate within an optimal performance limit in order to avoid quick deterioration of the battery cells [1]. LIB aging and degradation occur due to a complex interplay of mechanisms that ultimately lead to the loss of active Li inventory or active electrode materials. These mechanisms can include the decomposition of the solid electrolyte interphase (SEI), Li-plating/dendrite formation, cracking and loss of contact, leaching and deposition of transition metals, gas formation, and other corrosive processes [2]. Degradation during the LIB lifecycle increases the difﬁculty of state estima- tion [3], consequently, accurate estimation and prediction of state of health (SOH) and state of charge (SOC) is a major bottleneck in the safe and reliable usage of LIBs [4]. Moreover, limited observability of cell parameters, environmental and operating conditions and un- certainties during battery modeling makes accurate battery state estimation a challenging task [5]. Academic Editor: Sylvain Franger Received: 31 March 2023 Revised: 4 May 2023 Accepted: 12 May 2023 Published: 30 May 2023 Received: 31 March 2023 Revised: 4 May 2023 Accepted: 12 May 2023 Published: 30 May 2023 batteries batteries Copyright: © 2023 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/). Despite the broader scope of the term “state estimation” in the literature, in this paper, this term refers SOH and SOC estimation, and does not include state of function (SOF)/power (SOP). https://www.mdpi.com/journal/batteries Batteries 2023, 9, 301. https://doi.org/10.3390/batteries9060301 Batteries 2023, 9, 301 2 of 19 Typically, a model that describes all the dynamic processes at multiple scales is not feasible [1], and the models need to be tailored for their application. Physics-based (PB) models have shown signiﬁcant success in describing the battery cell behavior, but they can be complex and computationally intensive, making their integration in real-time applications impractical. Also, the assumptions underlying common electrochemical transport equations may break down at large driving forces, and even the advanced PB models have limited applicability in SOH forecasting [6]. Conversely, classical black-box neural networks (NN) have been widely used for battery modeling, but they may not accurately capture the underlying physical mechanisms and require a large amount of training data. One point of emphasis in [6] is that machine-learning (ML) models are unlikely to bring high-accuracy health forecasting transferable to situations beyond the available data without considering physical processes. Another research gap noted in [7] is that although data-driven approaches have achieved satisfactory results for SOC and SOH estimation [8,9], no data-driven models have yet been proposed to monitor the dynamic changes of the unmeasurable electrochemical states within LIBs. Thus, there is a need for a novel approach that combines the two models and taps into their advantages. This study aims to integrate the PB battery model and the ML model to leverage their respective strengths to estimate the SOC and SOH of LIB cells. This is achieved through the extension of a concept called physics-informed neural networks (PINN) to an electrochemical battery model. Speciﬁcally, the partial differential equations (PDEs) for Fick’s diffusion law are incorporated in the loss function to train the NN. p PINN aims to integrate governing physics into data-driven models to develop an accurate, interpretable, and physically consistent ML method to handle imperfect data, ill-posed problems, inverse problems, and generalization tasks [10,11]. When using the classical black-box NN, there is a possibility that only partial data may be available to describe a certain behavior, leading to poor prediction performance. This limitation moti- vated the development of PINN, which incorporates physical laws as an additional source of information in practical calculations [12]. Additionally, training NNs to estimate the SOC and SOH requires a large amount of experimental data for training, which leads to long training periods, high costs, and energy requirements for performing the experiments. Adopting a hybrid approach that integrates relatively limited training data with knowl- edge of the internal phenomena of LIBs is a pivotal step toward achieving a signiﬁcant breakthrough in the ﬁeld of battery modeling. The following sections in this paper will explore the theory, implementation, and validation of PINNs as a promising approach for solving PDEs governing the electrochem- ical processes of LIBs. By leveraging the inherent physical constraints and regularities of the system, PINNs can effectively learn from limited and noisy data, handle complex geometries and boundary conditions, and capture nonlinear and multiphysics phenomena that are challenging for traditional modeling techniques. Furthermore, PINNs can facilitate the integration of real-time measurements and feedback into the model to improve its accuracy and adaptability, enabling advanced battery management strategies and diag- nostics. Therefore, exploring the PINN for battery modeling is essential for advancing the state-of-the-art in battery research and development and enabling the deployment of safer, more efﬁcient, and sustainable energy storage systems. gy g y The fundamental equations that govern battery modeling are derived from the princi- ples of mass/charge conservation and electrochemical reactions. These equations become more intricate when degradation mechanisms triggered during different phases of battery life and operating conditions are considered. The ﬁve principal degradation mechanisms are lithium plating, SEI growth, cathode and anode degradation, and loss of active mate- rial [13]. Consequently, the application of PINNs to an electrochemical battery model that depicts battery degradation presents a challenging and captivating frontier, which could ulti- mately facilitate the discovery of previously unknown state/parameter estimation techniques. y y p y p q The methodology employed in this study entails training a PINN framework that incorporates the equations governing the diffusion behavior of solid-phase Li-ions in the Batteries 2023, 9, 301 3 of 19 3 of 19 electrodes of a LIB. This is achieved by integrating Fick’s law PDE into the loss function of a NN. This ensures that the underlying physics of the problem is considered instead of relying solely on data ﬁtting. Therefore, this approach is expected to provide valuable insights into how integrating PB and ML models can enhance the ability to forecast the SOC and SOH of a battery. The research question addressed in this paper is: How can the integration of physics- based and machine-learning models be optimized to estimate the future state of batter- ies accurately? The main contributions of this study are summarized as follows: 1. Development of PINN for estimating the SOC and SOH of three LIB cells operating at different temperature ranges; 2. PINN implementation is tested in different Python packages in order to verify the transferability of the methodology in different platforms. This allows for the wider adoption and practical application of the model in diverse settings; 3. The model incorporates the governing equation of Fick’s law of diffusion behavior of solid-phase Li-ions to train the NN. This improves the accuracy of NN training and enhances the model’s predictive power. The rest of the paper is organized as follows: Section 2 gives an overview of the state of research on hybrid models and speciﬁcally PINN for battery modeling. In Section 3, the formulation of the problem for Fick’s diffusion equation is described. Section 4 details the architecture and implementation methodology of our implemented PINN. Section 5 describes the main results including the SOC and SOH estimationHH. Here, the comparison of PINN trained solely based on data, solely based on physics, and a combination of the two is also presented. Finally, Section 6 concludes the paper by summarizing signiﬁcant remarks and providing insight on potential future studies. 2. State of the Research: Hybrid Modeling of Lithium-Ion Batteries Similarly, a hybrid algorithm that combines model-based Kalman ﬁltering (KF) and a data-driven relevance vector machine (RVM) is proposed in [17] to offer capacity prognostic results. A number of other combinations for hybrid models are also available [18–20], which claim to provide accurate predictions of the battery’s behavior. We observe that in the literature, the topic hybrid models does not necessarily indicate a combination of physics and a data-driven model. For example, the approach taken by [21] involves the combination of an extreme learning machine and a support vector machine to predict the battery capacity. The implementation of hybrid battery models poses several challenges that require attention. Firstly, it is important to ensure that the dataset accurately represents a range of possible battery usage scenarios. Secondly, obtaining an optimal amount of training data can be challenging as it may not always be readily available. Another consideration is the computational intensity of calculations, which can increase the cost of algorithm utilization. Finally, external factors such as temperature and humidity may impact the performance of LIBs and need to be taken into consideration [22]. Physics-informed ML offers new physical insight in dynamic battery modeling under limited experimental measurements [6]. Raissi and colleagues are recognized pioneers in the research of PINNs, and their contribution led to the development of Python packages for their implementation. PINNs are trained to solve supervised learning tasks while respecting any given law of physics described by general nonlinear PDEs [23]. There is a rapidly growing body of research in the ﬁeld of multiphysics modeling, with successful examples in ﬂuid dynamics [24,25], solid mechanics [26], optics [27], metallurgy [28], and thermo-mechanically coupled systems [29]. The main advantage of PINN is that the physical laws are integrated into the network’s loss function. Upon being properly trained, one is conﬁdent that the network adheres to the physical laws even for unseen data in the domain of training [30]. To gain insight into the latest progress in research, it is necessary to dive into the section of literature that is solely dedicated to PINN for battery modeling. There is a limited body of research on the use of PINN for battery modeling, and our review revealed that only seven articles speciﬁcally addressed this topic. 2. State of the Research: Hybrid Modeling of Lithium-Ion Batteries In this paper, the term “hybrid battery model” refers to a simulation model that combines physics-based and data-driven approaches to analyze battery behavior. For a systematic study of the state of the research, the following Google Scholar and Web of Sci- ence syntaxes were investigated: ((“lithium-ion battery” OR “li-ion battery”) AND “hybrid battery model”) and (“battery model” AND (“lithium-ion battery” OR “li-ion battery”) AND (“physics-informed machine learning” OR “physics-informed neural network”)). From the resulting search, articles were ﬁltered based on their relevance to hybrid ap- proaches for Li-ion battery modeling. We conducted a comprehensive search and critically evaluated approximately 50 relevant research articles, reﬁning our inclusion criteria and quality assessment process as needed. To the best of our knowledge, this represents a thorough analysis of the existing literature on hybrid modeling of LIBs. Observations within the literature suggest that over the past few years, there has been a gradual increase in research focused on hybrid approaches for battery modeling. As a perspective on hybrid models for predicting battery lifetime in [6], the different architectures for integrating PB models and ML models were studied. These architectures are categorized into two broad categories: sequential integration of independent models and hybridized PB and ML models. Here, the authors highlighted that the former category is more viable in the near term as it can be realized by integrating existing ML and PB tools without any fundamental changes, while the latter category (i.e., the hybridized PB and ML architecture) will require the development of new approaches. The majority of references cited in the following paragraph pertain to the former category. The idea of sequential integration was brought into practice in [14], where a single particle model (SPM) with thermal dynamics was integrated with a feed-forward neural network (FNN) through residual learning and transfer learning. This approach could achieve high accuracy for predicting the terminal voltage behavior of the battery under various C-rates. Another method employed in [15] is where a recurrent neural network (RNN) captures the unmodeled dynamics of an SPM. The predictions of the RNN and the Batteries 2023, 9, 301 4 of 19 SPM are combined to enhance the voltage prediction accuracy. An adaptive hybrid model is constructed in [16], which is a combination of an empirical model and a long short-term memory (LSTM) NN model to characterize the battery capacity degradation. 3. Formulation of the Problem The commonly used LIB pseudo-two-dimensional (P2D) models provide deep insight into the evolution of internal battery dynamics based on intercalation, diffusion, and electrochemical kinetics modeled by PDEs [15]. The PDEs are subject to the following fundamental laws of physics, which dictate their behavior [38,39]: 1. Fick’s law of diffusion determined the solid-state Li-ion concentration (cs) in the electrodes; 1. Fick’s law of diffusion determined the solid-state Li-ion concentration (cs) in the electrodes; 2. The law of charge conservation determines the liquid-phase Li-ion concentration (ce) in the electrolyte and in the separator; y p 3. Ohm’s law determines the solid-state potential (φs) in the electrodes; 4. Kirchhoff’s and Ohm’s laws are used to calculate the liquid-phase potential (φe) in the electrolyte and in the separator; 5. The Butler–Volmer kinetics equation describes the ﬂux density of Li-ions (j). The Doyle–Fuller-Newman (DFN) model is a popular LIB model. A reduced-order form of a DFN model is the SPM, where each electrode is represented as a single spherical porous particle while neglecting electrolyte dynamics [39]. The intercalation process and mass transport are modeled over spherical coordinates. Figure 1 depicts the classic SPM, showcasing the dynamic concentration changes during the discharge process. The PDEs of a SPM model are in the microscopic spatial domain, i.e., the particle radius. This results in one-dimensional spatial variations, as opposed to the DFN model, which is two-dimensional, as it also includes the variable x, which is the electrode thickness. The reason behind this decoupling is that intercalation reactions typically occur uniformly across both electrodes. During discharge, negative electrode particles delithiate at nearly uniform rates, regardless of their position within the electrode, and a similar behavior is observed during the lithiation of positive electrode particles during charging. This is why a single representative particle in each electrode (i.e., a SPM) can adequately approximate the behavior of the entire electrode [40]. EW 6 of 20 Figure 1. Illustration of a single particle model adapted from [41]. Particle represents the concentra- tion gradient due to solid–phase diﬀusion. 𝑅௣ା and 𝑅௣ି are the particle radius and 𝑐௦ାሺ𝑟, 𝑡ሻ and 𝑐௦ିሺ𝑟, 𝑡ሻ are the solid–phase Li-ion concentration of the cathode and anode, respectively. 𝑉ሺ𝑡ሻ is the terminal voltage across cell. Figure 1. Illustration of a single particle model adapted from [41]. Particle represents the concen- tration gradient due to solid–phase diffusion. 2. State of the Research: Hybrid Modeling of Lithium-Ion Batteries In [22,31], the authors employ PINN using Nernst’s and the Butler–Volmer equations to describe the behavior of a LIB under different conditions, which is then integrated with a RNN. The model can predict the voltage discharge curves of batteries subjected to constant or random loading conditions. Ref. [7] presents PINN for state estimation of the electrodes. Here, the authors used FVM to numerically solve the electrochemical–thermal coupled model, which is used to generate the training data for an LSTM network. In the work of [32], the lumped capacitance thermal equation is used in the loss function of the PINN to predict the temperature of LIB cells. Here, the network is trained with battery test data, and the heat equation is used to identify the thermal behavior of the cell. Similarly, in [33,34], the authors implemented PINN with the heat generation equation and LSTM in order to predict the temperature of an entire battery pack. The objective of our study is similar to that presented in [35], which aimed to develop a NN-based approach for the long-term health prognosis of LIBs. The authors introduced a dynamic sliding window LSTM NN based on the Kullback–Leibler divergence measure and demonstrated that incorporating physical information into the model leads to improved accuracy compared to conventional approaches. While their method requires a large amount of monitoring data and simulation results to train the network, these ﬁndings align with our hypothesis and underscore the importance of integrating PB models with data-driven approaches for accurate battery state estimation. This paper is distinct from the above-described research on PINNs for battery model- ing due to the following factors: Firstly, to the best of the authors’ knowledge, this is one of the ﬁrst few attempts in the battery research ﬁeld to predict the SOH of LIB cells using Batteries 2023, 9, 301 5 of 19 5 of 19 PINN. Secondly, unlike existing research, this study does not rely on the sequential integra- tion of independent models with a speciﬁc type of NN disguised as a PINN. Thirdly, Fick’s diffusion equation has been used to train the PINN. Finally, while previous applications of PINN have only used Dirichlet boundary conditions, here, the network has been trained using Neumann boundary conditions, which further sets this study apart from previous research. This work is implemented using the PINN package from Raissi [36] as well as the SciANN package from Haghighat [37]. 3. Formulation of the Problem R+p and R−p are the particle radius and c+ s (r, t) and c− s (r, t) are the solid–phase Li-ion concentration of the cathode and anode, respectively. V(t) is the terminal voltage across cell. Figure 1. Illustration of a single particle model adapted from [41]. Particle represents the concentra- tion gradient due to solid–phase diﬀusion. 𝑅௣ା and 𝑅௣ି are the particle radius and 𝑐௦ାሺ𝑟, 𝑡ሻ and 𝑐௦ିሺ𝑟, 𝑡ሻ are the solid–phase Li-ion concentration of the cathode and anode, respectively. 𝑉ሺ𝑡ሻ is the terminal voltage across cell. Figure 1. Illustration of a single particle model adapted from [41]. Particle represents the concen- tration gradient due to solid–phase diffusion. R+p and R−p are the particle radius and c+ s (r, t) and c− s (r, t) are the solid–phase Li-ion concentration of the cathode and anode, respectively. V(t) is the terminal voltage across cell. Batteries 2023, 9, 301 6 of 19 6 of 19 Li-ions move from the negative electrode to the positive electrode during discharging and in the opposite direction during charging. Lithium concentration in the solid phase, i.e., concentration of Li-ions in the active materials, follows the laws of diffusion [42]. The electrochemical reaction that takes place in the electrode materials of LIBs involves two kinetic behaviors during charging and discharging: the insertion and extraction of Li-ions and the transfer of electrons/charges that occur during the oxidation or reduction of the electrode materials [43]. When discharging the battery, Li-ions migrate through the electrolyte from the negative electrode to the positive electrode and generate an electric current, while when charging the battery, Li-ions are released from the positive electrode and return to the negative electrode. The battery cells used in this study are prismatic cells called Lithium-ion Power Cell LP2714897-51Ah-BEV, with a cathode of NMC-622 and a graphite anode. Reaction equations for the discharging process of this cell chemistry are as follows: Anode: LixC6 →C6 + xLi+ + xe−; (1) (1) Cathode: Li1−xNi0.6Co0.2Mn0.2O2 + xLi+ + xe−→LiNi0.6Co0.2Mn0.2O2; (2) (2) Cell: Li1−xNi0.6Co0.2Mn0.2O2 + LixC6 →C6 + LiNi0.6Co0.2Mn0.2O2; (3) (3) As the electrochemical reaction occurs in the electrodes, there is a difference in the distribution of Li-ions on the surface and inside the electrode, creating a concentration gradient that drives Li-ion diffusion [43]. Solid-phase diffusion dynamics are central to an SPM and are represented by Fick’s second law of diffusion. 3. Formulation of the Problem C = cs,j c0 ; x = r Rj ; τ = Ds,jt R2 j (8) (8) This results in adaptation of the corresponding governing equation, boundary condi- tions, and initial condition as given below: This results in adaptation of the corresponding governing equation, boundary condi- tions, and initial condition as given below: ∂C ∂τ = 1 x2 ∂ ∂x x2 ∂C ∂x (9) (9) Batteries 2023, 9, 301 7 of 19 7 of 19 ∂C ∂x = 0 at x = 0 and τ ≥0 (10) (10) ∂C ∂x = −δ at x = 1 and τ ≥0 (11) (11) where δ is the applied dimensionless current density deﬁned as: where δ is the applied dimensionless current density deﬁned as: δ = ± I(t)Rj ALjFDs,jajc0 (12) (12) C = 1 at τ = 0 and 0 ≤x ≤1 (13) (13) The SOC of the particles can be calculated by using the electrodes’ minimum and maximum possible solid-phase concentrations [47]. SOC = cs,j −cmin cmax −cmin (14) (14) Instead of estimating the theoretical values of the maximum and minimum concen- trations of the active materials, the stoichiometry value at 0% SOC and at 100% SOC are used for calculating the SOC across the entire discharge time period. Stoichiometry refers to the proportion of the active materials in the electrodes of the battery. At 0% SOC, the battery is fully discharged, and the active materials in the anode (x0%) is in its highest state of oxidation (i.e., it has given up electrons). In contrast, at 100% SOC, the battery is fully charged, and the active materials in anode (x100%) is in its lowest state of oxidation (i.e., it has accepted electrons). SOC(t) = cs,j cs,max −x0% x100% −x0% (15) (15) The concentration of solid-phase Li-ions in the anode of a LIB cell is directly related to the cell capacity. A high concentration of Li-ions in the anode results in a higher battery capacity, while a lower concentration results in a lower capacity. Overall, the concentration of solid-phase Li-ions in the anode is a critical factor in determining the capacity and performance of LIBs [38]. In the context of model-based estimation, SOH refers to the actual performance of a battery, normalized by the performance when the battery was new. In fact, any battery parameter that changes with usage can be used as an indicator for SOH [5]. 3. Formulation of the Problem Common metrics of interest for a battery management system (BMS) are capacity, impedance and internal resistance. In this paper, capacity-based SOH is studied. The capacity across the time period can be derived using Equation (15). Cnominal is the nominal capacity provided by the manufacturer, and Cmax is the capacity at the beginning of the life of the cell. C(t) = SOC(t) × Cnominal (16) (C(t)) C(t) = SOC(t) × Cnominal (16) SOH = max(C(t)) Cmax at 100% SOC (17) C(t) = SOC(t) × Cnominal (16) (16) SOH = max(C(t)) Cmax at 100% SOC (17) (17) Hence by describing the diffusion problem and presenting the set of equations above, we have established a foundation for estimating the SOC and SOH of battery cells. Hence by describing the diffusion problem and presenting the set of equations above, we have established a foundation for estimating the SOC and SOH of battery cells. 3. Formulation of the Problem The governing differential equation and its accompanying conditions are explained as follows [44–46]: ∂cs,j(r, t) ∂t = Ds,j r2 ∂ ∂r r2 ∂cs,j(r, t) ∂r (4) (4) Here, cs,j is the Li-ion concentration in the solid particles, Ds,j is the solid-phase diffusion coefﬁcient, r is the radial coordinate, j = p corresponds to the positive electrode, and j = n corresponds to the negative electrode. The boundary conditions at the particle center (r = 0) and particle surface (r = Rj) are: Here, cs,j is the Li-ion concentration in the solid particles, Ds,j is the solid-phase diffusion coefﬁcient, r is the radial coordinate, j = p corresponds to the positive electrode, and j = n corresponds to the negative electrode. The boundary conditions at the particle center (r = 0) and particle surface (r = Rj) are: ∂cs,j(r, t) ∂r r=0 = 0 at r = 0 and t ≥0 (5) (5) ∂cs,j(r, t) ∂r r=Rj = ± I(t) ALjFDs,jaj at r = Rj and t ≥0 (6) (6) Here, I(t) is the charge/discharge current, A is the battery sheet area, Lj is the thickness of the positive and negative electrode, aj is the speciﬁc interfacial area: aj = 3εj/Rj, where εj is the solid phase volume fraction of each electrode and F is Faraday’s number. The initial condition is given by: Here, I(t) is the charge/discharge current, A is the battery sheet area, Lj is the thickness of the positive and negative electrode, aj is the speciﬁc interfacial area: aj = 3εj/Rj, where εj is the solid phase volume fraction of each electrode and F is Faraday’s number. The initial condition is given by: cs,j = c0 at t = 0 and 0 ≤r ≤Rj at f ully charged state (7) (7) Equations (4)–(7) are now simpliﬁed via the dimensionless variables introduced in [45]. 4. Architecture and Methodology They are widely used and well-documented open-source frameworks for automatic differentiation and deep learning computations. Automatic diﬀerentiation is employed to calculate the derivatives of the NN 𝐶ሺ𝑥, 𝜏ሻ with respect to the normalized time τ and space x, applying the chain rule for diﬀerenti- ating compositions of functions [51]. In order to compute the derivatives required in the above equations, modern deep learning frameworks such as Pytorch and Tensorﬂow [52] are used. They are widely used and well-documented open-source frameworks for auto- matic diﬀerentiation and deep learning computations. We approximate the unknown solution 𝐶ሺ𝑥𝜏ሻto the Fick’s diﬀusion equation by a We approximate the unknown solution C(x, τ) to the Fick’s diffusion equation by a deep NN. Consequently, the corresponding function for PINN takes the following form, f (x, τ): We approximate the unknown solution 𝐶ሺ𝑥, 𝜏ሻ to the Fick s diﬀusion equation by a deep NN. Consequently, the corresponding function for PINN takes the following form, 𝑓ሺ𝑥, 𝜏ሻ: 2 2 f𝑓≔ 𝐶ఛ െ 2 𝑥𝐶௫ ൅ 𝐶௫௫ ; (22) Cτ −2 x Cx + Cxx ; (22) (22) (22) The shared parameters between the NNs, 𝐶ሺ𝑥, 𝜏ሻ, and 𝑓ሺ𝑥, 𝜏ሻ are learned by mini- mizing the mean squared errors loss function. The shared parameters between the NNs, C(x, τ), and f (x, τ) are learned by minimiz- ing the mean squared errors loss function. 4. Architecture and Methodology In this section, we will ﬁrst provide a concise overview of the PINN algorithm pro- posed by Raissi et al. [48,49] for solving non-linear PDEs. For a more detailed understanding of the PINNs algorithm, we recommend interested readers refer to [23,50]. However, for instructional purposes, we explain the algorithm by applying it to the problem explained Batteries 2023, 9, 301 8 of 19 m pro- d 8 of 19 m pro- d in the previous section (Section 3) accompanied by Neumann boundary conditions. C(x, τ) is the unknown solution derived in Equation (9). for instructional purposes, we explain the algorithm by applying it to the problem ex- plained in the previous section (Section 3) accompanied by Neumann boundary condi- tions 𝐶ሺ𝑥𝜏ሻis the unknown solution derived in Equation (9) in the previous section (Section 3) accompanied by Neumann boundary conditions. C(x, τ) is the unknown solution derived in Equation (9). for instructional purposes, we explain the algorithm by applying it to the problem ex- plained in the previous section (Section 3) accompanied by Neumann boundary condi- tions 𝐶ሺ𝑥𝜏ሻis the unknown solution derived in Equation (9) Cτ = 2 x Cx + Cxx ; x ∈[0, 1], τ ∈ " 0, Ds,jt R2 j # (18) unknown solution derived in Equation (9). 𝐶ఛൌ2 𝑥𝐶௫൅𝐶௫௫ ; x ∈ሾ0, 1ሿ, τ ∈ቈ0, 𝐷௦,௝𝑡 𝑅௝ ଶ቉ (18) (18) (18) (see Equation (8)) (see Equation (8)) (see Equation (8)) (see Equation (8)) Cx(0, τ) = 0 (19) 𝐶௫ሺ0, 𝜏ሻൌ0 (19) (19) (19) Cx(1, τ) = ±δ (20) 𝐶௫ሺ1, 𝜏ሻൌ േ𝛿 (20) (20) (20) Here, the sign of δ depends on charge/discharge cycles in the Equation (12). Here, the sign of 𝛿 depends on charge/discharge cycles in the Equation (12). Here, the sign of δ depends on charge/discharge cycles in the Equation (12). Here, the sign of 𝛿 depends on charge/discharge cycles in the Equation (12). C(x, 0) = 1 (21) 𝐶ሺ𝑥, 0ሻൌ1 (21) d l l h d i i f h NN 𝐶ሺ ሻ (21) (21) ሻ Automatic differentiation is employed to calculate the derivatives of the NN C(x, τ) with respect to the normalized time τ and space x, applying the chain rule for differentiating compositions of functions [51]. In order to compute the derivatives required in the above equations, modern deep learning frameworks such as Pytorch and Tensorﬂow [52] are used. 4. Architecture and Methodology 𝑀𝑆𝐸ൌ𝑀𝑆𝐸଴൅𝑀𝑆𝐸ே஻஼൅ 𝑀𝑆𝐸௙ (23) MSE = MSE0 + MSENBC + MSEf (23) (23) (23) Here, Here, 𝑀𝑆𝐸଴ൌ1 𝑁଴ ෍ห𝐶൫𝑥଴ ௜, 0൯െ 1ห ଶ ேబ ௜ୀ଴ (24) MSE0 = 1 N0 N0 ∑ i=0 C xi 0, 0 −1 2 (24) MSENBC = 1 Nb Nb ∑ i=0 Cx 0, τi b −0 2 + Cx 1, τi b ± δ 2 (25) MSEf = 1 Nf Nf ∑ i=0 f xi f , τi f 2 (26) 𝑀𝑆𝐸଴ൌ1 𝑁଴ ෍ห𝐶൫𝑥଴ ௜, 0൯െ 1ห ଶ ேబ ௜଴ (24) MSE0 = 1 N0 N0 ∑ i=0 C xi 0, 0 −1 2 (24) (24) (24) ௜଴ MSENBC = 1 Nb Nb ∑ i=0 Cx 0, τi b −0 2 + Cx 1, τi b ± δ 2 (25) (25) MSEf = 1 Nf Nf ∑ i=0 f xi f , τi f 2 (26) (26) Here, xi 0 N0 i=0 denotes the initial data generated in this example by the initial condition. τi b Nb i=0 corresponds to the collocation points on the boundary, and n xi f , τi f oNf i=0 represents the collocation points on the residual network f (x, τ). Collocation points are speciﬁc points at which the loss functions are minimized in order to enforce the governing equations. They can be interpreted similarly to nodes in ﬁnite element methods [30]. Collocation points should not be confused with data points. Data points refer to the input-output pairs of data that are obtained from experiments or simulations. Figure 2 shows the collocation points used in this study across the spatio-temporal domain for solving Fick’s diffusion equation. In such a continuous time NN, a signiﬁcant number of collocation points Nf are required to enforce physics-informed constraints. Although this poses no signiﬁcant issues for problems in one or two spatial dimensions, it may introduce a bottleneck in higher-dimensional problems as the total number of Batteries 2023, 9, 301 9 of 19 spatia collocation points needed to globally enforce a physics-informed constraint (i.e., in our case, a PDE) will increase exponentially [49]. location points needed to globally enforce a physics informed constraint (i.e a PDE) will increase exponentially [49]. Figure 2. Collocation points across the training domain, distributed randomly acr temporal domain and at the initial and boundary conditions. Figure 2. 4. Architecture and Methodology Collocation points across the training domain, distributed randomly across the spatio- temporal domain and at the initial and boundary conditions. Figure 2. Collocation points across the training domain, distributed randomly acr temporal domain and at the initial and boundary conditions. Figure 2. Collocation points across the training domain, distributed randomly across the spatio- temporal domain and at the initial and boundary conditions. Consequently, the loss on the initial condition is represented by 𝑀𝑆𝐸 forces the Neumann boundary conditions, and 𝑀𝑆𝐸௙ penalizes any devia diﬀusion equation from satisfaction on the collocation points. This examp encompasses all the key aspects of the PINN algorithm and can be extende Consequently, the loss on the initial condition is represented by MSE0, MSENBC enforces the Neumann boundary conditions, and MSEf penalizes any deviation of Fick’s diffusion equation from satisfaction on the collocation points. This example effectively encompasses all the key aspects of the PINN algorithm and can be extended to different PDEs requiring individual treatment of boundary conditions. encompasses all the key aspects of the PINN algorithm and can be extende PDEs requiring individual treatment of boundary conditions. Next, this section will provide further details of our utilized architectu q g y Next, this section will provide further details of our utilized architecture and the approach for interacting with inputs and outputs. Our methodology draws inspiration from the contributions of [53,54]. proach for interacti the contributions o 4.1. Experimental Data Purple curve is the cell temperature; blue curve is the temperature at the negative terminal; black curve is the temperature at the positive terminal. Table 1. Cell speciﬁcations and input parameters for the model. Symbol Value Unit Denotation Cnominal 51 Ah Nominal Cell Capacity A 1.843 m2 Electrode surface area x0% 0.028 Stoichiometry anode (SOC = 0%) x100% 0.794 Stoichiometry anode (SOC = 100%) y0% 0.914 Stoichiometry cathode (SOC = 0%) y100% 0.344 Stoichiometry cathode (SOC = 100%) L+ 55.5 × 10−6 m Cathode thickness L− 67.3 × 10−6 m Anode thickness R+ 3.28 × 10−6 m Cathode particle radius R− 6.72 × 10−6 m Anode particle radius ε+ 0.755 Active material volume fraction cathode ε− 0.7 Active material volume fraction anode cis,max1 36,129.55 mol/m3 Maximum theoretical solid-phase concentration of Li+ ions in the electrodes I −17 to 17 A Charge and discharge current input t f (I) s Time corresponding to the input current 1 i = p for cathode and i = n for anode. Symbol Value Unit Denotation Cnominal 51 Ah Nominal Cell Capacity A 1.843 m2 Electrode surface area x0% 0.028 Stoichiometry anode (SOC = 0%) x100% 0.794 Stoichiometry anode (SOC = 100%) y0% 0.914 Stoichiometry cathode (SOC = 0%) y100% 0.344 Stoichiometry cathode (SOC = 100%) L+ 55.5 × 10−6 m Cathode thickness L− 67.3 × 10−6 m Anode thickness R+ 3.28 × 10−6 m Cathode particle radius R− 6.72 × 10−6 m Anode particle radius ε+ 0.755 Active material volume fraction cathode ε− 0.7 Active material volume fraction anode 𝑐௦,௠௔௫ ௜ 1 36,129.55 mol/m3 Maximum theoretical solid-phase concen- tration of Li+ ions in the electrodes I −17 to 17 A Charge and discharge current input t 𝑓ሺ𝐼ሻ s Time corresponding to the input current 1 i = p for cathode and i = n for anode Table 1. Cell speciﬁcations and input parameters for the model. Symbol Value Unit Cnominal 51 Ah No 1 i = p for cathode and i = n for anode. 1 i = p for cathode and i = n for ano 1 i = p for cathode and i = n for anode. 1 i = p for cathode and i = n for ano (b) (a) (b) (a) Figure 3. Capacity check−up cycle followed by two sets of WLTP driving cycle. (a) Constant cur- rent and constant voltage charge and constant current discharge during the check−up cycle. proach for interacti the contributions o 4.1. Experimental Data the contributions of [53,54]. 4.1. Experimental Data This section describes the experiments conducted for the study. As men the battery cells used are the prismatic cells called Lithium-ion Power Ce 51Ah-BEV, with a cathode of NMC-622 and a graphite anode. Cell speciﬁcat marized in Table 1. The cell dimensions and kinetic parameters of the cell This section describes the experiments conducted for the study. As mentioned before, the battery cells used are the prismatic cells called Lithium-ion Power Cell LP2714897-51Ah- BEV, with a cathode of NMC-622 and a graphite anode. Cell speciﬁcations are summarized in Table 1. The cell dimensions and kinetic parameters of the cell are obtained from a trusted third-party source. However, the exact approach used to determine the internal cell parameters cannot be disclosed due to conﬁdentiality restrictions. Three cells were subjected to consistent environmental conditions within climate chambers with designated ambient temperatures (Cell 3–25 ◦C, Cell 5–45 ◦C, and Cell 7–5 ◦C). p p ( ) The experimental test procedure for each cell involved an initial check-up at the beginning of life, followed by cycling through 100 worldwide harmonized light vehicle test procedures (WLTP) at a speciﬁc temperature, followed by check-up test cycles after every 100 WLTP cycles, and then a ﬁnal check-up at end-of-life. The climate chamber was brought to an ambient temperature of 25 ◦C, and only once the cell reached a thermal steady state, the check-up procedure was performed. The check-up procedure involved a differential voltage analysis, a pulse test, electrochemical impedance spectroscopy, and a capacity check. Only the capacity check-ups cycles are used in this study as an input to the PINN. However, to give an idea of the cycling tests performed during the experiments, Figure 3 represents both the check-up cycle followed by the driving cycle of cell 3. 10 of 19 . 10 of 19 . Batteries 2023, 9, 301 Table 1. Cell speciﬁcations and input parameters for the model. proach for interacti the contributions o 4.1. Experimental Data Symbol Value Unit Denotation Cnominal 51 Ah Nominal Cell Capacity A 1.843 m2 Electrode surface area x0% 0.028 Stoichiometry anode (SOC = 0%) x100% 0.794 Stoichiometry anode (SOC = 100%) y0% 0.914 Stoichiometry cathode (SOC = 0%) y100% 0.344 Stoichiometry cathode (SOC = 100%) L+ 55.5 × 10−6 m Cathode thickness L− 67.3 × 10−6 m Anode thickness R+ 3.28 × 10−6 m Cathode particle radius R− 6.72 × 10−6 m Anode particle radius ε+ 0.755 Active material volume fraction cathode ε− 0.7 Active material volume fraction anode cis,max1 36,129.55 mol/m3 Maximum theoretical solid-phase concentration of Li+ ions in the electrodes I −17 to 17 A Charge and discharge current input t f (I) s Time corresponding to the input current 1 i = p for cathode and i = n for anode. Symbol Value Unit Denotation Cnominal 51 Ah Nominal Cell Capacity A 1.843 m2 Electrode surface area x0% 0.028 Stoichiometry anode (SOC = 0%) x100% 0.794 Stoichiometry anode (SOC = 100%) y0% 0.914 Stoichiometry cathode (SOC = 0%) y100% 0.344 Stoichiometry cathode (SOC = 100%) L+ 55.5 × 10−6 m Cathode thickness L− 67.3 × 10−6 m Anode thickness R+ 3.28 × 10−6 m Cathode particle radius R− 6.72 × 10−6 m Anode particle radius ε+ 0.755 Active material volume fraction cathode ε− 0.7 Active material volume fraction anode 𝑐௦,௠௔௫ ௜ 1 36,129.55 mol/m3 Maximum theoretical solid-phase concen- tration of Li+ ions in the electrodes I −17 to 17 A Charge and discharge current input t 𝑓ሺ𝐼ሻ s Time corresponding to the input current 1 i = p for cathode and i = n for anode. (a) (b) Figure 3. Capacity check−up cycle followed by two sets of WLTP driving cycle. (a) Constant cur- rent and constant voltage charge and constant current discharge during the check−up cycle. Cells are subjected to WLTP cycles until the SOC reaches 5%; (b) temperature variations during cycling and check−ups. Purple curve is the cell temperature; blue curve is the temperature at the negative terminal; black curve is the temperature at the positive terminal. Figure 3. Capacity check−up cycle followed by two sets of WLTP driving cycle. (a) Constant current and constant voltage charge and constant current discharge during the check−up cycle. Cells are subjected to WLTP cycles until the SOC reaches 5%; (b) temperature variations during cycling and check−ups. proach for interacti the contributions o 4.1. Experimental Data Cells are subjected to WLTP cycles until the SOC reaches 5%; (b) temperature variations during cycling and check−ups. Purple curve is the cell temperature; blue curve is the temperature at the negative terminal; black curve is the temperature at the positive terminal. Figure 3. Capacity check−up cycle followed by two sets of WLTP driving cycle. (a) Constant current and constant voltage charge and constant current discharge during the check−up cycle. Cells are subjected to WLTP cycles until the SOC reaches 5%; (b) temperature variations during cycling and check−ups. Purple curve is the cell temperature; blue curve is the temperature at the negative terminal; black curve is the temperature at the positive terminal. 4.2. Data Preparation This subsection sheds light on data preparation for training and validation. The difference in the magnitude of input parameters is problematic when computing using ML algorithms. The presence of larger data points in non-normalized data sets can lead to instability in the NNs, as the relatively large inputs can propagate down the network layers and cause cascading effects. To prevent this issue, the input data for PINN should be normalized. This involves converting the original values to new values and bringing most parameters into the same range. This allows the gradient descent optimization algorithm to converge faster and obtain a mean close to zero since the learning speed is proportional to the magnitude of the inputs. For our network, the equation (Equations (9)–(13)) resulting from the dimensionless parameters are used to setup the network. The input parameters required for PINN are also summarized in Table 1. 4.3. Architecture In this work, the output variable (i.e., concentration) is approximated via a fully connected FNN. The network has three types of layers. The ﬁrst layer is the input layer. Batteries 2023, 9, 301 11 of 19 11 of 19 Next, there are hidden layers in between. Finally, there is the output layer. Each layer can contain several neurons, and there is no connection between neurons inside a layer. Moreover, each layer is connected to the next layer to pass information from the input layer to the output layer. A neuron transforms an input x into an output as follows: Next, there are hidden layers in between. Finally, there is the output layer. Each layer can contain several neurons, and there is no connection between neurons inside a layer. Moreover, each layer is connected to the next layer to pass information from the input layer to the output layer. A neuron transforms an input x into an output as follows: z(x) = α(Wx + b) (27) (27) where α is a chosen nonlinear parameter called activation function, W is the weight matrix, and b is a correction term called bias. Each input x is given a weight W. The weights determine the degree of inﬂuence the neuron’s inputs have on calculating subsequent activation. The purpose of an activation function is to add nonlinearity to the NN, and a proper choice of the activation function is problem dependent. They should be differentiable so that the parameters of the network are learned in backpropagation. The backpropagation algorithm works by computing the gradient of the loss function with respect to each weight and bias in the network, starting from the output layer and working backward through the layers. The gradients are computed using the chain rule. Once the gradients are computed, they are used to update the weights and biases using a technique called gradient descent. In gradient descent, the weights and biases are adjusted in the direction that reduces the loss function the most by multiplying the gradient by a learning rate and subtracting it from the current values of the weights and biases. This process is repeated iteratively until the loss function reaches a minimum or a convergence criterion is met. Various optimization techniques can be employed to improve the effective- ness of the weight and bias updates in the NN. 4.3. Architecture The quations Batteries 2023, 9, 301 Figure 4. Architecture and loss functions of the implemented PINN for LIB diﬀusion problem. Figure 4. Architecture and loss functions of the implemented PINN for LIB diffusion problem. Figure 4. Architecture and loss functions of the implemented PINN for LIB diﬀusion problem. Figure 4. Architecture and loss functions of the implemented PINN for LIB diffusion problem. The above presented architecture diﬀers from previous architectures of PINN be- cause, here, the training is done on the basis of Neumann boundary conditions instead of the previously used Dirichlet boundary conditions. This is achieved by creating a separate serial input (current and time) for the boundary condition. The Neumann boundary con- dition is then solved in a separate FNN trained only on the basis of the calculated value of the maximum boundary condition. The above presented architecture differs from previous architectures of PINN because, here, the training is done on the basis of Neumann boundary conditions instead of the previously used Dirichlet boundary conditions. This is achieved by creating a separate serial input (current and time) for the boundary condition. The Neumann boundary condition is then solved in a separate FNN trained only on the basis of the calculated value of the maximum boundary condition. 4.3. Architecture Moreover, the hyper-parameters affect the network’s performance, such as activation function, learning rate, number of epochs, batch size, hidden layers, and neurons in each layer. The effect of the choice of hyper-parameters on the network is studied in [30]. The optimization methods and hyper-parameters selected in this paper are elaborated in Section 4.4. The PINN formulation for the diffusion problem of LIBs can be depicted in Figure 4. The network is based on Equations (4)–(26). The inputs for the PINN model are a spatial vector, particle radius (r) across a grid size of 100, and a temporal vector, time (t). The two inputs are then used to construct a mesh, which is a set of points in space and time used to approximate the solution to a given PDE. Along with this, the network also gets the concentration value from the measurement data. This is derived using Equation (15). The rate of change in SOC across the time series input as well as the stoichiometry ratios at 0% and 100% SOC need to be known for training purposes. Additionally, the current subjected to the cell is also given as a time series input used for calculating the molar ﬂux in Equation (12). The trainable set of parameters are {W,b}. Solid-phase concentration of lithium is the output of the network and is a function of the trainable parameters. Training is performed by minimizing the total loss. The solution to the diffusion problem using PINN is acquired by ensuring that the NN satisﬁes the diffusion equation at every point in the domain. This is done by adding a term to the loss function that penalizes any deviation from the diffusion equation. Speciﬁcally, the loss function is given by: Loss = ω1Lossdata + ω2Lossphysics (28) (28) Loss = ω1Lossdata + ω2Lossphysics (28) where Lossdata is the data-ﬁtting term that measures the difference between the predicted and the actual concentration, Lossphysics is the physics-informed term that enforces the diffusion equation, and ω is the weighting factor that balances the two terms. The physics- informed term is formulated by taking the gradients of the predicted concentration with respect to time and space and plugging them into the diffusion equation. The output of the PINN is the solid-phase concentration. Using this solution, the Equations (15)–(18) are executed, which yield the time-dependent SOC and SOH of the battery cell. 12 of 19 ion. 5. Results and Discussion 5. Results and Discussion D i 5 Results and Discussion (Right) Change in the Li+ concentration across the anode particle radius at the end of the CC discharge cycle. Figure 5. (Left) Change in the Li+ concentration across the anode particle radius at the beginning of the CC discharge cycle. (Right) Change in the Li+ concentration across the anode particle radius at the end of the CC discharge cycle. Figure 5. (Left) Change in the Li+ concentration across the anode particle radius at the beginning of the CC discharge cycle. (Right) Change in the Li+ concentration across the anode particle radius at the end of the CC discharge cycle. Figure 5. (Left) Change in the Li+ concentration across the anode particle radius at the beginning of the CC discharge cycle. (Right) Change in the Li+ concentration across the anode particle radius at the end of the CC discharge cycle. Figure 5. (Left) Change in the Li+ concentration across the anode particle radius at the beginning of the CC discharge cycle. (Right) Change in the Li+ concentration across the anode particle radius at the end of the CC discharge cycle. Figure 5. (Left) Change in the Li+ concentration across the anode particle radius at the beginning of the CC discharge cycle. (Right) Change in the Li+ concentration across the anode particle radius at the end of the CC discharge cycle. Figure 6. (Left) Change in the Li+ concentration across the anode particle radius at the beginning of the CCCV charge cycle. (Right) Change in the Li+ concentration across the anode particle radius at the end of the CC discharge cycle. Figure 6. (Left) Change in the Li+ concentration across the anode particle radius at the beginning of the CCCV charge cycle. (Right) Change in the Li+ concentration across the anode particle radius at the end of the CC discharge cycle. During discharge, Li-ions leave the anode and diffuse through the electrolyte and separator to the cathode. As a result, the Li-ion concentration of the anode decreases over time and radius. In reality, at a SOC of 100%, there is still a signiﬁcant amount of Li-ions intercalated in the cathode, which is why the anode does not reach the maximum theoretical concentration. The rate of change in anode concentration at the center of the particle, i.e., at r = 0, is almost constant, as per the minimum boundary condition (Equation (5)). 5. Results and Discussion 5. Results and Discussion D i 5 Results and Discussion Due to space constraints, we have included only the plots of the anode in this section. However, we would like to note that, similar to the anode behavior, the PINN used for deriving the concentration change across the cathode also satisﬁed the physical constraints. Due to space constraints, we have included only the plots of the anode in this section. However, we would like to note that, similar to the anode behavior, the PINN used for deriving the concentration change across the cathode also satisﬁed the physical con- straints 5. Results and Discussion Due to space constraints, we have included only the plots of the anode in this section. However, we would like to note that, similar to the anode behavior, the PINN used for deriving the concentration change across the cathode also satisﬁed the physical con- The following set of results describes how the Li-ion concentration changes inside a particle driven by a concentration gradient based on Fick’s law of diffusion. Figure 5 represents the anode particle at the beginning and end of a CC discharge cycle, where the SOC changes from 100% to 0%. Similarly, Figure 6 shows the anode particle at the beginning and end of a CCCV charge cycle, where the SOC changes from 0% to 100%. straints. The following set of results describes how the Li-ion concentration changes inside a particle driven by a concentration gradient based on Fick’s law of diﬀusion. Figure 5 rep- resents the anode particle at the beginning and end of a CC discharge cycle, where the SOC changes from 100% to 0%. Similarly, Figure 6 shows the anode particle at the begin- ning and end of a CCCV charge cycle, where the SOC changes from 0% to 100%. deriving the concentration change across the cathode also satisﬁed the physical con straints. The following set of results describes how the Li-ion concentration changes inside a particle driven by a concentration gradient based on Fick’s law of diﬀusion. Figure 5 rep- resents the anode particle at the beginning and end of a CC discharge cycle, where the SOC changes from 100% to 0%. Similarly, Figure 6 shows the anode particle at the begin- i d d f CCCV h l h th SOC h f 0% t 100% Figure 5. (Left) Change in the Li+ concentration across the anode particle radius at the beginning of the CC discharge cycle. 4.4. Training, Validation and Testing Data 4.4. Training, Validation and Testing Data 4.4. Training, Validation and Testing Data Input data for the PINN are split among training, validation, and testing data. Radial and time vectors are merged into a mesh {X,T} to represent each location at each time point of the particle. To solve the PDE, a predeﬁned number of collocation points over the entire mesh {X,T} is used. The data are split into 80% training and 20% validation data. Diﬀerent ratios, such as 70:30 and 90:10, were also tested, but a trade-oﬀ is reached with respect to Input data for the PINN are split among training, validation, and testing data. Radial and time vectors are merged into a mesh {X,T} to represent each location at each time point of the particle. To solve the PDE, a predeﬁned number of collocation points over the entire mesh {X,T} is used. The data are split into 80% training and 20% validation data. Different ratios, such as 70:30 and 90:10, were also tested, but a trade-off is reached with respect to the training time and performance on the test set. The entire mesh is then given as input to the PINN for the testing process. By monitoring the performance on the validation split during training, we deﬁne when to stop the training in order to avoid overﬁtting. Additionally, a termination tolerance of 1 × 10−5 is set to ensure that the training stops when the loss function improvement falls below the given threshold. p g The activation function utilized in this work is Hyperbolic-Tangent (tanh) function. Minimization of the loss function is completed by the Adam optimizer [55]. The initial learning rate is set to 1 × 10−3 and the ﬁnal learning rate to 1/100 of the initial learning rate. ‘ExponentialDecay’ learning rate scheduler is then used to gradually reduce the learning rate over time. In order to prevent overﬁtting, the ‘EarlyStopping’ callback function is used to monitor the training process. Adaptive weights are used to dynamically adjust the weights assigned to each loss term in the total loss function during training. This is implemented using the GradNorm method [56]. This is done in order to address the issue of gradient pathology, where loss terms with higher derivatives tend to dominate the total gradient vector and negatively affect the accuracy of the solution. Batteries 2023, 9, 301 13 of 19 of al e of 5. Results and Discussion 5. Results and Discussion D i 5 Results and Discussion At the particle surface, i.e., at r = Rp, the rate of change in Li-ion concentration depends on the molar ﬂux density, as per the maximum boundary condition (Equation (6)). When training the PINN, the loss function consists of multiple components, as ex- plained in Equation (23). The loss function measures how well the NN is satisfying the governing PDE of the problem. A decreasing trend of the losses indicates that the PINN is learning the governing physics, and boundary conditions and progressing toward a better approximation of the solution. Minimization of the loss function is affected by the choice of hyper-parameters such as learning rate, number of epochs, type of optimizer, and number of input samples. Therefore, selecting appropriate hyperparameters is crucial Batteries 2023, 9, 301 14 of 19 by the imizer, for optimizing the loss function. Figure 7 shows that around 500 epochs are needed to converge the optimization process. Measures taken to avoid the overﬁtting of data were already elaborated in Section 4.4. for optimizing the loss function. Figure 7 shows that around 500 epochs are needed to converge the optimization process. Measures taken to avoid the overﬁtting of data were already elaborated in Section 4.4. for optimizing the loss function. Figure 7 shows that around 500 epochs are needed to converge the optimization process. Measures taken to avoid the overﬁtting of data were already elaborated in Section 4.4. for optimizing the loss function. Figure 7 shows that around 500 epochs are needed to converge the optimization process. Measures taken to avoid the overﬁtting of data were already elaborated in Section 4.4. Figure 7. PINN training loss function. In order to verify the diﬀusion behavior across the anode, an approach described in the reference paper [47] is used. Here, a novel analytical approach to determind the Li- concentration within the electrode particles is derived and coupled to the ECM and im- plemented in MATLAB. The output of the analytics model is mainly used to verify the linear and quadratic behavior of the concentration curves, given the same set of input parameters. Figure 7. PINN training loss function. In order to verify the diffusion behavior across the anode, an approach described in the reference paper [47] is used. Here, a novel analytical approach to determind the Li-concentration within the electrode particles is derived and coupled to the ECM and implemented in MATLAB. 5. Results and Discussion 5. Results and Discussion D i 5 Results and Discussion The output of the analytics model is mainly used to verify the linear and quadratic behavior of the concentration curves, given the same set of input parameters. Figure 7. PINN training loss function. Figure 7. PINN training loss function. Figure 7. PINN training loss function. Figure 7. PINN training loss function. In order to verify the diﬀusion behavior across the anode, an approach described in the reference paper [47] is used. Here, a novel analytical approach to determind the Li- concentration within the electrode particles is derived and coupled to the ECM and im- plemented in MATLAB. The output of the analytics model is mainly used to verify the linear and quadratic behavior of the concentration curves, given the same set of input parameters. In order to verify the diffusion behavior across the anode, an approach described in the reference paper [47] is used. Here, a novel analytical approach to determind the Li-concentration within the electrode particles is derived and coupled to the ECM and implemented in MATLAB. The output of the analytics model is mainly used to verify the linear and quadratic behavior of the concentration curves, given the same set of input parameters. In order to verify the diﬀusion behavior across the anode, an approach described in the reference paper [47] is used. Here, a novel analytical approach to determind the Li- concentration within the electrode particles is derived and coupled to the ECM and im- plemented in MATLAB. The output of the analytics model is mainly used to verify the linear and quadratic behavior of the concentration curves, given the same set of input parameters. In order to verify the diffusion behavior across the anode, an approach described in the reference paper [47] is used. Here, a novel analytical approach to determind the Li-concentration within the electrode particles is derived and coupled to the ECM and implemented in MATLAB. The output of the analytics model is mainly used to verify the linear and quadratic behavior of the concentration curves, given the same set of input parameters. 5.2. State of Health Estimation 5.2. State of Health Estimation The PINN is executed ite y In order to investigate the diﬀerences between training the NN solely on data or solely on the underlying physics, as opposed to training on a combination of both data In order to investigate the differences between training the NN solely on data or solely on the underlying physics, as opposed to training on a combination of both data and physics, we conducted a specialized training procedure on cell number 3. y In order to investigate the diﬀerences between training the NN solely on data or solely on the underlying physics, as opposed to training on a combination of both data In order to investigate the differences between training the NN solely on data or solely on the underlying physics, as opposed to training on a combination of both data and physics, we conducted a specialized training procedure on cell number 3. and physics, we conducted a specialized training procedure on cell number 3. The procedure is that during training, the diﬀerent components of the loss function are selectively included in or excluded from the loss function (see Equation (28)). If the network is to be trained solely on the basis of physical constraints, the component of the loss function that represents the data is excluded (𝜔ଵൌ0 and 𝜔ଶൌ1), thereby resulting The procedure is that during training, the different components of the loss function are selectively included in or excluded from the loss function (see Equation (28)). If the network is to be trained solely on the basis of physical constraints, the component of the loss function that represents the data is excluded (ω1 = 0 and ω2 = 1), thereby resulting in a curve that represents ‘Only Physics’ (see Figure 10). Conversely, if the network is to be trained solely on the basis of data, the component of the loss function that represents the physical constraints is omitted (ω1 = 1 and ω2 = 0), thereby resulting in a curve that represents ‘Only Data’. If both components of the loss function have been included (ω1 = 1 and ω2 = 1), the resulting curve represents ‘PINN’. This approach allowed us to assess the relative impact of the available data and governing equations on the training process and to determine the extent to which each factor contributes to the overall performance of the network. 5.2. State of Health Estimation 5.2. State of Health Estimation The PINN is executed ite 5.2. State of Health Estimation 5.2. State of Health Estimation The PINN is executed ite The PINN is executed iteratively across all check-up capacity test cycles to determine the SOH of the cells. At the beginning of each new check-up cycle, the concentration value must be reset to the output value of the previous cycle. SOH is calculated by determining the maximum available capacity with respect to the concentration of Li+ in the anode, as per Equations (16) and (17). Figure 9 shows the measured and predicted SOH plots for cell number 5 and cell number 7. The RMSE of SOH estimation for cell 5 is 2.381% and for cell 7 is 1.176%. The change in temperature for cell cycling does not majorly affect the PINN performance for SOH estimation. The PINN is executed iteratively across all check-up capacity test cycles to determine the SOH of the cells. At the beginning of each new check-up cycle, the concentration value must be reset to the output value of the previous cycle. SOH is calculated by determining the maximum available capacity with respect to the concentration of Li+ in the anode, as per Equations (16) and (17). Figure 9 shows the measured and predicted SOH plots for cell number 5 and cell number 7. The RMSE of SOH estimation for cell 5 is 2.381% and for cell 7 is 1.176%. The change in temperature for cell cycling does not majorly aﬀect the PINN performance for SOH estimation. (b) (a) (b) Figure 9. (a) SOH estimation of cell 5 across 1100 cycles; (b) SOH estimation of cell 5 across 1000 cycles. Figure 9. (a) SOH estimation of cell 5 across 1100 cycles; (b) SOH estimation of cell 5 across 1000 cycles. (a) (b) (a) Figure 9. (a) SOH estimation of cell 5 across 1100 cycles; (b) SOH estimation of cell 5 across 1000 cycles Figure 9. (a) SOH estimation of cell 5 across 1100 cycles; (b) SOH estimation of cell 5 across 1000 cycles. 5 1 State of Charge Estimation 5.1. State of Charge Estimation 5.1. State of Charge Estimation Figure 8 shows the output plots of the estimated SOC compared to the measured SOC during a complete charge and complete discharge cycle. For every new check-up cycle, the PINN is trained for multiple cycles, and the change in SOC with respect to time is estimated using Equation (15). The quadratic behavior of the concentration due to the PDE in Equation (9) is responsible for the nonlinear behavior of the PINN SOC estimation. Figure 8 shows the output plots of the estimated SOC compared to the measured SOC during a complete charge and complete discharge cycle. For every new check-up cycle, the PINN is trained for multiple cycles, and the change in SOC with respect to time is estimated using Equation (15). The quadratic behavior of the concentration due to the PDE in Equation (9) is responsible for the nonlinear behavior of the PINN SOC estimation. Across the capacity check-up cycles, it is noted that the root mean square error (RMSE) for SOC estimation lies between 0.014% and 0.2%. W 15 of 20 Across the capacity check-up cycles, it is noted that the root mean square error (RMSE) for SOC estimation lies between 0.014% and 0.2%. (a) (b) Figure 8. SOC calculated from the estimation of lithium concentration in the anode: (a) Discharge cycle of cell 3 during the 1st check-up cycle; (b) charge cycle of cell 3 during the 1st check-up cycle. 5 2 State of Health Estimation Figure 8. SOC calculated from the estimation of lithium concentration in the anode: (a) Discharge cycle of cell 3 during the 1st check-up cycle; (b) charge cycle of cell 3 during the 1st check-up cycle. (b) (a) (b) Figure 8. SOC calculated from the estimation of lithium concentration in the anode: (a) Discharge cycle of cell 3 during the 1st check-up cycle; (b) charge cycle of cell 3 during the 1st check-up cycle. 2 S f H l h E i i Figure 8. SOC calculated from the estimation of lithium concentration in the anode: (a) Discharge cycle of cell 3 during the 1st check-up cycle; (b) charge cycle of cell 3 during the 1st check-up cycle. Batteries 2023, 9, 301 15 of 19 scharge l 5.2. State of Health Estimation 5.2. State of Health Estimation The PINN is executed ite 5.2. State of Health Estimation 5.2. State of Health Estimation The PINN is executed ite Hence, the ﬁrst set is u to estimate the SOH hile the second set is utili ed to forecast the SOH Table 2. RMSE of SOH prediction. PINN Only Data Only Physics RMSE 1.32% 3.48% 3.94% The results of this approach are shown in Figure 10, which shows the conver behavior of the network during training for diﬀerent combinations of data and ph based inputs. To quantify the accuracy of the predictions, we computed the RMSE fo of the three types of curves, as reported in Table 2. The ﬁndings suggest that the com tion of data and physical constraints yields improved prediction accuracy compa using the other two approaches in isolation. Speciﬁcally, the RMSE values associate the combined approach are lower than those associated with the purely data-dri hy i ba ed a oa he Thi u e t that by le e a i both data a d hy i By systematically varying the proportion of data and physics-based inputs used to train the network, we could quantify each factor’s relative inﬂuence on the resulting performance. By allowing the model developer to adjust the weights of the terms based on their level of conﬁdence in the available data and underlying physical principles, we get a model with high ﬂexibility to incorporate both data-driven and physics-based information. These ﬁndings provide valuable insights into the underlying mechanisms that drive the success of PINN and have important implications for the development of more accurate and robust hybrid battery models with the potential to generalize to new and unseen scenarios. physics-based approaches. This suggests that by leveraging both data and physics constraints, we are able to achieve a more accurate model. Table 2. RMSE of SOH prediction. PINN Only Data Only Physi RMSE 1.32% 3.48% 3.94% By systematically varying the proportion of data and physics-based inputs u train the network we could quantify each factor’s relative inﬂuence on the resultin The results of this study demonstrate the potential of using PINN to accurately es- timate SOC and SOH of LIBs operating at three different temperature ranges, even with limited training data and relatively low computational complexity and training time. 5.2. State of Health Estimation 5.2. State of Health Estimation The PINN is executed ite Additionally, to facilitate testing and evaluation of the model, the input data are partitioned into two sets such that the ﬁrst eight check-up cycles are reserved for training, while the remaining ﬁve check-up cycles are used for testing. Hence, the ﬁrst set is utilized to estimate the SOH, while the second set is utilized to forecast the SOH. The results of this approach are shown in Figure 10, which shows the convergence behavior of the network during training for different combinations of data and physics- based inputs. To quantify the accuracy of the predictions, we computed the RMSE for each of the three types of curves, as reported in Table 2. The ﬁndings suggest that the combination of data and physical constraints yields improved prediction accuracy compared to using the other two approaches in isolation. Speciﬁcally, the RMSE values associated with the combined approach are lower than those associated with the purely data-driven or physics-based approaches. This suggests that by leveraging both data and physics-based constraints, we are able to achieve a more accurate model. 16 of 19 g p performa Batteries 2023, 9, 301 Figure 10. SOH estimation of cell 3 across 1300 cycles trained until the 8th check-up cycle and afterwards until the 13th check-up cycle. Comparison of SOH trained solely based on data, based on physics, and a combination of the two. Figure 10. SOH estimation of cell 3 across 1300 cycles trained until the 8th check-up cycle and tested afterwards until the 13th check-up cycle. Comparison of SOH trained solely based on data, solely based on physics, and a combination of the two. Figure 10. SOH estimation of cell 3 across 1300 cycles trained until the 8th check-up cycle and afterwards until the 13th check-up cycle. Comparison of SOH trained solely based on data based on physics, and a combination of the two. Figure 10. SOH estimation of cell 3 across 1300 cycles trained until the 8th check-up cycle and tested afterwards until the 13th check-up cycle. Comparison of SOH trained solely based on data, solely based on physics, and a combination of the two. Additionally to facilitate Table 2. RMSE of SOH prediction. Additionally, to facilitate testing and evaluation of the model, the input data a titioned into two sets such that the ﬁrst eight check-up cycles are reserved for tr while the remaining ﬁve check-up cycles are used for testing. 6. Conclusions In conclusion, this paper proposes a novel approach for hybrid battery modeling using PINN and demonstrates its effectiveness in estimating the SOC and SOH of LIB cells. By incorporating the laws of physics into the training process, the proposed approach could produce adequate predictions even for unseen situations. The RMSE of SOH prediction using the developed PINN was 1.32%, 2.381%, and 1.176% for cell 3, cell 5, and cell 7, respectively. Additionally, the SOC prediction exhibited a variance of 0.014% to 0.2%. The ﬁndings of this study reveal the capability of PINN to enable ﬂexibility in battery modeling. Speciﬁcally, the ﬂexibility arises from the ability to select a data-driven model when sufﬁcient training data are available. However, in cases where the amount of training data is insufﬁcient, PINNs can still enhance the accuracy of the model by incorporating prior knowledge of the underlying physics. Therefore, this is potentially a valuable tool for BMS algorithms. g Future work includes applying this approach to other battery chemistries and optimiz- ing the PINN architecture for better performance. Looking forward, the proposed hybrid model will be integrated with the digital twin of a battery cell and battery module to reﬂect the actual aging and degradation of the physical battery. Using PINN for battery digital twin implementation to predict the SOH, we can learn the physical dynamics of a battery in terms of differential equations and automatically update the model parameters based on their solutions. This is a novel approach for implementing the model component of a battery digital twin that can learn the physics of battery aging and reﬂect it in the lifetime predictions made by the model. This study contributes to bridging the gap between traditional physics-based modeling and modern data-driven ML techniques, as shown by the performance of the proposed approach compared to a purely physics-based model and a classical black-box NN. Author Contributions: Conceptualization, S.S.; methodology, S.S.; software, S.S. and Y.E.E.; formal analysis, S.R.; writing—original draft preparation, S.S.; review and editing, Y.E.E., S.R. and K.P.B.; supervision, K.P.B.; funding acquisition, K.P.B. All authors have read and agreed to the published version of the manuscript. Funding: This research was funded by the German Federal Ministry for Economic Affairs and Climate Action (BMWK), funding code 19I21014C. Data Availability Statement: Data were obtained from project partners ElringKlinger and is not publically available due to third party restrictions. 5.2. State of Health Estimation 5.2. State of Health Estimation The PINN is executed ite This approach offers advantages in scenarios where the amount of available data changes adap- tively, such as at the beginning of a battery’s life when experimental data acquisition is extensive, but during operation after many usage cycles, experimental data is no longer valid or possible to acquire. Thus, for such scenarios, PINN is well-suited for application in a battery digital twin because automatic parameter updates are one of the main require- ments of battery DT implementation [57]. Further work will focus on extending the PINN approach and integrating it into a battery digital twin. train the network, we could quantify each factor s relative inﬂuence on the resultin formance. By allowing the model developer to adjust the weights of the terms bas their level of conﬁdence in the available data and underlying physical principles, w a model with high ﬂexibility to incorporate both data-driven and physics-based mation. These ﬁndings provide valuable insights into the underlying mechanism drive the success of PINN and have important implications for the development of While the accuracy of this approach cannot be claimed to always outperform existing state estimation approaches in all operational scenarios, this study provides a promising starting point for the systematic integration of PDEs of P2D models into NNs. Moreover, a limitation of this study is the assumption of a constant diffusion coefﬁcient due to isothermal conditions, whereas in real-world scenarios, the diffusion coefﬁcient changes with battery aging and affects the concentration behavior at the electrodes. Further exploration is needed to overcome challenges associated with developing hybrid battery models using PINN. Future research can also investigate different network architectures and their effects on integration with PDEs. Additionally, the proposed PINN can be expanded to include additional physical constraints, such as the heat generation Batteries 2023, 9, 301 17 of 19 equation or the Butler–Volmer equation. Finding the optimal balance between the ﬂexibility of machine learning and the accuracy of physics will be a crucial area of future investigation. equation or the Butler–Volmer equation. Finding the optimal balance between the ﬂexibility of machine learning and the accuracy of physics will be a crucial area of future investigation. 6. Conclusions Acknowledgments: We thank Alexander Fill from the Institute for Photovoltaics, University of Stuttgart for his support with the MATLAB analytical model used for veriﬁcation. The authors gratefully acknowledge BMWK for the funding resources and the project partners ElringKlinger AG and Huber Automotive AG. Conﬂicts of Interest: The authors declare no conﬂict of interest. g p p 5. Noura, N.; Boulon, L.; Jemeï, S. A Review of Battery State of Health Estimation Methods: Hybrid Electric Vehicle Challenges. WEVJ 2020, 11, 66. [CrossRef] g y g [ ] 4. Shu, X.; Shen, S.; Shen, J.; Zhang, Y.; Li, G.; Chen, Z.; Liu, Y. State of health prediction of lithium-ion ba learning: Advances and perspectives. iScience 2021, 24, 103265. [CrossRef] g p p J , , [ ] 2. Agubra, V.; Fergus, J. Lithium Ion Battery Anode Aging Mechanisms. Materials 2013, 6, 1310–1325. [CrossRef] [PubMed] 3. Zhou, L.; Lai, X.; Li, B.; Yao, Y.; Yuan, M.; Weng, J.; Zheng, Y. State Estimation Models of Lithium-Ion Batteries for Battery Management System: Status, Challenges, and Future Trends. Batteries 2023, 9, 131. [CrossRef] 3. Zhou, L.; Lai, X.; Li, B.; Yao, Y.; Yuan, M.; Weng, J.; Zheng, Y. State Estimation Models of Lithiu Management System: Status, Challenges, and Future Trends. Batteries 2023, 9, 131. [CrossRef] References Edge, J.S.; O’Kane, S.; Prosser, R.; Kirkaldy, N.D.; Patel, A.N.; Hales, A.; Ghosh, A.; Ai, W.; Chen, J battery degradation: What you need to know. Phys. Chem. Chem. Phys. 2021, 23, 8200–8221. [CrossR J.S.; O’Kane, S.; Prosser, R.; Kirkaldy, N.D.; Patel, A.N.; Hales, A.; Ghosh, A.; Ai, W.; Chen, J.; Yang, J.; et a g , J ; , ; , ; y, ; , ; , ; , ; , ; , J ; g, J ; ttery degradation: What you need to know. Phys. Chem. Chem. Phys. 2021, 23, 8200–8221. [CrossRef] y g y y y 14. Tu, H.; Moura, S.; Wang, Y.; Fang, H. Integrating physics-based modeling with machine learning for lithium-ion batteries. Appl. Energy 2023, 329, 120289. [CrossRef] gy 15. Park, S.; Zhang, D.; Moura, S. Hybrid electrochemical modeling with recurrent neural networks for li-ion batteries. In Proceedings of the 2017 American Control Conference (ACC), Seattle, WA, USA, 24–26 May 2017; pp. 3777–3782. [CrossRef] rk, S.; Zhang, D.; Moura, S. Hybrid electrochemical modeling with recurrent neural networks for li-ion batteri the 2017 American Control Conference (ACC), Seattle, WA, USA, 24–26 May 2017; pp. 3777–3782. [CrossRe y pp ian, C.; He, L. Short-Term Prediction of Remaining Life for Lithium-Ion Battery Based on Adaptive Hybrid M 16. He, N.; Qian, C.; He, L. Short-Term Prediction of Remaining Life for Lithium-Ion Battery Based on Adaptive Hybrid Model With Long Short-Term Memory Neural Network and Optimized Particle Filter. J. Electrochem. Energy Convers. Storage 2022, 19, 031004. [CrossRef] 17. Zheng, X.; Wu, H.; Chen, Y. Remaining useful life prediction of lithium-ion battery using a hybrid model-based ﬁltering and data-driven approach. In Proceedings of the 2017 11th Asian Control Conference (ASCC), Gold Coast, Australia, 17–20 December 2017; pp. 2698–2703. [CrossRef] 18. Zou, G.; Song, L.; Yan, Z. Lithium-ion Battery Remaining Useful Life Prediction Based on Hybrid Model. In Proceedings of the 2022 7th International Conference on Intelligent Computing and Signal Processing (ICSP), Xi’an, China, 15–17 April 2022; pp. 714–719. 19. Liu, D.; Li, L.; Song, Y.; Wu, L.; Peng, Y. Hybrid state of charge estimation for lithium-ion battery under dynamic operating conditions. Int. J. Electr. Power Energy Syst. 2019, 110, 48–61. [CrossRef] 20. Li, Z.; Fang, H.; Yan, Y. An Ensemble Hybrid Model with Outlier Detection for Prediction of Lithium-ion Battery Remaining Useful Life. In Proceedings of the 2019 Chinese Control And Decision Conference (CCDC), Nanchang, China, 3–5 June 2019; pp. References 5. Noura, N.; Boulon, L.; Jemeï, S. A Review of Battery State of Health Estimation Methods: Hybrid Electric Vehicle Challenges. WEVJ 2020, 11, 66. [CrossRef] 18 of 19 18 of 19 Batteries 2023, 9, 301 6. Aykol, M.; Gopal, C.B.; Anapolsky, A.; Herring, P.K.; van Vlijmen, B.; Berliner, M.D.; Bazant, M.Z.; Braatz, R.D.; Chueh, W.C.; Storey, B.D. Perspective—Combining Physics and Machine Learning to Predict Battery Lifetime. J. Electrochem. Soc. 2021, 168, 30525. [CrossRef] [ ] 7. Li, W.; Zhang, J.; Ringbeck, F.; Jöst, D.; Zhang, L.; Wei, Z.; Sauer, D.U. Physics-informed neural network estimation in lithium-ion batteries. J. Power Sources 2021, 506, 230034. [CrossRef] hang, J.; Ringbeck, F.; Jöst, D.; Zhang, L.; Wei, Z.; Sauer, D.U. Physics-informed neural networks for electrode n in lithium-ion batteries. J. Power Sources 2021, 506, 230034. [CrossRef] g, J.; Ringbeck, F.; Jöst, D.; Zhang, L.; Wei, Z.; Sauer, D.U. Physics informed neural networks for electrode lev lithium-ion batteries. J. Power Sources 2021, 506, 230034. [CrossRef] J [ ] 8. Li, W.; Sengupta, N.; Dechent, P.; Howey, D.; Annaswamy, A.; Sauer, D.U. Online capacity estimation of lithium-ion batteries with deep long short-term memory networks. J. Power Sources 2021, 482, 228863. [CrossRef] ngupta, N.; Dechent, P.; Howey, D.; Annaswamy, A.; Sauer, D.U. Online capacity estimation of lithium-ion bat 8. Li, W.; Sengupta, N.; Dechent, P.; Howey, D.; Annaswamy, A.; Sauer, D.U. Online capacity estimation of deep long short-term memory networks. J. Power Sources 2021, 482, 228863. [CrossRef] 9. Song, X.; Yang, F.; Wang, D.; Tsui, K.-L. Combined CNN-LSTM Network for State-of-Charge Estimation of Lithium-Ion Batteries. IEEE Access 2019, 7, 88894–88902. [CrossRef] 10. Karniadakis, G.E.; Kevrekidis, I.G.; Lu, L.; Perdikaris, P.; Wang, S.; Yang, L. Physics-informed machine learning. Nat. Rev. Phys. 2021, 3, 422–440. [CrossRef] 11. Cai, S.; Mao, Z.; Wang, Z.; Yin, M.; Karniadakis, G.E. Physics-informed neural networks (PINNs) for ﬂuid mechanics: A review. Acta Mech. Sin. 2021, 37, 1727–1738. [CrossRef] 12. Cuomo, S.; Di Cola, V.S.; Giampaolo, F.; Rozza, G.; Raissi, M.; Piccialli, F. Scientiﬁc machine learning through physics–informed neural networks: Where we are and what’s next. J. Sci. Comput. 2022, 92, 88. [CrossRef] ne, S.; Prosser, R.; Kirkaldy, N.D.; Patel, A.N.; Hales, A.; Ghosh, A.; Ai, W.; Chen, J.; Yang, J.; et al. Lithium ion: What you need to know. Phys. Chem. Chem. Phys. 2021, 23, 8200–8221. [CrossRef] 13. References A Physics-Informed Machine Learning Approach for Estimating Lithium-Ion Battery Temperature. IEEE Access 2022, 10, 88117–88126. [CrossRef] p 33. Pang, H.; Wu, L.; Liu, J.; Liu, X.; Liu, K. Physics-informed neural network approach for heat generation rate estimation of lithium-ion battery under various driving conditions. J. Energy Chem. 2023, 78, 1–12. [CrossRef] y g gy 34. Cho, G.; Zhu, D.; Campbell, J.J.; Wang, M. An LSTM-PINN Hybrid Method to Estimate Lithium-Ion Battery Pack Temperature. IEEE Access 2022, 10, 100594–100604. [CrossRef] 35. Sun, B.; Pan, J.; Wu, Z.; Xia, Q.; Wang, Z.; Ren, Y.; Yang, D.; Guo, X.; Feng, Q. Adaptive evolution enhanced physics-informed neural networks for time-variant health prognosis of lithium-ion batteries. J. Power Sources 2023, 556, 232432. [CrossRef] 36. GitHub. Maziarraissi/PINNs: Physics Informed Deep Learning: Data-Driven Solutions and Discovery of Nonlinear Partial Differential Equations. Available online: https://github.com/maziarraissi/PINNs (accessed on 30 March 2023). q p g 37. GitHub. Ehsanhaghighat/Sciann-Applications: A Place to Share Problems Solved with SciA https://github.com/ehsanhaghighat/sciann-applications (accessed on 30 March 2023). p g g g pp 38. Li, C.; Cui, N.; Wang, C.; Zhang, C. Simpliﬁed electrochemical lithium-ion battery model with variable solid-phase diffusion and parameter identiﬁcation over wide temperature range. J. Power Sources 2021, 497, 229900. [CrossRef] p p g 39. Jokar, A.; Rajabloo, B.; Désilets, M.; Lacroix, M. Review of simpliﬁed Pseudo-two-Dimensional models of lithium-ion batteries. J. Power Sources 2016, 327, 44–55. [CrossRef] 40. Brosa Planella, F.; Ai, W.; Boyce, A.M.; Ghosh, A.; Korotkin, I.; Sahu, S.; Sulzer, V.; Timms, R.; Tranter, T.G.; Zyskin, M.; et al. A continuum of physics-based lithium-ion battery models reviewed. Prog. Energy 2022, 4, 42003. [CrossRef] p y y g gy 41. Al-Gabalawy, M.; Hosny, N.S.; Hussien, S.A. Lithium-Ion Battery Modeling Including Degradation Based on Single-Particle Approximations. Batteries 2020, 6, 37. [CrossRef] pp 42. Tang, S.-X.; Camacho-Solorio, L.; Wang, Y.; Krstic, M. State-of-Charge estimation from a therm lithium-ion batteries. Automatica 2017, 83, 206–219. [CrossRef] ho-Solorio, L.; Wang, Y.; Krstic, M. State-of-Charge estimation from a thermal–electrochemical model of s. Automatica 2017, 83, 206–219. [CrossRef] 43. Physics of Electron and Lithium-Ion Transport in Electrode Materials for Li-Ion Batteries. Available online: https://cpb.iphy.ac. cn/article/2016/1806/cpb_25_1_18206.html#outline_anchor_4 (accessed on 30 March 2023). p 44. Liu, J.; Li, G.; Fathy, H.K. A Computationally Efﬁcient Approach for Optimizing Lithium-Ion Battery Charging. J. Dyn. Syst. Meas. Control 2016, 138, 021009. [CrossRef] 45. Subramanian, V.R.; Ritter, J.A.; White, R.E. Approximate Solutions for Galvanostatic Discharge of Spherical Particles I. Constant Diffusion Coefﬁcient. J. Electrochem. References Soc. 2001, 148, E444. [CrossRef] 46. Dey, S.; Ayalew, B. Real-Time Estimation of Lithium-Ion Concentration in Both Electrodes of a Lithium-Ion Battery Cell Utilizing Electrochemical–Thermal Coupling. J. Dyn. Syst. Meas. Control 2017, 139, 031007. [CrossRef] 47. Fill, A.; Bubeck, L.; Knauer, D.; Hemmerling, J.; Birke, K.P. Multi-dimensional model for electrical, thermal and mechanical simulation of a Lithium-ion Cell. In Proceedings of the NEIS 2022; Conference on Sustainable Energy Supply and Energy Storage Systems, Hamburg, Germany, 26–27 September 2022; pp. 1–8. y g y p pp 48. Raissi, M. Deep hidden physics models: Deep learning of nonlinear partial differential equations. J. Mach. Learn. Res. 2018, 19, 932–955. 49. Maziar Raissi\|Physics Informed Deep Learning. Available online: https://maziarraissi.github.io/research/1_physics_informed_ neural_networks/ (accessed on 30 March 2023). 50. Raissi, M.; Perdikaris, P.; Karniadakis, G.E. Physics Informed Deep Learning (Part II): Data-driven Disco Differential Equations. arXiv 2017, arXiv:1711.10561. s, P.; Karniadakis, G.E. Physics Informed Deep Learning (Part II): Data-driven Discovery of Nonlinear Partial ns. arXiv 2017, arXiv:1711.10561. 51. Baydin, A.G.; Pearlmutter, B.A.; Radul, A.A.; Siskind, J.M. Automatic differentiation in machine learning: A survey. J. Mach. Learn. Res. 2018, 18, 1–43. 52. Abadi, M.; Agarwal, A.; Barham, P.; Brevdo, E.; Chen, Z.; Citro, C.; Corrado, G.S.; Davis, A.; Dean, J.; Devin, M.; et al. TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems. arXiv 2016, arXiv:1603.04467. 53. Raissi, M.; Perdikaris, P.; Karniadakis, G.E. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. J. Comput. Phys. 2019, 378, 686–707. [CrossRef] 54. Haghighat, E.; Juanes, R. SciANN: A Keras/TensorFlow wrapper for scientiﬁc computations and physics-informed deep learning using artiﬁcial neural networks. Comput. Methods Appl. Mech. Eng. 2021, 373, 113552. [CrossRef] g p pp g , , [ ] 55. Kingma, D.P.; Ba, J. Adam: A Method for Stochastic Optimization. arXiv 2014, arXiv:1412.6980. 56. Chen, Z.; Badrinarayanan, V.; Lee, C.-Y.; Rabinovich, A. GradNorm: Gradient Normalization for Adaptive Loss Balancing in Deep Multitask Networks. Proc. 35th Int. Conf. Mach. Learn. 2018, 80, 794–803. 57. Singh, S.; Weeber, M.; Birke, K.P. Implementation of Battery Digital Twin: Approach, Functionalities and Beneﬁts. Batteries 2021, 7, 78. [CrossRef] Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). References 2630–2635, ISBN 978-1-7281-0106-4. 21. Chen, X.-D.; Yang, H.-Y.; Wun, J.-S.; Wang, C.-H.; Li, L.-L. Life prediction of lithium-ion battery based on a hybrid model. Energy Explor. Exploit. 2020, 38, 1854–1878. [CrossRef] 22. Nascimento, R.G.; Corbetta, M.; Kulkarni, C.S.; Viana, F.A. Hybrid physics-informed neural network modeling and prognosis. J. Power Sources 2021, 513, 230526. [CrossRef] Corbetta, M.; Kulkarni, C.S.; Viana, F.A. Hybrid physics-informed neural networks for lithium-ion battery nosis. J. Power Sources 2021, 513, 230526. [CrossRef] 23. Raissi, M.; Perdikaris, P.; Karniadakis, G.E. Physics Informed Deep Learning (Part I): Data-Driven Solutions of Nonlinear Partial Differential Equations. arXiv 2017, arXiv:1711.10561. q 24. Mao, Z.; Jagtap, A.D.; Karniadakis, G.E. Physics-informed neural networks for high-speed ﬂows. Comput. Methods Appl. Mech. Eng. 2020, 360, 112789. [CrossRef] g 25. Sun, L.; Gao, H.; Pan, S.; Wang, J.-X. Surrogate modeling for ﬂuid ﬂows based on physics-constrained deep learning without simulation data. Comput. Methods Appl. Mech. Eng. 2020, 361, 112732. [CrossRef] 26. Vahab, M.; Haghighat, E.; Khaleghi, M.; Khalili, N. A Physics-Informed Neural Network Approach to Solution and Identiﬁcation of Biharmonic Equations of Elasticity. arXiv 2021, arXiv:2108.07243. [CrossRef] 27. Jiang, X.; Wang, D.; Fan, Q.; Zhang, M.; Lu, C.; Lau, A.P.T. Physics-Informed Neural Network for N Optics. arXiv 2021, arXiv:2109.00526. [CrossRef] 27. Jiang, X.; Wang, D.; Fan, Q.; Zhang, M.; Lu, C.; Lau, A.P.T. Physics-Informed Neural Network for Nonlinear Dynamics in Fiber Optics. arXiv 2021, arXiv:2109.00526. [CrossRef] A.; Schmitz, G.J.; Apel, M. Theory-training deep neural networks for an alloy solidiﬁcation benchmark Mater. Sci. 2020, 180, 109687. [CrossRef] 28. Rad, M.T.; Viardin, A.; Schmitz, G.J.; Apel, M. Theory-training deep neural networks for an alloy s problem. Comput. Mater. Sci. 2020, 180, 109687. [CrossRef] 29. Harandi, A.; Moeineddin, A.; Kaliske, M.; Reese, S.; Rezaei, S. Mixed formulation of physics-informed neural networks for thermo-mechanically coupled systems and heterogeneous domains. arXiv 2023, arXiv:2302.04954. 30. Rezaei, S.; Harandi, A.; Moeineddin, A.; Xu, B.-X.; Reese, S. A Mixed Formulation for Physics-Informed Neural Networks as a Potential Solver for Engineering Problems in Heterogeneous Domains: Comparison with Finite Element Method. arXiv 2022, arXiv:2206.13103. [CrossRef] 31. Nascimento, R.G.; Corbetta, M.; Kulkarni, C.S.; Viana, F.A. Li-ion Battery Aging with Hybrid Physics-Informed Neural Networks and Fleet-wide Data. In Proceedings of the Annual Conference of the Prognostics and Health Management Society, Virtual, 29 November–2 December 2021. [CrossRef] 19 of 19 19 of 19 Batteries 2023, 9, 301 32. Cho, G.; Wang, M.; Kim, Y.; Kwon, J.; Su, W. Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. References MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
https://openalex.org/W3118766400	https://zenodo.org/records/1872511/files/article.pdf	English	null	Some Links between Natural History and Medicine	Scientific American	1,901	public-domain	6,783	PARASITES AS L1:'<iK';. But let me now pass to illustrate another kind of link-namely, in connection with parasites. Here nat ur!J,l history and medicine have over and over again 10ined hands, and it does not seem too much to say that, without the labors of zoologists in working out life-histories and inter-relations, the medical treat ment of parasites would have remained very empirical. The two illustrations which I take both concern mos quitos. As long ago as 1878 Dr. Patrick Manson called attention to the part played by mosquitos in spreading a minute threadworm parasite, well known as Filaria sanguinis hominis. His work has been confirmed by several workers, especially by Bancroft. Though the story has not been quite completed, it is almost certain that it will end as Manson and Bancroft predict. The story is this: The mature parasites live in the lym phatic vessels in man, and are by no means very small, measuring from 3 inches to 4 inches in length. They give rise to minute embryo filarial, which live iIi. the blOOd-vessels, swimming about when the host is sleep ing, resting when he is active, perhaps because of the difference in the caliber of the capillaries during wak ing and sleeping hours. They cause very serious dis ease, which I need not try to discuss. Now, when a mosquito bites an infected subject during the night it takes away with the drop of blood some embryo filarial. From the mosquito's stomach these embryos bore to its muscles, especially in the thorax, and grow with great rapidity for seventeen days. It used to be stated that the mosquitos only fed once and died within a week, but Bancroft points out that this was because the learned doctors omitted to feed the insects. By giv ing the introduced house mosquitos of Australia a lit tle banana to suck, Bancroft kept them alive for weeks -indeed, for a couple of months. It is believed, though it has not been proved, that no further change will occur in the young filaria within the mosquito un less it pass somehow into man, there to develop into the full-grown worm of 3 inches or 4 inches. Bancroft has suggested that a life-sentenced prisoner might be of some use to the world if he could be induced to swallow some mosquitos bearing filarial, on condition of receiving a free pardon. By J. AI(,["HU]{ THOMSON, M.A., F.R.S.E., Professor of Natural History in the University of Aberdeen. By J. AI(,["HU]{ THOMSON, M.A., F.R.S.E., Professor of Natural History in the University of Aberdeen. LET me thank you, first of all, for your kindness in giving me this opportunity of addressing you, which I can assure you I regard as a pleasure and an honor. As my work does not lie exa9tly alongside of yours. I thought we might find a convenient line of contact by considering for a little some of the links between nat ural history and medicine. But as the day has already doubtless been long to some of you, as to me, we may perhaps consult our mutual convenience by trying to observe the virtue of brevity, and be content with il lustrations of a subject which obviously affords ma terial for prolonged discourse. The links between nat ural history and medicine which I wish to-night to H lustrate are many and varied. Some were forged cen turies ago, and some of these have since snapped; oth ers are strictly modern; others, such as the role of mosquitos in spreading malaria, are at this moment in process of being formed and tested. I take a second illustration from the treatment of snake-bite. The venom in this case is well known to be of proteid nature, not alkaloid like that of toads and sa.lamanders. It is the secretion of a specialized salivary gland. (a) The natives of several coun tries have been in the habit of drinking diluted snake poison to render themselves immune; this is certain ly approached by the modern recommendation-e. g., by Prof. Fraser, of Edinburgh, and Dr. Calmette, of Paris, that an anti-toxin serum, prepared from snake venom, should be injected into the person bitten. (b) In many African snake medicines the bile of the snake is an important constituent; this is certainly ap proached by the recent evidence furnished by Fraser and Phisalix, that the bile of a venomous snake is an antidote, or even more than that, a preventive against the fatal effects of a bite. (c) An old Indian remedy for the bite of a certain snake is to chew the root of a labiate plant-Pagostemon puricaulis; this is certainly approached by the discovery of Phisalix, that the juice of the dahlia tubers, or of the fungus Russula, has immunizing properties against the vi per's venom. MOSQUITOS AND :lIALAIUA. That mosquitos play a part in infecting man with malaria is an old and widespread belief, to which re cent experimental work has given a scientific basis. An elaborate argument in favor of the popular belief was published by King in 1883, and since that date many workers have made contributions toward a so lution of the problem-Mansan and Ross in Britain, Laveran and Calmette in France, Koch and Pfeiffer in Germany, Bignani, Mendini, and Grassi in Italy. The question of priority is as difficult as it is tiresome. The general arguments in favor of the hypothesis, well summed up by Nuttall, are very suggestive: The malarial season usuallY corresponds with a period of warmth and moisture-conditions which favor the de velopment of mosquitos; malarial regions are mos quito regions; devices which protect against mosquitos -curtains, cities, drainage and so on-also protect against malaria; these are three of the many general arguments. But, suggestive as these arguments are, they do not prove the correctness of the mosquito-ma laria hypothesis. Experimental evidence was neces· sary, and it has been forthcoming. Following a sug gestion of Manson's, Ro", exposed malarial subjects in India to mosquitos, and observed that crescent-shaped parasites (Proteosoma) passed with the drops of blood from man to mosquito, and underwent a further development in the insect host. He also showed that only some of the many different kinds of mosquitos were suitable intermediate hosts for the malaria-para site. Furthermore, he succeeded in producing mala rial infection in birds-e. g., sparrows and crows-by means of infected mosquitos. He was led to the c)n clusion, since confirmed by others, that malarial in fection results from the bites ot mosquitos whose sali vary glands contain the parasite. J need not try to follow the numerous researches of other workers, but it is important to notice that Grassi and Bignani suc ceeded in infecting a patient with malaria as the re sult of the bites of a mosquito (Anopheles claviger). and in tracing the development of the parasites in those same m:osquitos after they had bitten a malarial patient. We have here a good instance of the con· firmation of general arguments by experimental evi dence. It is evident that the gentle art of using the leech is decadent. In The British Medical Journal for March 18, 1898 (I quote on the authority of Dr. PARASITES AS L1:'<iK';. It is difficult to believe that a prisoner would, in the circumstances, strain at a gnat. But how does the filaria normally get into man? It used to be supposed that the exhausted mos quito fell into water tanks, that the filarial emerged, and passed into man when he drank the infected wa ter. But there is no evidence that the filarial ever lib erate themselves in this way from their mosquito host, and Bancroft has shown that the young filarial put into water die in three or four hours. Therefore, water cannot be the vehicle of infection. It is now supposed either that man must swallow an infected mosquito or that he may put his finger into his mouth after killing one; or that, when an infected mosquito is bit ing man, the filarial, excited by the entrance of the warm blood, may burst into the mosquito's gullet and pass down its proboscis into man. This part of the story remains obscure. : This table shows that nearly 70 per cent of the world's pig iron production is now made into steel, leaving only a little more than 20 per cent to be made into the various forms and kinds of iron. And yet Bessemer's first patent was granted as recently as February 12, 1856, and in the same year Siemens Brothers obtained the.r first p .tent for the use of the regenerative gas furnace in the open·hearth process of steel manufacture. f The second class of ancient prescriptions includes those which may have something in them, for the ob· vious reason that the animals used contain certain potent or at least useful chemical substances. It is not surprising that decoctions of ants should have been found to have at least some effect, for ants COD' tain formic acid; it is not surprising that powdered pearls-a dose for the wealthy-might be of some service, since pe-arls are made of lime; it is not sur prising that the snail in its chemical complexity might have dietetic value. Has it not been called "the poor man's oyster"? Different countries le will be seen, make very dif ferent proportions of'iron and steel. France produces only twice as much steel as iron. Germany, on the other hand, makes five times as much steel as iron. TOADS AND SNAKES. Now we cannot have much interest either in the grotesque prescriptions or in those whose justification is obvious but there is a third set whose character affords fo8d for reflection. I refer to those in which there lurks an undeniable kernel of truth-reached we can hardly tell how-which has been justified sci entifically by modern researches. It has been shown that what seemed absurd (or worse) had sometimes a real justification. Two examples must suffice. Prep arations of the toad were prescribed for such mala dies as dropsy, and we now know that phrynin, the ac tive principle in the toad's skin secretions, has a pow erful influence on the pulsations of the heart; is very like the digitalin from the foxglove, and thus might really have a good effect in cases of dropsy from heart affections. As Dr. Hewlett points out in Science Prog ress, dried toads and toad ash, and toads in other forms, have been used medicinally since the time of Aristotle, and were recognized in materia medica till .the end of the eighteenth century. Now, since the al kaloid phrynin, found in the skin glands of the toad (and also to a slight extent in the frog) causes, when injected, contraction of the arterioles, rise of blood pressure, increased cardiac contraction, and so on, it is at least conceivable that it might be of some real service in ancient days. . ------,--- • Address 10 the North --;;;:itish B';'anch of the Pharmaceutical Society. Reprinted from Pharmaceutical Jou rnal. THB; LEECH. One cannot think of links between natural history and medicine without recalling, first of all, that a fa .. miliar animal-the leech-was used for so long (since before the Christian era) and so much (for all man ner of ailments), that its name, which is said to mean "heal " became that of the medical practitioner him· self. 'For people used to say, "Send for the leech," when we should say, "Send for the doctor," and leech ing was synonymous with medical treatment. Yet, as everyone knows, a reaction set in; the leech-the ani· mal I mean-and bloodletting went out of vogue; leechponds disappeared; and I suppose the trade in leeches is now relatively trivial. There is a record of a Breton peasant, in 1883, who had received from the surgeon half a dozen leeches to apply to some congested or inflamed part. Being ignorant of the method of application, he handed over the problem to his wife, who in this case was wrong, for, on a neigh bor's ad vice, she fried the leeches and gave them to the poor man. In spite of their bitter taste, he man aged to get outside of them all, and died thereafter. By J. AI(,["HU]{ THOMSON, M.A., F.R.S.E., Professor of Natural History in the University of Aberdeen. Let us note the line of discovery, doses of bile neutralize venom (Fraser); biliary salts and cholesterin have, like the intact bile, a neutralizing or immunizing effect (Phisalix); vegetable cholesterin will work like animal cholesterin, and even tyrosin from the dahlia has the same important effect (Phisa lix). Thus we come back to herbalist medicine again. OLD PRESClUP'l.'LONS. The first set of old prescriptions, which I select from .l<ʈernie's "Animal Simples," includes those which have no real justification: Rheumatic patients were. told to take a black cat to bed with them, for it is rich in curative electricity; the dust of a dried magpie was used in Germany as recently as 1880 as a cure for epilepsy; the skink lizard once held a place in Brit ish pharmacopmias as a cure for leprosy and much else; to stop hemorrhage the pa tien t was recommend ed to lay hold of a dead toad, "horror and fright con straining the blood to run into its proper place for fear of a beast so contrary to humane nature"; to lay an open pickled herring on the soles of the feet on go ing to bed was deemed a cure for swollen legs-to thrifty Scots it would seem sheer )y-a:stefulness; we would prefer an internal appliClibtfon. These are a few of the least extravagant of the many old prescrip tions which may, I suppose, be dismissed as without any justification. THE LION'S HEART A ND NEWER LINKS. Similarly let me remind you in a few words of an old·fashioned kind of remedy which, until recently, provoked only a smile from modern physician!'\. I re fer to the prescriptions that the coward should devour the raw heart of the lion, that the weakly should eat the heart of an ox, that the lethargic should dine on a ram's brains, that the jaundiced should try the liver of a fox, and so on at great length. Yet surely there is some approach to this in the modern treatment of those unfortunate subjects whose thyroid glands have in some way gone out of gear, with, the result that gOl ere or myxmdema ensues, for they are ordered to eat the thyroids of sheep or calf, or are treated with doses of thyroid extract. The results are apparently very effective; and the historical inference seems to me to be plain-that the old physicians were not such fools as they sometimes seemed. In former days, not very remote, a thoroughly quali fied medical man was expected to have a knowledge of natural history, developed along a line which is now but rarely followed. He was expected to know a large number of forms, not because of their scientific inter est, but because of their supposed practical value. In other words, he had to know the "animal simples," the uses of animals as medicines, uses which some times suggest gross superstition, but sometimes extra· ordinary insight. If we consult a compendium of animal simples, such as one by Dr. Fernie, we get two opposite impressions. We are amazed, on tʇe one hand, at the credulity which is implied both III those who prescribed and in those who swallowed many of the reputed medicines. We wonder, o> the other hand, at the justification which modern sCience has given to some of the old prescriptions. -,- But let me now give a few illustrations of what I would call newer links. The importance of having' other organisms besides man to work with is at once a trite and difficult question, but let me recall a recent instance which can offend no one. It is said that the importance of the thyroid is somehow bound up with the iodine compound (thyreo-globuhn), which it con tains. But has not the matter been complicated by Prof. A. Gautier's paper of last ye'1.r, in which he demo SCIENTIFIC AMERICAN SUPPLEMENT, No. 1313. SCIENTIFIC AMERICAN SUPPLEMENT, No. 1313. MARCH 2. 1901. MARCH 2. 1901. 21044 It seems to me that we may divide the old prescrip tions into three sets-those with no Justification, per haps 75 per cent; those with an obvious justification, perhaps 20 per cent; and those which have found some justification-by no means obvious-as the re sult of recent scientific progress, these forming the remaining five per cent. It seems to me that we may divide the old prescrip tions into three sets-those with no Justification, per haps 75 per cent; those with an obvious justification, perhaps 20 per cent; and those which have found some justification-by no means obvious-as the re sult of recent scientific progress, these forming the remaining five per cent. onstrated in the normal thyroids of dog, pig, sheep and other mammals, t3sides man, the presence of minute quantities of arsenic? Though there are insensible quantities in the skin, arsenic is absent from the or gans of the body, except thyroid, thymus and brain. In the thyroid it exists in combination with nucleins, along with the usual phosphoric nucleins. Does not this shed some light on the alleged advantage of ar senical medicines in treating diseases of the thyroid, and will it not lead us to translate "no health apart from the tnyroid," into "[,0 health without arsenic"? PIG IRON. STEEL. ---;-----c---- ---;-------- y care I Tons. ce;'t:9e. Years Tons. Icege. Country. United States"". 1899 13,620,703 34.56 Great Britain..... 1899 9,305,319 23.61 Germany"" . . ,,' 1899 8,142,017 20.66 France ... . . . .. .. .. 1 1899 2,567,388 6 51 Belgium .. . . " . '1' 1899 1,036,1'5 2 63 Austria·Hungary. 1898 1,427,240 3 62 Russia & Finland. 1898 2,222,469 5 64 Sweden . . 1 1 898 531.766 1 35 Italy . ...... ... " 1897 8,393 .02 Canada ..... " . ,, 1899 94.077 .2'1 Japan...... ..... 1897 57,67R .15 1899 1899 1899 1899 1899 1896 1898 1898 l8!19 1898 1899 10,639,857 5,000.000 6,290,434 1,554,354 729,920 880,696 1.494,000 265,121 122,954 94,667 22,000 39.25 18.44 23.20 5.n 2.70 3.25 5.51 .9H .45 .35 .08 Spain. . . . .. . 1 1899 295,840 .75 O1::::'1:"-,,::: I,oo:'" I "":::: ".:: MOSQUITOS AND :lIALAIUA. Fernie), it is stated that "leeches, on which the older physicians placed such reliance, seem to be coming again in to fashion." But we have known the ordering of a few leeches in a leading London hospital, not so many years ago, cause consternation among the dressers; a consultation had to be held as to which was Lhe "busi ness end" of the leech, and the combined intelligence of the staff was brought to bear on the problem "how to make him bile." PARASITES AS L1:'<iK';. In the above table the relation of pig iron production to steel production is shown, and it will be noticed that the United States converts a larger percentage of her pig iron into steel than any other country . . . • l p We give this week a diagram showing grall·mcally the production of steel in principal countries from 1873 to 1899, which should be compared with the diagram relaL_ng to pig iron production in the same countries which we published·on the 17th ult. This diagram is ,.the' work oC-the Washington Treasury Bureau of Statistics, and is based upon the most re llable figures. SOME LINKS BETWEEN NATURAL HISTORY AND MEDICINE.* By J. AI(,["HU]{ THOMSON, M.A., F.R.S.E., Professor of Natural History in the University of Aberdeen. CONCL USION. I have then illustrated some of the ways in which natural' history has been linked to medicine and some of the links that remain to the great advantage of both. But my illustrations do not cover a tithe of the real and possible kinds of links, nor have they sug gested at all that the deepest service of natural hiʉ tory is probably in heiping to lay the biological foun dation stones on which the stately edifice of medicine, both curative and preventive, is reared. I have mentioned a variety of animals-the leech, the toad, the snake, the mosquito, the bee, the sea ur chin, and so on-but the possibility of illustration is endless. Perhaps in the study of every animal one may find some suggestion of service in the study of man. As I was thinking of this, a Japanese waltzing mouse came dancing into the focus of my conscious ness. But it is as good an illustration as any other, though it may not at first sight seem promising. The mouse moves only sideways, neither backward nor for ward, and surely it becomes at once interesting medic inally in connection with defective equilibration and the like in man, when we note Von Cyon's discovery that it has only one of the usual three semi-circular canals developed. , It is estimated that 85 per cent of the sewage of Chicago is carried into the Illinois River. Dr. Jordan gives the following averages of the numbers of bac terial colonies found at various points on the canal, the Desplaines River, and the Illinois River: Bridgeport (practically Chicago), 1,245,000; Lockport (29 miles away), 650,000; Joliet (33 miles away), 486,000; Morris (57 miles away), 439,000; Ottawa (81 miles away), 27,400; La Salle (95 miles away), 16,300; Henry (123 miles away), 11,200; Averyville (159 miles away), 3,660; Wesley City (165 miles away), 758,000; Pekin (175 miles away), 492,600; Havana (199 miles away). 16,800; Beardstown (231 miles away), 14,000; Kamps ville (288 miles away), 4,800; Grafton (318 miles away), where the junction with the Mississippi River is effected, 10,200. q It is plain that in the solving of the practical prob· lem of malaria the medical investigator has need of all the assistance that natural history can give him e. MARCH 2, IIJUl. A large number of cells have passed into what may be called a state of irrecoverable fatigue-collapse. As Prof. Hodge says: "The nerve cells, in the course of the bee's daily work, gradually cease to be functional and die ·off, until no more are left than are sufficient for the necessary vi tal functions," His work suggests that there ar8 three grades of nervous fatigue in animals: (a l' that from which recuperation is usual, what one might call the normal daily fatigue; (b) that from which recuperation is possible but difficult, which one might call extra wear and tear; and (c) that from which there is no recovery, the final functional collapse of nerve cells as in the bee's brain, and in some forms of old age. The state of the nerve cells in the brains of some hibernating animals is also very interesting in this connection. I am not competent to follow the matter further, e. g., in application to man, but I think the suggestiveness of a contribution like this must be plain to all. At any rate, { recommend a re consideration of the little busy bee, but in a scientific rather than in a poetical light. The story of the bee's brain is a warning against over-industry, and one of the sections in the apology for rest. As my chief interest at present is not with malaria, but to illustrate the scientific method, I ask your at tention to another experiment. The story of the two volunteers on the Campagna is a fine illustration of scientific enthusiasm, and of confidence in scientific methods and results. What they have shown is that by avoiding mosquitos, or by getting mosquitos to avoid you, malaria is escaped. But it may be ar gued that this is only negative evidence. The ques tion is whether malaria can be produced in a healthy person by the bites of infected mosquitos. In a re cent number of The British Medical Journal it is stated that a consignment of mosquitos which had been fed on the blood of a malarial patient in Rome was· fe ceived in London in July, and that a son of Dr. Man son, who had never been in a malarial country since he was a child, allowed himself to be bitten, and there after suffered from well-marked malarial infection of double tertian type. THE BACTERIAL SELF - PURIFICATION OF STREAMS. DC:RING the construction of the Illinois and Michigan Canal, and especially when it was approaching comple tion and the sewage of Chicago was to be turned into the Mississippi River, many of the people of St. Louis, it will be remembered, were in a state of dread lest their drinking water, drawn from the Mississippi, should be dangerously contaminated as the result, says The New York Medical Journal. The flow has now been going on for some monUis, however, and the alarm appears to have been needless. Given sufficient time and sufficient distance, and flowing water, even water flowing sluggishly, will free itself from pathogenic bac teria. It is of no little importance to know what agencies are at work in any given instance to bring about this spontaneous self-purification. Instructive studies of the waters of various European rivers-such as the Spree, the Pregel, the Danube, the Limmat, the I8ar, the Rhine, and the Seine-have been published from time to time during the last few years, and some work in the same direction has been done in this coun try. The most recent and one of the most notable of American investigations of this sort has been con ducted by Edwin Oakes Jordan, Ph.D., of the bacterio logical laboratory of the University of Chicago, who contributes a condensed account of his observations to the December number of The Journal of Experimental Medicine. e Or, again, you are, I believe, professionally inter ested in the influence of chemical reagents on the complex substances which we sum up in the term protoplasm. But that is obviously one of the peren nial problems of biology. Let me recall one of the sensational researches of the last eighteen months, as the result of which Prof. Loeb announces that if eggs of sea urchins are transferred for a couple of hours to sea water to which magnesium chloride has been add ed (disturbing the normal proportions of salts), and are then replaced in their natural medium, they de velop without fertiliza,tjon and even form larVal. It is too soon to say much about it; but here, obviously, if it is only to find out the fallacy, is one of the many problems where biological and medical or pharma ceutical students may profitably join hands. Among the supposed causes of the gradual disap pearance of bacteria, Dr. Jordan first considers me chanical agitation and aeration. MARCH 2, IIJUl. Louis water works, from 1,800 to 69,000; at the same place (western bank), from 2,700 to 57,500; at St. Louis (tap water), from 340 to 2,790; at Jefferson Barracks, in water near the eastern bank of the Mississippi, from 1,000 to 69,000; at the same place, in the water east of the middle of the river, from 1,700 to 42,000; at the same place (midstream), from 2,700 to 56,500; at the same place, in water to the west of the middle of the river, from 3,200 to 91,600; at the same place, in water near the western bank, from 4,900 to 65,300. the morning-when they were presumably fairly fresh -with the brain' of bees who came last into the hive in the evening-presumably tired out-and demon strated what, fifty years ago, would have excited great wonder-the structural effects of nerve fatigue. that the mosquito's bite can be avoided. This idea led to the- erection of a mosquito-p,oof hut near Ostia, in one of the worst parts of the Roman Campagna, to serve as a cage for two volunteers who agreed to sub mit themselves to the test. Dr. L. Sam bon and Dr. G. C. Low, of the London School of Tropical Medicine, arranged to live from May to the end of October in this hut, situated in a region "where scarcely a per son spends a night without contracting malarial fever of a virulent type." No quinine or other drug was to be taken; the volunteers were to live in the mos quito-proof hut from an hour before sunset to an hour after sunrise, for the mosquitos feed only during the night. The experiment seems to have been quite suc cessful. On September 13 Prof. Grassi and other in vestigators visited the hut, and sent this telegram to Manson: "Assembled in British mosquito-proof hut, having verified the perfect health of the experiment ers among malarial-stricken inhabitants, I salute Man son, who first formulated mosquito-malarial theory. Grassi." Similarly, Dr. Elliot reports that the mem bers of the Liverpool expedition, sent this year to Ni geria, have been perfectly well after four months in some of the most malarious regions, and that they at tribute their immunity to the careful use of mosquito nets. f The same striking contrast is seen when the brain of a young bee, fresh from the comb, is contrasted with that of a relatively old bee. THE BACTERIAL SELF - PURIFICATION OF STREAMS. In so sluggish a stream as the Illinois, he says, there is not sufficient agitation to prove at all injurious to bacterial life; indeed, Meltzer's experiments point to the conclusion that moderate agitation is not detrimental to bacteria. Aeration of the water to the extent to which it takes place in a river of slow flow may also be dismissed as having little effect in promoting the destruction of micro-organisms. The immediate effect of dilution with purer water from underground sources or from tributaries must, however, be in the direction of dimin ishing the proportion of bacteria in a given quantity of water, and Dr. Jordan finds it difficult to under stand, as we do also, why this effect is sometimes referred to as not being a "true" purification. If, he says, a sample of water contains a hundred typhoid fever bacilli to the quart, and is diluted to twenty times its bulk with pure water, each quart will con tain only:-five germs, and, "apart from any influence the dilution may have upon the life of the germs, a purification of the water will have occurred to just the same extent as if 95 per cent of the typhoid bac teria had perished, and the danger from drinking a small quantity of such a water would be diminished in exactly the same proportion." Dr. Jordan summarizes a number of the best-known European investigations, but points out that, while their general purport points unmistakably to a lessen ing of the bacterial contents of the water, whether it is due to dilution, to sedimentation, or to the action of sunlight, great differences exist in the degree of apparent purification, depending upon the amount of initial pollution, the v810city of the flow, the season of the year, and other elements. Each stream, he re marks, seems to have its own conditions, and conclu sions drawn from an examination of any one cannot be applied to another; each stream must be subjected to detailed special observation. His own work, though nominally limited to the Illinois River and its tribu taries, really included observations at various points, ending as far away from Chicago as at Chain of Rocks, on the Mississippi, whence the St. Louis water supply is obtained. The results are recorded largely in tabular form. The tables. CONCL USION. g., as to t)1e many different kinds of mosquitos and their habits, or as to the life history of the sporozoon parasite, which is the real cause of the disease. A knowledge of the breathing habits of the larVal (famil iar to us in common gnats) explains the success of a petroleum film as a means of killing them off in marshes; a knowledge of the appetite which various fresh water fishes have for them suggests another method of destroying them jn ponds, and so on. On the other hand, since there is always action and reac tion between science and its corresponding art, it must be frankly allowed that it is the practical im- portance of malaria which has led the British Museum entomologists to hurry on at present with a mono graph on mosquitos. p Even in mv few illustrations I have probably made mistakes by " touching subjects which are beyond my field of work, but to pounce upon these is to ignore the whole aim of my address, which is to suggest that th8 various' departments of science-yours and mine and that of others-are all correlated. It is not solely by the aloofness of specialism and the pre-occupation of expert research, but by mutual interest and active co operation as well, that we make progress in under standing the one problem which, in some form or oth er is before us all-the Order of Nature. And it is th'is valuable width of 'scientific interest and sympa thy which an actively functional society like this is well suited to foster. === is effected, 10,200. No doubt the reader has been struck with the sudden and decided increase of bacteria at Wesley City, follow ing upqn a gradual and steady diminution from Chi cago. This is attributable to the large amount of or ganic refuse that finds its way into the Illinois River with the sewage of Peoria reinforced by the waste of manufactories, distillery slop, discharges from glucose factories, and the sweepings of extensive stock yards. The Wesley City specimens of water were taken from a point about four miles below this "outpour of pollu tion," which is said to vary greatly at different seasons of the year and at different hours of the day, thus accounting for the great irregularities and fluctuations in the number of bacteria observed at that point. MARCH 2, IIJUl. 21015 bacterial colonies in a cubic centimeter of water ranged from 225,000 to 1,850,000; at Lockport, also on tho canal, from 40,000 to 1,650,000; at the same place, bu t in wa.ter from the Desplaines lUver, from 1,250 to 34,000; above Joliet, in the same river, from 20,000 to 1,620,000; below Joliet, from 120,000 to 2,540,000; "t Wilmington, in the water of the Kankakee River, from 1,400 to 25,700; at Morris, in the water of the Illinois River, from 16,000 to 1,140,000; at Ottawa, in the water of the Fox River, from 450 to 34,500; above Ottawa, in tbe water of the Illinois River, from 650 to 130,000; at La Salle, in the water of the Big Vermilion River, from 1.400 to 62,000; at the same place, in the water of the Illinois River, from 700 to 228,000; at the same place, in the water of the canal, from 16,000 to 152,000; at Henry, in the water of the Illinois River, from 500 to 74,000; at Averyville, in the water of the same river, from 500 to 19,500; at Wesley City, in the water of the same river from 5,000 to 3,390,000; at Pekin, in the ʋ"water of the same river, from 5,000 to 2,030,000; at Havana, in the water of the same river, from 850 to 128,000; at Chandlerville, in the water of the Sangamon River, from 1,200 to 11,600; at Beardstown, in the water of the Illinois River, from 1,500 to 120,000; at Kampsville, in the water of the same river, from 340 to 23,500; at Grafton, in the water of the same river, from 180 to 97,000; at the same place, in the water of tbe Mississippi River, from 800 to 45,000; at Alton, in the water near the eastern bank of the Mississippi, from 900 to 39,000; at the same place, in the water east of the middle of the river, from 900 to 40,400; at the same place, in midstream water, from 700 to 27,500; at the same place, west of the middle of the river, from 700 to 25,500; at the same place, near the western bank of the river, from 650 to 37,500; at West Alton, in the water of the Missouri River, from 2,100 to 20,900; at Chain of Rocks, in the water of the Mississippi (eastern bank), from 1,300 to 113,000; at the same place (mid stream), from 1,600 to 68,000; at the same place, at the inlet tower of the St. MARCH 2, IIJUl. Mi.Cl'(}scopic examination of his blood shoW"ed the presence of numerous malarial para sites. This must be regarded as almost final proof of the hypothesis-shall we not say fact?-that ma laria is transmitted by mosquito bites. CONCL USION. Be yond Wesley City the gradual process of self purifica tion goes on as before. Taking the purifying process, then, as practically continuous, we at once find our selves interested in what Dr. Jordan has to say of the elements concerned in it. Before proceeding to present this, however, we must remark that he quite frankly points out that the observations are open to criticism on the score that the specimens of water examined were transported, and for the reason that they were not taken simultaneously from the different points. It seems that the bacteria multiply when the speci mens are not packed in ice, and that in ice-packed spe cimens they decrease in number at first, but subse quently multiply to a number exceeding in some in stances the original proportion. Allowance being made for these sources of error, which in some measure probably correct each other, it is not to be doubted that the general drift of the observations and of the deductions drawn from them may be taken as cor rect. PHAGOCYTES AND PROTOPLASM. Let me illustrate another perhaps less obvious link. In almost all animals, from sponges to mammals, there are, within or apart from blood fluid, wandering amce boid cells, technically called phagocytes. As you know, they perform many fUl1ctions both in health and disease. They form the bodyguard, struggling with invading bacteria; they surround and engulf i}" ntant particles; they help to repair wounds, to re generate lost parts, and so on. Phagocytosis has be come to the pathologist a word of comfort, and per haps there has been even some exaggeration of its im portance. What I wish to point out is simply that much of the working out of th"e theory of phagocytosis has been in the hands of zoologists, as Metschnikoff shows clearly in his "Lectures on the Comparative Pathology of Inflammation." The beginning of it was in 1862, when Haeckel observed that grains of indigo injected into the mollusk Thetys were surrounded and engulfed by amceboid cells. THE VALUE OF A CONTgOL EXPERIMENT. But let us now notice the value of a control experi ment. If malaria is due to the introduction of a para site from the mosquito to the human blood, then there should be no danger in a malarial district provided . ------,--- • Address 10 the North --;;;:itish B;'anch of the Pharmaceutical Society. Reprinted from Pharmaceutical Jou rnal. © 1901 SCIENTIFIC AMERICAN, INC MARCH 2, IIJUl. SCIENTIFIC A.\n:HICAN 8UPPLEME7T, No. 1313. THE BACTERIAL SELF - PURIFICATION OF STREAMS. thirty-eight in number, give each the results of a series of examinations for each locality, in most instarʊes between twenty-five and thirty, made a few days ap[,rt during the period from May to Decem ber. A VICTIHI OF OVER-EXERTION. A VICTIHI OF OVER-EXERTION. Finally, another illustration, which may seem at first sight almost further afield than the last. There are few subjects in medicine more important than nerve fatigue and nerve rest. Let me remind you of what I may call a natural history contribution to neurology. Is it not the case that almost from infancy many of us-especially those of us who are fond of laziness- 1Ia ve been qrged to consider the little busy bee which improves eiwh shining hour. A more intimate and critical knowledge of bees and their behavior entirely shakes our confidence in this exemplar of our child hood. I wish to lift only one corner of the seamy side by asking how the shining hour improves the busy bee. '" hen ever we ask this question we are struck by the fact that the worker hive bee has but a short life-often only for a few weeks after his industry be gins. It is a victim to oTer-exertion. With all its getting, it gets not wisdom, but foolishness, for its brain cells go steadily and surely out of gear. Hodge compared the brain of bees issuing from the hive in At Bridgeport, on the canal itself, the number of © 1901 SCIENTIFIC AMERICAN, INC

Subsets and Splits

No community queries yet

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